material/deep_learning/mnet_segementation.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "mnet_segementation.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HYzKnHXzN0lB"
      },
      "source": [
        "# 폐 영역분할"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ROQ8VnQXGeU4"
      },
      "source": [
        "다음의 2019 대한의료인공지능학회 Summer School에서 진행한 AI hands-on의 자료에서 코드르 가져와서 수정함.\n",
        "\n",
        "https://github.com/mi2rl/ai_handson/blob/master/KoSAIM_Summer_School_2019_Handson_03_Segmentation.ipynb\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bMHwQjK9OhLf"
      },
      "source": [
        "## 폐 데이터 다운로드\n",
        "\n",
        "다음에서 다운받은 데이터\n",
        "\n",
        "https://www.kaggle.com/nikhilpandey360/lung-segmentation-from-chest-x-ray-dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EAKPT_j5GITc",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "453a761e-59fa-4aac-c881-2ae3540a9cc8"
      },
      "source": [
        "!rm -rf Lung_Segmentation.zip\n",
        "!wget https://github.com/dhrim/hongik_2020/raw/master/material/deep_learning/Lung_Segmentation.zip"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2021-08-12 04:07:44--  https://github.com/dhrim/hongik_2020/raw/master/material/deep_learning/Lung_Segmentation.zip\n",
            "Resolving github.com (github.com)... 140.82.121.4\n",
            "Connecting to github.com (github.com)|140.82.121.4|:443... connected.\n",
            "HTTP request sent, awaiting response... 302 Found\n",
            "Location: https://raw.githubusercontent.com/dhrim/hongik_2020/master/material/deep_learning/Lung_Segmentation.zip [following]\n",
            "--2021-08-12 04:07:44--  https://raw.githubusercontent.com/dhrim/hongik_2020/master/material/deep_learning/Lung_Segmentation.zip\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 37648324 (36M) [application/zip]\n",
            "Saving to: ‘Lung_Segmentation.zip’\n",
            "\n",
            "Lung_Segmentation.z 100%[===================>]  35.90M   216MB/s    in 0.2s    \n",
            "\n",
            "2021-08-12 04:07:44 (216 MB/s) - ‘Lung_Segmentation.zip’ saved [37648324/37648324]\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "N8TLRtBfGsKj",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "58fa85a6-de8e-4799-d3ce-0262fcf33963"
      },
      "source": [
        "!rm -rf Lung_Segmentation\n",
        "!unzip Lung_Segmentation.zip"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Archive:  Lung_Segmentation.zip\n",
            "   creating: Lung_Segmentation/\n",
            "   creating: Lung_Segmentation/image/\n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0001_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0002_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0003_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0004_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0005_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0006_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0007_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0008_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0009_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0010_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0011_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0012_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0013_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0014_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0015_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0016_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0017_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0018_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0019_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0020_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0021_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0022_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0023_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0024_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0026_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0027_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0028_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0029_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0030_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0031_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0032_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0033_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0034_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0035_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0041_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0042_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0043_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0044_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0045_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0046_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0047_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0048_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0049_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0050_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0051_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0052_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0053_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0054_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0055_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0056_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0057_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0058_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0059_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0060_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0061_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0062_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0063_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0064_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0066_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0067_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0068_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0069_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0070_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0071_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0072_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0073_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0074_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0075_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0076_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0077_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0078_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0079_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0080_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0081_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0082_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0083_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0084_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0085_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0086_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0087_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0088_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0089_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0090_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0091_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0092_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0093_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0094_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0095_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0096_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0097_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0098_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0099_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0100_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0101_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0102_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0103_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0104_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0105_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0106_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0107_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0108_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0109_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0110_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0111_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0112_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0113_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0114_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0115_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0116_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0117_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0118_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0119_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0120_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0121_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0122_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0123_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0124_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0125_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0126_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0127_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0128_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0129_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0130_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0131_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0132_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0133_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0134_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0135_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0136_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0137_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0138_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0139_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0140_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0141_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0142_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0143_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0144_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0145_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0146_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0147_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0148_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0149_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0150_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0151_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0152_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0153_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0154_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0155_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0156_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0157_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0158_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0159_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0160_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0161_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0162_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0163_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0164_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0165_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0166_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0167_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0168_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0169_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0170_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0171_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0172_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0173_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0174_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0175_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0176_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0177_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0178_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0179_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0180_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0221_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0222_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0223_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0224_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0225_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0226_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0227_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0228_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0229_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0230_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0231_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0232_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0233_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0234_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0235_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0236_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0237_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0238_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0239_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0240_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0241_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0242_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0243_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0244_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0245_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0246_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0247_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0248_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0249_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0250_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0251_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0252_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0253_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0254_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0255_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0256_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0257_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0258_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0259_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0260_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0261_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0262_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0263_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0264_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0265_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0266_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0267_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0268_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0269_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0270_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0271_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0272_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0273_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0274_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0275_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0276_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0277_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0278_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0279_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0280_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0281_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0282_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0283_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0284_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0285_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0286_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0287_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0288_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0289_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0290_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0291_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0292_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0293_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0294_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0295_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0296_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0297_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0298_0.png  \n",
            " extracting: Lung_Segmentation/image/CHNCXR_0299_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0300_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0301_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0302_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0303_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0304_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0305_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0306_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0307_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0308_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0309_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0310_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0311_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0312_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0313_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0314_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0315_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0316_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0317_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0318_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0319_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0320_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0321_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0322_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0323_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0324_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0325_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0326_0.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0327_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0328_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0329_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0330_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0331_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0332_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0333_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0334_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0335_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0337_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0338_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0339_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0340_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0361_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0362_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0363_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0364_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0365_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0366_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0367_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0368_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0369_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0370_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0371_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0372_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0373_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0374_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0375_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0376_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0377_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0378_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0379_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0380_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0381_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0382_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0383_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0384_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0385_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0386_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0387_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0388_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0389_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0390_1.png  \n",
            " extracting: Lung_Segmentation/image/CHNCXR_0391_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0392_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0393_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0394_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0395_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0396_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0397_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0398_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0399_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0400_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0401_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0402_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0403_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0404_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0405_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0406_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0407_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0408_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0409_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0410_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0411_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0412_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0413_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0414_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0415_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0416_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0417_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0418_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0419_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0420_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0421_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0422_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0423_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0424_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0425_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0426_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0427_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0428_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0429_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0430_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0431_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0432_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0433_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0434_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0435_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0436_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0437_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0438_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0439_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0440_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0441_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0442_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0443_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0444_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0445_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0446_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0447_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0448_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0449_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0450_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0451_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0452_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0453_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0454_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0455_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0456_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0457_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0458_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0459_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0460_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0461_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0462_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0463_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0464_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0465_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0466_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0467_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0468_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0469_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0470_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0471_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0472_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0473_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0474_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0475_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0476_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0477_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0478_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0479_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0480_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0501_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0503_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0504_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0506_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0507_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0508_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0509_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0510_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0511_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0512_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0513_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0514_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0515_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0516_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0517_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0518_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0519_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0520_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0521_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0522_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0523_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0524_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0525_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0526_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0527_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0528_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0529_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0530_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0531_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0532_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0533_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0534_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0535_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0536_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0537_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0538_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0539_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0540_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0541_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0542_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0543_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0544_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0545_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0546_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0547_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0548_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0549_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0550_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0551_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0552_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0553_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0554_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0555_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0556_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0557_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0558_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0559_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0566_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0567_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0568_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0569_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0570_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0571_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0572_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0573_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0574_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0575_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0576_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0577_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0578_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0579_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0580_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0581_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0582_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0583_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0584_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0585_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0586_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0587_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0588_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0589_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0590_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0591_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0592_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0593_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0594_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0595_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0596_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0597_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0598_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0599_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0600_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0601_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0602_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0603_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0604_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0605_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0606_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0607_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0608_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0609_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0610_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0611_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0612_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0613_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0614_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0615_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0616_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0617_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0618_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0619_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0620_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0621_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0622_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0623_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0624_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0625_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0626_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0627_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0628_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0629_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0630_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0631_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0632_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0633_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0634_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0635_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0636_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0637_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0638_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0639_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0640_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0641_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0642_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0643_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0644_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0645_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0646_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0647_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0648_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0649_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0650_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0651_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0652_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0653_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0654_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0655_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0656_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0657_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0658_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0659_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0660_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0661_1.png  \n",
            "  inflating: Lung_Segmentation/image/CHNCXR_0662_1.png  \n",
            "   creating: Lung_Segmentation/label/\n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0001_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0002_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0003_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0004_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0005_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0006_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0007_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0008_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0009_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0010_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0011_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0012_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0013_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0014_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0015_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0016_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0017_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0018_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0019_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0020_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0021_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0022_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0023_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0024_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0026_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0027_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0028_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0029_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0030_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0031_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0032_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0033_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0034_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0035_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0041_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0042_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0043_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0044_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0045_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0046_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0047_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0048_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0049_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0050_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0051_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0052_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0053_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0054_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0055_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0056_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0057_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0058_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0059_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0060_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0061_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0062_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0063_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0064_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0066_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0067_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0068_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0069_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0070_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0071_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0072_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0073_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0074_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0075_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0076_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0077_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0078_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0079_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0080_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0081_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0082_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0083_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0084_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0085_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0086_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0087_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0088_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0089_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0090_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0091_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0092_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0093_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0094_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0095_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0096_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0097_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0098_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0099_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0100_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0101_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0102_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0103_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0104_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0105_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0106_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0107_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0108_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0109_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0110_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0111_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0112_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0113_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0114_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0115_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0116_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0117_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0118_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0119_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0120_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0121_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0122_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0123_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0124_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0125_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0126_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0127_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0128_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0129_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0130_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0131_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0132_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0133_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0134_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0135_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0136_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0137_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0138_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0139_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0140_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0141_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0142_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0143_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0144_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0145_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0146_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0147_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0148_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0149_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0150_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0151_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0152_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0153_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0154_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0155_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0156_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0157_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0158_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0159_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0160_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0161_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0162_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0163_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0164_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0165_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0166_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0167_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0168_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0169_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0170_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0171_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0172_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0173_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0174_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0175_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0176_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0177_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0178_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0179_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0180_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0221_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0222_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0223_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0224_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0225_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0226_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0227_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0228_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0229_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0230_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0231_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0232_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0233_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0234_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0235_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0236_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0237_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0238_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0239_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0240_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0241_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0242_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0243_0.png  \n",
            " extracting: Lung_Segmentation/label/CHNCXR_0244_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0245_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0246_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0247_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0248_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0249_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0250_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0251_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0252_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0253_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0254_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0255_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0256_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0257_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0258_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0259_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0260_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0261_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0262_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0263_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0264_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0265_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0266_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0267_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0268_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0269_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0270_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0271_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0272_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0273_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0274_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0275_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0276_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0277_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0278_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0279_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0280_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0281_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0282_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0283_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0284_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0285_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0286_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0287_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0288_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0289_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0290_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0291_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0292_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0293_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0294_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0295_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0296_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0297_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0298_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0299_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0300_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0301_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0302_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0303_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0304_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0305_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0306_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0307_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0308_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0309_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0310_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0311_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0312_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0313_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0314_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0315_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0316_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0317_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0318_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0319_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0320_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0321_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0322_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0323_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0324_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0325_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0326_0.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0327_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0328_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0329_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0330_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0331_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0332_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0333_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0334_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0335_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0337_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0338_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0339_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0340_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0361_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0362_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0363_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0364_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0365_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0366_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0367_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0368_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0369_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0370_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0371_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0372_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0373_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0374_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0375_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0376_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0377_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0378_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0379_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0380_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0381_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0382_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0383_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0384_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0385_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0386_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0387_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0388_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0389_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0390_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0391_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0392_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0393_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0394_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0395_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0396_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0397_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0398_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0399_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0400_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0401_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0402_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0403_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0404_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0405_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0406_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0407_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0408_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0409_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0410_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0411_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0412_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0413_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0414_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0415_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0416_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0417_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0418_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0419_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0420_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0421_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0422_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0423_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0424_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0425_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0426_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0427_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0428_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0429_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0430_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0431_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0432_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0433_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0434_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0435_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0436_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0437_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0438_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0439_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0440_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0441_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0442_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0443_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0444_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0445_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0446_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0447_1.png  \n",
            " extracting: Lung_Segmentation/label/CHNCXR_0448_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0449_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0450_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0451_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0452_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0453_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0454_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0455_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0456_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0457_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0458_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0459_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0460_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0461_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0462_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0463_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0464_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0465_1.png  \n",
            " extracting: Lung_Segmentation/label/CHNCXR_0466_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0467_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0468_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0469_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0470_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0471_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0472_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0473_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0474_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0475_1.png  \n",
            " extracting: Lung_Segmentation/label/CHNCXR_0476_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0477_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0478_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0479_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0480_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0501_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0503_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0504_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0506_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0507_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0508_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0509_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0510_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0511_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0512_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0513_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0514_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0515_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0516_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0517_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0518_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0519_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0520_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0521_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0522_1.png  \n",
            " extracting: Lung_Segmentation/label/CHNCXR_0523_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0524_1.png  \n",
            " extracting: Lung_Segmentation/label/CHNCXR_0525_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0526_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0527_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0528_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0529_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0530_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0531_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0532_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0533_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0534_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0535_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0536_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0537_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0538_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0539_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0540_1.png  \n",
            " extracting: Lung_Segmentation/label/CHNCXR_0541_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0542_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0543_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0544_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0545_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0546_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0547_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0548_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0549_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0550_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0551_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0552_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0553_1.png  \n",
            " extracting: Lung_Segmentation/label/CHNCXR_0554_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0555_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0556_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0557_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0558_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0559_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0566_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0567_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0568_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0569_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0570_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0571_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0572_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0573_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0574_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0575_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0576_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0577_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0578_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0579_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0580_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0581_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0582_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0583_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0584_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0585_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0586_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0587_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0588_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0589_1.png  \n",
            " extracting: Lung_Segmentation/label/CHNCXR_0590_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0591_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0592_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0593_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0594_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0595_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0596_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0597_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0598_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0599_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0600_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0601_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0602_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0603_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0604_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0605_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0606_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0607_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0608_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0609_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0610_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0611_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0612_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0613_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0614_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0615_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0616_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0617_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0618_1.png  \n",
            " extracting: Lung_Segmentation/label/CHNCXR_0619_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0620_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0621_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0622_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0623_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0624_1.png  \n",
            " extracting: Lung_Segmentation/label/CHNCXR_0625_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0626_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0627_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0628_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0629_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0630_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0631_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0632_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0633_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0634_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0635_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0636_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0637_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0638_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0639_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0640_1.png  \n",
            " extracting: Lung_Segmentation/label/CHNCXR_0641_1.png  \n",
            " extracting: Lung_Segmentation/label/CHNCXR_0642_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0643_1.png  \n",
            " extracting: Lung_Segmentation/label/CHNCXR_0644_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0645_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0646_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0647_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0648_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0649_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0650_1.png  \n",
            " extracting: Lung_Segmentation/label/CHNCXR_0651_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0652_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0653_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0654_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0655_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0656_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0657_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0658_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0659_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0660_1.png  \n",
            "  inflating: Lung_Segmentation/label/CHNCXR_0661_1.png  \n",
            " extracting: Lung_Segmentation/label/CHNCXR_0662_1.png  \n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AoxCIJscPmAC"
      },
      "source": [
        "## 풀린 데이터 파일 구조\n",
        "\n",
        "```\n",
        "Lung_segmentaiton/\n",
        "  image/\n",
        "    CHNCXR_0001_0.png\n",
        "    CHNCXR_0002_0.png\n",
        "    ...\n",
        "  label/\n",
        "    CHNCXR_0001_0.png\n",
        "    CHNCXR_0002_0.png\n",
        "    ...\n",
        "```"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZN8aC7EMQpDL"
      },
      "source": [
        "## 파일 보기"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IaPaC-FnQKU6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 529
        },
        "outputId": "99eb906b-b924-4e55-eaf8-2fff71f889d0"
      },
      "source": [
        "from IPython.display import Image, display\n",
        "\n",
        "display(Image('./Lung_Segmentation/image/CHNCXR_0001_0.png'))\n",
        "display(Image('./Lung_Segmentation/label/CHNCXR_0001_0.png'))\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GVdJ49RtQtry"
      },
      "source": [
        "## 데이터 로딩"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5ME_l5d-GuRE"
      },
      "source": [
        "import numpy as np\n",
        "import os\n",
        "\n",
        "from skimage.io import imread\n",
        "from skimage.transform import resize\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "IMG_WIDTH = 256\n",
        "IMG_HEIGHT = 256\n",
        "IMG_CHANNELS = 3\n",
        "\n",
        "data_path = \"Lung_Segmentation\"\n",
        "\n",
        "files = os.listdir(os.path.join(data_path, 'image'))\n",
        "# files = ['CHNCXR_0402_1.png', 'CHNCXR_0461_1.png', 'CHNCXR_0422_1.png', ... ]    \n",
        "\n",
        "\n",
        "raw_x = []\n",
        "raw_y = []\n",
        "for file_name in files:\n",
        "    # file_name = \"CHNCXR_0402_1\"\n",
        "\n",
        "    # input_file = \"./Lung_Segmentation/image/CHNCXR_0402_1.png\"\n",
        "    input_file = os.path.join(data_path, 'image', file_name)\n",
        "    # output_file = \"./Lung_Segmentation/label/CHNCXR_0402_1.png\"\n",
        "    output_file = os.path.join(data_path, 'label', file_name)\n",
        "    \n",
        "    image = imread(input_file)[:,:,:IMG_CHANNELS]\n",
        "    # image.shape = (256, 256, 3)\n",
        "\n",
        "    label = imread(output_file)\n",
        "    # label.shape = (256, 256)\n",
        "\n",
        "    label = np.expand_dims(label, axis=-1)\n",
        "    # label.shape = (256, 256, 1)\n",
        "\n",
        "    raw_x.append(image)\n",
        "    raw_y.append(label)\n",
        "\n",
        "raw_x = np.array(raw_x)\n",
        "raw_y = np.array(raw_y)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qhp-Lt0OWcxT"
      },
      "source": [
        "## 원시 데이터 보기"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MXnw8NwqTJd8",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "57713fd6-6994-4415-c82b-afd4bd367acb"
      },
      "source": [
        "print(raw_x[0])\n",
        "print(raw_y[0])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[[[ 75  75  75]\n",
            "  [ 37  37  37]\n",
            "  [ 28  28  28]\n",
            "  ...\n",
            "  [ 46  46  46]\n",
            "  [ 54  54  54]\n",
            "  [ 62  62  62]]\n",
            "\n",
            " [[ 73  73  73]\n",
            "  [ 35  35  35]\n",
            "  [ 24  24  24]\n",
            "  ...\n",
            "  [ 45  45  45]\n",
            "  [ 50  50  50]\n",
            "  [ 53  53  53]]\n",
            "\n",
            " [[ 72  72  72]\n",
            "  [ 32  32  32]\n",
            "  [ 20  20  20]\n",
            "  ...\n",
            "  [ 42  42  42]\n",
            "  [ 45  45  45]\n",
            "  [ 50  50  50]]\n",
            "\n",
            " ...\n",
            "\n",
            " [[  0   0   0]\n",
            "  [  0   0   0]\n",
            "  [  0   0   0]\n",
            "  ...\n",
            "  [ 42  42  42]\n",
            "  [ 42  42  42]\n",
            "  [ 56  56  56]]\n",
            "\n",
            " [[  0   0   0]\n",
            "  [  0   0   0]\n",
            "  [  0   0   0]\n",
            "  ...\n",
            "  [ 70  70  70]\n",
            "  [ 71  71  71]\n",
            "  [ 79  79  79]]\n",
            "\n",
            " [[  0   0   0]\n",
            "  [  0   0   0]\n",
            "  [  0   0   0]\n",
            "  ...\n",
            "  [109 109 109]\n",
            "  [111 111 111]\n",
            "  [112 112 112]]]\n",
            "[[[0]\n",
            "  [0]\n",
            "  [0]\n",
            "  ...\n",
            "  [0]\n",
            "  [0]\n",
            "  [0]]\n",
            "\n",
            " [[0]\n",
            "  [0]\n",
            "  [0]\n",
            "  ...\n",
            "  [0]\n",
            "  [0]\n",
            "  [0]]\n",
            "\n",
            " [[0]\n",
            "  [0]\n",
            "  [0]\n",
            "  ...\n",
            "  [0]\n",
            "  [0]\n",
            "  [0]]\n",
            "\n",
            " ...\n",
            "\n",
            " [[0]\n",
            "  [0]\n",
            "  [0]\n",
            "  ...\n",
            "  [0]\n",
            "  [0]\n",
            "  [0]]\n",
            "\n",
            " [[0]\n",
            "  [0]\n",
            "  [0]\n",
            "  ...\n",
            "  [0]\n",
            "  [0]\n",
            "  [0]]\n",
            "\n",
            " [[0]\n",
            "  [0]\n",
            "  [0]\n",
            "  ...\n",
            "  [0]\n",
            "  [0]\n",
            "  [0]]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uf0F31eBQ2S0"
      },
      "source": [
        "## 데이터 정규화"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-WA09ov8G19W"
      },
      "source": [
        "x = raw_x.astype('float32') / 255.\n",
        "y = raw_y.astype('float32')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "f9aAVB3HRMHB"
      },
      "source": [
        "## 데이터 나누기"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wBiEDjFBG8Lj"
      },
      "source": [
        "split_index = int(x_all.shape[0]*0.8)\n",
        "train_x, test_x = x[:split_index], x[split_index:]\n",
        "train_y, test_y = y[:split_index], y[split_index:]"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YAlIG-SDHGCk",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8a2b94d5-00b7-48de-8687-442ca8938acb"
      },
      "source": [
        "print(train_x.shape)\n",
        "print(train_y.shape)\n",
        "print(test_x.shape)\n",
        "print(test_y.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(452, 256, 256, 3)\n",
            "(452, 256, 256, 1)\n",
            "(114, 256, 256, 3)\n",
            "(114, 256, 256, 1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FcqjqeYpHL-7",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "557c0433-44c1-49da-e750-755cfdede42b"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "def plotTrainData(x, y, set_type):\n",
        "    print(set_type)  \n",
        "    for i in range(3):\n",
        "        ix = np.random.randint(0, len(x))\n",
        "        plt.subplot(1,2,1)\n",
        "        plt.imshow(x[ix])\n",
        "        \n",
        "        plt.subplot(1,2,2)\n",
        "        plt.imshow(np.squeeze(y[ix]), 'gray') # np.squeeze()는 (256,26,1)을 (256,256)으로 변환\n",
        "        plt.show()\n",
        "        \n",
        "plotTrainData(train_x, train_y, 'train')\n",
        "plotTrainData(test_x, test_y, 'test')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "train\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAC7CAYAAACend6FAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOy9e5Ckd3nf+337fr9Mz21Xs5JWoAjpnMQKEEIZcMllg4AcgzEYGWFbnBALO5hKTJzEVMkgH2HHcWInKlIQ5IAxwSQxYNnBhmNOEZ84LuLYRoYTQyEFdNud3Z1bz3RP36anu9/zx+znmadbs9rVane6d7efqqmZ6cv7/t7b93l+3+f7PL8gDENNbWpTm9rUri6LjHsAU5va1KY2tUtvU3Cf2tSmNrWr0KbgPrWpTW1qV6FNwX1qU5va1K5Cm4L71KY2taldhTYF96lNbWpTuwrtsoF7EASvDYLg0SAIvh0Ewc9drv1MbWqHadP7empXigWXQ+ceBEFU0mOSXi3ppKQ/l/S2MAy/ecl3NrWpHZJN7+upXUl2uSL3l0n6dhiGj4dh2JX0HyW98TLta2pTOyyb3tdTu2LscoH7dZJOuP9Pnn1talO7km16X0/tirHYuHYcBMG9ku49++9LgiAYyzgulpYKgkCM+VxjD4JAkUhE6XRakUhk6PP8HYlE7LUwDBWJRBSNRhWNRhUEgaLRqCKRiCKRiMIwVBiGtl1J6vV6GgwG6vf7kqRoNKpYLKZYLGafl6R+v69+v69er6d+v2/78+fB/x8Egfr9vnZ2dtTtdrW7u2ufY+wHff+5nk8/jou9B/yYzvV+GIaHdoON3tuHtd+pXZt2rnv7coH7sqRj7v+ls6/5AT0k6SFJikQiYSKROO9DevZ7kg4GgtH3DgKbUUDiNyDj/8aCIFA8HjfgBKg9SPM5/xONRlWpVHTLLbcolUopHo8rDEMlEgkD52QyqWQyqSAI1Ov1FI/HVS6XVS6XlUqllEwmVSgUFI1G1W631ev1lE6nlU6ntb29rXq9rk6no1qtpng8rmKxqFKppLm5OcXjcXU6HQVBoEajoY2NDW1ubqrVaqnf7w85CoCc1xOJhHZ3d/XEE0/oW9/6lk6dOnXg+eH4B4OBwjBUv99Xt9tVr9dTt9vVYDAYOjcYzorXD7p2z3YPPJtDHf1cr9c757aeo533vj67X7u3gyCYNm+a2ljscoH7n0u6OQiC49q7+X9E0t0X+uWDAPagz5zvvYPAe/S7oxEnv4meAXUAPRqNStIQKI1G8WwjEokok8kMfYftSVIsFhuK4KPRqMIwVK/XM8DvdrtaW1tTLpdTLBZTIpFQr9dTo9HQ+vq6arWa+v2+dnd3FYahWq2WEomEgfTu7q4ajYYikYgSiYS63a4dx2AwMHAH0HEGYRgqm82qUqkokUjYTIDjGDX2F4lEFIvFNBgM1Ov1tLu7q93dXZs58H0P+gc53HNdT+9M/Pk/aFuXwZ7XfT21qR2mXRZwD8OwFwTBT0v6Q0lRSR8Pw/Ab5/nOs/7GRiMz/9r5AN9v0//tgXv0ZxTQ/d/8Ho3y2X4sFlM+nzcgBcCDIDCw7PV6isVi5kgAvd3dXRtDGIYGugBtLBZTqVRSNBpVrVZTt9vVzs6OjQ8aBfDe3d1Vq9UywI9EIkomkzbOwWBgDqLT6RhYz83NaWFhQevr6+YYmHVI+7RSv983QOd/f0y9Xs9+2I+0D9LPZs9GH53rup/LmT8fu5j7empTG5ddNs49DMMvSPrCRXzvvO+fK9o7V8R/ELB7MAcsPd3iAd8D2uhv6Ai2CVgFQaBsNqtsNqter2cgDagCptIeELOPTCajWCymZrOpRCKheDxu+2J7yWTSnMb29rYajYbOnDmjbreraDQ6tP0gCAzYvcNgBgEYe4oGx9Pv9xWJRHT8+HE99dRTqtfr9hl/rsIwVDwe1+7urjkKIvXBYGCOzb9HRN/r9Q7MXVwo9eJtdDZwITTfc7WLva+nNrXDtrElVEftQqbmo9PwC9nGaIQImANugPcoqB/0GxvliwEVfkejUSUSCZVKJfselAogD//O53EGgDnAj9PJ5XIqFAq2j1gspqeeekrb29vK5XIG0uw7n8+r2+2qWq0OgW08HlcsFhtKwEqy/wHFaDSqbrerZDKpdDqtY8eO6YknnrBkrAd3tk/ehKgdcPdAj8OIRCKKx+Pq9Xra2dkxJyDtz5D8teJ3JBIxh8p4SRR7+sdvb2pTuxZtYsAde7bInYd1lAYZ/e7o1D0IAiWTSQOLg5QonnLxypbR5OlBiVeADLolFospm80ql8sNOYbBYGCgxWv8DTDGYnuXJJ1O2/uJREKFQmFIMdNsNo2m2dnZUTwelySL2hm3V82QXPTcP2CIOiYWi1kyNBaLqdPpqNvt6ujRo2q329ra2rLte8UONA5OFBD3Tmf0NWgqIno/o+I84hTj8fjQ+ffA7fc9Sv/s7u6q3W4/hztwalO7OmwiwH00qXnQ+/63dLAiZjRRN0pBeEqBH58s9UDu9+O/NzpW+HRANBaLmZrFf4fIOJFISJIpVPxrAGM6nVY8HlcqlTJwh58HULe3t7W7u2tcPO8BkLyeTqc1GAyUSqUUhqG63a4BOtEwYOhlmHxuZ2fHwHFxcVGxWEztdnvIuQHu0C3ePOfvZZ9E8TjHUSrMXzufhD7XfUOkPgrynv6a2tSuJZsIcJc0FI154IarhZ5gis77o/wtkbmnW0bBfDRiH43UGQPgNapRB5iJYHmdbadSKWUyGaNe/H75v9/vK5lMDnHRyCXT6bSSyaTx7yhkcB69Xk/tdlsbGxva2dkZmpUQzROdx+Nxdbtd5fN5xWIxcwjIMXnfOyG+S2IVsIzH46pUKqrX6+bc0MATKcPxj54jxu3P1+jfPmonN5HNZu27jM1TLqOOnfsIqsbvc2pTu5ZsYsDdq0kSiYSSyaRx1P6Bl4ajdZ/QRDFCROyTnz5KP3LkiDKZjL2/trZmShBsFHj8PuGUvfORhnl1uGEoCqJvIt7Z2VlTjZBw5TOAE9F3u902p7azs6Ner6daraZGo2GRaTQaVTabtQgXsPVOLBaLqVAoKAgCpVIpDQYDA3mSszhLouqdnZ2hGRPbwBFAC+3s7CgSiajb7do1Ivr3lBfUDOaTuN5BJpNJFYtFG2MQBGq1WkbrcN5weJ7uYWyjDnhqU7uWbCLAPQgClUolpdNpoyI8N85nfAQt6RlRGk4BcCK68/xwLBbTq1/9aq2trWllZUXXXXedIpGITpw4caBa4yAqyEeZHpg8iB70ebZJRA2njSOAe+d9xgxI7+7uamdnR+l02igLCoYikYhardZQ0tJH4oPBwByfT6ryfzqdNmci7Tswnxsg8Ykz8DmEaDRqx0I0T8TtaRh/bkapsmg0qkwmo3w+r2QyaecumUzaMXq9PceXTCYVj8e1s7Oj7e3tofuCMU5tateaTQS4x+Nxzc3NGZj4h56H2QMlDzxRMQnG3d3dIV21V03wfYDoxIkTeuyxx/SCF7xAlUpFJ0+etEh8NDFLZM3+2Y53PIAxPHYqlbLPok1PJBJWFARtEZ7VnQN8XpLJ3+l02ugWovPZ2Vn1+31tbm6qWq0ajZTJZGxWwgym2+2q0WgM6cupJPXJSF9FS1IVUAWMec87M69i4Rx0Oh37PM6XMeI4OI9U1pbLZWUyGZNuxmIx5XI5RaNRbW1taW5uzmYIHB/yz3g8rkwmo1QqpVqtZvsZVTpNbWrXik0EuPuIFR4YPhrA6Xa7Q3w1AAlozMzMqNlsWkEPkelB+vZ4PK7v/u7v1m233aZsNqtHH33UImtP6ZDs8+oTn3QdTQAyJkCa14jEC4WCgSXqDwAceoUk4GhyOZPJqFgsKgxDOycbGxuanZ0doiiIeqFFOp2OJXzZPqCII+V8AsLZbNY07clkUu122xwk22I7nirLZDKWfOX4cIzeaUqyKH1mZkazs7MaDAZqNpsG6iSkuSey2aztw/P6zGI4j8ViUZlMRpubm0Mzt6lN7VqziQF3En65XE6pVMq024AkQE1ECLCi6Egmk8rlcpqbm1On09H6+rrx06N9VFKplP7sz/5MTzzxhN7+9rdraWlJTz/9tEX5o4ndUS7ec+0ezL10zycsE4mEyuWyJJnG3VMixWJx6LhwbES4cM6AazQaVb1e1+bmpnq9nvWKkfYllKlUygAQ3pxtw9H7iJrZTrfbVbfbVaFQ0Pb2th07r/tZFA6E8xKLxZRMJo16YRwcJ1z/0tKSyuWytTuo1+tqNptDEkscj7Qv7xwMBnZ8fKZQKFhCnXNOqwachb9+U5vatWITAe489MViUYVCwRpk+QiZaJQEmgcQaS+xCkDm83nNzc0pDEO1221rsNVsNtXv9/Xoo48acP3xH/+xbrjhBm1sbFgEDH3hC4YAVc9ZA+Rw7kTUOJ50Om0Rt1foIO1jPz6xms1mlU6nlc1mhxLMRKuck52dHbVaLYuqvQqIGQiASxMweHOcC4A9yvdLezx/Nps1GoRoularKQxDU/AAunDs9K9JpVJKpVIql8tKp9MqFAp2bsIw1Obmpmq1mskUfcUs1BozGmnPuTA+nFalUhlyHNxLcPCxWEypVGoK7lO7Jm1iwL1QKCibzdpDDTjxP5xqPB5Xo9GwB7rT6Zjyg+gV+gGHMTc3p0QiYUC/sbGhSCSicrmsra0tbW5uGg3gI9jd3V0DL4CdyBQwYhyAI2AMuBWLRUv0AvioYgDP3d1dbW9vWxSODDKVSkmSisWiJJk0sdPpWHERfD3jgo7yxT+cs06nY5QKAI0iBoXKqAqoUqmoWq0anZTJZGwcPicCfw+FcuONN+qGG24wOmhra0utVstyI7FYzOgzHAj7APB5Hyedy+Uk7c0YZmZm7PzhEHA2fgZEzcDUpnat2cSAO+AKpzwYDOyBZcrNQ0sU6jXnRM4AoiQDahQciURCs7Ozuu6664zHbzQa6na7phlHTocypdPpGPgA5p46kGQRopdv5nI5VSoV44iDILAuiwA0x4PKhBkACcV8Pm/A6WcVzWZTW1tb9j/A2uv1lM/nTTkDvdNut20bgD10hc91IHFsNptqt9va2dlRKpVSOp222YDn0plJdDqdoXzJK17xCr3whS9Ut9tVrVYzfp1r1u121Wq1hpKqfsbkQXpnZ0exWEzlctmi8Xw+b7MR35seZ+VzLlO1zHgskUhobm5OknTfffdpfn5eYRjqgx/8oL71rW+p0+mMeYRXv00EuANsqVRKsVjMkmIk4ABgNM/I5PzDy9/dbtciNfh6r0DxSpp4PK5jx44pEono1ltvVbvd1tramk6fPq1Go6F0Om3aeWk4KYszYdu+mjKbzWpubk4zMzNGJzFuHNhgMLBSfoCU7UNheDAFvBuNhlqtlpaXl9Vut4dUMjgxchV8h3MB+BLZA3qeD2f2g2oGnntzc1OdTkftdvsZnDgyxEQioZe+9KW65ZZb1G631Wg0bBaSSCS0urqqMAxVq9VUr9dVr9etJTGJbmYNjCmRSFhfHWYFuVxOmUxG29vb5kyhp0j4en37uSqfp3bpjPvwB37gB3Tbbbfp+PHjesc73iFJQzOnN7zhDbrrrrv08MMPj2mk145NDLhXKhVls1lJ+5EcD+7Ro0cN3NFa12o1o0Kk/RJ+vj+qc/ZTdYCv1WppbW3NGnZlMhndeuutuuWWW7SysqInn3zSerh4isZz30SlVFLmcjnNz89bpClpqOS/2+1asRARPICM8/E91X3nSF+NC8DyOU9teJ377u6uzYrYD4laZj4AY6fTMaklOY1+v2+UGTy5T+Iy65ibm9Ptt9+u+fl5bW9vWxVsqVSyxUTa7bZWVla0vr6udrtt0Rv781w7ydF0Oq3Z2VmbYZTLZXOY6XRaOzs7pjriHKA24tpM7fJaPB7X/fffr7vuuksLCwtGn53rsx/4wAf0+c9//lIuojK1A2wiwB2VBVEi03HkevF4XKVSyW4GerfABbMNAJOIVtrv4dJoNIY46FgsZtFuKpUyCgaQedGLXqTrr79ep06d0vLysgEuAIwKhYU18vm8abVZOcknS8MwVL1eV7vdNo7dzzp8+Tw0EVE8x0wHRWgHwJDIlFWb/BhJQBLNQ3vRArjZbBoHDw1WLpeVz+e1s7OjRqNh57dSqVhbAqJlchcvetGLNDMzYxJKIuxWq6VHH31UZ86cUb1eV6PRMNBFfgmXz0yp2WxK2mtxvLCwYODvWzKgZc/lckPXwTvK0RqJqV16e/nLX673v//9es1rXnPB9FelUplek0OwiQB3aa8RFtwuqhN6g3iqQdq7OfL5vFKplFZWVgwcebhHVS9EiL7gBiVLuVw24ECVAnWQSCT01/7aX9Ott96qU6dOaWVlxSgEn3iE62c7RNVUjIZhqEajoWq1aglXvs8PswyvwvHVm+QhOE5mJ+jYU6mUbde3Ewa0+Tx8fqPRMNrGNwnr9Xqq1+taXFxUNpsdosc4bziAYrGomZkZA33aA9AT5sknn9Rf/dVfqdVq2T58DQO0VSaT0WAwUKvVspWlUqmUFhcXh6gfonCS5Tg7f2y+HYVX20zt0tvLXvYyfe5zn9PRo0ef0/eKxaK+53u+R1/+8pcv08j2ZoN//a//df3AD/zAM977y7/8S33xi1+UtN/q+mq0iQB3uPJer6dCoTDET/vofXd312RuKCFisZhOnz4tScZtw9X6KkboiHa7bb83NzfVaDSUSqU0OzurXC6nYrGobDZrwFer1ZROp7W0tKSlpSXt7OxodXVVrVbLkqjsGxkjALmysmK0Drp72gED7sxYULwwy/CFUUTIzDioVAWoOb5UKqVKpWKKGAqIvKKFsdVqNW1sbAzRQmjjY7GYlpeXNTc3Z6s9FYtFS2TSrqBUKhllRXUwM6hvfvOb+va3v22zMTh6Cq2gV7yShwR0KpVSPp9XPp83B1kqlcz5cb4Bb64p2/cJX0/XTe3S2Ctf+UotLi7qwQcffM7ALkn5fF533HHHZQP3N7/5zfoX/+JfKJ/Pa3Z29hnvo5hrtVq67777tLKyoq985SuXZSzjtIkA936/r2azOdSn5MiRIwaEADMyQqLAZDJpfDw0B4DZ7/c1Pz+vSqWiwWCg5eVlnT59Wp1Ox6L5fr9vSbnV1VXNzs5qYWFB119/vfL5/FAbWbjdbDarm266yThtJIw4le3tbS0vLxvoRiIRk1t6WR+A6nuuZLNZU6d4nbdvgYADy2azxt1Le5EKIMp2KaYiot3Z2dFTTz2l5eXloYIt+HPoGfrEeFqj2WyqVCrp+uuv1+Lioo2JAjEoon6/r1OnTmltbc1mG4PBwLT+8Xhc1113nalmcC60Geh2uzp27JgSiYTNSLrdrh0b53lzc9MSsr7+QZLlVdD0TymA529HjhxRsVjUAw88oNe85jVGF16slUolC0IuhcXjcd1www164IEHdOedd1rR4EFG4CBJv/M7v6PNzU196Utf0s/8zM9YoHg12ESAO1w22ufNzU2FYailpSUlk0kDCQpSvOIlnU5rfn5e0WjU+OFodK/t7vz8vAHv7bffrmKxqMcee8xoFd85MRqNmgql1Wppbm5ON954o0qlkiUcoS6IvKPRqK22VK1Wdfr0aa2trVkJPpQC4AnYI8EEPAGyer1uzgeJZDqdVrlcNo5/d3dXxWJRm5ubarVaymQydtwAM6oXD+qtVktPPfWUNjY2TAYKd02y2YN1JBLR2tqaBoOBcrmcZmZmjN+G62aftCje2trS6dOn1Ww2jVrjvOHAMpnMUKESMthisWizKGYz/O0VPJJUq9VUq9UskvfXEurNt3aYgvvFWzab1Xve8x69+93v1pEjRy6ZrPQnf/In9au/+qt68sknn/e2giDQe9/7Xn3wgx98RnHjhVi5XNZdd92lG2+8Ub/+67+uT3ziE1cFXTMx4A59wMMahqFWVla0vb2to0ePGgVDUy6izHQ6rVwup1wup/X19SElC06Aop8bbrhB3W5XTz755BCQ4UB86f/a2pparZZmZ2c1Oztr7WfRoqPD7/f72tjY0GAwsKXt4vG46vW6AStJVPT5RJpQGYwXmgKZJEVbXrboE5kcn0+Weu2317+vra1Zx0RJxq+T1GUmMdqDp1QqaX5+XqlUyhbpQBJJCwH6wrB9cgS+ARn7BWgBeCSwOzs7yufzz8g98He5XFaz2dTa2pp6vZ7S6fQQXYWzZX9se6rIuHh729vepl/5lV+xmdalNF8n8ny38773vU8///M/f1HA7u1v/+2/rdtvv13vfe979WM/9mN65JFHnvf4xmkTAe6SDFwTiYRFdyhJVldXbare7/eVy+UMxGkHm0wmrZqS6TjVqwBGrVbT0aNHtb6+rkajYQVGRIhw9b7Yx/epWVxctNkDETiRum9VAD8diURUr9cVi8W0ublpWnYKf9CGIwFlJoI0kyXi4LnptkiyF/rEOw2UMUTx9FchiuccopgBiOHKpX01Ujab1fz8vH2HJmLSnnNAuUKrXegXHIFfBIQfJJh+sW6un78X2CfOl4StJIvKuX5sl2pY324Bxzm1C7cgCPTWt75VH/nIR6w6ehItHo/rZ37mZ/T+97//kuVWksmkbrvtNn3uc5/T29/+dn3961839daVZs8L3IMgeFLStqS+pF4Yhi8NgmBG0n+SdKOkJyW9NQzDzfMOxCXKAFdej0QiWl1dNcUEwJ9Opw24UdkQ/fkWBLSRZftM+6EWiBJp8kUiU9rvJx6NRm1pu3K5rF6vp2q1qkajMSTHJMqm8KlUKimTyWhtbc3oE2YSJIW97DOdTmtxcXFoNgOAeaewublpuYpR+oEov9PpGNA3m82hiLxYLBpYA4j00yfS5ocKVI6Ta/Od73xH/X5f2WzWOFRopkQiYZWu/rr6fkEUr/mWCETh8PB+eUAAHVkr54hZEZ9FHQXYXwy4X8p7+0qzH/qhH9LHP/7xia8ReMELXqB/9s/+2SWZAYzajTfeqP/23/6bPv3pT+vee++9ItfhvRSR+/eGYbju/v85SV8Ow/CXgyD4ubP//9Nn24CvHqUak4ZcvrhnZWVFS0tLlpwDLJiiU37f7Xa1vr6uVCplQIACh6k80SogkUwmDWyl/X7tzCQAoTAM9fTTT2tlZcUqQ3EygCmUCRI9dPdQFmwTjTgOgYZb3KxeVkjikWgXwENGCG/uK1CRRNbrdfX7fTsOaC0/48G5IvUkGuY3eQmUP0gW4/G4Ob2FhYWhhC6FXFtbW0PcOcliuHqULjhp34ohFoup1Wqp1WrZ7IVj99eJPjM4F+gt7qlx3dtXmhWLRf29v/f3Jh7YJem7vuu7LguwY5FIRHfffbfuuOMO/cN/+A/1uc997rLt63LY5aBl3ijpjrN//6ak/1fneQCIJj3XSnTHA02k12g0TN7Ea75plbSng6c3DLI4aU8CBWB6/hxpH4APqECxABDRaFS1Wk2nTp1SvV63pfnS6bS1LCbihwZBIZLNZk3iiB5eko0NsGo0GioUCrZARz6ft5wCEb/XbrN/tPa0SyYBK8laDgPQ0F20F2Cm0W63FYvFjAqilgDA3draGpJPQmVJ0sbGhoIg0MLCgp2rINhbYcvLMDkvpVLJOl8iZfS93/19wDJ+zLR89S4OgfNAspcENvfOJbLnfG9fSZbP5/XhD39Yr33ta8c9lAuyd73rXZd9H5FIREtLS3rooYc0GAz0u7/7u1dMO4vnC+6hpC8FQRBK+mgYhg9JWgjDED3RGUkL59sIvLekZyQsoUR85Nhut1UqlYb6nafTabVaLQOixcVF48pxAJLMWUgycBrt9IgDyOfzBvQkMgF2FCg0JNva2lI2mzWdvF9XlGIjSdZquNVqqVAoWM8UEryj0SdafXhoIm8cSzS61+IYnhww9v1w0MQTUVNhSj/1arVqenLAkmQw7QZ2dnZUr9fNAZLM5fxJUrPZVKPR0MzMzFDCtlwum2R0FLh9GwXfMA6nS7sJZnTMbGKx2FAPeACebflWExcJ7pfk3r5SLJ1O69/9u3+nt771reMeykTazMyMPvnJT+ruu+/W5z//+XEP54Ls+YL7K8MwXA6CYF7S/xMEwbf8m2EYhmcfjmdYEAT3SrpXkjXKOvu6pP0FGog0mXqzIHOj0VCpVJK0512LxaKBJpFtpVKxxB5AjELF0xhEvCRR2R5JRkD09OnT1kKAKBrgaLfbqtVqOn36tLXPnZmZGcohoGqBEqnVagb2mUxmCIxoG8yxA4w4uXK5/IzFMyQZJYMDI0HMuYXfJ4ewtbVlyhofcTMT2djYsIQx/d99D3vaCGBUivr++2EYWqWrryImN4BzYLuM1S94Th6FWRYzHpLC3Ad8FyWVXwj9OdolubevBCsUCnrooYf0wz/8w+MeykRbLpfTO9/5Tv3RH/2R5c8m2Z4XuIdhuHz292oQBA9LepmklSAIjoRheDoIgiOSVs/x3YckPSRJR44cCX1CkKiNBCe9R4iu6aJItEpzLICJAqVsNquZmRmtr69re3vbok6SbL4TJdGgny1I+x0rT548qTNnzhjXDGjRIx7+ORKJGF3QbDYteYmjSqVS2t7eNj4bkMRZEJl6h0MUinxQknHt/OCoyBsA1ihT/FhxKMxypH0HS4K12+1qa2vL6C7yIMwA8vn8EGXGPtDix+Nx6y0TBIGdAxqKMaNh3Lzm8xXQOTh28heSLFFOroJx4gS5H3CQ47q3z+UAJsl+7dd+TXfddde4h/Gc7I477tDtt99+6Pt9wxveoNtvv11/8id/cuj7fq520eAeBEFWUiQMw+2zf79G0v8l6T9LukfSL5/9/Xvn2xYACtjxwMOh+t4t0CuSrC0vOulCoaBUKqX19XWLEnd3d23tUhKbOATftwQ6w9Mw0h5NtLW1pRMnThgo7+7uGhjD61Os43XsJHnb7bZRDUTozWbTaCE/OyGBCAUSjUZt2bher6dMJmPRN4lcqAu/vB+RNvQQNBJLEkJl4FykPcoILTv7o9ulTzBLsvPgm3whDd3d3dXi4qJmZmbMUcORM6uBsvFLCvqqXGm/Va9X+fheQ1BVUFp+XCwOQu5lXPf2pNtLX/pS3XnnnYXWCUUAACAASURBVIe+3z/6oz/SxsbGRX9/YWHhWatQL5cFQaD7779fr3nNaya+b9HzidwXJD18Fhhikj4dhuH/HQTBn0v67SAI3inpKUnnJfFGe8d4bhlg83plOFUoFyJKIlIqUzc3Ny3S8xQNYDKqyiA6BawTiYRqtZpWV1eNUqEvC2Og9wt9Y/guPVr8Yh1eJw5nXKvVrD85enfoJfbDknfQUr4FA5EqAIoeHWAERHGa6MoZJ2Oq1+uKRPYW7PALYEAJQffgkHiPBDHFYvD71WpV+XzeFgAhArcbL7a/Ni5VwpxTSUOcPsfFcfIeTpNzxn6oAMYJIoMdx709yUbjr6WlpUPf93//7/9dtVrt0Pd7KexVr3qVXve61+kP/uAPxj2UZ7WLBvcwDB+X9F0HvL4h6fue47aGOFKqUVmmjkZfgBR0AnwsGmm/LBtUBjQDssNOp2PFMFA/KFpI3EUie4ss1+t1W6EJTTlJWxqHQdEQeeZyOVObEJVmMhnlcjnl83mjbJBc0t+G48FBsDDHaAMxnxCGxvCqFcCSMaH/5/xB80BZ0O4A50LC2vPvRMM4QK9ogR6CkoGLnJ2dNaD17QD4HP30/fg9LcM94fvb+OZkfI5ZFLw8oM//fqGIcdzbk2o33njj2ID9SrdEIqF//I//sf7wD/9woiugL59I9DkawA63XiwWTfoG6MIvS7KpNslNXqcqc21tTd1uV4VCQZlMxkAPWSB6cF81mUqlTPHSarWMS4eu4ELCfQPUvlo2Go1qZmZmCIChTQA+kot0W7z55puVy+WeAUTQFMj6cExeDw7w4gSk/Y6JRPk4ynw+r1KpZOeEmQySRXjqcrlsnTILhYJisZiNj5kPah4P2lBhuVxO3W7XZjnkHjz94q8p0kqoI4AZZ8S5xPH568asgVwAsyRoHBzd1PYtGo3qHe94xxTYn4e95CUv0etf//pxD+NZbSLaD8CDE7XncjmLIKmOJHrzja0oUfcAI8lW/YF2QIfOghvJZFLz8/PW/0XSUK9xgL3T6VjfczTs6LuZLfA9FBy+2Ihomv4rPioHlHAO5XLZFgyhJQH7oPrS95KRNBTde5qJ30hEffSLk9ve3rYCL857r9dTsVgcorqI2AHTubk5i+YlGQU0GAw0Nzc31DYBR8B58C2BURBx/FAnqHu89FXarxvwCV7Ol1/sBIeI0/NU0NT2gP3+++/XP/2nV7Y8/6mnntLq6qrm5+fHsn+UM//lv/yXiVXOTETkDnVRKBQ0MzOj3d1dnThxQsvLy9ra2hpqvOVlggCZ17u3Wq2hCI6ITtor0iCKp61wqVSyKBRArdfrFv23222Llknowf2SbASAvZQRygdwJYFJlAm/T4IQegIVSKlUskIfaCOic+gOr933MkDOKTMRL4HkO0Tz9G3xgD1qAL9PdJIPYUZw7NgxVSoVex+uu16va3V1dWjFJ/h2ks/eKTF2AJoZAdvleNHaSzLQ57OdTkf1et0SqlPbt6WlJf3UT/3URdFVl9Lm5+ef1xj+9E//VN/4xjcu4Yieu73hDW/Qv/k3/8baB0+aTUTkHolENDs7q3Q6rdXVVa2urlo5OgVIKFr8DQEI+ISsp2iY+o/SKTiEMAytUAng2N7etmXu/CIQKGNKpZJRJDTtQkFCNIq0EaeQSCRs0W+iZzj5mZkZo3xwNLFYzPq1w88TiRPNNptNi7C9w+Mz7Bv1DaoRpIqU9LMsHw4KhQxFSD4CRkbqJZnFYlFhGBp1BY9OxM7MglkORVu0HeCzvs0B113aT5T6ZCvOkXV1AXcAvdlsWg0Cs4Op7QH77/zO75gTHqe9853v1D//5/9cjz/++EVvwyvnxmX33HOPotGofuInfsIKFSfFJgLckcedPHlSa2tr9kACZFxEQJQoE0AC0KhY5IdIFLqEaA+QQI4In4xCZWVlxcDP0ztQCsgMAW6Ai8Svf4/kIa0TKI4KgsDWFKXTpbTXaAwQZuaB7pxyfaJy5H/MPKR9h8YYAEsoDNoM0PAMeSUznWg0qrW1NW1ubppjpS0BqyQxg8EBcY5Q2zBDoec6tBRKoXK5bOcUWoY8Atec/z19xfiYgTAToCslC3hAlQH6U9uzD3/4w3rxi1887mFI0iWhy+67776xyDhH7e6771Y0GtW99947URTNxID7iRMntLm5OTQV9z1lvOxutKGVX6jBR5keyAFk6BuibN8eACqmWq1aNE0rAECWaT4Kj0ajYRSElxBCD7B9yu/b7bZ1riQSjkQi2tjYULPZVKVS0fz8vEXO8NleR8+5AOA9rcGPr/SElqHBl19T1vdfYcm7zc1NbWxsDIEykTZad8//02aB6SmLXAPIzICQSFLY5fMXRO9E2YzJR+dcR2ZeRO/1el2nT5+2xDXf4z651iP3IAh0991363u+53vGPZRLat/4xjf04IMP6qd/+qcvqlDtUlkkEtHb3vY2dbtdvec979H29vbYxuJtIsB9Z2dH1Wp1SI9NEhVg9xEeSTyA00cBgBkRNv97dQYJTXhntr28vGzSQGgbABZ5I9tGZ+4LmjxPTNWm3ycac6SYUBrS/hSTRbRTqZQKhYIpYXwR1yjVwI3teWxpDwgpiGo2mzazkDQk36SwiYTz5ubmUKsGaB+ieBRA6PDn5uZULpetNYOkoYIunJi0p27a2tqyxdB9jxsfsQP2XD9mQWjufYKVmRbXn+P3fe6vZcvlcvqX//JfTnRv9ouxdrutn/3Zn1U0GtXf//t//2LbTFwyu+eeexSJRPSOd7xjIgKKiQB3X13ou0NygohUiUp90QsRKtGyr3D0xUk84ES6/rOxWMxWKqLvCUDIzACH0u12Va1WrWBKktEaAKIvgwdA/XF6OgGVjC/a6XQ6tsQdLQHg0/1iHOzLR/c4H5KKzWbTFtYgl8DSgzgKaY+iikT2Oj/6Qi++CzXkcyDQNEEQ6PTp00PFRtK+w4Iv94nWZrOpmZkZ62vPzIoZDQogryzyY4KqWVtbG1rvdZSHvdaBvVAo6GMf+5gWFq6aHmdD1uv19I/+0T/S6uqq3va2t+nWW28d63he//rX62/+zb+pr371q2MdhzQhahlJRrmgzZZkChFpn6MjmpNkU3MAjfe9lM6rM4hA6SHjwaJWq1k/dq+xJhnLikxbW1tGxfjkZBiGpp6RNDRrANxQlhQKBXMWRPfFYtGcF8DFdv3xwdf7aSgOChBsNpsG6iQv4/G4yuWyjh49avkFCscGg4Gq1apWV1ctIQqtBOiy2Idf1BonA9998uRJ612Do0IW6ReyJvFKQljal1T6iN/nVrgfcC60GNjY2DBHjxP3NBldNq9VW1xc1Jvf/OarWhLa7Xb1wAMP6Ad/8Af1zW9+c6xjqVQqesMb3jDWMWATEblLsqk3YOx16z7yA8iIEIl4PRDzvrS/EIhvBIaeHQDZ2NhQvV43MIK3ZZZAszCSkSQ7faTIeEk4Um3quzMChnyGYwDA5ubmjIJhIQvkhr7/jLSv/KFYh3MEiOIsR88tcssTJ07Y+FutlqLR/dWjmLXweZLTVNxyHgF8QFnSUAXvzMyMnW/vfCXZuczn85bXGJ2t+fwF54kZBs7P5yO4/iikUFxNwhR5XPYjP/IjVzWwe3vsscf0pje9SQ8//LBuu+22sY3jta99rR588EFVq9WxjUGaIHD3Wm1JprIAEHmPRKgv5vHqFUAA89Ez3/fl65K0tbVloO4pgEajoXq9bm1xiar9ohKjenZWMvL6cp/wJcHIuKGhUNUcOXJEs7OzisViBvAAsqQhKsa3Lwb4+SGfwMwGnpzePcePH9fa2po2NjbsGPgc6huKoLwaJhaLmdoG58HY4M5xFKyGRQdJSXYucAD5fN5AmHPP9eNYMWZKALmvQmUm4XMl1zKoY6961avGPYQD7etf//plSTwC8L/7u787NormJS95ifL5/BTcJVkiTdKQwsFH34CLX8eT0nmScP4zozppwMODoCQDSd+EDAUIVI0kAx2fH8hmsxZ1RiIRA0HAi/a9o7QKEbPvBAkV5Dlrqmlp2EXfc6/95n8vhfRJZ/IYUFhw6PV6XZ1OxyJ0vgfoxuNxlUolc2Akg3u9nunvcQJUukYiEaN5GFM8Hlez2VQmk7EkebfbtZwFzoe2zZ6nx9n4YjTAm3PFOSDZ7FsXePXS1CbL/uAP/kBra2uXZduPPfaYfuiHfkj/43/8DxUKhcuyjyvBJoZz95JFIkQPzDzgPmrj8/47/oGXNATkzAJ8gRR6ds+fE7FjRLW+1S79xZlFoK0epU4SiYRyuZwKhYJKpZLK5bK1RCDy9cU7fgYC+PlEs1cQoe9ntuJ5foCWTpKcYxKztEo+SF9O9S2acpwDTqZcLqtSqdjCG6idUA3hEP218gVOOEC4e79IyqjO/aBkuL+uVKtynxCxexrtWgX322+/XS960YvGPYyx2IkTJ8amOUcaOW6bmMhd0pCqxCc94V0BG3TXfgrvJY6A+EER7qgO3Fdy9no9ra2tDfV88V0WASEfaUNJwD+z+DTOAg09Mw0i6nw+r2w2q3w+b0lCxtfpdIYKi5BB4iy8KobZBHkFjhkQ9Vw0zoBzTKthPuc5+Xw+r3a7PSSzZDlBuHkPrhwzn6M/PteCCJ5jxxH64izvtL2T4n+O0zcs49qPBgCcB67BtWjHjx+/ZpuDNZtNPfDAA/rIRz5y6PsOgkCveMUrDn2/ozYR4C5pCKR88Q4Pu9e07+zsDAE/AACAjlI8vqDJSyhp4YuG+syZMway0A18NhaLmRab1+CxfdRIJNtsNo3z9lQCrYLh3X0RkW9PgENgf7QB8Bw0jsv3u8cR+upOonqvK9/e3raIl4Ijaa/9QrFYNF38wsKCnT9qAFAZcd4ikYhV2HINcIhcI6gyFvdgpSboFo6BBC7XyWvWcajeYXOsnH+v8feVr9eaxeNx/ZN/8k/GPYyxWTqd1k/91E+Nbf+lUknFYnGsPesnBtwlDVVfei7W88jxeNwWhfBaaPqjewPM2DYOATqCha47nY7x6+xvVFEDF4y0ECcD5YCjYBw4BLhmuj22222l02mLeH3/eVoMl8tl475LpdJQRS6OB87bN//CoDdG+6/giPxSegA/xwenjrRzZ2fHNPd+1SOibr7rnaJfYxWH4J1iv99Xo9FQLpez78PZQwGxbZydj+p9cppgwMsevRME/K81S6fTuuGGG8Y9jLFZNBod66zlla98pV796lfrs5/97NjGMBHg7iNqnxwkQcaDzgOez+ct4pU0xDcDvoDVKA9NNA3Y7e7uan19fQhI+v2+qUoajYbROIVCYYiSCMPQtO8oVnAMKGiQ+kHvQLF4CR90CLTF2trakCMDAEcTpB7AkCt6esb3m/eRs9d+cw5yuZyazaay2az6/b5SqZQqlYr6/b7JG1dXV63PTLFYtDVsvWpJkoE21JqfTXFtY7GYtre3lcvlhqqJvcbd3x+jSiAP2r4AyheN+ZnMtWbvfe97tbi4OO5hXNP2wAMP6OGHHx5bf6OJAHceamk/seqrMH13Qvjb0eQiDzQ93n0PeC9b9DI7ep1w8kmSMmsAyBlHOp226Jsk7GCw17UR2ohoHtvd3VU+nx9yPp524PhJ1mI4Fa89H12MwgPhQQAGHUJHSL9qTKfTsXVGsdEFNQBzFrkGQJE6sm06a+KQ0KIzdmYSXq2ETn9zc9N0/RSwcS39MXGNfK4BJ+kTqh7cJQ39fa1YMpnU8ePHh+7DSbK1tTU99NBDl3Uf73nPe8bebuGmm27SBz7wAX3wgx8cSwfLiQB3SUNyPcDa93kB2FF+jPYGJ+KHP4aW8EU3JCeJYkkOMq1HW047XaJEHAsLXPjCJRwG3DM6bwqKiMZjsb31V+liCXWRyWSsRa9PnrJvFsTGPDByXJ7XH9V6S7LZD2BM90u09UTPW1tb1j+H/IJPSIZhaCDMviks8pw/+yfZ7FVKzGQoiBoMBqrVaioWi+ZcPG+P+WjeLz7C/jxHPzqOa81uueUW/eiP/ui4h3Gg9Xo9ffSjH9Xy8vJl3c/Ro0fH2kxM2gs03/e+9+m3fuu39Oijjx76/idGCgn/zEPPAwqfS0RM5OblcVAcgJnnqD1Vkc1mjWemKpVoOJ1OG3iwWAavNZtN65Toi2fYD8AO7QOIYvRPoe0Aqx8hS2QlJHTsaNvL5bJmZmaGCn84JzgWaV8pxGyGH8AZR8N54vsA+OiiFr7BGY7Wr+9KFD0Y7C0diJy0VqtpY2PjGU3AGCvXllmVX8CaJmQ4Xq6dXwXKc/Y+UcpvroG3azGh+sADDzzjPEyKPfDAA7r//vuvmRlVLBbTv//3/17XX3/94e/70Pd4DuOBJmL2DaZ8NMbndnZ2rLPgKB/rl4YDGABOFCRksQEEH7nzuXg8rlqtplqtZqoPvwg3lBF0BNWbyWTSpoQ7Ozu20PdgMFCpVLLKUuiFI0eOaGVlRc1mU+l0WuVy2VaJkvY5dfj80Qpbr4mH+sEBApieG89kMuaIstmscrmcksmkNQgjSewVJzg070R8TQHv+Ug+m83aPiSZfJUI3VNw9LX3XTV9spfj8RH8KCXH66PnZVLpictl3OuTZidOnNDHP/7xy85Bv/CFL5wInTn2t/7W39Ktt96qp59++lD3e15wD4Lg45L+D0mrYRj+72dfm5H0nyTdKOlJSW8Nw3Az2HuKHpT0ekktSe8Iw/CRC9iHAVgYhlbc4nuqA9wAlJdAelCDjvFA4Je/A4BIJI62HQDcq9WqLWiBRBHQZCk/CpSi0ahF35Isyib5y5iZCVDN2u/3lU6ntb29bbJCtO/QHz5vQAfGUQpCklFRjAsQ9f3vAcJcLmdqGCiiRCKhmZkZa/37rW99y1oM+LVj19fXtbOzY7QSMyGfOI1GoxaJk5wm4saxSHpGR89Op6NyuWz6ehyqvz+QRHLMbMPnI7jOB3H3h31vH7a9+MUv1t/4G39j3MM40D796U/r5MmTl30/iAEmye655x596UtfOtRZ5IXM3T4h6bUjr/2cpC+HYXizpC+f/V+SXifp5rM/90q6oAqCUW0yC2M3Gg2buvOwEolCIxwUqfppPcAI6Pt1VgEcuGeiz52dHTUajaFkbhAEKpVK1me9VCppYWFBpVJJyWTSQIcxwHNL+xWoNOny3DWfzWazKhQKmpubU6VSMWcD745U0NNNABuOC9Dj3KDyYV3WTCZjElLGRZTH8ZdKJc3Pz2t+ft4W3yBxvLu7q2KxqLm5OVs2kAIk75xR0UCfVKtVq07lentOnPH4HIufnfhrPWqeooEC8tTUedZQ/YQu87192Hbs2DEdOXJk3MOY2oi97GUvO/QZ5HnBPQzDP5Y02gHnjZJ+8+zfvynpB93rnwz37E8llYIguKA7Dd01XRfX19e1vb1twOsTZ1466XlX3x4WkCX559sIoxAh4ue9wWCgZrNpNMxoHxP09UeOHFGhUDAFi+eyScYyXhb58Jy8H28mk1Eul1M6nVapVFIul1M2m7VZgKcpKJLyTsRz2zgKFqnAyaG1J4Lm2Ee7b/ombddff/0QSDD+XC5nMwuOh+ZfVLdGo1EVi0VTAfX7fW1vb5vCSNpXvdCvBuULPeNJnnvKjesNdeT7BHG8Pqo/nx3WvX1YFgSBXvjCF457GGO3m2++edxDmAi7WM59IQzD02f/PiOJlQCuk3TCfe7k2ddOa8SCILhXexGQKpWKyfXoF06hzdnPDkkHfeTn1TIAJtEe70nDER7b3t3dtUKlRqMxFK0D6CT/AEcSpr4acrRNQhiG1qqXBKGXbkYiEZVKJePnu93uEPWDc/DgDRhStSrtAy4cvrRf6Uvk61VDAKafNfhWDF7/vrm5qVwupyNHjmhzc9OSqVS3otmnoItx+v7tFJ4xVqgwet4wRk+jIBP1NNIoJeNnCRwjbZY9fefbEozr3j5Mi8Vi+gf/4B8c9m4nzt773veOewgTYc87oRqGYRgEwXMmksIwfEjSQ5J04403hl537lUsPtnpG36N8s6jU34f4ROdwkUDPjiFVqulzc1NSfuac/hcFC0+8SrJJIwAjgfMWq2mdrttSUkoHcCn2+0qlUrZghV8B8BjexyHj1KhY/w5wIF5RzIazQPyXmfeaDTUbrctucuYWeM1kUioVCrZ+QOckWyiSffy1Wg0asVebANwxZmyX9+AjO+TM/BUnaSh2dtoVO6dFOcTB+C3cRH36PO+ty/m+1Ob2qWwiwX3lSAIjoRhePrs1HT17OvLko65zy2dfe28hjIGvt0XMWUyGdNtewphMBhYgyof4UrDlakk/4hyAZBIJGLyPR8tplIpA6fRoiTA1y9AEY/HtbGxIUlGiwCyVKv2+32VSiVLnm5vbxt9QQ6g0WhY75VRVYgvgvLUitfyErH6GY6v2hytzPX5Bl/shJPpdDqqVquqVqsGmL45G8lSnJxf7xVqCSoGWaaPtlutlql1cDicM74zSsP5WZqvB/BO3S91eBEJrEt+bx+W3XDDDUb7TW1qFyuG/c+S7jn79z2Sfs+9/uPBnr1cUs1Ncc9pcN1bW1uq1WoWgfIQ+2IkaT9S89NuokceZj7vI8pYLGZdIFmPFMrBt9WlPwpgCFVETxS2y1Jvp06dMsc02hYAZ8XsgzGj6YeTRwPOrMLr99nfaHsBP5PAKeHEOK+MHcqJVrucB/8/bQcymczQbAcHwsIl1WpVtVptaIEMSXZMvkvkzMyMKpXKkIoJGqvX66ler1s1rrSfvIUO8/eILzbDGTA2P4PxdQ/nSageZJf03j5Mu+eeezQ3NzfuYYzdLsciIFeiXYgU8j9IukPSbBAEJyV9QNIvS/rtIAjeKekpSW89+/EvaE8q9m3tycX+zwsZxGCw18QLLtcnMwFcz3MDUl4vS1RJlOejeKI7wI7VhVZXVy3aA0ySyaRtl4ifMUp7ALa5uWmqFxwH3RaZAdB1krERtdLPHX7byy+p4GR1IZ/sBczh3A/SuPsiLugm+Hvf2ZGkbKfT0fb2tgqFgsJwr6L09OnT6na7WlxcNGqnXC7buqicH4C52Wwqn88rmUxaxO87UVIxHJzVvUP/4EB2dna0vb2teDyumZmZIWfMzIvjB+yDYL9BGedXkoH/aAJ2nPf21A7f7rvvPr3uda8b9zDGbucF9zAMz1UN8H0HfDaU9O7nOojBYKCtrS1bIMMvukGS1CcHfYUqwOmLYjxfSxSIg+Chp9kX3wFM0LEHZ/XxFP+wiEej0bB9wrN7hYePumkeFo/HTbHC72azaSsaccwAFk4KXbu0X+TlZX+8zt9e7++3y3veIYxSIOQXoJWazaa1UqAaN5/PW892HDD69tGI3M+UuJ6RSMRmAhwH369Wq8rlcjZ2rjXX0VcGo62X9vu+e16de4Rg4FwAfxj39mHZ0tKS3vWud417GBNh4+jjMok2ETXKvV5PzWbTptC+/aznrUlGYjz8XgIJ6Ph+M1AMFNnU63XFYjFls9mhAhsiP1Qevv0sUj6ADq6Znu9Ep0TxAJdfmBoVEFw1MwrGEIvFlMvlTGnDMnZ+m6N0FA6JWQTbxDnQ7mA0EetVLP1+X8ViUbOzs5qdnbUOkeQn4L/7/b4lu+kkKclon+XlZZ06dWqIZgGEcVRcD2gormEkErG6BgCZ2ZRXDqHC8U3MoLkOUhhdS2XuMzMz4x7G1M5hCwsLetOb3nSo+5yI9gMk+aT9ylOiQd9THaAHdKR9hYiP1r0kETqCaJNe5QC3XxYOh+A7E25ubhpn7sGUxB00hN9nqVQyYPIVo2jEcVSVSsXGT5Tv2xgTtZMTANBY0MLLA0d1/3DwntrguCVpZmZG/X5fGxsbSiaTWlhYMKVLo9GwWQWAiuOltQISTugjQBo1UhiG1n7AK39833dmMShqyIHMzc3ZdfROiGQykbl/zatqfB7C3xtXs915551Dealr2dbX1/XII4/oxS9+8biHYpbL5Q59ycOJAHevhIGHpcTdUzD+QSWhxkM+GAyGCoYkWZTIsnVQDTgLom9kiUS9vV5vqGOkN19qD1gRXc/Pz1vkT9sCaQ8QW62W5ubmlM1mbUYCUKIcYZFuAJxxEpV63b7XevtGad68RBH6yHeapEMkoNrtdnXs2DFbRabRaNg2q9WqafEp1uIcA96ZTMZmQCRyaXdMToJkqJcqeoqF6+j73PgCLq4/15XXfKsBzhs02bVgb3zjG5+rnv+qtTNnzugrX/nKRIG7tNcCmHzbYdhEuHo01FAgKDlGIy7vAHy0iPleJdI+5UBkR3MqKBBoGhQi8Xhc7XZbp0+fNkfAmABtACefz1sPi0qlorm5OYuUUZ34JCjjglKgm2IYhlpcXLQEJvvrdDr2Xfh4OG1PyXgKSNqPXL1TgNbxKhq2u7Ozo1qtZpTKd77zHWtrXCgUVCwWzTHgdAFVfph5MQsimua812o1S74yw/DFZx74oc8A53g8blWvXiXkk9mcW+4NT1354OBqtVQqNZVAjtinPvWpsS2QfS77sR/7Mc3Ozh7a/iYC3AFNuHe/Yj2ROaAMwPF5Pw0fdQh0I+TzvuGVT+gFZ4uMBoOB1tbWrO0BVAEREZ8l4l5YWLDS+yAITO3jwY7/6cGC1BGlR6vVso6QjBcKg+On6hPNOL+9VNO39OVcsC1pX2HC+QBIaSOAo1lbW9NTTz2l5eVl491brZaBNvsoFAq2vdF2DozLV+12Oh1tbGxoY2NDQRCYM+VaeEqJc45CijH7hDV8uz9uT5EB7KMBwNVo3/d936c77rhj3MM4r7385S9XuVw+lH395V/+pb7whS8cyr6eix3m7GoiwB0te6/XM67X86Y+YXjQe7zv5ZJw3jiFzc1N24Yvs8dZrK+va2NjwyJFtsUP0/uZmRmVy2WTFaZSKVOB+G3CJaM9h3rBaeTzeS0sLGhubs6Srb5DJQlVFrHwjo2omKiVxKuPlr3CyPPxlxehpwAAIABJREFUfm3YdDqtQqGgcrlsckhJJnus1+tWZUuCuVAo2IpVyWRSlUpFS0tLmp+ft5mTT2riNDAWD6dKN5VKmZqJlsse1FHejNYt+FmUV1Z5xdS1AOzS/vmedHvVq151aN0au92uvvjFLx7Kvi7U4vG4fv7nf/7Q9jcxnDtURK/Xs26EHiR4wL3Mj2pKX7Luo2u4V2m/fJ39ED0HQaDHH398qBUASo5Wq2WUABSJb4uA8ob3y+WyotGo1tfXbVs7OzuqVqvmBJaWlpRMJlWr1SzxypJ3JEqRAfJ9wMwrQGjv65OM0n7iEeD3wA53TS4BR+h165IMyNPptDV0k2SJZt6ndTCSyVKpZJQT4xpteyDJ6Jx6vW7ngM9TkYyj8lWwPrrHoUr7yWdfB+F7y0xtMuzhhx/W6dOHV/f15S9/WY8++qhuueWWQ9vn+ez6669XKpUyEcfltImJ3Le3ty1J51/ngfal5J5zB+A8ReHVIYChB7kwDC0qrlarVsBDZA8Hf+TIER07dkyzs7MqFApDhVHwu4A/AP3000+rXq/b/81m05zPzTffrEqlYtEv4O170QBUROA+KoVa8WA4KgP1qiFAHoD1Pc6Z8dCvPZFI6LrrrjPHRuVqo9HQmTNnjA7y+yc/ggPg3FUqFStsYtuMPZ/PK5fLqdFoqFarqVqtWkIah+NnZj6KhwbidS+b5HhH1VJcq6vVIpHIFVOw89hjjz3jGb+cduLECX3qU586tP1diN155516xStecSj7mojInSImkpAUCPmkKdwzGml6nEQiEePSverCF8PwfUACtcg3vvGNoWQp30USyBQfhY3v/uiTn9ApOAycANRQuVzW0tKSjh8/bkVQfI91VWOxmAqFgjkPOj3iyDBP/4xW4+IEiHal/apN6JhOpzNEG0n7NNJgMFA+n1ez2bTiJpLbzWbTZkORSMQap8XjcdXr9aGqUl/lyzli7NIeR88iJq1WSydOnNB11103lBRmhkQk7jXrnpPn/hlNrF4rEXskEtHf+Tt/Z9zDOK+tra3pX//rf33o+/3Upz6ln/iJnxjLMncHGdh0GDYRkTsAMBph8YDyoI8m3jz4+OpLAAg6AMAjORuLxbS8vGzyPxposdRcNps1eicajQ4tRQcFA43h+WDoBKpSy+WyNdFKpVJDhUEA02jRDn3QfY6B4wHoiex9m19pn8byoIhahjFzTOQWWMt1fn5+qAgrlUqZpBPqpdlsWq/95eVlK2jKZrPKZDImYyTZG41GlcvlbObDjY08kmvSbrf1+OOP2/J8UGjMKpi54fhHE7Y+t8C59MqgqY3fPvzhD2ttbe3Q9/vkk0/qzW9+s06cOHH+Dx+SHVax2cSAO2Dm5X0+asNGO0N2Op0hTnp0W6OKDIAdTTwSP3q8EH2yPaJ2aB2aZaH4CILAVmbq9/va3NzU9va20Q7tdnuIZmFsaOk5/nQ6bclV+GZfoekTo35mIu13jcRRjdIbzHpG10P1AJ9IJDQ3N6f5+XnNzs4arUJUTGtezgX5huXlZeuImc1mNTc3NyTDRP9eKBS0uLioXC5nFI5vztbr9bSysmKgzvh9fx+vhBmNfg5y8P4cXK3GPTHJ9vjjj+u3f/u3x0aP/cVf/IXe9KY3HfoapueyX/iFXziU6H0iwF0aVkEAgjz4vusgESjvee21f+hxEl7pEo1Gtby8rK2tLQNo+sdkMhnj7/ksNAsa4lHlju8mCa+NVp4xHz161BQh9ETp9/smt0TOVy6XbYbA7AEVDBG6PzeYz0cwdqJd/+CP6t35O5lMamZmRvPz87rhhhs0NzendDqter0+5DT897PZrMJwf9GTbrerM2fOqFqtanNz05KdKH4YA3JKQL3dbhsQ4zjPnDkz5OA5Ho6N2YuXfHoqxuvgfSR/tdq73/1uXXfddeMexrPa7/3e7+mb3/zmWMfw1a9+VW9+85v1hS984RmFiYdt1xQtIw13/RstQvH8NglGpuck1niPzwO+WCaTsbJ6KINUKqVyuWwtDXAe/f5ea1+SohsbG1ZxSj8VaBJmD74aMplMqlAoaGZmxkCa1yXZakhsA0lkNpsd6nfji7JGozOfHPbAPkrR+O6KgDp8O8eDJNGX8tNPhyrbRqNh8kNklCSTicJpruZbKDA78g7ZK5ik4W6OjUZDq6ur5qh53T+QOHj24XvU4BRwAFc7uFPhPKkWhqE++tGPjnsYkvYi+De+8Y36xV/8xUNZqHvcNhEJVWm4bauPTokWfcKTKA3Q9FNv/zD7niOSrFf8YDAw0KKs3evVg2Cva2S73bbIH1kkrW1brZZxyui//ZjpqbKwsKByuWwyP6o1Ad5YLKZSqTQkCRwMBkMNwzyQ+5mJT17izLxChMjVJxs5L9Ad9LUPw9CUPUePHjXHhfafXjzw55w33+uemQzaeAq3fItjHIe031K52WwONTZjZaz5+Xn7Ti6Xs8jfSzk5Ln9f+HzF6Exnaodrf/InfzIWrv1c1uv19Au/8Av64he/qK985StXdT+eiTgyzyF77tQ/pPDGXs/tIzRfGcnrvsXv6urqUOk7YEAkTzTfbrfVbDaNdvDackm22tD8/LwqlcpQ4Q6tCGZmZjQ3N6disaiZmRnlcjnjlUneSlK5XLbPkIz10k/2y7GzgLhPEI/SUR78pf1Wwr7oCWqJSJ6GZZlMRvl8XouLi7rppptMtsh2oVMoNCMqhxunQdrm5qbq9bo2Nzd15swZra2tWVtmaW8WxfGij/d94Ol1M+qU/OzDz+o41tF8iQf7q9EymcxEabhHbWdnR7/xG7+hanV0DfLxWhiGeuSRR/T+979/7BTN5bSJAHdpnyP3MjYABRDwkboviAEs/fTUSwTRY4fhXsdGGl2RWISe6Ha7ttIQXD4J3SAIrKfK9va2JUOp8pT2VyLyiUhAk3J9om3GFQSBFhYWhpQwXt0ymkz1AMb4peEl9TgXvpJ1NELxxUmevqC1wuzsrBYXFzU3N2c9aBKJhLLZrHZ2drSxsWHFWtBYkuxcMqNot9uqVqtaXV0dmiFxXD7pSh8bjscDNN9jlsFncFJeFupncj4Bf7XZ3NycfviHf3jcwzin/a//9b/0W7/1W+MexoG2u7urD33oQxOlornUNjHgLu0nVYlcpX2ahkQk3Dag4vloIrZUKjXEz1M4QWUl/WB8v5pGo2GJVkm27ByOBt03za+gE/xi26hmKpWKZmdnTV6YSqU0MzNjETIOJggCLS4uGv1CIhSA8olar/zxSVyOnwpaolqfSPW/+RtH4p2BtK8xZ8w33XSTlpaWlMlkLAG6tLSkQqFgMymcWrPZtD75fsbDOaUoalTiidNiuUHyGFTxEdH7VgR+ljZamOVnfldzZDbJ9vjjj+uuu+6a6IUz6vW63vKWtxxq1exh2sRx7l7i5/lyD06AgbT/QHvZn48I4WjhvD2QrqysGJ8Ov+6rHwGnnZ0d5fN5ZTIZ07oT5XqJJYVUaMSJnjc2NtTtdrWxsWHLzlWrVeuhTjIXcPeFS556GE06o9v3DtEvbOKjebbpF/720a7vj8M5oPDphS98oZrNptbW1tTr9XTmzBllMhnNzc3ZAiRco42NDWWzWXN6kqxfTK1WUyKRUKfTUaFQsJwI+8EYB3kOtoNj9xWrzH5wSj4B75PcUztcmwSFzIXYI488oi9+8Yv6u3/37457KJfcJgbcpeGIkv8BKADaR6JEqT6JinxwtG+K365vkwu4ezAA7HAERNoY0aaXKVLgk8vldOrUKaNn8vm8gejKyooikYglEGlQxrERkfvksW+ANlq44/MLjItIlgjXL7TB+fJ5DOga/mZ70eje4iDJZFKzs7N6wQteoHg8rlarpXq9rlarZQBN5E4imuIujoE+8s1mU+l02hbkXlxctHExjjAMjdZpt9sG+kGwtwYr+RBJQx0i/bGzHdQ9V6u95z3vmciE4Gc/+1ndd9994x7GBduHPvQhveUtbzF69WqxiQD3UeqAhx31xCiQA2xe3+2lgPDVmUxmSCFCIyqUFqySTpRLab6kIalgJBIZ6vvCPnu9nq2mxGLTkkzmSFTa7XaVSCRMsknh02jjfp8sJur3x02E6itTPUB7p+j5dE+/8JrPb3juvt1u2yyHdV7Rwkt7PCp1BZKGlstjpakwDK3id3RNWQ++GxsbqlQqphIiYcsxcm5wzujr/exuVCnjE/Me7K9Gu+WWWyaugKnRaOihhx6yNtBXgn3ta1/TZz7zGb3zne8c91AuqZ3X7QdB8PEgCFaDIPgr99r9QRAsB0HwtbM/r3fvvS8Igm8HQfBoEAR3XsggfORNlCoNF+14WgLqQpL1UaHQxj/cJAKpEkWWF4/HtbW1ZaspUcQEVyzJFr+mPa2PBBOJhEWXZ86c0crKira3tw2UpT0+r1qtamtry1Qt+Xzeol0SxRyHbyfgk6qjCVGfKPXKH/+3pyh4zydh/XmUhusK2D9FSGEYqtFo2DksFAo2G2m1WkMJy3a7rUKhoEqlYkBMkdjo9WZW5LtxEuFTx0AS1evavTM/CLipfvXFV+eyw7i3rzX7yle+ov/6X//ruIfxnO2BBx44NIf01a9+9VCS/Bcyp/uEpNce8Pq/CsPw9rM/X5CkIAhuk/Qjkv63s9/5cBAE562wAGi89NEnV3nYfTRG5OipFt9IKggCpdPpoW6R/X5frVZLq6urajQatoCH54bL5bIVNqHjhqdGz07LzlqtZq0D6B9D87NarWaAS0tcKl0zmYwlVH1PFByQ59YBT37O1VvFSwDt4rriHmlfAulB3YOk7/gYiUSUy+VUKpWszQKJ6TAMDXgBZxwY+v+FhQUdO3bMzhfbpCEaEX6r1dL6+ro5R5LGOG2fHCZn4mWQo+arUi8gav+ELvO9fS1Zq9XShz70oYlOop7LlpeX9W//7b89lH19/OMfnwxwD8PwjyVdqFD1jZL+YxiGO2EYPiHp25JedgH7GAJuX43K+0SuUCBeRcL7YRhaUQ4l60RvqG7QX/vl4dgOAASHDNhTBQg4QqlASUDzQHXUajUDJpQ5gFQulzMaxGvnoSP8OPy54Tj9OLwGns8ChN7pSXrGtkfL+aGcAFjeRypKrxmoFmY9g8HAKlODIFC73dba2po2NzdtWcPRDpTQXPTVbzQa2tjYsKQ2slUftXPcOHCvm2dGBR3lz8uzAfxh3NvXkn31q1/V7//+7497GBdlvV5Pn/zkJ68q5czzycb8dBAE/9/ZqS1rZ10nyQtHT5597VnNUzHSPjD5h9dTNIClL9zxiUOiPgCCbZ86dco07HQ0pIsj4AnQAfCj0sNoNGoLaPA5FBydTse035FIRLOzswqCwBYfAejK5bKperxuHccyWnjke8KQh8B4zXPPzBhQzYwmGTmnnHffpGz0czjJ3d1da1WQy+Ws6MhLNGOxmK2/Wq1Wrd1Do9GwxCyO8SCu//Tp05b38ElwHCEVqpIsugfYmd15+ul5yCAv2b19Ldmf//mfj3sIz8u+/vWvT9zqTc/HLhbcPyLpBZJul3Ra0q8+1w0EQXBvEAR/EQTBX0iyB5FokGgW4PQg5XuS2IGcjfJozsXDTWTaaDRMl86KT/wQBXrFDdFovV7XmTNndObMGet+WK/XhypNZ2ZmdPz4cVUqFZVKJS0tLWlhYcH4dcbX6XS0ubk5lKwlB+ATwdJ+AtgXMKFIwal4CgUghPf3lA3OQdpfZIQoHI4enh1H5iN6xkjOolKp2Hvsg/YESBdxeqPFWZwTrhX0GnLUEydODMlBcYJenuopLM6tl8DyvYtMpl7ye/tasd/4jd8Y9xCet33oQx9SrVYb9zAuiV0UuIdhuBKGYT8Mw4GkX9f+9HRZ0jH30aWzrx20jYfCMHxpGIYvlfabR3l6xssZD4rKSJp6WgagYoUgVDAnTpwwUAc8fBMyvyg3evTt7W3TcFO5SYIWR+GVHSxokc1mVS6XDbwjkYg1IpP2gJvkKp8nOuW4ATPyAr41MX3s/bqrnodmGwdx+qPKGe8kfIIXp8CYs9msSqWSUqmUAfRo615UQYlEQqVSSUePHlUmk7GeOltbW7ZYCatTkbtgW7u7u9ra2hpaSNkvPMJ19/kY7h3f6tjnbsZ9b0/tyrGvfe1r+vznPz/uYVwSuygpZBAER8IwhJx6kyTUBv9Z0qeDIPg1SUcl3Szpzy5km3DfPPS+gRjRGtGZjwQBahJ1kqzKUdqL8tfW1obK45vNphqNhilW/EzhoOQdAEh1qnci7XZbtVpNkUjEep7TRKzX66nZbJquPRqNamZmxlrhesUPjsvLLYk8icyhnVjLFPO5B983xucq+N8DpKeFiL5xEqNOk/OM1pxzT7Mx9pvP521hkmg0aueGcVSrVVvFiUpY8g1sp1arDckgSW77gjUct08m+/PgC5mei12Oe3tqUxuHnRfcgyD4D5LukDQbBMFJSR+QdEcQBLdLCiU9KeldkhSG4TeCIPhtSd+U1JP07jAML2i1BOgXFCFQFL6kfJQz9ok2v7CFtA94SBEBF3rGoGKRZItsUyAzGAys/4xfSJu+7PD78XhcpVJJiURClUpFuVzOaB62hQMAyEcpGM8v+8Zh/I/Ra96DN8flWxUfpJrBAHMPpkTd8OpeTkoS1Df0QhFUKpVUq9UMqINgr/cOoIyGn9kM+0FaSasB5JA+N5JIJKyugOvM2HBQzByI+L2aCGDnvI773p7alWUPPvig3vKWt5i67Uq184J7GIZvO+Dljz3L539R0i8+14GEZwtfiOR4QL3mncjdR/CAGICZyWSG1jxFrpjNZrWxsWHcM5EniVBf2OMVGrlcTrlcbigqBkypXGVFo3q9bhRNvV7X/Py86vW60um09ZUpFApW5AR37iNjX3TEjMSDFlpwqAoUP5wPf3xEtRwXTpOf0UZt3inAieM8mC0B8IuLi1pfX7f1VhOJhOr1+tCSgpxD3/qB8TIu3waBGQ0OAFoIzf329rZSqZR16/RUHWDuE8WeihrnvX05zOdRpnZp7YknnlCn07niwX1iapeRsfnI0XOoPLyA0WhFKr8BB99+FvAFNJHiIU2kyyEtddkOoIJut1AoKJvNqlAoaHZ21hxNuVy25Ovu7q5VddLz3FM/6+vrWltbsyjX9zn3Mwy4dZ/4xalJGnKAflbj5ZA+d+E5aXh8NOpw7aNOQ9pP7HLu/UIkhULBxtNut5VKpZTP521MrG9Lv/daraYwDG1BEq5fr9dTMplUuVxWJpMZkjySuB6lkDzdhPkK1au57YAkfe/3fq++//u/f9zDuCqtWq3ql3/5l8c9jOdtEwPu0vBCHYCSBzBAlwcYIIM+SKfTRpmkUimjDfhupVKRJGvdG4ahqWeI8OhRjpNpNpvWBphIdXd31yR70WhUJ0+e1MrKyhDg9no9ra6uamNjw/qat1otA2xPJ3CcvOZ5bq+G8coXxuurSr0ixvfK8edvlItmP4wNB+NlisxqKFSCe7/uuutspkW16vb2tgErNBXVwPzQVMw75Xw+r0QiYa2RvewVuonz6hPAns7zx8Y1fK4J1SvFfF+jqV1aC8NQv/mbv6knnnjismz/Z3/2Zw/lvpyI3jKSbMrPDwCDeYmbL7rxS7qVy2VTT/jCGbZDZSRySU8dwH3DqTcaDVtlKR6PG9WTSqWMplhfX1er1RqKdmkbDJ3gddkoZ/xC3P4BhbIA1JmFjCZ62c5oQpFjBehGz6OXSyJdBBj5DK952oiZh7Sf+A7DUAsLC1pbW9PGxoY1BvOJUcady+XMaVAjMBgMbDZFRTBL9KVSKVvFiVmG1+ozk2JcHtg9LXO1AvvULr+dOXNGH/vYx/TBD37wkm/7+PHj9pxeTpuoyF3S0HTaJ8qIWL38zfc1QfLoI2EecP5Hd01LWqJVKi57vZ4V4dAugGi6UCiYCgY5JDQMYFWtVk0jS5TL2GdnZyXJkogoRdCye5mn54mJ3D3fD0B7cMdBef33aOTqFTLScA8bn9OQ9hpANRoNU66M5htwxvl83iJtJJ1cG9obk7vw4wPo0dDzfRwLlbAcC9/jPAHo3BMcL7TTRWrcrwgLgkDf/d3fPe5hXPX2mc98Ro8//vi4h3HRNhHg7sGMKM3TCjzIg8F+Z0FfaCPJim1QzqDC8NWLg8FAjUbDom06PkLLEKmSFKTlLZp4QH1jY0NnzpxRo9FQtVpVvV7XYDBQs9m04icAJ5/PG3VAxA1owbmzb4qIpH1tN0DoF/Lw581H5p6n9oCH8/CR7UHqHLbjE6xcE09PeeqF+gBoJSpwOU5yC1AvFE7hmOnx4/MKHCtj4Bi8k/NqGK975/i9o7zaLBKJ6O1vf/u4hzFkq6urVhNytdhjjz2mz372s5c9wr5cNhG0jAciD4SjfDsPMdEjem/AKZlMmmJD2gcnKAK4XuR60AUUCVFQlEqlzMkQrUIHIPlj5SGicHqc+9cXFhYMsNrtto1L2s8PMG4fOQPyzFp8FM/5YDbik6c+uej18n47nGs/U/CUF5SNnxn5Bmp+HVeAnH3v7u7a4t/5fN76zVDwFY/HNTc3Z60h6LjZaDRs0W2i/9EKU1+chcP3/WPILRC9cyxX6oM5SVav1xWGoT7zmc/oz/7sYGn///yf/1Pf+c53Dnlkl98+8IEPaH5+Xj/+4z8+9HxdCTYR4I55WZuXO0rDKzQRofpolApJwDEMQ9XrdTUaDeOPKWRCIw3dAJiy0hLtBzz9gVwSiWK1WlUYhtZYLJFIaGFhYSgxWywWlc/nhzpDJhIJk0UCUJ5eYM1VT60Q+XIevLYd2ghunCpRwNBvR5IlGn2Uy/aIvj3F4yktXyHL+U6n0zpy5Ijy+bza7bYSiYQlULl2uVzOmoJtbm5adW4mkzFarF6vG+2FDl6SSSy9Tv6g2QrHyDn199TUzm+9Xk+f/vSnrfYD63Q6+qVf+iW12+2h1bWuFet0OvrJn/xJnTx5UnfffbduuummcQ/pgm0iwB1AJOrqdrvKZrNDNAGJxFHOWNp/wLnxoDwajYaBx+rqqknqiOyCIFCxWLTvsDwezcSgYbrdrkWG6XRa+XxeN9xwg1Wcsmg0ssJSqaRsNmvFT6O9TwAnAEySUqmU0TZ+NuKjapwL9BM/RLqjihm/LxLPSEQZC0DILMKX7uMUfQIbB4vUEz49nU6r2+1qbm5O29vbevrp/7+9cw+x7Kr3/HdVddWpOlXVXd3pTqcf0aQ1g8n8M1cuF8OEARGZGf8xIxjvGIyOgQySgZFcEL0xEAXRVidDBuRihh5UEhINJkRkItzEwMwgN76IeYwmN8Yek5g29LMep049utb8cc5nne9edaq7KjdVdfrU+kJR5+xz9t5r7b3Pd/3W9/dYf0zFwwifHBholQienZ2taO9EL0nS5ORkJZkKn0RuyfNceBvZxrXoV1nm05/+tA4d+qfXLFteXtZvfvMbfe1rX9NLL72k5557rjIwFrQwPz+vu+66S8eOHdPHPvYxfeQjH9E111xTKZHRi+gJcgeQh0sOrjM7qSPheLje8vJyWq8T673ZbOrEiRMVxynWbL1eTzHwkpJ2nmfL1ut17dy5UwcPHkxkQugeksrAwEAiemYA3VZvgqjoF8sCOkHlafRswzfAtcglFtepPaSRa5On4xOtk+vbHnHi14eZBVU0vUib+w2Gh4d18OBBnTt3Tq+99prOnDmTwiXxZaDHEznEYHzq1CldfvnlaRBmMD979mzFMndfDIMcn/e7dXnFFVdUDIP1YnFxUb///e/11a9+VT/4wQ9WWOsF3XH8+HEdPXpUR48e1fvf/34dOXJEd911l975znduddO6oqfIHTDtz3+0LiXwwyZUEUnB9XZqm3jJAZyZXtsFosGqhFAmJiZ06NChVA+GNU9ZiQknrluwkJzr6V7dkc88KYiQSY9j92gYrFYWH3Frlr4zS5A6cpWHibrfguvI+bhezC74Dm0YGRlJPoKZmZl0bQlbnZ+f186dOxOpcj0OHDigyclJnTp1Si+++GKKSOK4RNTgxJaUwlMZmFkRK09iy9vt9W3c+iyaexXnz5/X5z73uUtuKbxew1NPPaWnnnpKv/rVr3THHXfo5ptv7jlNvifJ3X+oHg7JjzhfKJsfs8eMY6VTwZHFndGRPQZ+fn5ep0+fTmQ5OTmpvXv3pvC9PXv2aHx8vJLu7rHYEBahjVjKIQTV6/VUBpeoD9pO7DvWse+HFeyaMp9h2ft18OtEJBEE7Rm/UtXq931ZDi+EkBygTvLU0IGk0eZZaUmSpqenkxN2dnZWIyMj2r17t44cOaLTp0/r+PHjmp6eTkv1cT/n5uZSdmqz2UwVL72MgveV/1wTBq98dtJv2L9/v2677ba3tO9Pf/pT3XHHHXrhhRcqz1XBW8czzzyjW2+9VS+88II+9alP6T3vec9F9xkZGdHBgwf12muvbWjbeorc3YHp+rlrvVLHGuMBdeuVaA+pVboX65/PR0ZGUriir9KEM/Dw4cO64oorUj0YrPTz589r586dSXdnHVTXfXFyQkp8h77xn8+clPL4dNe+vWwA+3tST57sBfl7tIgnQfm5GBi9TZK0a9euNLjOzc1VooekTlkC2u5SGNFB7vzdt29fen/8+HE1Gg3V6/V0HxnIuJdTU1OJ2JHf6A9tZGbCNeReemhnv2nuw8PDuuKKK9a1z9TUlL74xS/q4Ycf1okTJzaoZdsXi4uLOnr0qO6//3499NBDuuGGGy74/SuvvFI333yzjh49uqHt6ilyB0gsLs24pOByAokwvhpRCEGnTp1KoYlIC1jQWHcsELG4uKirr75a73rXu1J0CxY3ZAOZQew4dzm21CJIQippLxanL6uHRewWqBMy+zBTcT+BW9IQMyTrmrlUzVZ10neC7lavxY/hzmefUeGcpnAaDlFkJe5Ho9FIS/WxktKOHTv04osvrpiFIbewVmsxFXoIAAAgAElEQVSj0UgzptzS5LngGaBtPgAVSI899pi+/e1v6yc/+Um5LhuM119/XR/96Ef1yCOP6Prrr9/q5vQOuTvpYIG5I9DXCfXoE7RmiAoHKOtwhhB02WWXJXmG9H6pJSHs3LlThw8f1rXXXqvx8XFJHbnE9V0GExyGOG5pExq4V1rkNTMJT8uXOo5Tt8hxWrqPwa1arzUD4fqsgP38M1/Vyn0YyFBcOyxhZBfkFe4FTk4fMBh48Ee4cxcZjMGamRW5BC+99FKy4NkHwmfGgGxGqQIGNv7yhCuPtNmuaDab+vnPf64777xTzzzzTAoBLth4nDhxQjfeeKM+8IEP6Fvf+taWRtT0DLnnyImIHzKaM0RAiCSfI4VA7lKrXjvJSpLSIHDVVVdp//792r17d5IAIBG+hwTisgvwMLw89JDPCcuUVCE8T/snVp3IEScsX0CDQcRJLi854OUKJKWsUY8skarx3xC2R6K4k5L259E3tBELm/7jP5CUCopxnzhmvV7Xtddeqz/+8Y86efJkksG498vLyzp58qSOHDmihYUFTUxM6NSpUxWpybOY6ZvLd/0oy3giXDecPn1aX//61zd8yl+wOt588009+OCDWlhY0He+851kNG42eobcXRvG2sRiRHNGf/XFnwnnw7olbhqy27FjR4oK8HVBDx06pFqtpn379iXHqTtHGTDc0Vir1VLNcc4NmTKzyGuyu3XPYOBWuYcjYsGj03uYIWV3XaJy8iI6Jk/d9/dIJ/glPMQUJyohh8we/B4QJkpCF/vSZ2rD4GzlGgwODqZ7grXP8Q8dOqSJiQmdPHlSUmuWgjN8cXExlQ3mWnlWKm3KCT9/pvoJX/7ylyuLuDgWFxd1/fXX6+WXX97kVhV0wyOPPKKlpSXdf//9Kwh+M4yOnpm7emc9U5X3LjNAqJAqDjckAixtJBRAPDm1yA8ePKhdu3ZpbGwsLaDhBOnrizKAeGq+a+p5DLvLL8getMt9Bq6FS9WqjD4zcOscuAMRgsOqpt0ck7b4uQkjZQCESAlRPHfuXLK6iTwiM3V+fj5FtMQYdebMmRTH7o7UkZGRis+BNhFxND4+rssvv1yHDx9O99PDOr3Ew2rJXXlNHZep+g2rxbc3m0194Qtf0CuvvLIi2atgaxBj1I9//GP94he/WPHZ7bffnooJbhR6htyB/yC94qI7V92xx/ex5KWOY47KghDb2NiY3vGOd2jPnj1pWTwILi/cJXVImZA/qZO+7yF4UscJ6UTq+jd980WoXRt2qQNr3gmb4yAN+QyAY+SSjuvfvi+DFm2hpj36Ou9ZLhDnNitMueOXcgTz8/M6c+aMpqenUw4BzuehoSHVarXkaGYGJLUGm3q9rnq9rsnJyXQdFhcXU47C4OBgimZyiQ45y2WxvBzBdsDCwoLuvvtu3XPPPSXEscdw/vx5feITn9Czzz5b2X755ZdveD3+niL33OJyInVnax7GBwFgBZMIg4XJeqj1el179uzRgQMHkkXp8otb0FjMED4kwmAAueVZsy4RePKSDzzeX49PZ0aATMOxIeE8BNIHndypms96gA8EkGaz2Uwk7U5fpBFi1r2EMWGlHM+zbLkOkLgvnzc4OFhZIJxKkWNjY5V1aLH0iZxBjkOWIlvXfTIM/N7nfrTeczzxxBP6xje+sW0Gs0sNr7/+uu6///5Nvz89o7l3A1a3Z03iNEOKce0XC3NwcFBTU1MpPb7RaGjfvn2VpeGwKMl+5BhOXMgIWMoQKeTlcdfdLOU8K9QjgiB8ZA0HESMu1bhTFccx792J6tfKJR3O6Q5ZiJy+e74AUgqvKUHMdwcGWkXLvNgaSVxIZOfPn09VHmdnZzU6OroimYoBddeuXVpYWKiUUPZZDQMp18ITsbwUgbRyoOtnnD9/Xl/60peKFNPjeOCBB/SZz3xGV1999aads2fJ3XV3LDJ3IkJSHinBj3l0dFTNZjOtDDQ42FqlaXJyMu07Ojqqer1eCVsMoVMuwOPO86QjqZNAkxdaQv6AgN15SlvcdyCpQvhSZ0Umt0zdwuccvtKSO5Bde+fPnahecdHLJuO/cI3fCdKvE+8ZRDg+1r8neNFH9HsGDGYGSCw7duzQ/v37deLEibSdEMtc/uJ4nlzVTZLod8v9hz/8oZ5//vmtbkbBRfCnP/1J9913n77yla+k38TExITeeOONDTvnRck9hHClpO9J2i8pSrovxnhvCGGPpO9LukrScUk3xRjPhNav6V5JH5LUkPSpGOOv19Molxaw3j2hB4sTa9NjqXfv3p10Y8L0KPKF9kvBMJc98nh2yIptubzh1qS/h9ByWcC1d0+T94gfjgOZeigmg40PcMTa0468tIFLFN1IGv2dY2NZe0KYyx4e4cNxuvk+CDXlWrgEJCnNqPI1UJHOLrvsssrgtrS0VJFhvO1+fZGN/D5dyHLfimf77cTU1JSOHTtWasRcIrjnnnt07bXX6pZbbtHw8LDuvvtuffzjH9+w861Fc1+S9DcxxuskvU/S7SGE6yR9XtKTMcZrJD3Zfi9J/1bSNe2/2yT93Voa4iTEeyd3T9zJrVKv5wKxo5cPDLRWAIIA3fGKFUhSkseM43zktQ8Ebr3mUS2uxft2+sQxkDx4DSm5hp5r2BB77oT16B0vt+ARR7QVOGFzXeiLW8mSKgNebrXzPZek6CN9Q2MfGRlJpO4JXsvLy0n7HxgY0OHDh1MGMBE3Xp+HwdEX7eD9Oi31TXm2NwpnzpzRk08+uZVNKFgHFhYWdPvtt+tnP/uZpM7sf6NwUXKPMb6BdRJjnJb0W0mHJH1Y0nfbX/uupBvbrz8s6XuxhX+QNBlCOLCWxuRWFtEYTOHzabdLCefOnUsyS7PZTAWoWBSC6BB3oPoAgFWP5Zmn4juJuNThyUM+CLhliRPRBw/3lNMWz77EoUm/PUJI6iQeMeDRNtrr2rS3wyUXf8+2PAPWLf78vZM694bjUpeHB9h1eRzc+DsYBOjX2NiY9u3bl/wlzEIIdfUwUl+WsZvufCGy38xneyNw7NixorVfYpiZmdE3v/lNTU9Pb/i51jV0hBCukvQXkp6WtD/GiGB0Qq2prdT6cbxqu73W3vaWxCUcc5A22aYQEPrt7Oys9u7dqxBCWgxiaWlJk5OTSVuGAD3lnpBJlxCQKdxKzQk/t97zmYcTr1u8ucbvTlmOBdnlFQ59oOE/VqzPaPyYrs8zM/AIHi+X4DKTX4Oc1F3y8VWxfND0hCmOSXTNzMxMkm7Yp1arpZr7IQTt3bu3svrV6Oio/vCHP1TyAFxm8pDXfKa0FmzFs/1PwSuvvKKHH3647x3G/YhHH31UzWZTk5OTG3qeNZN7CGFc0g8lfTbGOJXpmjGEsK6nLIRwm1pTW47hn6Vt/OA9soOYa1L7WV6OqTz14CcnJ7Vr166KRe71X1xXlzryg5fMdQKlbW7hQtRSpzqhF8PCciUCBjBAsZ/Hr7Mtj/7wc+TbvMSvR+dg1bvlDjH6oJTLNH58l8O4L3yP/2SlkqnbbVZA21nsw2cW9Jvl3Kihz/l82T+uJzMcSV1ndmvFRj/bG4FHH31Uv/vd7zbyFAUbiMcff3zDz7Emcg8hDKn18D8QY3ykvfnPIYQDMcY32lPTN9vbX5d0pe1+uL2tghjjfZLuk6SBgYHolmH2vZSsAtn7gtKQFIQB8c/Pzyer3fXbdn9SyKPHaTvpSR0CdhnCna65Ve7/fQBgQBkYGEgzCLe4ndS5DrnV7ta4E7cPLn5Ot6K5LnmIJOfItT+2eb9pmzt0PfTSB07O5dfTpSf0eEoNSJ0BijISWOu+fz6Ico+QsHz7WrHRz/Z6B4a14OzZs7r33nvf7sMW9Bku+ksILcY6Jum3McZ77KMfSfpk+/UnJT1m228JLbxP0jmb4q4LkAnTd596u6UJGZE+PzQ0pNHR0VRWoF6vr3CEurXoSUpYoV7h0C1MP99qUTF87oTtEo/r2h7DzncZQNya92P4tfEBBgKFJCFawhNpn0enoJfnA4FLP/ksIW9L3kfPwM2/4zIMZZUJPyXxiVWf0OVjbC2azfE9QQ2fi4dw5ue8wPO1Zc/2W8X3v/993XTTTXr11Vcv/uWCbY21WO7/UtInJD0XQnimve1vJX1N0g9CCLdK+n+Sbmp/9j/VChV7Wa1wsf+w1sY4QQJivgmfIzaaBTFYes51c19ogyp6XiOmXq8n4vOYdEjNQwqlajVEyBbyc0vZSdv74xKOk2qun7tUQgSNL0jhxOuE7jMA/8zb7dm3ToIcw9vAAMe9cKdxPuPIrXsfcJFx2N9nKW7l0wf3i4QQ1Gw2U7XJpaUl1Wo1TU1NSdKKEFkqV+bP0kWwac/224UHH3xws09ZcIniouQeY/w/klb7pXygy/ejpNvfaoNcAuB1XiGSH7RLDyyjR30S6phgwUKUFAjLrWd3KubJRbzOLXHXrPNwQSc3J7vcYeuDCrMTt8IhPgYSHxQ8PDLX87v1ixT+3OpnH/cBuD+BdkDMXqWzG6k7PIqFaBf/nhfCIvIFK54Zhyc/IcFwXdxy5950e55WedY29dkuKNhM9GSGaq7t8gPnj/U9sS7RWxcWFtRsNhOJO5EzALgztVt0iVviXt7AiRy4o8/hx+TPydCta+CLbvB9SNvb6olVkG/uSMz16IGBgcr6rz4geHt8IMi3Dwx0yiGji5Mhm1vv3n/a4SF7PjtipjU4OKjTp0+nfbHs5+fnU7ZrrVarzBgYxEiYymdMBQXbGT1VOEyqShkAy8wdjU4YfBf5Znl5OSXM8EfBL7dKvdiVSxcQvFQt2sVrPve67jlxoqHnljXfdWlGUqWwFrMNHwC8rk2uafvxODfk6G1jH5eDnAjzASefffg94nvMoryCZ7fz8J/75kSMXEaVTvdpNBqNikxG7D++BPchFBQUdNCTljtwS9aXinOZhhV+3Folosbrm/AdLF5IQeoMDr7NZw+e5u5SRu4jcDklD2N06xYJx7czGGAhsy/7++AG2bmlLHUKcbll65IP5/OwRJeKfNbiUlNuxTMIMlj67CTvf+678AHQZw/4T2KMOnv2bMVhOjExsUJT92ubL63nrwsKtit68heQR4VIqmQixtgpOuVWG2QDSUhK6evuZHQCk1auAiWpQs6c15f+c7nHJRskBbeC8+P7sXPNmhK79Bfr25Oo8oHDz5PPfNwvwX+Ox+s8AsnBNgYK2sV53dfgMyW/f1jt+WpVeYy7l2RgYRAseU/U8nuBVLfa81NQsF3Rc5Z7TniAqTgOVX7oTj6uVccYk/7O+26rAXEeyNqdplJVa8/DKUEeMuiOWu+Xa/xOzN30aPwGLhuF0Mr89PN0S5jyZKXBwcFK0S2XkaSVWn8OJ3wGLa9qybGIRnIr2mdGXH+3ql3G4fvIT1j5REOxiPZqf37PvC+F6Au2K3rOcu+m7UJu6Ole34WFJmZmZjQwMJAyFSFJn/pDNu40zCsnehtyicKTaXLyzvVm/+uma7sMw+ce346MxLk9Fj8nXB/0fHbjfXWt38nVZaY8ht39GvTfv0OfGUA9zJH9+D77e0goVncua1HBk9dTU1OanZ3V2bNnU389W9kH9IKCghZ6jtxXg1c+hOT5W1paUqPR0PDwsObm5lJUhTs1naghgjxrNY/h9pDDnPBc6gA+ODgJdpM6XBOmLWjp9NUHKg8DdUuf7e78lDo6tFvxXlbBB0ja53Vl6A/7+mDlWr5HzXAMrqOf32cqrvG7bBNjTCtnjY6OqlarpRnb+Pi4Tp06pcXFxVRegvZ7GGQ3Sa+gYDui52SZHE6YONgIeSRahXKwntZPFUZPioHw+R41Z9yB6SSL3MBnMcaULMWxnMjc0ZoTOscmnBAyy+u+AD7zgYX9pGosutQZ/Dwz1dsyNDSUEr98IMj1f49kWc3Cx1r2cEeXeXjvshnH89cUF+PY1KeJMaa6MmQde30h2sksxWdQ+XNTULBd0TPkvppe6iRJKV+sNhJbJKXFnOv1uhYXFyuLM0OSLP/m+i+Eki9Q4YlOtMWJ3NvpTkUnNI8wyQnOCbNbuj5RMfl2l35oM5azL7yRX0upo8nTJydbju378h7pJY8e8lmNOzuZdTCrkJRKKjMo+n+I3a8Ta9yyhurZs2dTuCUOdT9+QUFBFT0ny6zmCAshVBZxxlKHZBYXF1Ptb370XrUQQnCNPSfTPP7av+NOzQvt5zIHx/CKlujSOdHSd5dC2OaDhGv/i4uLaYELkFv3uf+BAcvP723ivVvDrv37PXIdnT4ODLQWrybngHs2OzurmZmZipOVgm7uC5GUEsN8FjUzM5MGD2SZbpEy/uwUDb5gO6NnLHdwoR/m8nJrxZ6xsbEKIUCAjUZDS0tLyWqHrCSl8LluES8uqXjIobelm37v8o1LHU6w1K9hux8jHyzc+crswqNScgkJmSV3PDMroT/+uZ8vH0D8HuRRQ3k0EJ/l+rz7QxqNRppVLC931m6t1+srBhH8Kcw+uK87duzQmTNn1Gw206zLrXxerzZIFoIv2K7oGXLnh7iak1LqlP/1rEj+S63VfhqNhsbGxiqFwbAg3dKEENyJmEszfn4nstySzqUY1/hdAuE87gzMLe08BtyduxzHLW2uR271orFz3lwjJ2zRnZC5XMPg5SGQ9J/2uqMZKx3/CLMsT7rykEYGAvbnOz7oUd/d/RvuXGat1fxZKSjY7ugZcu9G7N3QbDaT9upSwODgYJJmICG3AKWOtMLCGb7UnVQNS3Qt2Qce4NudfN1qzDNJ/TUE7rq2a8iu9/PeBw9JFQlK6sw6uBbdnK601/Vuj4jxtnk/8AF4nHse+kli2dzcXEp28rh3nKg4vNnma59yHu6b308kLr/3Lsvk94h9Cgq2I3qG3KXV63/zWlKa7u/evTuRkKREZhAeEgDkhF7LtB/SkTqZke7sRNpwwgMu6/Adtx4hTLfoPQ7eY9OBx37TRo/hpvYM/YG0GaRos/eVgcAdmO7IdT3fBzCXa3wA8+vF8btFxSwvL6dopYGBgRSeKilZ8n69aDfRUH4tXB7yMr8MIt3CIFcbkAsKthN6itwvBixOftQQFlacZ3NitRK54Vo7hOHWs4ck8hkEAclwTM6Rx847sXez2L0ejrSySBrnX21Q4xhs43iesZsPMujWvtAJx/NZQ65b+/4ck4GTNjN4+OzBE8aIVx8aGlrh+KWyJOdxKcuvUwitUNfp6emKnLO0tJR0+G4E7te+oGA7omfIfTWtPXewEo0xNzdXITaXH9DiWXQZ65DjOHEROZLLLJJWkCHt8YHE28cAQDvzRCF32gK3jiFaZhEue3hdG5AnVLkE5duRajzO3v98kPEZSzdHq5/L2wEgZo5dq9XSMagV44Ps4OBgyls4f/58qvroA5PPPHxNXRbQ9utZUFDQQs+EQq53Ck25Abd08yqGUsc6zxegdonDz+sO15w42c/j2X0QcmvfP+9WojfX9tnuhAyJSZ26Lrnc0u36cW7+u4TifgC/ft4Ob6t/19vvRb8YtIhQQpKBvEdGRiq14z2cEwesXz/0egaafKaCE31ubu6Cz4jPqAoKtht61nIHuTzBj73RaFSiV/LoFMr9ukyQOw49GkWqhip2I2+3zH0m4bKNt8Wljwv11SNTcF4S5ugDktRZLtAHofx4WOAeReODDu99m1dz9FmGx7HnIaJSx4k7PDycpBciYkZGRtI1GR0drQy0fo25P1TDhLTHx8cr9XL8P87bfFDKtfZ89lFQsF3QM+QurZ62L63MWiXszuuBS0oW38LCQlqE2bM2nUSxwD2pxy1eiBYixHqWOpozBONauGvztC8nfp8N5LIS+1L+VmoNVp6u79avE5oTM9cij0X3c9P3fLAA+XX3wmCci/b6eQYHB5OuHmNMRM921+lpB7IMx0Zz5xgeQz87O1u5H97e3BAoKNiO6Aly7xbdsBrJg+XlVkVIvocmvWPHjpQhSRw4sd9YfF5x0S1hrLxu8ezejjyMMCcRPxbfz19z/Fwm4dx5DXqPhfdr5qGduW/B24UT1x3DwNvDuXxG46/d0ufctAMC5zw+E1heXtbExERFTvLBcHp6Wo1GQ81mU0NDQ9qzZ4+mpqZSlI2HilI8LLfY8+cln/UVFGwn9ITmnk+duzlWHRBbo9FI75EyICasXKb6voB0HvGRa7OQUS7B5Lp67uB0sve+5Kn93gdvR55165E5XhWT1agY4HAw42Dk+J6U5QW3/DqijTt5534A2pSHP7rlPDjYWklpfHxco6Ojqaibf4d7QBgqztKTJ09qZmYmRcFg2S8vL+vNN9+syDDIN3Nzcxe0zHOfQUHBdsNFyT2EcGUI4akQwv8NIbwQQvjP7e13hxBeDyE80/77kO3zhRDCyyGEF0MI//qtNs5/mN1+oGfPnq2QmdQJYWTtzYGBAc3Pz+vcuXOV6BWO6Q5K/rsT1MMMPezPLdRujkWOn/sCPJLEt/G9PDSTpB8idGgD2jQp/j7rcHnG48vza+VJRfk1zqsv+uzKr5eHY/p/9vWYffwk9GtmZkZTU1OprIDUmrHs2rVLc3NzKRkLvwPnJbLmQhb5WhypW/lsFxRsNNYiyyxJ+psY469DCBOSfhVC+Pv2Z/81xvhN/3II4TpJfy3pn0s6KOmJEMI/izGurMuaIdeOu+ntgEiLRqORVuvxiJBarZZqyVDEysMXnZydvLpVVfTszJzovN1IFN4WjuVZpVJnBSSPzXfCR3/mu0TPSKrEgOfRPj748BkhiC7HuKzUzUrPQzHZ3g25BOSyDtfCl8XzDNoYY/In+CIlMcZUaAyLntesLEWb8/bTpnxG1gWb9mwXFGw2Lmq5xxjfiDH+uv16WtJvJR26wC4flvRQjHE+xvgHSS9L+qu1NGa1qXROpmB5eblS75sY6HwdUknJ4uMYrotzzrz+ip+bY3TT33NLFuJ1onbnY+5I5Ry5RJJb9k5YaNru0MyPT/88K9S/5xa8W+X5veB/HmKZO4BduqHd+TYcowx27MuATG7C0tKSpqenk+XPNq9fkw/+68VmPtsFBZuNdWnuIYSrJP2FpKfbm/5TCOHZEML/CCHsbm87JOlV2+01XfgHc0FcyLEaY6f2iJeB9YiLpaWlVHiqW2o+lq6n5kvVJKCc9PJjoP96BAhE5GUOQB7iiC4PMftxvOxt/n0nesIf8yQhdxZ7dJAPbt309FzPZkbjMgzHcYc17fMYfO4b94dSxcgxy8vLGhsb0+TkZJK4qCE0Pz+vZrOZnK1c124W+WoS3lrIfyue7YKCjcSayT2EMC7ph5I+G2OckvR3kt4l6V9IekPSf1nPiUMIt4UQfhlC+GX7ff65um3PnZWQABEyEMj8/LxmZmYS+btU4HHWXmtd6lib/gfRufYrKclCMzMzOn36dKo57qn+bul6WCVkx2eeJUu/Qwiq1WoaHh5WrVZLCUIkCfGfzyFZ9qG9HmOeryzVzXmaS1d5oTauJ4ODa+au7UsdOWpubk5TU1MVJ7jXYx8aGtLc3JxmZ2dT4bGFhYVUBx6HsVeS7Pas+IC8Vmz0s11QsBVYUyhkCGFIrYf/gRjjI5IUY/yzff7fJf24/fZ1SVfa7ofb2yqIMd4n6T5JGhwcXPWX6D9UtybZDqFLSuVhR0ZGtLTUWlcVh+rS0pLq9XolEckJLk+I4jWfQ9Kh7QxEAsJClpQkoXq9Lqka786g4MSUW9Fgx44dGhsbS/ozqxLRlvY1U61WS3KTR+NA+JwjhJB0baz5EEJlQOk2Q+n2l18jz9blOnrcuztBuT4xxlSb3uPsiXhicQ+s9qmpqVRPyBdp4R7mz4v//+AHP6inn35aq2Gjn+0QQgnVKdgSrCVaJkg6Jum3McZ7bPsB+9q/k/R8+/WPJP11CKEWQrha0jWSfr6Wxqz2g02NzWKzY4xJcmGZPaxD19tdguE9TjqICXjERzcLnjA8LEnO6xULc0mDczqBIrcQbeNSC9KEt9tr15PU4/4Ezo+FjSQFEbKACdcXUvU4eydpd+7miUJ+Xo+/p/84fv2+Dg8PV8oQcE39PnB++uKzL+7paqtp+fX2GdC73/1uHT58eLXnbdOe7YKCzUa42PQ1hHCDpP8t6TlJmJ1/K+nfqzVtjZKOS/qPMcY32vvcKenTakUjfDbG+PhFzjEt6cW33ItLD3slndzqRmwSeqGv74wx7ss3lmd7Q9AL93uz0At97fpsS2sg981ACOGXMca/3Op2bBa2U3+3U1+7Ybv1fzv1t9f72hMZqgUFBQUFby8KuRcUFBT0IXqF3O/b6gZsMrZTf7dTX7thu/V/O/W3p/vaE5p7QUFBQcHbi16x3AsKCgoK3kZsObmHEP5NaFXYezmE8Pmtbs/bgXbK+pshhOdt254Qwt+HEP6x/X93e3sIIfy3dv+fDSG8d+tavn6E1Ssr9mV/14N+e7bLc32J9Xe1TMTN+JM0KOn3ko5IGpb0G0nXbWWb3qZ+/StJ75X0vG37uqTPt19/XtLR9usPSXpcUpD0PklPb3X719nXA5Le2349IeklSdf1a3/XcV367tkuz/Wl9VxvteX+V5JejjG+EmNckPSQWpX3LmnEGP+XpNPZ5g9L+m779Xcl3Wjbvxdb+AdJk1mGZE8jrl5ZsS/7uw703bNdnutL67neanLfTlX29sd2lqOkE5L2t1/3zTUI1cqKfd/fi2C79LPv7/Ol+lxvNblvS8TWPK6vwpTCysqKCf3Y34KV6Mf7fCk/11tN7muqstcn+DPTtPb/N9vbL/lr0K2yovq4v2vEduln397nS/253mpy/4Wka0IIV4cQhtVawuxHW9ymjcKPJH2y/fqTkh6z7be0ve3vk3TOpn09j9UqK6pP+7sObJdnuy/vc18811vt0VXLy/ySWpEFd251e96mPj2o1iIPi2ppb7dKukzSk5L+UdITkva0vxskfavd/+ck/Vw9wkcAAABxSURBVOVWt3+dfb1Branps5Keaf99qF/7u85r01fPdnmuL63numSoFhQUFPQhtlqWKSgoKCjYABRyLygoKOhDFHIvKCgo6EMUci8oKCjoQxRyLygoKOhDFHIvKCgo6EMUci8oKCjoQxRyLygoKOhD/H9TT7/EckntrgAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "test\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAC7CAYAAACend6FAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOy9e5BkWXXe+518P+vR1Y95exg82GEGM4gxQgbzCBkkYcmIGMcIiBAjCQEB2EbEtX0BPSw0doDC6KKw5YcYWRYSQtLIwwBhydLVle8NfGVzMUFIlkAwYGA0w/R0T3VVVuU7qzLP/aPqt+s7u7Onu2e6O7O7ckVUVFXmyXP2eeS31v7Wt9ZO0jTVwha2sIUt7Nqy3KwHsLCFLWxhC7v0tgD3hS1sYQu7Bm0B7gtb2MIWdg3aAtwXtrCFLewatAW4L2xhC1vYNWgLcF/Ywha2sGvQLhu4J0ny3UmSfCVJkq8lSfKey3WchS3sStriuV7Y1WLJ5dC5J0mSl/SwpFdJekzS/5D0hjRNv3TJD7awhV0hWzzXC7ua7HJF7i+S9LU0Tb+epulI0m9Keu1lOtbCFnalbPFcL+yqscsF7jdKetT+f2z/tYUt7Gq2xXO9sKvGCrM6cJIkb5X0VkkqFosvLJfL2t3dVZIkYZuYMuL/NE2VJMlZ75/rcxfz+rn2669x/KcaYz6fV7lcVqFQUD6fVy6XU6FQUC6XUz6fV5Ik4SeXy4XfvMe+/FzZnnEmSaLJZKLJZBKOm8vllMvlwmv8sL/JZKLxeKzJZKLd3V2Nx+OwDfva3d3VcDjUZDLJHJPjnu/vc/3v1y3ebtp1n3bseB/++WmfS5JE3W5Xg8Ege+DLaP5sS3rhlTruwg6npWk69dm+XOD+LUk32/837b/mA/qIpI9I0vXXX5/edtttWl9fV6lUCgDkwATwjMdj5XK5AEZ82QGinZ0dSQqv2fHC3+PxOPydJEnm/9iBsF+O5UAa75fx7uzs6Pjx43r2s5+ttbU11et1VSoVHT9+XNVqVdVqVcViUcViUZVKRaVSSfl8XvV6XcvLy6pWqwHk0zTVeDxWqVTSaDQKwIvDKJfLyufz2tnZCWMuFova2dlRt9vV7u6uRqORdnZ2tLOzo06no263q06no+3tbbXb7fBev99Xp9PR+vq6Hn74Ye3s7KhYLAaH4Y6Ia5fP5yXpLIfFNoVCQWmaKp/PK03T4MC45vl8XpPJRMViMbNf7nG5XM6AOvvCgZVKpXA8HwvjLpfL+v3f//0LemgvwM77XO8/B+HZTpJk0bxpYTOxywXu/0PS7UmSPEt7D//rJb3xqT7gUaODqAM4gM92uVxOOzs7yuX22CU+Kyl8+aU9sGBWwGtxRDzNEfi2/j6vTTuWA2yxWFSpVFKhUFClUjln5M1nRqORCoWCxuOxdnZ2VKvVMpFqkiQql8sqFosaDofBkeTzeR09elT9fl+j0Ujj8Tg4An773zs7OyF6j8+f1/nfz5txcL25roA1wA84S9LOzs5ZMxF3iDjP3d3d8Jo7kuFwGPZVKpUy++A5YEbE3z5rwdlfIrvo53phC5uVXRZwT9N0N0mSvy/p9yXlJf1ymqZffIrtMwCey+UCyMRRs4NO/DeRYQwA56JQHNymReN+fAdABzyO5wAvSZVKRYXC3uV1agZjTIzd6RTAOUkSVatVjcdjtdttdbvdTDSfy+VUrVaVJIkGg0EYT7/f1/b2dgA3nMXOzk6gYgBlzovImnM/F80x7br49eIcY/rEZzx+PD7PucfXk2icc/FnhX2MRiOVSqXMvR2Px2eN+5naxT7XC1vYLO2yce5pmv6upN+90O2ZpjuI8kXf31+GqgEU4v8dgGNQj6PzGMRjisaBK+blMY/YfZtSqaRisRhAPZ/Pq1AoZADHo+Y0TVUsFjMRLPuHiuHzOzs7AcyIcLvdrobDoSqVSsaJcBycCOfq0TDXlXHG9+Nc3Df3x4FeUnBq0/IR7gA4TrwfdzTj8TiTh+A4xWIxc++ZyY3H4+BMJZ11LZ6pXexzvbCFzcpmllB1AwTiqT3Axnse2fHj//v+PKo8F0jHxv5jCobPAR5xVDotaoc+4bdHqvH+J5OJRqNRoGGgTwaDQZgBODDCQ+MUer2e8vl8oC0Gg0FmtrO7uxt+Oz/tXDqA79G80y9+bdjeZ0fOxTvo+nZsE1NaDuJsw/8cP3biPCseCHBsdxLT7uXCFnYYbG7aDzgYOzB59B4DO19ip2gwjzjPFXVLB6DNe7GjcL7bAQvzqJNtSqWS6vV64IILhUKIZgHUacDIeyRUfSbCvpyamEwm2tnZUavVUrvd1nA41GAwCM7Bz8V/A+LMJvyHaxErZfwaxVG8m9+vaa/H98VB3/fl0XvsLGInEb8+zQEvbGGHzeYicpcOONkYBKZF4C71iykSknsOyti0qNnfj0HdHc25tvHfTq+Uy+WgZPEoeVrSMQbDQqEQFDSFQiEkTSWFv0nAEt2Xy+UQuXtS8anGyXaVSkX9fl+FQiHMMtjex+v3x++TX3/nzuMZgJvfo3PdY8YQj9+vF9fXP+fAz7ksbGGHzeYC3KdFxFL2y8wU24ERmsEje+eUY918DNRObUwDEHck7himRf++z2q1GiR6xWIxRMTQIX7OgBhJTxQf1Wo1bDMej9Xv99Xr9YKEEblfsVjU0tKSyuWyJIWE6TRaxWcBnm/Y3d0NElT4ak8Ax7x77GSxeHufLfg1jp3ONCCPt49nHhybe8OYPTnNbGlhCzuMNjdP/zSQdtCPde/xdH4ajx0DTAwuMUDF4OXbujk4xceRpKNHj4Zomt+AMZpu+HvOF6AG8PmfCBcnUSqV1O/3w2v1el3ValXlcjkoYwA3ZjF+rTgeUXs+n1elUgmgSOTuNJJfqzjX4dct5uOdanJncq7r6xG7Pws+i4pnH7xeLBbDWGJe/qnyLAtb2LVqcwHuMU0B6MX8dwwIMVjzRY4TozHAT6Mr/L1pCTwfp49pGnB5kRLRJEoXp2j4vPPyqFeSJAmATRKViL5arWo0GoUkKlG7R+bsh8gcDh6n4scFHKmqrVQqQbnDmJ37ju9d7OyIouPr51G335dzUTe8B+0SOyhecxWS7+dcM8KFLeww2FwkVJ3bnhaZS9PldR75xhF4DMh88c+VaJsG0tP4+GmvOZe+srKiZrMZCpeIhgEg9uHnRgUqAIwD4H9mLpPJJBT1QO84xYITkBQ4+1jFwnY+brh2fpMMdmcWR+Lx/fPtiJq5ln4f4tcw/vYxczyXt/o9mnaffF8xDbewhR0mmwtwd/Noy7+YzqU70DqgTAMyLObUp70X/3+u/cagBJilaaqVlZVAlQCcABbRvEsq2QdyxcFgoH6/r62tLW1tbYV9VyqVcHwHZxzjeDwOPxhVqX4sT+w6sAPq5XJZKysrOnr06FkqpLhOYNp19NmPXx8H9DjZ6+9zHXiNH87bcxaMycfJb6ilBbgv7LDaXNAyUjZJh/kXPNYxO8Xgn/co81xKl/i1CwGAmNd3vp5xVqtVnThxIihXvMcJfHrMQ/v/3l5BUojSK5WKRqORhsOhRqNRiOBJ1MaRcFyI5MDozmpaQRE0z/Hjx/X444+fFbVPo63ixKorVOL7Cs0TO1R3oLHzjGdeMd1C4ZJH/FBPi6Tqwg6rzU3kHgMI5jxqPAWfJnmMk2/++fNx+P56LOebxt96Qi+fz+vWW2/V6upq4NyhNuCz48Ihj7x9bAAgrQXYvl6va21tLUP1UCSFE5F0ltPxfQJ2SZKE8dH/hv2VSiUdO3ZMJ06cyPSa8WSoAynm5xZH5Bxz2rj89fg6TEuyci35m/dd4pnL5VQqlc7J5S9sYde6zdWTn6ZppnGYpLOAOaYXpCyw++fORaXEn/PXpyVrXbHhBg2Sy+V0yy236NixY0EG6QDsgBpHpa5ZRw7JLIAOkZPJXgUrwEuFai6XU61WC6oaAHiahtxBF5B3pwDYkyuoVCq69dZbdfPNNyufz2d60sS5jVtuueUsnnsaP+6fcWfpTolr6jM2d6LnmmX5/qlqjp3LwhZ2mGxuwD3maTH/0jqdMe1LHvPD06L6ePtY6RHz7f43/LdHsqVSSc961rN00003aWVlRY1GIyN/hHMnmnSQ5VjuqCi7r9VqISp1IOz1emepiWhRAPCikIkTkSRrvRo1VuuglqlUKlpeXtYtt9yi22+/XcePHw9JXjpL4oh/6Id+KMg8Y3rMr683g/NchecMzvV5fwbY3hO3MbUVPw8LW9hhs7kkJB0Qpk3ZzxVFx/vwIqdpQB0n/6QsEPJ/nDSFhllbW9OJEyd07NgxLS0tqVarhf7sUB1ExnFnyBi0x+OxarWaarWaGo1GBvicwikUCur1eoFzZ+ZQqVTCNeh2uxkFEeN3eaKkMDZvVsa51Wq1wGVXKhUtLS3pxIkTevLJJ9VqtdTv98OshrECzj4z8uib12IOPo7epyVlec9nJZIyeQSah3mu5hK3/F3Ywq4amwtwd7Dl/2m/MQetOKHK374gRMzBs820SP2povU03Vs8otlsam1tTSsrK1paWlKj0Qj6cwqK6OXulaQObH5O4/FY1Wo1jAv6wxcekaRer6eNjQ0Nh8MA6sPhMNBAJF49oeoAOy0pjc7dE5AUNgHuOKtGo6GVlRX1ej21221tbGxoe3tbkkJfG7phulPyQiq/F1h8H7h/XCOqdv1+eMWyt5yIHcKi/cDCDqvNBbhLZ8sPnVI4V1LOIzxXzrAf31/sOHw7/3HpnTuAWq2m1dVVLS8vB6kjKyzVarVAaRB5wrd73xNpeg7Bq0FRxOAg3CnBJw+Hw7A9vd9jyiPOOcQ0lc+KXGLofDy9anhtMtnrQ0M+4OjRo+p2u1peXtYHPvABffnLX9Yv/dIvhRkF+5sWeXshVXw/OBcMysmpOfbnn49nA/SsX9jCDqPNDbjHoDoNED3CjPlYT6Q+1bJ5fizfxqNC/oYmAdSp3CSpCbCjOPEEZTyL8CrPmDLhPcZdKpU0HA61u7sb9lkul1WtVlWr1dTtdrWxsRHA2XvokPh0qsMB1oGf43nCl6iY8/QIn/eofq1UKmo0GvrkJz8ZxvHt3/7tGg6H6na72t7e1u7urvr9fqZvTjyWuEDNE83cb96HNvJEt1NvjNFnCudKwi5sYdeyzQ24xxG0K1/iKN2jP953x+CvYa7E8O39yw+IrqysBGD3BCPRuVMX0h4Y06bXZY7eQjemYzxyZTyj0SgT8S8vLwcVC06j3+8rl8upXq8HbtyBDacApYKShjFI0xuB+XUmkofuQcHDa1BGNBvj2Lu7u6rX6wHofe1WesyPRiN1u12NRqPMvQf0SQQ7TRTPrOKktAcEfj2d2lnYwg6bzQ24Y568jJUTsbwOi6P+OLqPKR4HAmiGer0ewByO2UvyAVxkiACfR54euTstwnlhAL1H7PyPhJIWAD6bgLZBsZIkSXBG3W43s9iJNxGLHUqc1IR6cZ7cgRUH4/vBaXCf4m6ULmkE+P1+uKP1WdpwOMwsDUjU7zSLU0zMBvyZYH+eTF7Ywg6bzQ24+4r2mEeaDuAO2L7wNV9mBynXcwPq0CkoW6AhiEx9IQuKYbza1KNzQN4BhySk0x1EkvGMBPAl+meMLKMXgyfn7PvjHFl31ZOJvjTfuVoksy/UJpwX50gkzpiLxWJYNIQxxTMi/83xMc7f8xE+u3CZIzmG7e3tzGIkvoyeL2Xoz8605O3CFnZYbC7A3QEJI0Ljiw41wI/rxxuNRqYKlIQmQOVdBYmMHRSdivAydl9Jif+9da+DPJ+PdeQOQpwrNEQMatAs9Xr9rEh3OByq1+uFKBbr9/vqdDoql8uBj4eS8f2WSiV1Op3MNXYVjUfFGM4HHpvI3ykPHA0JV+lslQ73j/t6vhkWf9MRM01THTlyJCSPR6OR2u22tre3M86Jz/pztaBmFnZYbS7AXTpIrsU0jJRdpanRaAQgozkXAA2IO8jH5ekeefs2jMFXInJ6wlUwvCcdzDhiR+HJW++dDpVSq9UyyVDOHa4ah8Lr/X5f7XY77JP8ANcG+oKxeLtgAJ6OkdIe/eE5Cq4DziO+TgC0OxafieAg3HE5qDIT4VhxMtydMPkMXsMBco7D4VD9fl+bm5va3NzUxsZGODYzjjivsLCFHTZ7RuCeJMk3JbUljSXtpml6V5IkRyT9lqRbJX1T0j1pmm4+1X4cCIgMAbrxeKxyuaxjx46FCB2ALZfLAfQBFAcI76FuY84kZT05Fy9S4a852Dl9EYO9gy7AzzngdABnEqhxIng4HIbqV5Kko9EoA6A4hWq1GiiTfD5/VhdIxpamqarVaojCSc76otnOvbM/T2KmaRrGkSSJ/uIv/kL9fl8vf/nL9Zf+0l/SF77wBfV6vcz5cE5xkzPG1Wg0lM/ng6wUIyIfj8dhnJw/qqFqtaqjR4+q3W7riSeeCF00WVkqTqpfjF2qZ3thC5uVXYrI/ZVpmq7b/++R9Idpmn4wSZL37P//v59vJ14oxBe7VqvpxIkTWl5eDknOSqUSQIfkm0eXTqlM6w2OpM+Ts85d+/9xoy9Xk3iU77QMEapL/srlshqNRjhupVLJODTG51SPg5sk1Wo1NZvNAF44Fj9PVCjurEg4onrByAl0Op2Mo0NT7kAMDcX5MbbHH39c1WpVjz32mF7+8pfrK1/5SlDBuFP1+0DFa6PRCLMdzyfEfD0FW4yJXjoUb+3s7IRrs729rZMnT2bOE6f7NO2SPNsLW9gs7HLQMq+V9Ir9vz8q6f/Reb4AMSVTLBZ1ww03hOXqSqVSULIAlAAbn0/TVIPBIER68MtOO0jZ6lbpIBHodEJMEbBIdVzF6eDKuOPoH/4dAAXwnHICfD1KRmfOvuDEt7a2MkljgBKpIefiyVJej9sfMIMoFAoaDocZJygpOAe2B5y5FlTk7uzs6JOf/GRYvJvzJtlMLqFWq4XXvG871yjWwLvaZmtrK9BMjLNcLgf+vVgsqlaraWlpKdA1Psu4RHbRz/bCFjYre6bgnkr6P5MkSSX9YpqmH5F0Ik3Tk/vvPyHpxIXsaDKZqFwua21tTWtra2o0GqHys16vq16vZ+gDp0aQNBLhSweqi+FwqE6nExbBcAkgkkKXUbryAoAZDAYBHL2CU1ImcvcCpnw+H2gQHIB0MEPxiNsTs4xnaWkpdIREK45aBCnicDhUuVwO44tnGYyPXMA0oMYxEcVLe1G9zwA8QYrzkaQXvvCF4VoPBoNwfCgW6Jbl5WUVi0UNBoNwzeOcAvej2+2G5mhw6MViMTgS1DP8XS6XVa/Xw+Lh9ME5evSoWq1WSLrO8tle2MJmYc8U3F+apum3kiQ5LukPkiT5sr+Zpmm6/+U4y5Ikeaukt0pSo9HQiRMntLS0pKWlJeVyudC3ZTzeW0BiaWkp07+EpCHyQemgBS/AVSgUAi8LSG5tbWkwGGSKfQaDgaQD0IuLpODFiUadQ94/z4xGPUn22hW4ggZduCtQiM7dAD6nm6SD/iuAogOipIwGHxoDgAXYPakLEENh5HI5VavVMAOA8vAqWu9D4zMCnBVyTqSmS0tL4VzIIzBOl2i22211Op1MLx2cSb/fD/fKo3C/rjxDq6ur6nQ66vV6oUVCv9/Xf/kv/2Vmz/bCFjYre0bgnqbpt/Z/n06S5CFJL5J0KkmS69M0PZkkyfWSTp/jsx+R9BFJuvnmm9PbbrtNOzs7ajaboXWupNAMK01TLS8va3V1NUSEHj37ND6uVgXMcBKTyUTtdlu9Xk+DwSAADNEq1EYcxXO8mLpwvhq6wJO9vo0reZIkyejkXWboycydnR31er3werfbDfuDeoiVLTgXV9DE66rCj7ucMJfLqd1uT1UWQT3FsydAHWqEYyNddBWLJI1GI/X7ffV6vRDJ53K50JvHJZGAfKVSydxbumiyNCD7bTQaoeXB5uZmaBkxq2f7XA5gYQu73Pa0wT1JkrqkXJqm7f2/Xy3pZyR9WtK9kj64//tT59tXPp8PC1PwhXW1TKlUChF7kiTqdrva3NwMgOR69FKppGazqUqlEoBBOuhLQ9n8kSNHtLa2FiSG5XJZnU5H7XY7o1H3aNlfh0IB/ADOUqkUxk+UCsAQ8UsHWnCcEioVnwE4tdTr9dTr9TLXjPNlxsJMo1KpnCUNxBEwZi/uQs0zGo3C9ltbW+G8nYqRFAAe+oUcCCDrzgInt7u7G5wpjc888cu1I5r31r6eP2DGkqZp6MbJLMOvJcnX9fX1jFrqSj/bC1vYrOyZRO4nJD20/wUsSPp4mqa/lyTJ/5D0QJIkb5b0iKR7zjuIQkHPec5zwpeQxllE4t1uV2maqtfrBcBxOoNEKOqKfr+vJEkCb08kDTg4eCwvL2tpaUm9Xk+tVkvdblf9fj/QE9A7HIe/pQMViRdUraysBLrF5ZHICGMKiCSsFx65oodtvYhLyjZCg8phHNAygBzXzvMUnL9LOekb02w2VSqVtLW1FaSVbONNzIjYvS8Mx0VVg1OKHZMntj1JjtP0ZmNxErpYLAbKxytkfbYBRbWysvJ0FDOX7Nk+LPaX//Jf1rd927eF/3/v934vtINe2GzsaYN7mqZfl/T8Ka+fkfSdFzWI/aQp8kfAcDAYqNVqBbBwSsNlj5Iy0SVAj2oGfp4kJYlOioic4x8MBqGjYb/fD7QH48Q8CepKjWq1mknOSgrab7TllO4TWTebzRBZ+2zBNfFEp8wICoVCoEBcoUMRFNG1R/6M25uK0XkSMM7l9vrXcE7lcjnMCuDTnQrB4eBgmAkwIyLZK2VX08IJQbFRB+DVv71eL6PuofjLVVI4aj6Hs8XZUAx1MXYpn+1r1a6//notLy/rx3/8x3XjjTfqpptu0u233x7e/9znPqcnnnhCb3/72/X4449f9vGUSiXddttt4f8f+IEf0Mtf/nJJ0mAw0E/8xE+o1+vpiSeeUKvVuuzjmQebiwpVojeADBBg4YszZ85kpuiAlje28gpMStRdOjcajXT69OlQOBMn/ADfer2uZrOpY8eOhZ4mKG3iUndJIXnoFATOBAcDvxxr0CWFiB5w43OTySRU404mEx05ciQDiL6MX7VaDW12JYXr0O12w+u0LhiNRgEAJQWZIslsFuVgTNBiXB9v3UCCGApsd3c3JDRR+ACsMYcvHTRag6P3HArVx/1+P7N2LNffaSjPVTht4wVaC3vmViwW9R3f8R167Wtfq1e/+tV67nOfe85r+6IXvUjSnhP4/u///ksO8Mw0v+d7vkcvfelLtba2pje96U3h/Xhc3/3d3y1J+k//6T/pTW96Uyh4u5ZtLsBdyvZh99JzwKPdbgeu1qfp3GS47RhkvNqV/Q0GgwDY1WpVzWYzAKJH1NVqVaurq9rd3dXW1lbYJ9G+c+ne671YLKrb7erUqVNqtVpB7SEdUCiMm/VWvbMh4yQZ6XpzePh+vx8KgrgujAFV0NbWllqtVtgPn2fmwxeAmQUVrsViUY1G4ywlz2QyCTRNsq+xxzxBCn3ix/EfQN7vfZz09FkC+5KU6WHjMzlvqkZugeu4APdnbt/1Xd+ld7zjHXrlK1+pZrN5wZ/7G3/jb+jBBx/UPffco0cfffSSjOXOO+/UG9/4Rt17771qNpthFbOnMp6B7/3e79XDDz+sD33oQ/qP//E/6utf//olGdM82tyAu6RMQg0ZH1FqvV4PdIkn+eC56dVChHnkyJFQxcq0HbpG2gOJ0WgUKIt6va6VlZXwoEDbEEXSWpcZAQ5jOByGpOJkMtFgMNDW1pY2Nja0sbERolv4aEna2dnJ8PDQTsvLy8GpEZnjbJx/57NeSMU12dnZ0ebmptbX1zNgG1NWLot0eSHXHd6ee8Ksg2Qtx+MaMjacHnQPvd995sU4uL9xNO8Vta7O8XwD18edBOPJ5/MhZ7JYQ/WZ2W233aZ/82/+jV7ykpcEBdvF2otf/GI98MADuvvuu59RBL+0tKR/8S/+hV7zmtfopptuelr7SJJEx44d08/+7M/qDW94g173utfpm9/85tMe0zzbXIC766hJ8qFu8cTjkSNH1Gw21e12g1wPxQRJVu9gSHdFzAFCUqZoaDAYaGNjQ/V6PST32OdwONRwOMzwwyT7ADQKpRg3iVPp7EW3GR9jAEg9Uu/3+4GW8eRokiQhZ8D5OE3kWn7pIJnreQrMawVYmBu6xRfaoPUD44UP39raCrMKjkFk7cejpQHn7hXCnBeVyJIyDsMllBRKIbvEwXDNuffM/Liui8j94i2fz+tZz3qWHnroId1xxx3PeH8vfvGL9clPflKvec1rtL6+fv4PRFapVPSLv/iLev3rX/+Mx4Ldeeed+vSnPx0Anu/ktWJzAe6Swpeb5CJAxpecLzEURrPZDNEk4AYAO8A4mMV8rJTtB87xxuOxjhw5EhQXngSEupAUQJZmVtJBr3WiT2/dS6TsUkocFHTKxsZGWD8VwMfJSXs9Zjhev9/P9MkZjUaBBnJNO04gjniZMXhRFtsS0TN+6WAtUwqSXNXjs4FisRh06bQ24B7hAEi+up6fcwWkJWWcIKtdcT1xQLHz5B7j7Bd2cVYoFPTBD35Qb37zm7WysnLJ9nvXXXfpVa96lX7jN37joj/7j//xP9Y991x6cdLznvc8ff7zn9f999+v973vfZlg8Gq3uQD3fD4fosa4AyNcNF98ACkGTqIzkniAAIDqU/q46tIX83Ae2xOCJP08UibCdLkeUS7JQGgJpyZIDLL8HBJAwLPVaqlWq2Xkg0SnRMO07MXhsC+WusOBwcP7OD2KJjHNNeJz3sKYFg7kCOJiK288htQUWoaEszcf8ypd7oePg/1wv5kFcJ2YSeH0/B7zTPiMbxG5X7hVKhW9//3v17ve9a6MOuxSWJIk+gf/4B/od37ndy5KJnnjjTfqLW95y9ORtF6Qrays6N3vfrfG47He//73Z2bdV7PNBbhL00FXOmheBRWCAwDUiczy+byazWYoVKKcHb4Z8PXmYOwDZ4JCg306Vz0YDAIXjmPhIYBqYJ+cA5EkNIKkQPWgxWedVCJRrkGr1dLa2lqIjLe3t0MRUL1e12g0CnSU0w9ewIQWvdPpZK4DjoVrQiI41kZwK4wAACAASURBVMFvbm4GR8BsaGdnJ9QNDIfDcN0AW09ek7QmiodiYeaAughwhoIiIerSRhYid/0/n+O6elGYzywW4H5hVigU9P73v1//5J/8k8t2jO/4ju/Qd37nd+qhhx664M9UKhXdeOONl21M0t65v+c975Ek/eRP/uQ1EcHPBbj7NNsTniT3iISJeAeDQaaLIJGit42VlPkskRwOAaCiyhMaRDrofw4X7EVP0BLsH6CXlOHMGQeFPNLBYiBUhAKYRMn1el2tVuus6kyUNcwEer1eGDM8NZJGB2cH8TRN1e12wyzCk8WMrdPphGvnlZ5o/XEoUC90hPSI3fMk3Etf99WB3aNsPo+2npkMzgAn71p/n1n5OcfKnAW4n9+gYt71rndd9mO9+MUv1qc+9alM/ut821+pe/iP/tE/Uj6f13vf+96rntKbC3DHAEW4bnTZ/X4/VE0CkN5kiqShc8uFQiGoTyik8YgOoPfIHSO6Rx7JMQqFgrrdrjqdTgAmqAfpwCnABSdJEmSX7ngAcySYgJkfk6QmzsdpB2gO1oHlmP1+PwCvpKCzb7fbgZ/nGN6Ot9/va2NjI6iLAGCqUZkFUAHMOOjzw3Hz+Xz4vFMwAK03IsPpeH6CHIqvrOXqII7tUkuuqecGnDrz1acWNt0uJxUzzd74xjfqJ37iJy4Y3H/wB3/wioF7oVDQu9/9bp06dUo/93M/d0WOeblsLsA9TlpiHkUCBg6ERJQACwk6wNwrI12RAsdLsRPbeXKV/bk0sd1uB6AEpGiT68cjaSlJ9Xo98OP8ED3DiaOK8WSjpACWPgaP8nEqcO2SguZ/NBrpzJkzarfbmQ6TgB194XndFS986XAYaZqq3W4H/pwWu4VCIVTlkhNhZgKwkyCFZvEkKqCOQ2ZmBLXE61BG/kxAMznP7tJOl8ou7Nx2JaiYq80KhYLuvfdePfjgg1e1THJunnyiOFc8EKX5cnOU5VcqlUzhD0lRombpbLqH13EWnoxzgI8VHYA4BUE7OzuhjQDAydg9SUgTMSgUjoljQCECXYLTOH78uOr1enBaTvVIBz1uoHcajUaogE3TNMgU2R+fzefz4XUHQm9SxnZOfXB9e71eSK7SQwdlj2vTOUccKNfV7ymJUdfs8xrXC7B3msqT50Tsfj3dYUD/LGy65fN5ffCDH9SP/diPXdHjNhoNveAFL9DnPve5K3rci7HnPe95euihh3T33XdftYVOcwPunhiLV9ABGMbjsbrdblDWANhE5rlcTtvb25lI2RObAItLIiVlAB6wQ2UCFdFqtdTr9UKlrJfyO5fuPeUBWueWaX0QK3lQu0wmE21tbQXQhILp9XoZEOXzXi3qnDwcN9uhwvGWDFwzd3qMB+CHGsJpcd22trZCpL68vCxJoRWCq3zYn++bPEXsVPya0dffpY+YzzLiH8yv0cLONqdirvQ1WllZ0fd93/fNNbhLezr43/qt39JLX/rSkGe7mmxuwF3KLo5N9OvKGL7oTN3hsom0UVk4FcA+ARWAjeSsOxVXZvixKOPHcQB0/Ca69RJ4GnE5FTSZTELUS+TLeTr9wkyhUCio3++HMTuN4uadEePkLgVHg8EgKHUkhXyB0yGx03PFgCejOedut6szZ85oMpno2LFjGW4dp+szFe4xf7vCyO8jET/3yFU8TrtMi+adj3eHvbADKxaL+umf/ukFFXMBdscdd+j1r3+9fu3Xfu2CcwTzYnMD7j6Vl7JfUkAXMPAvtxcgsfCGd3IkSedTe0AETtYldd4vZTKZaH19XZubm2HxaU+MevRIrxdP6uJsHFRxPJ1OJ6wpOpnsLUbhTgPA92rWJEkyXScdKIm+PYfANSPBirrGQZFcBtdNOujP4t0WXUHk94mK2Fwup6NHj4aEbZwTwQBc/6I4+KKxj+Wq7qBdNuvPiM/SzhXRH3bL5/P6wAc+cMWpmKvVqIz97//9v+vhhx+e9XAuyuYG3KVsNaJHox79ARREywABJf9EoF656Hy4Vz661IlmWRxvd3dX6+vrgVt3mSY0yjROGfkkCVL/HDMGHIrz6qh6yD2MRqMgBeUccEbuVNgv58Axx+NxKDhCaeINtbiG/GbpQs6fxKdTKk7z+P0Zj8dhcY+lpSXt7Oxk1DXe7IuxT0uce12BO0/uP+fK9ePae5Tu1yO+5wuTnvWsZ+mHf/iHF9fkIqxcLut973uffuiHfmjWQ7komxtwd+pFyq7SAzi7ZBG1CeXzXvhCBOxRn5QFYI/mSNAS/e3u7ur06dOhWyLHlg74XmgNjs9+XQUSyzYxB9xGo5FZucn13B65cn5UwPpqRjg+LwxiDAAx4+RaAvJeECQpowSCDqE4jIWz/Ty4jtBXksL26ODjCtY47xFXHrozpiUzjs0pGigbH4tz84xrEbnv2W233aZPfOITOnLkyKyHctXZq1/9ar3sZS/TZz7zmVkP5YJtbsBdyrbyxQB4qlYBM48EAT9P+rEfnITztXEyDmqEKHl9fT1oqgEmQNrXPAUAHXg9b8A+vbCH30TKzFLoPDmZTILCo1QqhcUpOGdUMZIyhVB+rWgG5m0GpIMVnXCCHsH7LAYQhs7xBDdVsV4sBRCTHyAZnSRJ6Obp19Kjc3fmvMY98ipaErD+Wcbn+QJ/Jtw5Lkz6hV/4BT3vec+b9TCuSrv++uv10Y9+VH/1r/7Vqya5Ojfg7rypf4EBLBphOX/uFIFH7JIylA3m6guO6TSJJG1ubmo4HGaStYAUS88hpZQOlDaMD0eD1p4Fm/lBhx5XY+IAGAsJyd3dXTUajUy1qDsDpyy83QLXjui20+loa2srOC3249Ew4E3ewWdOXG+kihwf5wIYUzRFdEifnlgRw/3wcbqKB1D2v3224bp+aLTY4fvPYbYkSfSjP/qj+lt/62/NeihXtd100016xzveoQ9/+MOzHsoF2dyAO19yV01AL6B7jisT+VK7/DBWZQB+gLDL+iihB8S2t7fV6XQyqzgRTXvPciJKT9LC+dMaYXd3V91uV+vr68rlcmGZPBKr1WpVnU5H3W5XzWZT+fzeIuG0XXBngSPxdgN+HWik1e/3w3ZE0ryHfJMfrjlRL/QFWnZJYXbhC6FwzXz2QosGHCJ5Cq61yyMdxLlufr88mgeUmTm4Agr1D/cRDj7m9q/2EvJLYcvLy7rvvvuedj/2he1ZoVDQj/zIj+jjH/+4Tp06NevhnNfmBtylg2k6X3T+ho6RdBYoQFHEHK5HsFA6HrFDX3gkSJESgERzMNfHLy8vq9frhTVWXeroBTUAC+dCT5ThcJhRkzDm0Wikra2tICms1+uZIhwHYEmBvnGuntd7vZ7OnDmjXq+XcXx+7lwn9slsyRO4vV4vzCJQ9ywvL4f74ucwHo+Dfn9nZ0e9Xk8rKysZrh6n6HSJz6Z8hoC5OolzwIkmSRJmEqzIQx+bc+3vsNnKyop+7dd+TcePH5/1UK4Ju+OOO/R3/+7f1f333z/roZzX5gbcvVQcMCZKBRQATKbksUqC11x1QbTuBqgAnGjBmfZT9o+MEopnd3c3FDF1u91Qwi8drNzEceHoAd24EMiLiXwZvo2NjUCdQLuUy+WwZCBAiWLHrx9UELr2WLrJOH12gwOCooFL9/J/HABFZIPBILOsIOOAF6fFsEfhrqn3Qibun1Nw7qin0SpOvUkKtA+tEHBqTlkdVjt+/Lj+zt/5O4uk8iW0e+6556oA9/M2SE6S5JeTJDmdJMmf2WtHkiT5gyRJvrr/e3X/9SRJkn+ZJMnXkiT5n0mSfNuFDIIvNeX+1WpV9Xpd4/FYvV4vUCXQC/DX0xKogIR3jSSC85WePIIkSiUJSNJROlDlbG9vq9Vqqd1uh6ixVquFYzuVAP8PpSNlk5I+VpKZfAbZZavVCsDqahhPBkMNAeLQR1AgXlzFj/fL4Rp5G2XpoHgJR+d93HF6tDigLbA3K5N0lj7eQZvrFCtc4oKnaXRbnJjlPWY+w+EwSEx9RjKrZ3vWdu+99171wM7C7/Nid9xxh+66665ZD+O8diHd739F0ndHr71H0h+maXq7pD/c/1+SvkfS7fs/b5X0by9mMKywVCgUAnC02231er1MV0MHM8A0ljeSjIu/4M7TTkvkoZWHrtnc3NSTTz6pfr9/1spHnoB0RzGZTDIJWfjzOBnILITmXIAxr1HQ4yDoLW8BXkmhz3pMOTFO/vdkMPQF75HA5Zrk8/kA0oCoV5TieHF43g/GF9Lm3voshnuBueNi9hDfNykL7tKBU+W4m5ubGo1GoW3DeexXdIWe7VnYs5/9bP29v/f3Zj2MZ2w/+ZM/OVczsOuuu05vetObnjJwmAc77+jSNP2MpI3o5ddK+uj+3x+V9P32+q+me/ZZSStJklx/IQMpl8th6bzTp0+r3W5nJG6AmnQADt4/haicCN0BiRmBdNAfnUgVGSQRKe9DuxC5I+djn07BSMr0o6lWq1peXtby8rKOHDkStOwssn38+PHAT2Pj8TiodEh+kiAlykVpA1BCWQDCnBcSSmY/Ozs72t7eziy8wbVgduA0DA7JNfye2IUSwungjIiwvIWAO1Epq8v3PADHdurGi8YYg9NMjMN18+PxWBsbG9rd3c0skjLNrtSzPSv763/9r+s5z3nOrIfxjG0epYd333333HccfbqjO5Gm6cn9v5+QdGL/7xslPWrbPbb/2klFliTJW7UXAenGG29UpVLRxsZGZtUgBwVoGKdaiOR9UYe4qMW7H3okC2BKe4U7ADsROq0MiCQBSWid2IEkSaKlpaWwf5dKSgpcNA2+iPiZRRQKhcAVe4Ts+QVAnBWg/JxwVL6gBgnNnZ2d0IERKokOmiw6TZtdl0Y6D841Zz9xYlRSmK14cpWEqvdy93sQSy39NU+Au3LGHavnVvxebW1taXV1Nah8ZvVsz8oKhcJVV1F5Ndnq6qruvvvup7Ue7JWyZ+x60jRNkyS5aDlCmqYfkfQRSXr+85+frq+vq9vtZoBNyvYiAXxi820cuB2U2G7/2AEoAfZCoaAzZ86EpKTvFzBynr9Wq4XSdkA/Vn9wbKJHKJ18Ph9aDXiRki/A4T3hvWAKB+GVok59NBqNENHm83l1u90MvQJ9Q4LTpZXSgbwU8OczTn9xTdDt8zmAnQh+eXk5o993hzGNZ5+m/PHrB+B7onQacEOvbW5uam1t7Wlzzpfi2X46n78Ulsvlrgpe+Gq1arWq5z//+XMN7k+XNDrFlHT/9+n9178l6Wbb7qb9157S0EZjHlXzP+DuvOz+8UN06dy2G1QLnyWKhmpALYPCAxD2RKa0V5Bz9OjRUFREJ0pPWLok0ikEaBWUN/V6PYyDcQOU9GivVquZJloAPNQNjgKqptFoaHl5OTgHXodXh67xHudxEtM7QTK2er2uarWaWQ3LpYlcM6dgRqOR2u12RntOTsB/4nvouRN/DijgOheo43Digqt2u32x3OglfbZnZe985zsX8sfLbD/yIz+iW265ZdbDOKc9XXD/tKR79/++V9Kn7PU37SsLXixpy6a45zS+lJIyCpgYCOCGPVJ1Dbon1+IeNXz5oSA8AkVlEYMD+2k2m1pZWdHS0lJ4z0v4ScAyRqSSrVZLGxsbOn36tE6ePKmTJ09qY2MjrOjk5+sOAhqFDo9Evz7umBohWerXh4iZilMAPl71qdFohH1DC/n+uG71el1HjhzRyspKoF2YzTBW+uLw2V6vF8ZDNSnG+fIMeJOvGPzZp4+Jz3kexJ8dSSFXchF2SZ/tWdmxY8fmlhPe3NzUgw8+eMHbN5vNuVT8HD16dK779Jz37idJ8huSXiHpaJIkj0n6p5I+KOmBJEneLOkRSffsb/67kl4j6WuSepJ++EIH4k2s+KLHU3iSjui/AViPPJ2DZl8O1HRaJGJHV+4zAQCOSkzAGyoEysKjdS+AiouGiPA9QkWyyFqvg8EgJDA90epae/hyb1PsLQm4FtASnU5Hg8EgVKoS+frC3DGXzRJ+UDC+LB6zkiRJwpJ7NBoD2Dlflvprt9tqNpsZB4OUVMr2EWLW5ZQN26Hb55w9gmd7Twg7HXeuKtUr9WxfaVtZWdG3f/u3z3oY57Rut6svfvGLF7z9fffdN5fgniSJ7rvvPn3f933frIcy1c4L7mmavuEcb33nlG1TSe98OgNxzjWeYnvkTfm8q1/4PMlAB9f4GL5Ix/b2dmaqz+dJLjoN5K1yiU4ps/cFuH2RDKLdbrcbom3yBnDZUAccQ1IAesZLLxpPNHplKWuMMlYqODudjjqdTijywRlCB3HtfMFvnES73Q7OEDqGyNodJyXtLN3nCW6uA0VPyb4Kh/HEbQhwWM7vA9IkeaWseobfMV3HM+DR/JTn9Yo821fa1tbW9IpXvGLWw7hkdgGS1pnZPI9tbuZtrobxaM6/6NLeF5vug3z5vZjFQT3m7bkRfOnb7XZm/zT58kjYKzYp3nEKhXEPh8MMuJDERDXjCVF3WuQbvMAK0EexQzLSAd0jX+gUFvymZ41H614568eJuXOcBefmY3DOnmvoCVPUP3Hi1Hv1cF08mmY24/eQ8boj8/vLfeNaerGTO/zDaCx7uLDLb3feeade9KIXzeWSgXMD7hhf1mm/iVz7/X5oJuZT8jiZ6pEuVIK0B/rb29uZBmJeYYrOXFLgjZkxAGQuUZSyC2Uw3slkr4WAA79XwRIZ83miXKJ/JI3MGhgrn/drRnTbbrfD/w6+ruJx4HZ+XlLGkcDpE/GjZ4eSoeka1208HmtlZSUAOuDO8XwxlNghe+QNWPusyWkYB+1pyqY4IX/Y7L777jvU538l7cSJE7rhhhtmPYypNjfgzpd+2lSb//0LPh5nVxry30SF8MrTmk95QhRaA2qiWq1qNBoFvTvqD8bg+QGoFi+6AaSIOOv1ekjYonJZWloKFA3gLynw/BRDOYcc0xiudPFomKSs925xh4NTjHvBs8C1pDAGT7DiKKFVer2eqtVqSJSyDRJPjst4mA1BM0HFMAafiXlEzhjckcfnxeteQXtYQT5Wiy3scNpcPPkAuS8k7bwq4OJfWkDSt2E76SDCg6tFVkg0i94bYAcYnbMmivbqVMbm44FqkA4iX5KeROcnTpzQ2tqajhw5ErorXnfddSEK9giUAifOgQpbL8nnOG6cP9cgn99byq9Sqaher+vo0aO67rrrdPTo0VC9CZgyTs9JuPMDKEnqkqhttVqhgZqrd3A4mHPo3lJhGrB7At3pm3iNWvYbR+1xIv4wWaVSyawDfC3Y/fffP9f38s1vfvNcOtS5iNz5QjrP7ZrnabK3eCrvX3qnc9iP87AkZJMkyXQQHAwGGZkiyT9P9MVL3DlVA2hRus+YkVOi1CFpiwLHQTqWNjJ2nJJr1AE7rp2koHvHabnShWicFrm+NizXejgchgSrUznMiHq9Xobm4fqTvygUCsEBMAsiAnfzxKwDvP8f30d3gJ634DWna5i1HTbe/VWvepVe9rKXzXoYl9T++I//ODPrnje78847M4zBvNhcgLuUXSrNAQuP6JE120NrONDHCT2AwsvciappdSDtKUb6/X7oYc5YKpVKSJgiw2SMaM8bjUagJpxCAHBc3w1dAqV05MgR7ezshIKf0WgUEpaMgWQuETz8N5WxXBd4cFfz8HmnbwB+l0a6mufo0aNBzUPbAJLMTr14VI9j2N7eliTVarXgXCqVSoYewwl74ty/uJ4s9mSqPyPukPxzfu/9WTosFl+TebT77rvvrHWFF3bpbW7AXTp71XrkgD7tjotgyuVyBvjjh9sjTwCM3u1EmltbWwGU2K9LMvmcR9WNRiMjJ+S3r8fqMweifEB9NBqFRmMA3+7urprNptbW1kLbYxLHPvNwmkTKRr5QSLQv8AUyWFs1l8tpaWlJS0tL6vV6waExIwEc2Z4VpFz14m2COU/09Ti5lZWVANDu/Px+uwOf9jw4RRNz7Zg7Cgfywxq9z7s9+eSTsx7CJbV8fm8VtfX19VkPJWNzAe5E054E8wjRwcuB2r/w7hjcXF7p9AKggmwQyR3RpksdATNkgIVCIbMQh3PxUlZxQjdGlqwjt1CtVkN/GcaF/hwnEKs/iJadV0e1Q8LXo146UdKON5fLhQW2jx07pqWlpVCezyyACJmxIpfM5XJaXl4O5xUnqblursLpdDpaXV0NAO8UkXPjOGV3lMwsfHueBS9Ocmcfq6MOW9R+Lds838frr79e//Af/kP91E/91KyHkrG5AHf/EgJy/A14s12sGIEqcS33NKAnSnfApE0tdEWa7q1i5Fw+6hE6PuIcyuVyACCnR8bjcViL1RU5gDOrBbXb7SDpBEQ5F5JiRJ6SQoTvSVZXzDhYkgRmgRPknDgwzj2Xy6nVaoViLrTqzEpwPlxTnAqcPqsycd6xsqnb7YYKXMZHtS7msxwMKsjPi23R87u+3qk7fwac5jsMliSJ7rjjjlkP45LbI488oo985CN6+9vfPuuhnNPmcXY4F+AOcLiE0HXnnkjz5KDLJ/msdECr+JcdICDqYwEQIkKSix6ZOtABOCzmwfGJnHu9XtCYk6yNWyFIB8vNkbis1WqhEySAjZbc9d3uWGInReQNcONwyBN4RM+solgsqtVqhaiaa8W4UNpw/lxXjNlAuVwO+QJPdmOdTkeFQkFra2sZR+dRuztm7jOUjQM7a7nS5C1J9mSmS0tLwQnyfLji57CAe6FQ0Nve9rZZD+OSGzPghV2czQW4S9lScoA+7iXiU3c35549qebASiIS8ItXXOLz+Xw+yCadw2c7b5HLgiKuIgF4net19QegyEIWLCYNtVEsFrW6uprpGQ+AezTrEbs3HJOUaYOA4/JZAP3V+YwrX9ieWcxoNNLS0lJIjHr0z4yjVqsFTp4EqBeY+WLZ8T2M6RPnz51iYVk/HBfniVNeW1s7q7r5sMohr0WbNyVKbPMohZwbnbtXJLpSY2trS6dPn1ar1QpLz/lnvNe4d8FzoHdg3NnZCf3GY269VCppaWkpJDqlPWDa3t4O5fzdbledTkcbGxshGQnNAGD6uqTw3nF/dqJ6fyhIIA+HQ1UqlfC/67v98z4rIWHJe3EC2Mv7AXB64qB55/qgAuJ6djodbW1taWNjIyRM48IllkisVqvhvnjyM27pHDsun5WRfAb0u92u2u12JpJne6K6ra2tjPNjGwf7hV299vM///M6ffr0+Teckb397W/XzTfffP4Nr6DNVeTOlxe+lvUwUavA3x45ciSTNKP3SZxQxFzBgkwQgGN7LyQC8DmuL39H8ZQvSycpgCdROU4F6SEyw1qtltHlF4tFNRqNwIGnaRr6u3POHtX63869VyqVTBWoR7GehPYlCr1SlVmSpEDPoLoBNLkO5AWoTgVkKXLyZCuzGRKtLp2MI3jumz8LzLLcucV0y2Qy0fb2dug9j3m+ZmFXt7VaLZ06dWpue9Qzu50nmwtw58ssKWi9z5w5E1riAgasBTqZTLSyshKoDLTncbToGnAAD6AgsoZWcHWG5wCI2nmfZB7mgMh54ESkgwUvUMvQe4XqV1ZicgUPD4m32/Uuik4XOXjhUHB4rgWHHyfSxglBp9BqwGkql0d67/xutxtmVo1GI7MYN+Nme5wUNJu05zyhpDy/4gBP5O59bZhpOdXCvWWGhZqJ52DevnALe3q2s7Ojn/mZn9EDDzywcNgXaHMB7tJB86vxeG+BY76oUjZinUz2Goclyd6apd5DJU6u8jdFO+wLAPMFpH2NT2mvhW273Q4qE6d/yuVyiCo9cUc/GEAXUKOjIgDPeXnDMgDJK0OJkmNePVb+SNkiL0AeEGbpPc7fE5HSXlSEZJLo32kUp3u4fr44B4lXrjnnwHJ+Ll1EVeM5AygmngG4e5Q+OGZ3vvFsJk1TtVotFYtFNZvNcG19Zalr3f7KX/kroTr4WrTf+Z3f0c/+7M/q7rvv1u233z7r4cy9zQW4A5rj8VitViv0N/ckqitjqL4sFAqBopGmSyCdl3YJHk2yUM2w//F4HGSKNMhy0EdN4wAOp+6FOgC6q3vgwUkIxuN2AIXKIYFLIZF00EEypqBwdCzTR9Mzp5VIFtfr9RAJA6Q4Qumg+IuEJZGz97ThmpFYXl1dzfTUh6ZxBYvXB8SyVkAa0MfpxufpKiGSu7x35swZpelen3mc92GxH/iBH9Da2tqsh3HZrN/v673vfa8+9rGP6b/+1/+q1dXVWQ9prm0uwF1SkLmtr68HgPIoD6NyFbAG5PnC4xScumA/qEi8vS6ABWXTbrczy8S5o3DdOtWbPj4cEZQEESzARpKWCB4duKSMM1heXg5RP71tAD+XeLqahCQoTpIFMpilANLQQOQTUOqwP69ABUDr9XqgkQB9b8KG+mZzc1PValWNRiMTYdMfnh/oMN8HRuTPDIXXuIdOS/nn2I/POHBWC8XM/NhXv/rVi1qFaZp9+ctf1oMPPqgf/dEfvUSjujZtLsAd2d3GxkYAF28/gMGFSwe87XA4DJz5tCImkoYO8pVKRb1eL7MfSWHVIl9gQ1IGxIrFYma5O7hkwFE64JyZYZCgHA6HoXDJI3L089Al0B61Wi0ze/GEccw74ty8ghWlCZE0RV9ra2tBgikdSCNZTQkAhdKAwqrVaqEwyhPCsUolTVMdP348ky/wFgsskxh/hvvhFJrPfLyojde93TKOYWNjI+wLh7ew+bBvfOMbevjhh5/RPsbjsf7Df/gPeu1rX6tjx45dopE9M0uSRC94wQv0zW9+c9ZDCTYXUsjJZBJ4dlehAErxdJy/iTShHwBhj+hcZhcXAPGeS+04nh+X6HV1dTWU4Pu2koJToDp0e3tbGxsbevLJJ7WxsREKpjjmYDAIMwmalsF5oxAhuuV8adQV97QB9Lx/u3QwQ8BhlsvlsKAv146/ue7ValXLy8vByXgBUz6/16d9aWkpOF6/DvDlqFekgxWoROjc8gAAIABJREFUcI6xhNPvJ+0fHKz9nvt99Pdjvf9gMNCpU6e0ubkZFmVZ2LVl/+2//Tf92Z/92ayHESxJEr3lLW+Z9TAyNheR++7urp588smQfPSITjq7MpWoFTCDgoDKwOKydBbGgMfN5fZWZPKCJldxpGkaaAbXxHuC0CPp4XCora2t0J6AKL9UKmltbU2NRiMkCeGm4aupCKUyFUfEfkmyehQfK3twSsgiO51O6DLJtUnTNFPdyTnQA8ZXoJIUnALvOQVUKpUCl+8yTmYI7XY70+GSvENMk7gzdbB2eg2A5lyRanrRlSfd2+12mNV5ruVaNb47h8keeughvfKVr5z1MObWzhu5J0nyy0mSnE6S5M/stZ9OkuRbSZL88f7Pa+y99yZJ8rUkSb6SJMl3XcggvF/ItCjdgU1SBvyJ+Ih0nev1SNmlioDp9va2Njc3A9jBvaOXPn78uJaXl0MkvX9+IYJFTkj0vbm5mQH+XG5vXda1tTXV6/UQkUoKBVokbcfjcSiUYtwc1z/nGnCOg5MD+LywCiB0iakXWkkHgMmsgQU4tre31Wq1Mn1y+OGeNBqNIIeEz2d2REsGn1nE+vT4nFDEMFOIVTHuKJgVeFKV4+zs7Ogb3/iGTp48OdNn+0rZzTffrHe84x2zHsYVtT/4gz+Y9RDm2i7E1f+KpF+Q9KvR6x9O0/RD/kKSJH9N0uslPVfSDZL+ryRJnpOm6VPOix2giOz4onovEpfkOf0ANcPqRQCMc60e3dIN0lU5/CRJouXl5UzlZS6XU7/fD4CZz+d15syZwKNLe33f19bWztKqk3wESImMt7e3NRwOg3LFzw8KQzqYnXC+nlTEyC0AahyPawEwkiim3bDvD1UNvWV4D5qFBU6gabwXfMyhA/6TyUSdTkflclnNZvOsKmSPzh2ocQJeU+B0m+cfPCDAOW1vb2t9fV1pmoaZ1Kye7StlUIfzbGma6ud//ucv2f4IVq5lhdAzsfNG7mmafkbSxgXu77WSfjNN02Gapt+Q9DVJL7qAY2S+7PyN1C1uMUBkJylQGN4vJk6wAR5I62hBAJAQ9VarVTWbzUzrWK/KlPZA+YknnghdF2kFjGNaW1sLNAbqm1arpd3d3SBD9F7y9Xo9UCY4Bm/uhcTSO0vyvidPAXJmN64PJ+npjlKSut1upv3CtGrQ5eXloDEnN4EDgZrhuNVqNejyXfMOPSQdNHBDg06kDZXjnLqULRJzOg5nxX0i0dvtdnXmzJmQZ/AE+Sye7YUdWJqm+upXv3rJ9vf1r39dv/qrsV9eGPZMEqp/P0mS/7k/tUVweqOkR22bx/Zfe0rzLypfdKdkPHLnNemAF+ZzAJR/3vuMS3tgsbm5mXEacOvww4A9yct8Pp9J+LrcEfldt9tVv9/PUC3OLaMsOXr0qE6cOKHrrrtOKysrkg405ShJSFbG0bz/8JpTTh6Ju3MgAUr3SdeZe2SMc2NlKZyb685R/HjhmN+TYrGo5eXlcM5cc2+vwLiJ8CUF7t6LjuIZizsdP18caa/X02OPPRaauPnM5SLtkj3bCzuwL33pS2EJy0tlf/InfxJmzwvL2tMF938r6dmS7pR0UtLPXewOkiR5a5Ikn0+S5PNUUsb6ZudonUd3Pt0jfgf2eNrPb6969ISppBD5e2TY6XTUarXCgtmAHfuEyyaC5jxQ19xwww267rrrQmKPYiQKn7y1ryc97Tqd1YZA0lnA7w6QqL5YLGZ6wEwmk8CNswi375PompkP8khe5z6Q46DVr9NOzAKI6tk/1x3H5IVQfMbpMefo2c7pHK8sHgwG6na7OnnyZMifcO084r9Au6TP9sV+9lq13d1d/fIv/7JOnTp1Sff7wAMPLNoBn8OeFrinaXoqTdNxmqYTSffrYHr6LUneGu2m/dem7eMjaZrelabpXaVSKYCIR2ZOr8SJOI/knKP15llQH1iSJBk+mSpGAJLoGRXI+vp6iMSprPSFJarVqqrVqlZXV3X06FGVSiUtLy8Hagfem9Xo2f/W1lYAYI9MoW3QwRN5A55w3k7JQNn4OWPw4bVaLdA/JIJxCDi3ZrMZuHOfsUgH1Bf5EM4D5weVk6YHPXO4vn79Udv4OL0lQaxGcnNFDY6HiL3f7+v06dM6c+ZMpihNuvg1RS/1s33BB36Gds8992SK/ebN1tfX9a//9b++5PsdjUb6xCc+ccn3ey3Y03oakiS53v59nSTUBp+W9PokScpJkjxL0u2SPne+/QF4gLMDdPxld3qG153W8KIdT7RJCtSApABk0AuSwvqi6+vrarVamdJ3AMPliLVaTTfffLNuvPHGAJpQCUTDKEm8NQH8MJROkiQBhKl8BfhdbkmE66X6Ht0yvlKpFKJ1Zgj1ej3TGsATytAX8Oe1Wk1HjhzRyspKGH+cI+BejMdjbW1taX19PbO4N+Oq1+thhgGnzr0D6Jnt+H32GQqOiP+hblApbWxsaH19PSOj9efkYuxSP9tXyl7xildc7AzlitrnPve5qU77mdp4PNav/MqvzHU74FnZedUySZL8hqRXSDqaJMljkv6ppFckSXKnpFTSNyW9TZLSNP1ikiQPSPqSpF1J77wQNUGSJFpfX8/0Lyfi8qjat495+Fj14WXnJBbhtGk1gBOR9gp+6FnuvV88CZmmaQBqmoTRKwbwS9M0VJZWKhXV6/UAXP1+P0S6S0tL4dxQojQajeAkAGpXn3A+rpYh6sfhoJrh2lDt6s7SF++AUnLAZj9E6yySPS2q5lohYVxdXVW5XA7H2NnZCQlmEtU+W3EQ99+cZ5xod6kjVNnjjz8errfvx+mqWT3bC9uzj33sY5etidtnP/tZffnLX57bdsCzsvOCe5qmb5jy8r9/iu3/uaR/fjGDINJ79NFHA+0AuDPN9pJ6l885F79//AAsgKRHft790JOsLPjAtp5odF55ZWUl01WSKBynhDOAyvEiKVel+PaMi+Ny3tN4dsCVa8BYnJoiQuZaeOJ4c3MzSBx7vZ5WV1eVpqmWlpZCOwdJGb482S8UA6w7nc5UhzsajbS1taW1tbXQLpiiMFod+7hc/ojjYV9x3oR7xf1gqb0nnnhC/X4/qJlcdXS+ROqVeLYXJn3lK1/RF77whct6jN/8zd/Uy172sst6jKvN5oakQ9HyyCOPZDo1+pffy+s9KndelS+3qzy84tOjQMBxc3MzrOTjxTGoQyqVio4cOaJjx46FpCnj9TH2+32dOXMm9IAnh+BJTyJYX1ibKlDGIx0UbuHsoE3YDgfItYNi4rx8ZsO18VkHkfpwOAxL2KH6IUoHSKW9xUxwtFTsOh3kiho6M9JqGScBvSMdyBmJztkPDm5aEh2HCc+Ont0Ty+4UXc2zsNnZb//2b+t//a//dVmP8alPfUp//ud/flmPcbXZ3Dz5fOldR97r9TJNr/jhy+6RHfsAEGIawUHPJXpnzpwJwO4SQqJqkpHpvpbateGA/2Sy10uFBUZ8ZkG06/JKpyNoOoZM0ROonjj0/AJg7QlnzwU46Pvshl7nXjDmeQl3UFTbetLSHc7KykpYVUo66GrJ7IFVtPgM98cdG44xvp8+Zr/HOG2anj322GMB0DlOnJNxCmthV95OnTqlX/qlX7rsx3n88cf10Y9+9KkK1g6dzQ24OwicPHky06ERsPYvvnTAPcc9ZdwAL//CTyZ7C11vbGyo0+kEwEcNQmKw2WyGIh6sXq+H6tF+v6+dnR21Wi11u92zFB2sQ0rXxeFwqGazGc4DeSB92yeTSSjhj6kpj0LjhlsAGuOME6wUaKGMwcG4U3SQJSrGyRI142wBW1Q2zv97o7FWqxXUN54YZuzcb1fPEOX7deQ1lz62Wi1tbm6Gz3CtOAfUPochcr/jjjv03Oc+d9bDOMvSNNX999+vRx555Ioc71/9q3/1lO0mDpvNxZPvX2hoj29961sBPJ2TjakHB3VPfBLhoyuHSyZqRRLpnyWazOfzWltbCwtNsIwcycrhcKjNzc1A5yCf9Ii5VquFxCrgCPfuWvRarabl5eWMygYJJecDJeVJUacuYommAzdJXVr21mo1ra2tBSDM5XLq9XqBU3f9+XA4DP1laA8MwHLNqtWq6vW6kiQJ98qTwS6TZL+xE/JoyyN1d+yAO8VKjz/+eBiD78tnNU9D435V2q233qpbbrll1sM4y9bX1/VzP3fRZQJP23q9nu67774rdrx5t7lpI+f0Si6XC8CJekTKrs3JlxewjIENgPftaOtLi13Mm3Wtrq4GoGZWwIIZADTJPO9wGHdazOfzoc0v7X1XVlaC5JExetUoMwLvo8J5xRSLO7g4uQmYuuooVqgsLS2F8TFTcXqHa5ckSbhWJF69UEw64PLb7XYAZV4HkF0dxP7jGYgrcXBaDuwUjG1tbanT6Zw1w/FZDM/RYQD3ebUPfehD2trauqLHfPTRR9XpdEJh4mG2uYncY458d3dXjz32WCjrJ2KMOwDGkkgpu5AFoAUXPJnsLaYMQHlvFNYB9aZYgC5dHKlWdZkmv4m4UYiwn2KxqNXV1QDuSPaQGXqilOsB0MK/O0gD6s41S1mO2Tl4iqLy+XxoR0BLAk+cQiVB3fAZ19kTiXvVLAC9uroaHJe3RvCFUVC9kNx2ummaugUnR4K73+/riSeekJTtZc84HORJXi8Afjb2F3/xF+dVLF1q+8//+T/rs5/97BU95rza3ETukjJAnSSJNjY2tLW1pXq9HrTrzsVCAbjFCbVYSgcP7F96KikdHJEs5vP50IrX2xu4ZBPqB5CmO2Wz2QyaemgdPksE32w2Q9GSz0SkA07aAdSvlQMZ58Lr7gykg37f7PvIkSMqlUp68sknA/3hvDvJVvTrgDQrSeEkoGNIvJZKJa2srGTkp0my14a42WyG83EZJK9xzzwRzXhoakY7CF/ExKmkuB7gSoPLlbYkSfSSl7xk1sM4y06ePKmvfe1rMzn2j//4j+tv/+2/PZNjz5PNReQuZTtDSgdc6unTpzPce6yqiH87uPFFp0Ogr9okHSyxt7KyEmR7o9FI7XY78OObm5shYcqCGnGXRrjspaUlHTt2TPV6Xc1mMyzHFwOxJ0DZBz+ulSdalZSJTv1vp0+8m6QnUwF4xs1sBB7+6NGjofUBoOl5BB+35z7OnDkTFDHY7u5uJtHKedMZEuB2+sylqV5jgFNGXTMcDkMlojsOp1+YzcVR/bVq+Xxeb3jDNLn+bO0rX/mKPv/52bTW+dM//VM98MADMzn2PNncgLtPnT3yWl9flzcW82jPaQyP0gA9j3xzuVxYOAIKpVarqdFoBE43n89rMBgER7C9vR0KdqSD3iiAHvr0lZUVXXfddTp69KiazWZQoziwEhn3er2ggZcU9OIO0PF5xefq5xlfQ5wGjoLj4gQAcH68mjZNU1UqFTWbTVUqlcDJOw3ksycUMSyy7RF/qVTK8J6obeiDw37cYtUMzg1gp9UA91VScIKeY/BrwXVd2OGxfr8/E0po3mwuwD1OBjptMhgMdObMmdCRMU7mxck/j2rhsVF+eMdHmmi5SqdQKATq4Mknn8zMGBgT/+fzea2urqper6vRaKharYakoUec/X4/s5Zqt9sN4O4RPX1mvHkavLZLIT3adeD1Y6K9h9bg/Pv9fvgtHfTL4RpNJpOwgEitVtPq6qoqlUpI+HoCudfrhX1vbW1pMBiE5mNcf2YHKJSgueKiNL/3Tos5pUZxFCtiAeLeRZTjxhWvCzt89oEPfECtVmvWw5ipzQW4S1kJm5T9orPq0Wg0Cl96zJNmTsWwPwDE+VtfIs853fF4HFQ6LCxdLBZVr9dDebukQLvgQOgxQ6IVjvnRRx/V+vp6iJhLpVI4Tqwzlw4clYN2XJDk1E2s9/fr6LkLABzQQ8lDLoHt6YY5GAzCMSis8pWe/Bw5HxZAcWqFZLJLH71egRlFfN/dgZPshS7z5KvTQa4g8s6ZzsFfi/aWt7xFN9xww6yHMXfWarWuqAxzHm2uEqoO8HxZC4WCNjc31el0wlJtAIdH8C7/c3AHtLy5l7cYdgXK9va2+v1+oB+QJhL1kwClaIp9eUMkqjNJLjpVAL2Ew5Cyy+jR1pdr4SsIuQqG8+K37yNWn3CdvAWuLwIO0I9Go5BY9mpUScGh5HK5sHwd1BY9aKBPer1eaKOMxLRSqWQoGU+cM0b+9r7u3GtmG5ubm2Em4y0lOGen83AwTuFci3bixInwzCzswCaTib7+9a9PFV0cFpubyH1awtB5dBJyvtCGF914NSLyP+8vMxqNVK/XValUzlpdCNkiqpZcLhfWEx2NRqEFrqtD0Lz76kS9Xk+nT5/W7u6uVlf3FvDxalASuwDUZDIJvLbnEly/7Tx0LOvzSD4GfqJsVwCRI/CkKi2B+/2+Wq2WkiQJDdHg4onMc7lc6Lznswu26fV6oZWwpEwuAwWRR+2xDt3vuSeMR6ORNjayq+ERnceOy58h6KaFHU777d/+7cvesGyebW7AXVLmiyllKYb19fVMxP5UNI5HnZKChh6AIRKnY+TGxoY2NjaUpgdFPEgam81mAHbntgFU1hUlYdvv9wNo+6pP3l/Gz8/pFFY4ApTiylx+e2FP3Afd6Qxe9/wEx3JnSB95Im+knahUoEUATNr6ur4eoB4Oh9ra2tL29nYAW19rltmCR+5Or/h957qxELKfD/SSPzOMwYvY4n0v7PDY7u6ufuZnfmbWw5iZzcV8xaM0N6cYYs6dSBJ1Bl9qAIMvvidDOZbLKlutVgAx6SBqxlnAtRPhk4hN01Tb29uaTCah/J5KVBKmTms4BYHFXR49ove/cSiegMW8tS00BooZXgMsncbw80MiySxEUuDSq9WqarVaqMAlUr7uuuu0vb2dWaEKh4PTIh8RtyTAeXmEHjslL2ajcZl0ULjEPriuUBMuheVeL+zw2iOPPKLBYBBWQztMNjdhjQOU861Yq9UK+vNpa6VKWWcgZfuNAMgALjI+KBpAg17hhUIhROtI95rNpprNpnK5nDY2NjQYDEL1KREv4yHKh+rxxbeRU7LGKWCOubSP9rz8uL7el+Nz4CQyh7Kh0yTnJynDVVMliwHkrHTk8kyOnaZpqEhFxeT3ZTQaqdVqBefi43fnPC2xzL3np9frBbkl27mTxPx9b9+wkEIeXvviF7+oj33sY7MexkxsbsBdUgaU+FJ64pBI0Ze/kxSAFKfgtI3v23lkuk5CdThNgbTR9dqA0HA41KlTp0IZfr1eD8VRAK3TIPD3OJZKpRJa5bKdR++MlfG6E3Nqxh2I0xB83o3rAU/Na05zVCqVEPHjYOnhniRJeN/VLET2yB99ZoH+HEfJmKgE9n44nHtMqUDJdDqdzH30a8RY4+UH2fe1nGy84YYb9La3vW3Ww5hrS9NUDz/88KHMvcwduHvEHvPrp0+fzujVHdjcIcR8fHwMX6tVUliCbzKZaHl5ObTdXVpaUpruVVdubW3pzJkzYa3QXC6nI0eOqFwuZzpOEpECcA7oSCrjBa8pivL2BdJBQtLB1KmnGNT83F1eyW/oKxwoYI8qyVsnIxlliTxJmc6WdMccjUYh8vdZA2NjdSsidm//G+cbuGf8kKh+8sknw/112sbP2T/njiKWzl5LViwWdezYsVkPY+7t3/27fxeS/IfJ5oZzB9SdD3Zgh9P2JfI8UnPtuFMzDoRsR5+abrcr6aDUfmlpKQATTgAayKNv6BSUN5IynRP5oR0uET06cE/8ElF7ZO38tEehriRhW64BzoHzd7CTDuoBODcko1w3WjEwW4FmYRzeQI3PM2MpFAohWc2i207f+LXze+GzMjh/7iHXHJmmz6y4Fs6te7dQ5/CnOfiFLeww2FyAu3R2EZM3lAIoKGSB22Z7V4bEURpg0e/3g7oFGqZWq6nX6ymXy2Xa8KKJHw6HIeEKN53P7y18kSR7xVWoTDyRCT8PZy8dJDbjXiksRE3rA7hvwHuanC8us+c3DhKQpPAolox6krVUKqlWq6nT6ahYLIakcFxPgDMg3wDYo5kvFAqq1+vK5XKZNVaXlpYC/cX94Z5x/7guOBunn8htkJvgWSDBHDtz9u3J8Gs1ch8MBvrmN7+pZz/72bMeysLm0OaGlpkmaXTAT5I9fXm73Q60gHPMcYQLuBAt8j/ADaATsftnAbnt7e3A0wPGRIjtdjuj64aOaDaboWUuwC0p9JQhb+CFR5Iy9IEXSXni0ykc6YDT57o5dUPkzdJ5tG/wSl3URDRE83YN0sFKU/l8PiRZKf+n1wvnwixnaWkpgDxKHEkhMeuKHa5BTLH5+Lvd7tRKW7+nPmPzCJ/n5lq1U6dOHdpk4cUYIojDZuc94yRJbk6S5P9OkuRLSZJ8MUmSd+2/fiRJkj9IkuSr+79X919PkiT5l0mSfC1Jkv+ZJMm3XchA+IJ6Qi6maIimoWbiaTdf9mmNxeiB4mCYz+d13XXXSVKI7qF+aBpGkVGj0QhROslYolxJoSukpBDBEzmfPn36rBWb4oSmq2XgxCVlirbiJDOA5u0IYnWIc9T0uiFvwTV1Xt9zDjgxnALRMhExETS0DOog6gN4n/PgnBkzr8V5AklBW0/xkucXuMacH/d6GoXlz8GUZ+6KPNsLm629973vDd/Nw2QX4s52Jf1vaZr+NUkvlvTOJEn+mqT3SPrDNE1vl/SH+/9L0vdIun3/562S/u2FDMS/4LHEUTqQ4K2vrwdQjWkUSZnpvn+5d3d3Q/k/YMjC13RETJK9IhwaVOFkUMsQ7QNay8vLKpVKoaiH6J3zOHnypDY2NsLxpGwpP+AO3zyZTIJc0B2b79MBnGvliUycokfETuHgIDk+MxOoIvh8Xwu13+9nomSib5qg4QSZXe3u7oZ2wpwfdA37dG6c65ymaQB1HNE0nbqf97SiML/v54ncr8izvbDZWaPR0K233jrrYczEzgvuaZqeTNP0C/t/tyX9uaQbJb1W0kf3N/uopO/f//u1kn413bPPSlpJkuT68xzjLE7WE2Ie8dGsiqZfHu17YjbuYQLvi5qFVr/ovkmmsn+iykajEagNl07CVTcajUBbAIz0OmcFIufr3cbjcZgp+PqqXn0K3eFcvevp3RlKCioWZITw+XFk77z21tZWZjESbwfM+q5eFct1rlQqga7a3t4O+6RZGGobQNs18yRvGbvPJHDYTzzxRIZawcH69fNEOdviuM4nf7sSz/bCZmsvfOEL9b3f+72zHsZM7KKIqCRJbpX0Akn/n6QTaZqy1PgTkk7s/32jpEftY4/tv/aU5hGcqy2cRgGYSIoy5QfMPdKHKwYI4epdLUJUSDSKLptons6PqFzy+XygdlZXVzWZTMLqRCQq4aEBJOfHAUdJmejdk4Pe8tdpJ0808hnPPTht4clQv7acq3Qg2+R9rg+tBzifQqGglZWVoFunIKtQKASNO+Not9uhaIprjfND2uicvlM1fs44VyJ9fya4Fv5MuFMgmXquZPS57HI+2wubnf2zf/bPZj2EmdkFg3uSJA1JD0r6sTRNt/29dA85LkqSkCTJW5Mk+XySJJ/3rorSgeLBIzPnqdfX10MPl/19HZzQPkiyqDXRuWugy+VyAGnkgyRP6ZniYEtLX5yLKzYAfWSBRMWSAr1BtO1LwwGQaMypNsXxsL6qtwfg2L6ikrc2AGQBcpdKOoh6ToLFvhmj1xn4eOHhy+VykJBC70gHuQHPh5Dw5RrgoHxGFuvW+dtbDvBM8BycqxqX/52Sc4f/FM/iZXu2L+Zz15J96UtfmvUQ9LrXvU533XXXrIcxM7sgcE+SpKi9h//X0zT9xP7Lp5iS7v8+vf/6tyTdbB+/af+1jKVp+pE0Te9K0/Qu7/dNotO283FIUojcnaJwbbdvC49Oa9pKpRJokG63q8lkok6nE3j8crmsG264QY1GI0SxaNp9NsH+UdbQPtijZMDaI3b6reTz+RDVJkkS5JXsAwD2YznweksCXxoP8PQ6AK6LyyQ98t/e3g7nPh6Pz9Ljsz8cEs6M8ySZnMvlQsKZ+4kD9J420kEzt5ia4X6hLOIZ8HPg72l0Tfy8nE/nfrmf7ac8+DVok8lEjz32mD784Q/PeihaXl4+lD1lsAtRyySS/r2kP0/T9P+wtz4t6d79v++V9Cl7/U37yoIXS9qyKe5THeesL6gnEz1yQ4boPKuDukeETn34ohNE5adOnVK/3w8Rb61WC8k8kqWSgnMAqAAc6AyOC2+N8/EFQSqViqrValhf1TlpZhT877QT14TzcPWMc8sAOtcReoP2xfDq3uK40+lod3c3OAmAH2dHToLZQrVaVaPRyETlS0tLYbbBtad5Gvth9gMwx0onZh/kGuglw/13SsbvazzD4xqcJ5HKs3JFnu2r3eJc1lP9fPzjH9dznvOcmS2OjRWLRb3qVa+a6RhmbRdSxPQSST8o6U+TJPnj/dfeJ+mDkh5IkuTNkh6RdM/+e78r6TWSviapJ+mHL2QgHtW5ThkQkw5Anyk7Ko6YX/YeKEmShCKker0eGnyVy+UAdKg7qMCErwesyuXyWdJMpJAkB1G5MI5arRbeS5JEjUZDKysrKpfLobeMpOBwOIb3fx8OhyGa9miV6+B8sl8np1f8c7wOFYUaxWcXUCieIJYOesLs7Oyo2WxmlDXSniqBqlKP6lEkcX1xJFApnjdxp0y5uPPnnJ9Xzfqzg8Nxisa7Zk6xK/JsX622vb2tP/qjP9If/dEf6dd//dcv6DNQprO2fD5/aFUy2HnBPU3T/1fSucKg75yyfSrpnRc7EL6c0xQyUnYhBlZnIrqFK/cImhJ6wIfIEwrCgcRB0hNxUDD0NqcfeZqmwbGUSiUtLy+rXq+HiJ4WwL4wRqVSCVp59p0kSYZPdwWMdEAr4DC4PjG4A2Bw954oBWx9X07XAIYe8XNsZJCoffr9fojwl5aWMgldxoQDIM/hskqcJclZxuiR4neKAAAYK0lEQVTJYhzLmTNnMvecffMZf2Y8oeqBAOd4rij+Sj3bV5Pt7OzoG9/4hn7qp35KJ0+e1Gc+85lZD+lp2WAw0Ic//GHdddddmY6nh8nmpv2AfxH5m9fjL26SJIFOIboFwLya0YtkAGa2Q5+dy+XCCkvuLGj0Fa+zSmKUJfcAbAp7vDEY4A0g+kPGeRLlOpc/jZ7yc4+vW5w8pEjIo21X1XifeDh0qkeZ5QD2KIKgXQBTnKlP2X0//NDymCjau076mCWF4yM59VmKyxw5Z08yc104L68NuJCk6mG3yWSiP/mTP9H999+v+++/P0MJXq324IMP6t3vfrf+5t/8m7MeykxsrsDdQf1cIMaXmcRdr9fLrPAD+MZRWy6XC8qXM2fOBHoA6gQtO7MAmoIBUqhW1tbWVKlUQrk+BVDe2dHb/vI5V7FAFen/b+9qY+Oq0vNzxrE9Y8eOHX8Ruwm4sBWwIgIUtis16t9W/NmK5MfyZzcq0vbHIrGlQqJdCYoQEq1aIJWqhI8EUqnt/gCkrhAJ2TRINCQb1URpNm2UXWjzQeQ4eBzbM7bHH5nTHzPP8XOPx7Hdxr7jO+eRLM/HvTPnnHvned/zvO95D+a3ziNZkeT5p6l9ABwZ831fl+cY0Vtubm5GfX29y0NnMFo1e5I2M4w0u4jPp6amnITEIHZTU1MkR5/ppUq209PTyGQykb5zgZjGEaiXFwoFjI2NRQLUs7OzkYVm7Ktq7zSAqs3rSt+AxXHy5Ens378fH3/8MbLZbNzNuWOw1uKZZ57B+++/j7vvvjvu5qw5qobc6dmpzq66uerdQKnuyY0bN9DV1eUWNdErpHdHnVk3hFatnR69EhazWEgqdXV1aG9vd5F35nkzi4bSjNZkJ8EwCEv5gIQFIEKuhUIBmUwGQLRgGElPPXodCxbvAuaDqT7ZMwiaTqddoJILikjEQHQPW/V6gWiZARKpVmrcuHEjstmsk8FUWtPZiRIwyVmzeHhdmJbKTActFKb3gPZTyxAA0Wqgywmu1hpUfjl69Chu3rwZd5NWBQMDA9i9ezcOHTqE++67r6Ykmqoid32sWjN/pL78QBJgjrmSn2ZSMPBnynndOp2nDMNjGOybnZ1FU1MT2tvbXTYIy9qqTq7nAHDFvdSzJMHRS1b5SYPELS0tETlCF+vwOSUXXX3Lx7owSlMLqfmT6IH5omCUm3K5HBoaGiJVIYH57BTONHK5HBobG10BMRq1TCYTqXIJRMsq6PXj9/oZLpRlGExVz5yzAt34Q+8Z7bcGtjWOEFAa03PnzuHNN9/E22+/vexFXusZAwMD2L59O/bs2YPe3l4AQF9fH5566ikA0ay8JKEqyN3XRX3Pj8cojDEYHh5GLpdDS0uLCySqJ6o6MwBXVre+vt7p6fSoi8XSakqgRAgdHR3o6OhwhKkbUlA7prbP/0xn1O+j4VCphAFM7Yt65LrQR40UP0Nz+yk96AIhf5UmjR0NA1CSa/i53CeV6Yt+vXwNdt66dQu5XM6lO3IVK8eTpRloEHhdxsfH0dnZ6TKM/NRHGtl8Po8bN264MdX7g/3Q8aLBUyPHMebYBs29hHw+j6effhofffRRouSX5eDWrVs4cOCAe97Q0IBXXnkFANDf349nn30W/f392L59e1xNvOOoCnL3pQp9nT9g9dZIOpOTkxgdHXV7eZJUGDgl2dLrtda6mum61R4A5+11d3e7HHRm3DQ0NCzYqYjEQxKiV8/PYTvpSfI5EN3bld9NItbjeaxKM+qRqm6t3rafG66eLI2QeiuUqOrq6jAxMeFiENpezSzKZDLI5XKuYiRnJM3NzS5IzXFpampyBA7AxUZ4jUjsTG1lcTclZTVOBK8p4yuaDsn7RcexljEyMoLjx49j7969OHHiRNzNqQrMzMzg6tVSJYmrV6/is88+w7Zt2/DAAw9gz549eOyxx7Bt27Z1vU1jVZA7EE2F9DMdlOD4OlCyxkNDQ+jp6UFra6sjH5KTEjsLZNXV1WFqagptbW3IZDJuEQ899ba2NlcKgJ9BTV2nbiQNkhr1aNX6ucjHl5hUf+Zn0ZtlYFa1Yl3so+do4NIfH9X+9TzVpnl+fX29S5kE4DJcSOiMZeguTIxXjI+PY2Jiwl0TTcXUmY4xBvl8Hh0dHc5j97N4pqamXIlfkrQaVJ3d6H2jfdN0UR23WoS1FtevX8fu3btx6tSpMINZAleuXMGVK1dw9OhRGGPw8ssv47nnnlu3BF8VgqROuVWG0dQ+lQd0Wj4yMuK8PdVudbUm88AZBGQmzNTUFDKZDDZv3oy77roLnZ2drp4LA6fU00nUTGmkV0otW71bat1K2vT02R4SpKYJ6rHMiuFiKq0nw+/Xtug46uwBmCdF3ZqPrzNVku1i4JJ91YVBnPEwDTSdTqOnpwddXV3OiOqGIrrLVCqVcpIYUyT1ms7MzGByctLtuerfBxzXSkZS+67FyGpdkjly5AgeeeQRnDx5sqbHYaXg/fPSSy/hoYcewhdffBF3k/5PqApy97Nh/B8tEN3AQs8ZHx9HLpdzOw2RZP3FNboRBz1rTuv7+/vR3d3t0gZJ7vTg/XrsqVQK6XR6AbHyv9ZkVyJXOUR/bPSG6SFrILRSNotvtHSsVJLR9Ep/az6Sqe4Ty8wizXdX40bDw8Asyyd0dnait7c3kh/PVbjsGwuUkaDpudOLZzvGxsYW3BvaHz8zRmMTfraSSlW1hqGhIbz66qsYGhqKuynrFjMzM7h48SJ27dq1Lgm+KshdCZw/YD+Q6gddgfkf99WrV11wT7V1Pk+lUq6KJAmUpHHPPfe4sgD01HXBjhIvAKfj6xJ6nf6rZ64kr6RD2YRlc2kkKGPQo9V+qpSjZK7k5mvMep4uEKJ3qxtikOCLxSLS6bTTsjOZTGShE/vV0NDgFnBxxWpfX58zKCR6jjcwX+uGwVbOEij5jI2NuT1tCRo1XleV71SHZz9pBHjda9F7z2azeOKJJ9bt6tJqw+XLl7Fr1y6cOXPmtsdV28KvqtLcNTsEiGrNqr3rD7aurrQ7UzabdTKJpsupFk1iJLlt27YN3d3dC4iaQT9d5ETDwEVKKiUpkfgESwJS/ZxETAIlubNPfN9foOOPCcdJ0yM1UMrzdHMP9a55LjVyfhfLDdPY+nn31CD5HtMiW1pasHXrVgwPD0dkIc6WeL20LATbNT09jcHBwci48JrrbEQ9dA2i+mOrq3CTjAsXLriFfMTx48dx6tSpGFuVPFy+fBlPPvkkzp0751ajK4rFIl544YUYWrY4qsJzJ3QKruSsGip/4Dye5HX9+nU39ae0oUE13fiB6Xo9PT3OGGgancow9NhVg6d3rkSpBMvjff2c7dWVqHysWS/st5+KqCtLgfkcdD5WfVwJUPV71d5VUydJap66nqf90D7q9zQ3N6OjowPt7e3uOBpKfjavn+bRT09PY2xsDPl8PmIY/fTHxTzxxdJlK832koYPPvgA33zzjXuez+exd+/exPc7Dly6dAkHDx6s+F42m8Xw8PAat+j2qBpy9xebqAzhp8QpeMy1a9cwMTHhtHcWvSJIKDMzM6ivr0dfX5+TQdQDpMeukooxxskP2k7KCiRAHs/j/Kwf1eHVILD/JDD2S8dCJSf1zNkOgu2m3KOauf7gfVJXeUOlHgZQNfagQV2NUQBAJpNBe3u7K/GrAXDm1ltrXakHVvjMZrNO59frqsbMN9iaFeMbfJ3ZJR26z+yRI0fw+eefx9ia5GJmZgYHDx7E4ODCKs/79u3D119/HUOrFkdVkLumK/oZMyQD9eAIJaVCoYCRkRG3TR49RK3Tzs/YvHmzKzGg3q8fxCRRqMfqe4Oqp5OsK3mXbDcNh3qi7KNmxKhsQs9eA6Mq0/hePz1kzjQIGhedDam3n0qlHMGSuNl3BpA5HnzOEsZc+WpMKUWyo6PDtYNt4vlacIxxAG7M4Y+dPveNvN4favQZsPXLMSQRc3NzTg6Ym5vDiy++GHOLko2BgQH8+MfRwqCDg4N44403YmrR4qgKcgei+2H6ecqL/eD9BT2UZjQlUhfv0ANvaWmJeKE+yROaaqhkS6jH7nuImoaoqYQq1xSLRRfgVKlDc905k+BuUFr6wO+D/8c+LGYQ1DBpEFjz37X9mr7J89kGrUnP3aaamppceqXuNMUAKvufy+XcqlQ/a0rvD398/eM04Mo+qDOQVLBcw3vvvYevvvoq5tYkH59++mkkWP36669jdHQ0xhZVRtWQu++5a3aEeuyL/XhTqZSreEhS1IyKdDrtAoWUGRhMZIkAQglMV6Zqjji/k0SrmRsqr/A4lWrYNnqXwPzG0MwB1x2dtOwwiVHJVWUlkqdKWJq7r0aHhM02kZRpTPxVuf7xvgSlQWF68Fy9ysqQ7I/2LZ/PR+QbjrNv6H0S11mISje8Fv5ML6k4f/48Tpw4gQMHDrgV1wGrh9HRUezfvx+3bt3CxYsX8eGHH1ZljKOqsmX43y+YBWDBtNsndmuty3lnkStKACQi1o5Rb93/HiU/GgAN5Ko2r3KMfz7hyzQkOK72ZPs1eMrXNH8bQIQAWbGS5/geu2bEAHAZP6rP8jv4/fx8rsrVcecYpNPpiCbPPuqaAlaQZOE1FmvbsGGDi4NQMmMKpF77Sro7s2x8b53tUplJA8K1kOd+6dIlvPPOOzh9+nTcTakZHD58GGfOnMEnn3xStbOlqiB3n5gIPxVSidcHvXBq7iQPesJcXakbTlCKoZer3qEGO3XWoFkumoLH45R8Kmncmr6on6eBQZ/IlOT1T1fh6vj4Xq3q9SzNQKOgskuxWHQePo2SGhEALpvGr9uum3RwMVRjYyM2bdrkzmftGd0YZGJiAqOjo5ibm3OGw7/mGhzVflW6dzTewse1gEOHDsXdhJrC6Ogo9u3bh2PHjsXdlEVRFeQOzGez+NkPSviqsfvZH5QIcrncAqLkVnGNjY3Oe+eyePXs+X2pVCpSF6Wurs7VYqcWzeOUVNkO9e799qvH6cs/mjKpe4v6OrcaERoxtssnNebnq7buS10amKamz7HgGKkMpq/xOO03P59ZO9oXGluVZbhJiG/AaYBI1v4Y00DrrEHHimMTELAaePfdd+Nuwm1RNeROotJptHrVPIav+8eQFG/evIl8Po+2tja3QYUxBul0GrOzs8hkMk7rJkEwXVA9RNXTSewkPd3FifClGfVqNdDnk7sSNTCfV095odIiKJ050PBoWqYGTElwzHyhJ+4XOuM4aDyDEo4aMdXfVQtXiYvnMkOGaZHsN3duKhQKTkrTgHQl/VINuy9T+R69H3OoRj00IGC1URVzVnpoukJVpQz98717DV5yuj85OYmpqSnMzc25xUvUkTUjRL1PYJ44VI7xCaJS1okG9TRzRqUf9XyV/NVAaC65asfUyrUNfqaObyg0+4b9YN+YjqmZQMzEodHyjRD7ozMiXyLRMdN+sxwyvWjNU6cG7wd71eCx7X6AVT9L5Sdtu58BFRBQK1iS3I0xW40xnxpj/ssY85/GmGfKr/+lMeaaMeZs+e9xOefPjTFfGmMuGmP+YKnvUHLnc5KZrzvzfX8aTjBnmp4hvcSxsTFHeppVokFRICpR+NUllSRJXn42hx8EVkPFz9d+68yDnjWNgp7ja8zad7bLf0+DwTozYB81R52evW+s+FlKru7mSaUWXDt+9uzsrAtIs/Qys4C4krhQKCCbzS5YMKbGQsdZX9M28Hpw3Dgzo3FZzHNfi3s7ICAuLEeWmQPwZ9baM8aYFgBfGGN+UX7vdWvt3+jBxpgHAXwfwLcB9AI4Zoz5HWvtomkLJAglNCBaO0XT2/gagAWea7FYRD6fd9vG6UbXzABR7Vy9bJKlEjYzP/jcJ2d+ngZpSVD6mMf7BKyvsZ0kQ/VYSZo0GFonnp6+Hu8HE9UA+K/72UMcex6r56pBBLDAwPE6aookDSz3uiUB0/hyzPWaq0TGz2fZCB0HP+PHl780O6gCVv3eDgiIC0uSu7V2EMBg+XHOGHMBQN9tTvkegJ9Za6cB/I8x5ksA3wFw20pGSt5KhuphS5sicgZ1ZJLJ5OQkCoUCJicn0djYiFwu51akFgoFRzy6gbUaB/3zpRO2FYhKFrqBs39MJVKvRML0NpnK6WfvUNbgd5IoK8kjqpMr6VdqD8dUx5vQxyqtqCH2ZTUaTl63ubk5V46ZlR8nJibcTk4aTNV2Mditffa1fTU4Wu2T5/v9VKzVvR0QEAdWFFA1xtwD4BEApwH8HoCnjTE/ADCAkgd0E6Ufxy/ltK9x+x8M6urqsGnTpojOrFNtYGHetmbMqCfJACO9dv5pgJR6sxsE0Z1VtlB9XL9DSd1/jWSqmrRmlRDqmapG7vfLz/bQz9C0SW4xqDEJ9fYrtZXf4QdW5XpXlHnUcJHE1SA2Nja6WQjbxHrxs7OzmJiYcLMrAG6rPW0TiV3r6fiaP8HMJ50tFYvFiPFeCqt1bwcExIVlk7sxZiOADwD8xFo7bozZB+BlALb8/28B/PEKPu9HAH4EAOl0Gvfee+8CKYI/Vg2eaX44sHAj5FSqVH2wtbUV3d3daG1txcaNG9Ha2oqWlhZH7MVi0ckyGmT0iVvJWUlRj1uKwAmSkp8Saa11xsZaG6k576+A1dozjY2Nzpu21kZSIX3DoobH95B9aYpjrs9VtlLpQzcWYVE2Xx5hzjuvG8d9amoKzc3NkY3NOabpdDoyzmo0NfWT5/CxBtf5uj/zqnAvrtq9HRAQF5ZF7saYepRu/n+01n4IANbaIXn/bQAflZ9eA7BVTv+t8msRWGvfAvAWAGzYsMGeP38eXV1d7v3+/n50dXUtWPnI5dUkJ+Zlq9RAArHWuuXtxhhMTk46IleP35j5ZfT0UCsFDst9jchC+p1qnHis6tV6Tnd3N3p7eyOGhOTNY5jK6ctAuurVmPmURZVWZJwjBCnXLNJeNU70trW9amDVyPFYyi7M7NGAJuu95/P5SDbRli1b3PaIfhYMv5uGjkZRx5ZZTTyOhhGYT2etr6/H2bNn/dtPx2FV721jTMjDDIgFZqkpqyn9mg4BGLHW/kRe31LWLGGM+VMAv2ut/b4x5tsA/gklLbIXwL8C+Nbtgk61+AO4//77sWPHDmcYSFyUEioRK4AFBM1jfE3fJ3kfKrmoAdCYQSUjAUTlHQ12ajva2trw8MMPO7JnqYGdO3fC2lKRt+HhYVd+gJ+hRoXGTjX1SqmrPN6XuGhIXnvtNVy5cmXBYIR7OyAJsNZW/KEvh9x3Avg3AL8CQJb5CwBPAngYpanrJQB/Ij+In6I0jZ1Daap7eInvyAG4uMy+JAGdAKqrsv/qoRr6ere1tst/Mdzbq4JquN5rhWroa8V7G1gGua8FjDED1todcbdjrVBL/a2lvlZCrfW/lvpb7X2tihWqAQEBAQF3FoHcAwICAhKIaiH3t+JuwBqjlvpbS32thFrrfy31t6r7WhWae0BAQEDAnUW1eO4BAQEBAXcQsZO7MeYPTanC3pfGmOfjbs+dgDHmoDHmhjHmvLy22RjzC2PMb8r/28uvG2PM35X7f84Y82h8LV85zOKVFRPZ35Ugafd2uK/XWX914cla/wGoA/AVgN8G0ADgPwA8GGeb7lC/fh/AowDOy2t/DeD58uPnAfxV+fHjAA4DMAC+C+B03O1fYV+3AHi0/LgFwK8BPJjU/q5gXBJ3b4f7en3d13F77t8B8KW19r+ttTMAfoZS5b11DWvtZwBGvJe/h9JqSJT//5G8/g+2hF8CaDPGbFmblv7/Ya0dtNaeKT/OAWBlxUT2dwVI3L0d7uv1dV/HTe59AK7K8yRX2eux5VWOAK4D6Ck/TswYmGhlxcT3dwnUSj8Tf53X630dN7nXJGxpHpeoNCXjVVbU95LY34CFSOJ1Xs/3ddzkvqwqewnBEKdp5f83yq+v+zEwFSorIsH9XSZqpZ+Jvc7r/b6Om9z/HcC3jDH9xpgGlLYw+3nMbVot/BzAD8uPfwjgX+T1H5Sj7d8FMCbTvqpHubLiAQAXrLWvyVuJ7O8KUCv3diKvcyLu67gjuihFmX+NUmbBT+Nuzx3q0z+jtH3bLEra21MAOlAqEfsbAMcAbC4fawD8fbn/vwKwI+72r7CvO1Gamp4DcLb893hS+7vCsUnUvR3u6/V1X4cVqgEBAQEJRNyyTEBAQEDAKiCQe0BAQEACEcg9ICAgIIEI5B4QEBCQQARyDwgICEggArkHBAQEJBCB3AMCAgISiEDuAQEBAQnE/wLyPTYN+t3gAQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rmbvBqVeWh18"
      },
      "source": [
        "## M-Net 모델 정의"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oLVmgCbuI2Vy"
      },
      "source": [
        "# from tensorflow.keras import Model\n",
        "# from tensorflow.keras.layers import Input, Conv2D, BatchNormalization, MaxPooling2D, Conv2DTranspose, concatenate\n",
        "# from tensorflow.keras import backend as keras\n",
        "# from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "# from tensorflow.keras.callbacks import ModelCheckpoint, LearningRateScheduler\n",
        "\n",
        "# def unet(input_size=(256,256,3)):\n",
        "#     inputs = Input(input_size)\n",
        "    \n",
        "#     conv1 = Conv2D(32, (3, 3), activation='relu', padding='same')(inputs)\n",
        "#     conv1 = BatchNormalization()(conv1)\n",
        "#     conv1 = Conv2D(32, (3, 3), activation='relu', padding='same')(conv1)\n",
        "#     conv1 = BatchNormalization()(conv1)\n",
        "#     pool1 = MaxPooling2D(pool_size=(2, 2))(conv1)\n",
        "\n",
        "#     conv2 = Conv2D(64, (3, 3), activation='relu', padding='same')(pool1)\n",
        "#     conv2 = BatchNormalization()(conv2)\n",
        "#     conv2 = Conv2D(64, (3, 3), activation='relu', padding='same')(conv2)\n",
        "#     conv2 = BatchNormalization()(conv2)\n",
        "#     pool2 = MaxPooling2D(pool_size=(2, 2))(conv2)\n",
        "\n",
        "#     conv3 = Conv2D(128, (3, 3), activation='relu', padding='same')(pool2)\n",
        "#     conv3 = BatchNormalization()(conv3)\n",
        "#     conv3 = Conv2D(128, (3, 3), activation='relu', padding='same')(conv3)\n",
        "#     conv3 = BatchNormalization()(conv3)\n",
        "#     pool3 = MaxPooling2D(pool_size=(2, 2))(conv3)\n",
        "\n",
        "#     conv4 = Conv2D(256, (3, 3), activation='relu', padding='same')(pool3)\n",
        "#     conv4 = BatchNormalization()(conv4)\n",
        "#     conv4 = Conv2D(256, (3, 3), activation='relu', padding='same')(conv4)\n",
        "#     conv4 = BatchNormalization()(conv4)\n",
        "#     pool4 = MaxPooling2D(pool_size=(2, 2))(conv4)\n",
        "\n",
        "#     conv5 = Conv2D(512, (3, 3), activation='relu', padding='same')(pool4)\n",
        "#     conv5 = BatchNormalization()(conv5)\n",
        "#     conv5 = Conv2D(512, (3, 3), activation='relu', padding='same')(conv5)\n",
        "#     conv5 = BatchNormalization()(conv5)\n",
        "    \n",
        "#     up = Conv2DTranspose(256, (2, 2), strides=(2, 2), padding='same')(conv5)\n",
        "#     up6 = concatenate([up, conv4], axis=3)\n",
        "#     conv6 = Conv2D(256, (3, 3), activation='relu', padding='same')(up6)\n",
        "#     conv6 = BatchNormalization()(conv6)\n",
        "#     conv6 = Conv2D(256, (3, 3), activation='relu', padding='same')(conv6)\n",
        "#     conv6 = BatchNormalization()(conv6)\n",
        "    \n",
        "#     up = Conv2DTranspose(128, (2, 2), strides=(2, 2), padding='same')(conv6)\n",
        "#     up7 = concatenate([up, conv3], axis=3)\n",
        "#     conv7 = Conv2D(128, (3, 3), activation='relu', padding='same')(up7)\n",
        "#     conv7 = BatchNormalization()(conv7)\n",
        "#     conv7 = Conv2D(128, (3, 3), activation='relu', padding='same')(conv7)\n",
        "#     conv7 = BatchNormalization()(conv7)\n",
        "    \n",
        "#     up = Conv2DTranspose(64, (2, 2), strides=(2, 2), padding='same')(conv7)\n",
        "#     up8 = concatenate([up, conv2], axis=3)\n",
        "#     conv8 = Conv2D(64, (3, 3), activation='relu', padding='same')(up8)\n",
        "#     conv8 = BatchNormalization()(conv8)\n",
        "#     conv8 = Conv2D(64, (3, 3), activation='relu', padding='same')(conv8)\n",
        "#     conv8 = BatchNormalization()(conv8)\n",
        "    \n",
        "#     up = Conv2DTranspose(32, (2, 2), strides=(2, 2), padding='same')(conv8)\n",
        "#     up9 = concatenate([up, conv1], axis=3)\n",
        "#     conv9 = Conv2D(32, (3, 3), activation='relu', padding='same')(up9)\n",
        "#     conv9 = BatchNormalization()(conv9)\n",
        "#     conv9 = Conv2D(32, (3, 3), activation='relu', padding='same')(conv9)\n",
        "#     conv9 = BatchNormalization()(conv9)\n",
        "\n",
        "#     conv10 = Conv2D(1, (1, 1), activation='sigmoid')(conv9)\n",
        "\n",
        "#     return Model(inputs=[inputs], outputs=[conv10])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mjhRLU5tcW09"
      },
      "source": [
        "copy from https://github.com/adigasu/FDPMNet/blob/master/test.py"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UT6fo6ZgcfxT"
      },
      "source": [
        "from tensorflow.keras import Model\n",
        "from tensorflow.keras.layers import Activation, Input, Conv2D, BatchNormalization, MaxPooling2D, Conv2DTranspose, concatenate, Dropout, UpSampling2D\n",
        "from tensorflow.keras import backend as keras\n",
        "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
        "from tensorflow.keras.callbacks import ModelCheckpoint, LearningRateScheduler\n",
        "from tensorflow.keras.optimizers import SGD\n",
        "from tensorflow.keras import backend as K\n",
        "\n",
        "\n",
        "def getFPDMNet(patchHeight, patchWidth, ipCh, outCh):\n",
        "\n",
        "    # Input\n",
        "    input1 = Input((patchHeight, patchWidth, ipCh))\n",
        "\n",
        "    # Encoder\n",
        "    conv1 = Conv2D(16, (3, 3), padding='same')(input1)\n",
        "    conv1 = BatchNormalization()(conv1)\n",
        "    conv1 = Activation('relu')(conv1)\n",
        "    conv1 = Dropout(0.2)(conv1)\n",
        "\n",
        "    conv1 = concatenate([input1, conv1], axis=-1)\n",
        "    conv1 = Conv2D(16, (3, 3), padding='same')(conv1)\n",
        "    conv1 = BatchNormalization()(conv1)\n",
        "    conv1 = Activation('relu')(conv1)\n",
        "    pool1 = MaxPooling2D(pool_size=(2, 2))(conv1)\n",
        "\n",
        "    #\n",
        "    input2 = MaxPooling2D(pool_size=(2, 2))(input1)\n",
        "    conv21 = concatenate([input2, pool1], axis=-1)\n",
        "\n",
        "    conv2 = Conv2D(32, (3, 3), padding='same')(conv21)\n",
        "    conv2 = BatchNormalization()(conv2)\n",
        "    conv2 = Activation('relu')(conv2)\n",
        "    conv2 = Dropout(0.2)(conv2)\n",
        "\n",
        "    conv2 = concatenate([conv21, conv2], axis=-1)\n",
        "    conv2 = Conv2D(32, (3, 3), padding='same')(conv2)\n",
        "    conv2 = BatchNormalization()(conv2)\n",
        "    conv2 = Activation('relu')(conv2)\n",
        "    pool2 = MaxPooling2D(pool_size=(2, 2))(conv2)\n",
        "\n",
        "    #\n",
        "    input3 = MaxPooling2D(pool_size=(2, 2))(input2)\n",
        "    conv31 = concatenate([input3, pool2], axis=-1)\n",
        "\n",
        "    conv3 = Conv2D(64, (3, 3), padding='same')(conv31)\n",
        "    conv3 = BatchNormalization()(conv3)\n",
        "    conv3 = Activation('relu')(conv3)\n",
        "    conv3 = Dropout(0.2)(conv3)\n",
        "    \n",
        "    conv3 = concatenate([conv31, conv3], axis=-1)\n",
        "    conv3 = Conv2D(64, (3, 3), padding='same')(conv3)\n",
        "    conv3 = BatchNormalization()(conv3)\n",
        "    conv3 = Activation('relu')(conv3)\n",
        "    pool3 = MaxPooling2D(pool_size=(2, 2))(conv3)\n",
        "\n",
        "    #\n",
        "    input4 = MaxPooling2D(pool_size=(2, 2))(input3)\n",
        "    conv41 = concatenate([input4, pool3], axis=-1)\n",
        "\n",
        "    conv4 = Conv2D(128, (3, 3), padding='same')(conv41)\n",
        "    conv4 = BatchNormalization()(conv4)\n",
        "    conv4 = Activation('relu')(conv4)\n",
        "    conv4 = Dropout(0.2)(conv4)\n",
        "    \n",
        "    conv4 = concatenate([conv41, conv4], axis=-1)\n",
        "    conv4 = Conv2D(128, (3, 3), padding='same')(conv4)\n",
        "    conv4 = BatchNormalization()(conv4)\n",
        "    conv4 = Activation('relu')(conv4)\n",
        "    conv4 = Dropout(0.2)(conv4)\n",
        "\n",
        "    conv4 = Conv2D(128, (3, 3), padding='same')(conv4)\n",
        "    conv4 = BatchNormalization()(conv4)\n",
        "    conv4 = Activation('relu')(conv4)\n",
        "\n",
        "    # Decoder\n",
        "    conv5 = UpSampling2D(size=(2, 2))(conv4)\n",
        "    conv51 = concatenate([conv3, conv5], axis=-1)\n",
        "\n",
        "    conv5 = Conv2D(64, (3, 3), padding='same')(conv51)\n",
        "    conv5 = BatchNormalization()(conv5)\n",
        "    conv5 = Activation('relu')(conv5)\n",
        "    conv5 = Dropout(0.2)(conv5)\n",
        "    \n",
        "    conv5 = concatenate([conv51, conv5], axis=-1)\n",
        "    conv5 = Conv2D(64, (3, 3), padding='same')(conv5)\n",
        "    conv5 = BatchNormalization()(conv5)\n",
        "    conv5 = Activation('relu')(conv5)\n",
        "\n",
        "    #\n",
        "    conv6 = UpSampling2D(size=(2, 2))(conv5)\n",
        "    conv61 = concatenate([conv2, conv6], axis=-1)\n",
        "\n",
        "    conv6 = Conv2D(32, (3, 3), padding='same')(conv61)\n",
        "    conv6 = BatchNormalization()(conv6)\n",
        "    conv6 = Activation('relu')(conv6)\n",
        "    conv6 = Dropout(0.2)(conv6)\n",
        "    \n",
        "    conv6 = concatenate([conv61, conv6], axis=-1)\n",
        "    conv6 = Conv2D(32, (3, 3), padding='same')(conv6)\n",
        "    conv6 = BatchNormalization()(conv6)\n",
        "    conv6 = Activation('relu')(conv6)\n",
        "\n",
        "    #\n",
        "    conv7 = UpSampling2D(size=(2, 2))(conv6)\n",
        "    conv71 = concatenate([conv1, conv7], axis=-1)\n",
        "\n",
        "    conv7 = Conv2D(16, (3, 3), padding='same')(conv71)\n",
        "    conv7 = BatchNormalization()(conv7)\n",
        "    conv7 = Activation('relu')(conv7)\n",
        "    conv7 = Dropout(0.2)(conv7)\n",
        "    \n",
        "    conv7 = concatenate([conv71, conv7], axis=-1)\n",
        "    conv7 = Conv2D(16, (3, 3), padding='same')(conv7)\n",
        "    conv7 = BatchNormalization()(conv7)\n",
        "    conv7 = Activation('relu')(conv7)\n",
        "\n",
        "    # Final\n",
        "    conv81 = UpSampling2D(size=(8, 8))(conv4)\n",
        "    conv82 = UpSampling2D(size=(4, 4))(conv5)\n",
        "    conv83 = UpSampling2D(size=(2, 2))(conv6)\n",
        "    conv8 = concatenate([conv81, conv82, conv83, conv7], axis=-1)\n",
        "    conv8 = Conv2D(outCh, (1, 1), activation='sigmoid')(conv8)\n",
        "\n",
        "    ############\n",
        "    model = Model(inputs=input1, outputs=conv8)\n",
        "\n",
        "    # sgd = SGD(learning_rate=lrate, decay=decay_Rate, momentum=0.75, nesterov=True)\n",
        "    # model.compile(optimizer=sgd, loss=my_loss)\n",
        "\n",
        "    return model"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "raJk70zBWm7M"
      },
      "source": [
        "## 모델 컴파일과 학습"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gGlqId9BJkxm",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "11027e38-8596-451d-ba9b-4ac82cb18671"
      },
      "source": [
        "# 모델 생성\n",
        "# model = unet()\n",
        "model = getFPDMNet(patchHeight=256, patchWidth=256, ipCh=3, outCh=1)\n",
        "\n",
        "# 모델 컴파일\n",
        "# model.compile(loss=\"mse\", optimizer=\"SGD\", metrics=['accuracy'])\n",
        "\n",
        "lrate = 0.1\n",
        "decay_Rate = 1e-6\n",
        "\n",
        "def my_loss(y_true, y_pred):\n",
        "    l1_loss = K.mean(K.abs(y_pred - y_true))\n",
        "    return l1_loss\n",
        "sgd = SGD(learning_rate=lrate, decay=decay_Rate, momentum=0.75, nesterov=True)\n",
        "model.compile(optimizer=sgd, loss=my_loss, metrics=['accuracy'])\n",
        "\n",
        "\n",
        "# 학습\n",
        "hist = model.fit(train_x, train_y, validation_split=0.2, epochs=20, batch_size=8, verbose=1)\n"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/optimizer_v2/optimizer_v2.py:375: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
            "  \"The `lr` argument is deprecated, use `learning_rate` instead.\")\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/20\n",
            "46/46 [==============================] - 24s 131ms/step - loss: 0.0927 - accuracy: 0.9207 - val_loss: 0.4093 - val_accuracy: 0.6372\n",
            "Epoch 2/20\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0468 - accuracy: 0.9574 - val_loss: 0.4630 - val_accuracy: 0.6228\n",
            "Epoch 3/20\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0388 - accuracy: 0.9641 - val_loss: 0.3873 - val_accuracy: 0.6176\n",
            "Epoch 4/20\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0398 - accuracy: 0.9628 - val_loss: 0.3539 - val_accuracy: 0.6444\n",
            "Epoch 5/20\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0327 - accuracy: 0.9691 - val_loss: 0.2066 - val_accuracy: 0.7973\n",
            "Epoch 6/20\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0309 - accuracy: 0.9706 - val_loss: 0.4373 - val_accuracy: 0.5604\n",
            "Epoch 7/20\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0285 - accuracy: 0.9728 - val_loss: 0.1693 - val_accuracy: 0.8324\n",
            "Epoch 8/20\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0273 - accuracy: 0.9739 - val_loss: 0.5896 - val_accuracy: 0.4075\n",
            "Epoch 9/20\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0271 - accuracy: 0.9739 - val_loss: 0.0699 - val_accuracy: 0.9324\n",
            "Epoch 10/20\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0255 - accuracy: 0.9755 - val_loss: 0.0699 - val_accuracy: 0.9314\n",
            "Epoch 11/20\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0247 - accuracy: 0.9762 - val_loss: 0.1459 - val_accuracy: 0.8563\n",
            "Epoch 12/20\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0246 - accuracy: 0.9763 - val_loss: 0.0728 - val_accuracy: 0.9285\n",
            "Epoch 13/20\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0241 - accuracy: 0.9767 - val_loss: 0.0379 - val_accuracy: 0.9631\n",
            "Epoch 14/20\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0233 - accuracy: 0.9774 - val_loss: 0.0494 - val_accuracy: 0.9515\n",
            "Epoch 15/20\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0235 - accuracy: 0.9771 - val_loss: 0.0325 - val_accuracy: 0.9682\n",
            "Epoch 16/20\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0230 - accuracy: 0.9776 - val_loss: 0.0396 - val_accuracy: 0.9613\n",
            "Epoch 17/20\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0225 - accuracy: 0.9781 - val_loss: 0.0523 - val_accuracy: 0.9487\n",
            "Epoch 18/20\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0230 - accuracy: 0.9776 - val_loss: 0.0375 - val_accuracy: 0.9632\n",
            "Epoch 19/20\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0225 - accuracy: 0.9781 - val_loss: 0.0305 - val_accuracy: 0.9701\n",
            "Epoch 20/20\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0219 - accuracy: 0.9787 - val_loss: 0.0665 - val_accuracy: 0.9344\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pRvmJPY6UrSP"
      },
      "source": [
        "## 학습시 로스와 정확도 보기"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DrqmWivCUrSQ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 283
        },
        "outputId": "2ada1ce9-87e6-4429-89a1-296b2b753c96"
      },
      "source": [
        "fig, loss_ax = plt.subplots()\n",
        "\n",
        "acc_ax = loss_ax.twinx()\n",
        "\n",
        "loss_ax.plot(hist.history['loss'], 'y', label='train loss')\n",
        "loss_ax.plot(hist.history['val_loss'], 'r', label='val loss')\n",
        "loss_ax.set_ylim([0.0, 1.0])\n",
        "\n",
        "acc_ax.plot(hist.history['accuracy'], 'b', label='train acc')\n",
        "acc_ax.plot(hist.history['val_accuracy'], 'g', label='val acc')\n",
        "acc_ax.set_ylim([0.0, 1.0])\n",
        "\n",
        "loss_ax.set_xlabel('epoch')\n",
        "loss_ax.set_ylabel('loss')\n",
        "acc_ax.set_ylabel('accuray')\n",
        "\n",
        "loss_ax.legend(loc='upper left')\n",
        "acc_ax.legend(loc='lower left')\n",
        "\n",
        "plt.show()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VPpiU7yhUrST"
      },
      "source": [
        "## 테스트 결과 보기"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kPiPkfxtWlfr"
      },
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "def plotPredictions(model, train_x, train_y, test_x, test_y):\n",
        "\n",
        "    ix = np.random.randint(0, len(train_x))\n",
        "\n",
        "    input_ = train_x[ix:ix+1]\n",
        "    mask_ = train_y[ix:ix+1]\n",
        "\n",
        "    preds_train = model.predict(input_)\n",
        "    preds_train_t = (preds_train > 0.5).astype(np.uint8)\n",
        "\n",
        "    plt.figure(figsize=(10,10))\n",
        "    plt.subplot(1,3,1)\n",
        "    plt.title(\"train_x\")\n",
        "    plt.axis('off')\n",
        "    plt.imshow(input_[0])\n",
        "    plt.subplot(1,3,2)\n",
        "    plt.title(\"train_y\")\n",
        "    plt.axis('off')\n",
        "    plt.imshow(mask_[0][:,:,0], 'gray')\n",
        "    plt.subplot(1,3,3)\n",
        "    ret = model.evaluate(input_, mask_)\n",
        "    plt.title(\"Prediction: %.4f\" % (ret[1]))\n",
        "    plt.axis('off')\n",
        "    plt.imshow(preds_train_t[0][:,:,0], 'gray')\n",
        "    plt.show()\n",
        "\n",
        "    ix = np.random.randint(0, len(test_x))\n",
        "    input_ = test_x[ix:ix+1]\n",
        "    mask_ = test_y[ix:ix+1]\n",
        "    preds_test = model.predict(input_)\n",
        "    preds_test_t = (preds_test > 0.5).astype(np.uint8)\n",
        "    \n",
        "    plt.figure(figsize=(10,10))\n",
        "    plt.subplot(1,3,1)\n",
        "    plt.title(\"X_test\")\n",
        "    plt.axis('off')\n",
        "    plt.imshow(input_[0])\n",
        "    plt.subplot(1,3,2)\n",
        "    plt.title(\"test_y\")\n",
        "    plt.axis('off')\n",
        "    plt.imshow(mask_[0][:,:,0], 'gray')\n",
        "    plt.subplot(1,3,3)\n",
        "    ret = model.evaluate(input_, mask_)\n",
        "    plt.title(\"Prediction: %.4f\" % (ret[1]))\n",
        "    plt.axis('off')\n",
        "    plt.imshow(preds_test_t[0][:,:,0], 'gray')\n",
        "    plt.show()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vl3OZ7I0UrSV",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 438
        },
        "outputId": "e03b95e6-5048-44f8-b933-cb4ec0c3a18b"
      },
      "source": [
        "plotPredictions(model, train_x, train_y, test_x, test_y)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "1/1 [==============================] - 0s 22ms/step - loss: 0.0157 - accuracy: 0.9846\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "1/1 [==============================] - 0s 23ms/step - loss: 0.0797 - accuracy: 0.9204\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "tzdSWI67VIWl",
        "outputId": "37d8eff2-a82b-4006-a574-c8bacb600519",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "source": [
        "hist = model.fit(train_x, train_y, validation_split=0.2, epochs=100, batch_size=8, verbose=1)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/100\n",
            "46/46 [==============================] - 5s 105ms/step - loss: 0.0217 - accuracy: 0.9788 - val_loss: 0.0637 - val_accuracy: 0.9370\n",
            "Epoch 2/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0217 - accuracy: 0.9788 - val_loss: 0.0286 - val_accuracy: 0.9718\n",
            "Epoch 3/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0215 - accuracy: 0.9790 - val_loss: 0.0316 - val_accuracy: 0.9691\n",
            "Epoch 4/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0213 - accuracy: 0.9792 - val_loss: 0.0428 - val_accuracy: 0.9576\n",
            "Epoch 5/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0210 - accuracy: 0.9794 - val_loss: 0.0353 - val_accuracy: 0.9652\n",
            "Epoch 6/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0215 - accuracy: 0.9790 - val_loss: 0.0325 - val_accuracy: 0.9679\n",
            "Epoch 7/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0214 - accuracy: 0.9790 - val_loss: 0.0256 - val_accuracy: 0.9748\n",
            "Epoch 8/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0208 - accuracy: 0.9796 - val_loss: 0.0259 - val_accuracy: 0.9746\n",
            "Epoch 9/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0209 - accuracy: 0.9795 - val_loss: 0.0861 - val_accuracy: 0.9169\n",
            "Epoch 10/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0214 - accuracy: 0.9790 - val_loss: 0.0239 - val_accuracy: 0.9764\n",
            "Epoch 11/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0207 - accuracy: 0.9797 - val_loss: 0.0396 - val_accuracy: 0.9607\n",
            "Epoch 12/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0206 - accuracy: 0.9797 - val_loss: 0.0613 - val_accuracy: 0.9392\n",
            "Epoch 13/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0208 - accuracy: 0.9796 - val_loss: 0.0245 - val_accuracy: 0.9758\n",
            "Epoch 14/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0203 - accuracy: 0.9800 - val_loss: 0.0407 - val_accuracy: 0.9598\n",
            "Epoch 15/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0201 - accuracy: 0.9802 - val_loss: 0.0322 - val_accuracy: 0.9682\n",
            "Epoch 16/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0202 - accuracy: 0.9802 - val_loss: 0.0241 - val_accuracy: 0.9762\n",
            "Epoch 17/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0200 - accuracy: 0.9803 - val_loss: 0.0386 - val_accuracy: 0.9617\n",
            "Epoch 18/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0202 - accuracy: 0.9801 - val_loss: 0.0250 - val_accuracy: 0.9753\n",
            "Epoch 19/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0202 - accuracy: 0.9801 - val_loss: 0.0230 - val_accuracy: 0.9772\n",
            "Epoch 20/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0199 - accuracy: 0.9804 - val_loss: 0.0232 - val_accuracy: 0.9770\n",
            "Epoch 21/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0199 - accuracy: 0.9804 - val_loss: 0.0655 - val_accuracy: 0.9347\n",
            "Epoch 22/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0197 - accuracy: 0.9806 - val_loss: 0.0262 - val_accuracy: 0.9741\n",
            "Epoch 23/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0196 - accuracy: 0.9807 - val_loss: 0.0233 - val_accuracy: 0.9769\n",
            "Epoch 24/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0198 - accuracy: 0.9805 - val_loss: 0.0251 - val_accuracy: 0.9752\n",
            "Epoch 25/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0195 - accuracy: 0.9808 - val_loss: 0.0478 - val_accuracy: 0.9524\n",
            "Epoch 26/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0198 - accuracy: 0.9805 - val_loss: 0.0407 - val_accuracy: 0.9608\n",
            "Epoch 27/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0195 - accuracy: 0.9808 - val_loss: 0.0264 - val_accuracy: 0.9739\n",
            "Epoch 28/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0193 - accuracy: 0.9810 - val_loss: 0.0252 - val_accuracy: 0.9752\n",
            "Epoch 29/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0193 - accuracy: 0.9810 - val_loss: 0.0260 - val_accuracy: 0.9743\n",
            "Epoch 30/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0196 - accuracy: 0.9807 - val_loss: 0.0276 - val_accuracy: 0.9727\n",
            "Epoch 31/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0193 - accuracy: 0.9810 - val_loss: 0.0248 - val_accuracy: 0.9755\n",
            "Epoch 32/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0193 - accuracy: 0.9809 - val_loss: 0.0250 - val_accuracy: 0.9753\n",
            "Epoch 33/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0193 - accuracy: 0.9810 - val_loss: 0.0288 - val_accuracy: 0.9714\n",
            "Epoch 34/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0193 - accuracy: 0.9809 - val_loss: 0.0228 - val_accuracy: 0.9774\n",
            "Epoch 35/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0191 - accuracy: 0.9811 - val_loss: 0.0219 - val_accuracy: 0.9783\n",
            "Epoch 36/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0194 - accuracy: 0.9809 - val_loss: 0.0229 - val_accuracy: 0.9773\n",
            "Epoch 37/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0191 - accuracy: 0.9812 - val_loss: 0.0245 - val_accuracy: 0.9758\n",
            "Epoch 38/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0190 - accuracy: 0.9813 - val_loss: 0.0851 - val_accuracy: 0.9151\n",
            "Epoch 39/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0189 - accuracy: 0.9814 - val_loss: 0.0231 - val_accuracy: 0.9771\n",
            "Epoch 40/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0190 - accuracy: 0.9813 - val_loss: 0.0264 - val_accuracy: 0.9738\n",
            "Epoch 41/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0189 - accuracy: 0.9813 - val_loss: 0.0236 - val_accuracy: 0.9766\n",
            "Epoch 42/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0189 - accuracy: 0.9813 - val_loss: 0.0223 - val_accuracy: 0.9779\n",
            "Epoch 43/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0187 - accuracy: 0.9815 - val_loss: 0.0228 - val_accuracy: 0.9775\n",
            "Epoch 44/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0187 - accuracy: 0.9815 - val_loss: 0.0340 - val_accuracy: 0.9665\n",
            "Epoch 45/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0187 - accuracy: 0.9816 - val_loss: 0.0219 - val_accuracy: 0.9784\n",
            "Epoch 46/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0188 - accuracy: 0.9814 - val_loss: 0.0226 - val_accuracy: 0.9776\n",
            "Epoch 47/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0188 - accuracy: 0.9815 - val_loss: 0.0655 - val_accuracy: 0.9347\n",
            "Epoch 48/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0187 - accuracy: 0.9815 - val_loss: 0.0270 - val_accuracy: 0.9733\n",
            "Epoch 49/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0188 - accuracy: 0.9814 - val_loss: 0.0220 - val_accuracy: 0.9783\n",
            "Epoch 50/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0186 - accuracy: 0.9816 - val_loss: 0.0219 - val_accuracy: 0.9782\n",
            "Epoch 51/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0187 - accuracy: 0.9816 - val_loss: 0.0218 - val_accuracy: 0.9784\n",
            "Epoch 52/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0189 - accuracy: 0.9813 - val_loss: 0.2239 - val_accuracy: 0.7759\n",
            "Epoch 53/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0189 - accuracy: 0.9813 - val_loss: 0.0345 - val_accuracy: 0.9660\n",
            "Epoch 54/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0186 - accuracy: 0.9816 - val_loss: 0.0242 - val_accuracy: 0.9760\n",
            "Epoch 55/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0185 - accuracy: 0.9817 - val_loss: 0.0240 - val_accuracy: 0.9762\n",
            "Epoch 56/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0188 - accuracy: 0.9814 - val_loss: 0.0857 - val_accuracy: 0.9144\n",
            "Epoch 57/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0184 - accuracy: 0.9818 - val_loss: 0.0234 - val_accuracy: 0.9768\n",
            "Epoch 58/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0186 - accuracy: 0.9816 - val_loss: 0.0216 - val_accuracy: 0.9786\n",
            "Epoch 59/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0184 - accuracy: 0.9818 - val_loss: 0.0243 - val_accuracy: 0.9759\n",
            "Epoch 60/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0184 - accuracy: 0.9818 - val_loss: 0.0722 - val_accuracy: 0.9307\n",
            "Epoch 61/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0208 - accuracy: 0.9795 - val_loss: 0.3043 - val_accuracy: 0.6961\n",
            "Epoch 62/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0190 - accuracy: 0.9812 - val_loss: 0.2807 - val_accuracy: 0.7182\n",
            "Epoch 63/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0190 - accuracy: 0.9813 - val_loss: 0.0244 - val_accuracy: 0.9759\n",
            "Epoch 64/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0186 - accuracy: 0.9816 - val_loss: 0.0685 - val_accuracy: 0.9318\n",
            "Epoch 65/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0184 - accuracy: 0.9818 - val_loss: 0.0239 - val_accuracy: 0.9762\n",
            "Epoch 66/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0185 - accuracy: 0.9817 - val_loss: 0.0238 - val_accuracy: 0.9764\n",
            "Epoch 67/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0183 - accuracy: 0.9819 - val_loss: 0.0243 - val_accuracy: 0.9759\n",
            "Epoch 68/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0185 - accuracy: 0.9817 - val_loss: 0.0327 - val_accuracy: 0.9675\n",
            "Epoch 69/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0185 - accuracy: 0.9817 - val_loss: 0.0223 - val_accuracy: 0.9779\n",
            "Epoch 70/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0182 - accuracy: 0.9820 - val_loss: 0.0217 - val_accuracy: 0.9785\n",
            "Epoch 71/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0182 - accuracy: 0.9820 - val_loss: 0.0218 - val_accuracy: 0.9783\n",
            "Epoch 72/100\n",
            "46/46 [==============================] - 5s 105ms/step - loss: 0.0183 - accuracy: 0.9819 - val_loss: 0.0218 - val_accuracy: 0.9784\n",
            "Epoch 73/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0182 - accuracy: 0.9820 - val_loss: 0.0213 - val_accuracy: 0.9789\n",
            "Epoch 74/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0183 - accuracy: 0.9819 - val_loss: 0.0836 - val_accuracy: 0.9168\n",
            "Epoch 75/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0186 - accuracy: 0.9816 - val_loss: 0.0214 - val_accuracy: 0.9787\n",
            "Epoch 76/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0182 - accuracy: 0.9820 - val_loss: 0.0222 - val_accuracy: 0.9780\n",
            "Epoch 77/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0182 - accuracy: 0.9820 - val_loss: 0.0322 - val_accuracy: 0.9681\n",
            "Epoch 78/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0181 - accuracy: 0.9821 - val_loss: 0.0223 - val_accuracy: 0.9778\n",
            "Epoch 79/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0181 - accuracy: 0.9821 - val_loss: 0.0239 - val_accuracy: 0.9763\n",
            "Epoch 80/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0180 - accuracy: 0.9821 - val_loss: 0.0217 - val_accuracy: 0.9785\n",
            "Epoch 81/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0181 - accuracy: 0.9820 - val_loss: 0.0219 - val_accuracy: 0.9783\n",
            "Epoch 82/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0179 - accuracy: 0.9823 - val_loss: 0.0213 - val_accuracy: 0.9789\n",
            "Epoch 83/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0180 - accuracy: 0.9822 - val_loss: 0.0232 - val_accuracy: 0.9769\n",
            "Epoch 84/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0179 - accuracy: 0.9823 - val_loss: 0.0237 - val_accuracy: 0.9765\n",
            "Epoch 85/100\n",
            "46/46 [==============================] - 5s 105ms/step - loss: 0.0179 - accuracy: 0.9822 - val_loss: 0.0363 - val_accuracy: 0.9640\n",
            "Epoch 86/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0179 - accuracy: 0.9823 - val_loss: 0.0217 - val_accuracy: 0.9785\n",
            "Epoch 87/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0179 - accuracy: 0.9823 - val_loss: 0.0226 - val_accuracy: 0.9776\n",
            "Epoch 88/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0178 - accuracy: 0.9823 - val_loss: 0.0233 - val_accuracy: 0.9769\n",
            "Epoch 89/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0179 - accuracy: 0.9823 - val_loss: 0.0900 - val_accuracy: 0.9101\n",
            "Epoch 90/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0179 - accuracy: 0.9823 - val_loss: 0.0222 - val_accuracy: 0.9779\n",
            "Epoch 91/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0179 - accuracy: 0.9823 - val_loss: 0.0213 - val_accuracy: 0.9789\n",
            "Epoch 92/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0178 - accuracy: 0.9824 - val_loss: 0.0233 - val_accuracy: 0.9768\n",
            "Epoch 93/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0177 - accuracy: 0.9825 - val_loss: 0.0243 - val_accuracy: 0.9759\n",
            "Epoch 94/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0177 - accuracy: 0.9824 - val_loss: 0.0208 - val_accuracy: 0.9793\n",
            "Epoch 95/100\n",
            "46/46 [==============================] - 5s 105ms/step - loss: 0.0176 - accuracy: 0.9825 - val_loss: 0.0233 - val_accuracy: 0.9769\n",
            "Epoch 96/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0176 - accuracy: 0.9825 - val_loss: 0.0311 - val_accuracy: 0.9690\n",
            "Epoch 97/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0177 - accuracy: 0.9825 - val_loss: 0.0601 - val_accuracy: 0.9401\n",
            "Epoch 98/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0178 - accuracy: 0.9823 - val_loss: 0.0211 - val_accuracy: 0.9790\n",
            "Epoch 99/100\n",
            "46/46 [==============================] - 5s 103ms/step - loss: 0.0176 - accuracy: 0.9825 - val_loss: 0.0215 - val_accuracy: 0.9786\n",
            "Epoch 100/100\n",
            "46/46 [==============================] - 5s 104ms/step - loss: 0.0176 - accuracy: 0.9826 - val_loss: 0.0229 - val_accuracy: 0.9773\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bzJvWsZClnAS",
        "outputId": "04057ca1-408e-425e-b60e-dcc61faf2d21",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 438
        }
      },
      "source": [
        "plotPredictions(model, train_x, train_y, test_x, test_y)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "1/1 [==============================] - 0s 29ms/step - loss: 0.0358 - accuracy: 0.9645\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "1/1 [==============================] - 0s 25ms/step - loss: 0.0135 - accuracy: 0.9866\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x720 with 3 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BZdGFkq7loZs"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}