{
  "metadata": {
    "kernelspec": {
      "language": "python",
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python",
      "version": "3.7.12",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "colab": {
      "provenance": [],
      "machine_shape": "hm",
      "include_colab_link": true
    },
    "gpuClass": "standard"
  },
  "nbformat_minor": 0,
  "nbformat": 4,
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/yxmauw/cxr-multilabel-clf/blob/main/enet-colab.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Setting up environment"
      ],
      "metadata": {
        "id": "kNdVjbIkwTnV"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# mount google drive\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive', force_remount=True)"
      ],
      "metadata": {
        "id": "RQDSDTzNwTRQ",
        "outputId": "8a3f90fa-3621-4ef8-a33a-78fbefe4c85b",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mounted at /content/drive\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!git clone https://github.com/yxmauw/cxr-multilabel-clf.git"
      ],
      "metadata": {
        "id": "7LVf0IqtxO28",
        "outputId": "b21853a5-e504-40ad-f4d0-5c06c8efe602",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Cloning into 'cxr-multilabel-clf'...\n",
            "remote: Enumerating objects: 63, done.\u001b[K\n",
            "remote: Counting objects: 100% (63/63), done.\u001b[K\n",
            "remote: Compressing objects: 100% (61/61), done.\u001b[K\n",
            "remote: Total 63 (delta 28), reused 0 (delta 0), pack-reused 0\u001b[K\n",
            "Unpacking objects: 100% (63/63), done.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!mkdir ~/.kaggle #Make a directory named “.kaggle”"
      ],
      "metadata": {
        "id": "gKoUXV34xws6"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "!cp ./cxr-multilabel-clf//kaggle.json ~/.kaggle/ # Copy the “kaggle.json” into this new directory"
      ],
      "metadata": {
        "id": "vQ9JVUJqx8RD"
      },
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "!chmod 600 ~/.kaggle/kaggle.json # Allocate the required permission for this file"
      ],
      "metadata": {
        "id": "5Q13Q3d3yEtl"
      },
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "!kaggle competitions download -c ranzcr-clip-catheter-line-classification # download dataset"
      ],
      "metadata": {
        "id": "gIWBx5k-yIYI",
        "outputId": "a853e922-85da-4609-d269-ef34880bec92",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading ranzcr-clip-catheter-line-classification.zip to /content\n",
            "100% 11.7G/11.7G [06:38<00:00, 38.3MB/s]\n",
            "100% 11.7G/11.7G [06:38<00:00, 31.5MB/s]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!unzip ranzcr-clip-catheter-line-classification.zip #unzip folders"
      ],
      "metadata": {
        "id": "2oGiv_1SyxFW"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Data"
      ],
      "metadata": {
        "id": "L3RKSB9ZxE_F"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np \n",
        "import pandas as pd "
      ],
      "metadata": {
        "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
        "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
        "execution": {
          "iopub.status.busy": "2022-10-28T02:46:40.023210Z",
          "iopub.execute_input": "2022-10-28T02:46:40.024111Z",
          "iopub.status.idle": "2022-10-28T02:46:40.047013Z",
          "shell.execute_reply.started": "2022-10-28T02:46:40.024018Z",
          "shell.execute_reply": "2022-10-28T02:46:40.046119Z"
        },
        "trusted": true,
        "id": "6HJ9QhyevJlw"
      },
      "execution_count": 8,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "train_df = pd.read_csv('train.csv')\n",
        "display(len(train_df))\n",
        "display(train_df.head(3))\n",
        "train_annot_df = pd.read_csv('train_annotations.csv')\n",
        "display(len(train_annot_df))\n",
        "display(train_annot_df.head(3))"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-10-28T02:46:44.475857Z",
          "iopub.execute_input": "2022-10-28T02:46:44.476564Z",
          "iopub.status.idle": "2022-10-28T02:46:44.724851Z",
          "shell.execute_reply.started": "2022-10-28T02:46:44.476517Z",
          "shell.execute_reply": "2022-10-28T02:46:44.723861Z"
        },
        "trusted": true,
        "id": "q_tcXmTovJly",
        "outputId": "e024c922-b5c6-4ddc-e4f1-102f7cb40ee6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 421
        }
      },
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "30083"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "                                    StudyInstanceUID  ETT - Abnormal  \\\n",
              "0  1.2.826.0.1.3680043.8.498.26697628953273228189...               0   \n",
              "1  1.2.826.0.1.3680043.8.498.46302891597398758759...               0   \n",
              "2  1.2.826.0.1.3680043.8.498.23819260719748494858...               0   \n",
              "\n",
              "   ETT - Borderline  ETT - Normal  NGT - Abnormal  NGT - Borderline  \\\n",
              "0                 0             0               0                 0   \n",
              "1                 0             1               0                 0   \n",
              "2                 0             0               0                 0   \n",
              "\n",
              "   NGT - Incompletely Imaged  NGT - Normal  CVC - Abnormal  CVC - Borderline  \\\n",
              "0                          0             1               0                 0   \n",
              "1                          1             0               0                 0   \n",
              "2                          0             0               0                 1   \n",
              "\n",
              "   CVC - Normal  Swan Ganz Catheter Present  PatientID  \n",
              "0             0                           0  ec89415d1  \n",
              "1             1                           0  bf4c6da3c  \n",
              "2             0                           0  3fc1c97e5  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-73905c72-a1c6-442e-aedd-4da3ee948c64\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>StudyInstanceUID</th>\n",
              "      <th>ETT - Abnormal</th>\n",
              "      <th>ETT - Borderline</th>\n",
              "      <th>ETT - Normal</th>\n",
              "      <th>NGT - Abnormal</th>\n",
              "      <th>NGT - Borderline</th>\n",
              "      <th>NGT - Incompletely Imaged</th>\n",
              "      <th>NGT - Normal</th>\n",
              "      <th>CVC - Abnormal</th>\n",
              "      <th>CVC - Borderline</th>\n",
              "      <th>CVC - Normal</th>\n",
              "      <th>Swan Ganz Catheter Present</th>\n",
              "      <th>PatientID</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1.2.826.0.1.3680043.8.498.26697628953273228189...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>ec89415d1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1.2.826.0.1.3680043.8.498.46302891597398758759...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>bf4c6da3c</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1.2.826.0.1.3680043.8.498.23819260719748494858...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3fc1c97e5</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-73905c72-a1c6-442e-aedd-4da3ee948c64')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-73905c72-a1c6-442e-aedd-4da3ee948c64 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-73905c72-a1c6-442e-aedd-4da3ee948c64');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "17999"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "                                    StudyInstanceUID             label  \\\n",
              "0  1.2.826.0.1.3680043.8.498.12616281126973421762...      CVC - Normal   \n",
              "1  1.2.826.0.1.3680043.8.498.12616281126973421762...      CVC - Normal   \n",
              "2  1.2.826.0.1.3680043.8.498.72921907356394389969...  CVC - Borderline   \n",
              "\n",
              "                                                data  \n",
              "0  [[1487, 1279], [1477, 1168], [1472, 1052], [14...  \n",
              "1  [[1328, 7], [1347, 101], [1383, 193], [1400, 2...  \n",
              "2  [[801, 1207], [812, 1112], [823, 1023], [842, ...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-f87c356b-8a6f-4c1a-8017-062938a9a512\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>StudyInstanceUID</th>\n",
              "      <th>label</th>\n",
              "      <th>data</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1.2.826.0.1.3680043.8.498.12616281126973421762...</td>\n",
              "      <td>CVC - Normal</td>\n",
              "      <td>[[1487, 1279], [1477, 1168], [1472, 1052], [14...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1.2.826.0.1.3680043.8.498.12616281126973421762...</td>\n",
              "      <td>CVC - Normal</td>\n",
              "      <td>[[1328, 7], [1347, 101], [1383, 193], [1400, 2...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1.2.826.0.1.3680043.8.498.72921907356394389969...</td>\n",
              "      <td>CVC - Borderline</td>\n",
              "      <td>[[801, 1207], [812, 1112], [823, 1023], [842, ...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f87c356b-8a6f-4c1a-8017-062938a9a512')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-f87c356b-8a6f-4c1a-8017-062938a9a512 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-f87c356b-8a6f-4c1a-8017-062938a9a512');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# value counts\n",
        "train_df.drop(columns=['StudyInstanceUID','PatientID']).agg(['sum'])\n",
        "# unbalanced dataset"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-10-28T02:46:59.167135Z",
          "iopub.execute_input": "2022-10-28T02:46:59.167596Z",
          "iopub.status.idle": "2022-10-28T02:46:59.208167Z",
          "shell.execute_reply.started": "2022-10-28T02:46:59.167559Z",
          "shell.execute_reply": "2022-10-28T02:46:59.207260Z"
        },
        "trusted": true,
        "id": "5huaW8WQvJl1",
        "outputId": "5dcae458-0bfe-4dd5-aaab-bbbebfda5ab8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 197
        }
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "     ETT - Abnormal  ETT - Borderline  ETT - Normal  NGT - Abnormal  \\\n",
              "sum              79              1138          7240             279   \n",
              "\n",
              "     NGT - Borderline  NGT - Incompletely Imaged  NGT - Normal  \\\n",
              "sum               529                       2748          4797   \n",
              "\n",
              "     CVC - Abnormal  CVC - Borderline  CVC - Normal  \\\n",
              "sum            3195              8460         21324   \n",
              "\n",
              "     Swan Ganz Catheter Present  \n",
              "sum                         830  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-1d5519f2-ca9d-4063-b99f-34450f88f419\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>ETT - Abnormal</th>\n",
              "      <th>ETT - Borderline</th>\n",
              "      <th>ETT - Normal</th>\n",
              "      <th>NGT - Abnormal</th>\n",
              "      <th>NGT - Borderline</th>\n",
              "      <th>NGT - Incompletely Imaged</th>\n",
              "      <th>NGT - Normal</th>\n",
              "      <th>CVC - Abnormal</th>\n",
              "      <th>CVC - Borderline</th>\n",
              "      <th>CVC - Normal</th>\n",
              "      <th>Swan Ganz Catheter Present</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>sum</th>\n",
              "      <td>79</td>\n",
              "      <td>1138</td>\n",
              "      <td>7240</td>\n",
              "      <td>279</td>\n",
              "      <td>529</td>\n",
              "      <td>2748</td>\n",
              "      <td>4797</td>\n",
              "      <td>3195</td>\n",
              "      <td>8460</td>\n",
              "      <td>21324</td>\n",
              "      <td>830</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1d5519f2-ca9d-4063-b99f-34450f88f419')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-1d5519f2-ca9d-4063-b99f-34450f88f419 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-1d5519f2-ca9d-4063-b99f-34450f88f419');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-10-28T02:47:01.769145Z",
          "iopub.execute_input": "2022-10-28T02:47:01.769618Z",
          "iopub.status.idle": "2022-10-28T02:47:02.507132Z",
          "shell.execute_reply.started": "2022-10-28T02:47:01.769578Z",
          "shell.execute_reply": "2022-10-28T02:47:02.506044Z"
        },
        "trusted": true,
        "id": "VBczRJs9vJl2"
      },
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# value counts\n",
        "train_df.drop(columns=['StudyInstanceUID','PatientID']).agg(['sum']).T.sort_values(by='sum').plot(kind='barh')\n",
        "plt.legend(loc='lower right');"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-10-28T02:47:04.180989Z",
          "iopub.execute_input": "2022-10-28T02:47:04.181680Z",
          "iopub.status.idle": "2022-10-28T02:47:04.493106Z",
          "shell.execute_reply.started": "2022-10-28T02:47:04.181644Z",
          "shell.execute_reply": "2022-10-28T02:47:04.491889Z"
        },
        "trusted": true,
        "id": "ePQF1z29vJl3",
        "outputId": "e043571f-f8b6-4084-9dfe-055082040594",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 265
        }
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "len(train_df.drop(columns=['StudyInstanceUID','PatientID']).agg(['sum']).T)\n",
        "# num of classes"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-10-28T02:47:07.675004Z",
          "iopub.execute_input": "2022-10-28T02:47:07.675678Z",
          "iopub.status.idle": "2022-10-28T02:47:07.695859Z",
          "shell.execute_reply.started": "2022-10-28T02:47:07.675642Z",
          "shell.execute_reply": "2022-10-28T02:47:07.694954Z"
        },
        "trusted": true,
        "id": "RA8cTCndvJl4",
        "outputId": "ad5fb681-9460-4846-c56e-0adcc336c1b4",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "11"
            ]
          },
          "metadata": {},
          "execution_count": 12
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Create Datasets"
      ],
      "metadata": {
        "id": "4ckn3ZfUz4T_"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "import cv2\n",
        "import numpy as np\n",
        "from torchvision import transforms\n",
        "from torch.utils.data import Dataset"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-10-28T02:47:09.612626Z",
          "iopub.execute_input": "2022-10-28T02:47:09.613355Z",
          "iopub.status.idle": "2022-10-28T02:47:11.485435Z",
          "shell.execute_reply.started": "2022-10-28T02:47:09.613293Z",
          "shell.execute_reply": "2022-10-28T02:47:11.484503Z"
        },
        "trusted": true,
        "id": "RAfp_UQivJl4"
      },
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "class ImageDataset(Dataset):\n",
        "    def __init__(self, csv, train, test):\n",
        "        self.csv = csv\n",
        "        self.train = train\n",
        "        self.test = test\n",
        "        self.all_image_names = self.csv[:]['StudyInstanceUID']\n",
        "        self.all_labels = np.array(self.csv.drop(['StudyInstanceUID', 'PatientID'], axis=1))\n",
        "        self.train_ratio = int(0.85 * len(self.csv))\n",
        "        self.valid_ratio = len(self.csv) - self.train_ratio\n",
        "        # set the training data images and labels\n",
        "        if self.train == True:\n",
        "            print(f\"Number of training images: {self.train_ratio}\")\n",
        "            self.image_names = list(self.all_image_names[:self.train_ratio])\n",
        "            self.labels = list(self.all_labels[:self.train_ratio])\n",
        "            # define the training transforms\n",
        "            self.transform = transforms.Compose([\n",
        "                transforms.ToPILImage(),\n",
        "                transforms.Resize((400, 400)),\n",
        "                transforms.RandomHorizontalFlip(p=0.5),\n",
        "                transforms.RandomRotation(degrees=45),\n",
        "                transforms.ToTensor(),\n",
        "            ])\n",
        "        # set the validation data images and labels\n",
        "        elif self.train == False and self.test == False:\n",
        "            print(f\"Number of validation images: {self.valid_ratio}\")\n",
        "            self.image_names = list(self.all_image_names[-self.valid_ratio:-10])\n",
        "            self.labels = list(self.all_labels[-self.valid_ratio:])\n",
        "            # define the validation transforms\n",
        "            self.transform = transforms.Compose([\n",
        "                transforms.ToPILImage(),\n",
        "                transforms.Resize((400, 400)),\n",
        "                transforms.ToTensor(),\n",
        "            ])\n",
        "        # set the test data images and labels, only last 10 images\n",
        "        # this, we will use in a separate inference script\n",
        "        elif self.test == True and self.train == False:\n",
        "            self.image_names = list(self.all_image_names[-10:])\n",
        "            self.labels = list(self.all_labels[-10:])\n",
        "             # define the test transforms\n",
        "            self.transform = transforms.Compose([\n",
        "                transforms.ToPILImage(),\n",
        "                transforms.ToTensor(),\n",
        "            ])\n",
        "    def __len__(self):\n",
        "        return len(self.image_names)\n",
        "    \n",
        "    def __getitem__(self, index):\n",
        "        image = cv2.imread(f\"./train/{self.image_names[index]}.jpg\")\n",
        "        # convert the image from BGR to RGB color format\n",
        "        image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)\n",
        "        # apply image transforms\n",
        "        image = self.transform(image)\n",
        "        targets = self.labels[index]\n",
        "        \n",
        "        return {\n",
        "            'image': torch.tensor(image, dtype=torch.float32),\n",
        "            'label': torch.tensor(targets, dtype=torch.float32)\n",
        "        }"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-10-28T02:47:13.898477Z",
          "iopub.execute_input": "2022-10-28T02:47:13.899027Z",
          "iopub.status.idle": "2022-10-28T02:47:13.912689Z",
          "shell.execute_reply.started": "2022-10-28T02:47:13.898986Z",
          "shell.execute_reply": "2022-10-28T02:47:13.911765Z"
        },
        "trusted": true,
        "id": "KdLACwrLvJl5"
      },
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import torchvision\n",
        "torchvision.__version__"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-10-28T03:05:44.948332Z",
          "iopub.execute_input": "2022-10-28T03:05:44.948713Z",
          "iopub.status.idle": "2022-10-28T03:05:44.955478Z",
          "shell.execute_reply.started": "2022-10-28T03:05:44.948681Z",
          "shell.execute_reply": "2022-10-28T03:05:44.954402Z"
        },
        "trusted": true,
        "id": "v1ofiKiavJl6",
        "outputId": "0ecc26fd-3a56-437f-ca65-b18197b3ae8a",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        }
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'0.13.1+cu113'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(dir(torchvision.models))"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-10-28T03:42:01.322455Z",
          "iopub.execute_input": "2022-10-28T03:42:01.322825Z",
          "iopub.status.idle": "2022-10-28T03:42:01.328721Z",
          "shell.execute_reply.started": "2022-10-28T03:42:01.322792Z",
          "shell.execute_reply": "2022-10-28T03:42:01.327749Z"
        },
        "trusted": true,
        "id": "utBbFn0_vJl7",
        "outputId": "9ee13f10-8c60-40a3-f6e6-2672810eecd2",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "['AlexNet', 'AlexNet_Weights', 'ConvNeXt', 'ConvNeXt_Base_Weights', 'ConvNeXt_Large_Weights', 'ConvNeXt_Small_Weights', 'ConvNeXt_Tiny_Weights', 'DenseNet', 'DenseNet121_Weights', 'DenseNet161_Weights', 'DenseNet169_Weights', 'DenseNet201_Weights', 'EfficientNet', 'EfficientNet_B0_Weights', 'EfficientNet_B1_Weights', 'EfficientNet_B2_Weights', 'EfficientNet_B3_Weights', 'EfficientNet_B4_Weights', 'EfficientNet_B5_Weights', 'EfficientNet_B6_Weights', 'EfficientNet_B7_Weights', 'EfficientNet_V2_L_Weights', 'EfficientNet_V2_M_Weights', 'EfficientNet_V2_S_Weights', 'GoogLeNet', 'GoogLeNetOutputs', 'GoogLeNet_Weights', 'Inception3', 'InceptionOutputs', 'Inception_V3_Weights', 'MNASNet', 'MNASNet0_5_Weights', 'MNASNet0_75_Weights', 'MNASNet1_0_Weights', 'MNASNet1_3_Weights', 'MobileNetV2', 'MobileNetV3', 'MobileNet_V2_Weights', 'MobileNet_V3_Large_Weights', 'MobileNet_V3_Small_Weights', 'RegNet', 'RegNet_X_16GF_Weights', 'RegNet_X_1_6GF_Weights', 'RegNet_X_32GF_Weights', 'RegNet_X_3_2GF_Weights', 'RegNet_X_400MF_Weights', 'RegNet_X_800MF_Weights', 'RegNet_X_8GF_Weights', 'RegNet_Y_128GF_Weights', 'RegNet_Y_16GF_Weights', 'RegNet_Y_1_6GF_Weights', 'RegNet_Y_32GF_Weights', 'RegNet_Y_3_2GF_Weights', 'RegNet_Y_400MF_Weights', 'RegNet_Y_800MF_Weights', 'RegNet_Y_8GF_Weights', 'ResNeXt101_32X8D_Weights', 'ResNeXt101_64X4D_Weights', 'ResNeXt50_32X4D_Weights', 'ResNet', 'ResNet101_Weights', 'ResNet152_Weights', 'ResNet18_Weights', 'ResNet34_Weights', 'ResNet50_Weights', 'ShuffleNetV2', 'ShuffleNet_V2_X0_5_Weights', 'ShuffleNet_V2_X1_0_Weights', 'ShuffleNet_V2_X1_5_Weights', 'ShuffleNet_V2_X2_0_Weights', 'SqueezeNet', 'SqueezeNet1_0_Weights', 'SqueezeNet1_1_Weights', 'SwinTransformer', 'Swin_B_Weights', 'Swin_S_Weights', 'Swin_T_Weights', 'VGG', 'VGG11_BN_Weights', 'VGG11_Weights', 'VGG13_BN_Weights', 'VGG13_Weights', 'VGG16_BN_Weights', 'VGG16_Weights', 'VGG19_BN_Weights', 'VGG19_Weights', 'ViT_B_16_Weights', 'ViT_B_32_Weights', 'ViT_H_14_Weights', 'ViT_L_16_Weights', 'ViT_L_32_Weights', 'VisionTransformer', 'Wide_ResNet101_2_Weights', 'Wide_ResNet50_2_Weights', '_GoogLeNetOutputs', '_InceptionOutputs', '__builtins__', '__cached__', '__doc__', '__file__', '__loader__', '__name__', '__package__', '__path__', '__spec__', '_api', '_meta', '_utils', 'alexnet', 'convnext', 'convnext_base', 'convnext_large', 'convnext_small', 'convnext_tiny', 'densenet', 'densenet121', 'densenet161', 'densenet169', 'densenet201', 'detection', 'efficientnet', 'efficientnet_b0', 'efficientnet_b1', 'efficientnet_b2', 'efficientnet_b3', 'efficientnet_b4', 'efficientnet_b5', 'efficientnet_b6', 'efficientnet_b7', 'efficientnet_v2_l', 'efficientnet_v2_m', 'efficientnet_v2_s', 'get_weight', 'googlenet', 'inception', 'inception_v3', 'mnasnet', 'mnasnet0_5', 'mnasnet0_75', 'mnasnet1_0', 'mnasnet1_3', 'mobilenet', 'mobilenet_v2', 'mobilenet_v3_large', 'mobilenet_v3_small', 'mobilenetv2', 'mobilenetv3', 'optical_flow', 'quantization', 'regnet', 'regnet_x_16gf', 'regnet_x_1_6gf', 'regnet_x_32gf', 'regnet_x_3_2gf', 'regnet_x_400mf', 'regnet_x_800mf', 'regnet_x_8gf', 'regnet_y_128gf', 'regnet_y_16gf', 'regnet_y_1_6gf', 'regnet_y_32gf', 'regnet_y_3_2gf', 'regnet_y_400mf', 'regnet_y_800mf', 'regnet_y_8gf', 'resnet', 'resnet101', 'resnet152', 'resnet18', 'resnet34', 'resnet50', 'resnext101_32x8d', 'resnext101_64x4d', 'resnext50_32x4d', 'segmentation', 'shufflenet_v2_x0_5', 'shufflenet_v2_x1_0', 'shufflenet_v2_x1_5', 'shufflenet_v2_x2_0', 'shufflenetv2', 'squeezenet', 'squeezenet1_0', 'squeezenet1_1', 'swin_b', 'swin_s', 'swin_t', 'swin_transformer', 'vgg', 'vgg11', 'vgg11_bn', 'vgg13', 'vgg13_bn', 'vgg16', 'vgg16_bn', 'vgg19', 'vgg19_bn', 'video', 'vision_transformer', 'vit_b_16', 'vit_b_32', 'vit_h_14', 'vit_l_16', 'vit_l_32', 'wide_resnet101_2', 'wide_resnet50_2']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from torchvision import models \n",
        "import torch.nn as nn"
      ],
      "metadata": {
        "id": "_bjmGalO9UHo"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def model(pretrained, requires_grad):\n",
        "    model = models.efficientnet_v2_s(progress=True, pretrained=pretrained)\n",
        "    # to freeze the hidden layers\n",
        "    if requires_grad == False:\n",
        "        for param in model.parameters():\n",
        "            param.requires_grad = False\n",
        "    # to train the hidden layers\n",
        "    elif requires_grad == True:\n",
        "        for param in model.parameters():\n",
        "            param.requires_grad = True\n",
        "    # make the classification layer learnable\n",
        "    # we have 11 classes in total\n",
        "    model.classifier[1] = nn.Linear(in_features=1280, out_features=11)\n",
        "    return model"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-10-28T04:24:13.656142Z",
          "iopub.execute_input": "2022-10-28T04:24:13.656782Z",
          "iopub.status.idle": "2022-10-28T04:24:13.663143Z",
          "shell.execute_reply.started": "2022-10-28T04:24:13.656746Z",
          "shell.execute_reply": "2022-10-28T04:24:13.662004Z"
        },
        "trusted": true,
        "id": "m_obBk7lvJl8"
      },
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install torchmetrics "
      ],
      "metadata": {
        "id": "3okxQXjv0jYw",
        "outputId": "10d86872-257a-4706-bdf4-a8d4f60997cf",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Collecting torchmetrics\n",
            "  Downloading torchmetrics-0.10.2-py3-none-any.whl (529 kB)\n",
            "\u001b[K     |████████████████████████████████| 529 kB 14.3 MB/s \n",
            "\u001b[?25hRequirement already satisfied: numpy>=1.17.2 in /usr/local/lib/python3.7/dist-packages (from torchmetrics) (1.21.6)\n",
            "Requirement already satisfied: typing-extensions in /usr/local/lib/python3.7/dist-packages (from torchmetrics) (4.1.1)\n",
            "Requirement already satisfied: torch>=1.3.1 in /usr/local/lib/python3.7/dist-packages (from torchmetrics) (1.12.1+cu113)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.7/dist-packages (from torchmetrics) (21.3)\n",
            "Requirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in /usr/local/lib/python3.7/dist-packages (from packaging->torchmetrics) (3.0.9)\n",
            "Installing collected packages: torchmetrics\n",
            "Successfully installed torchmetrics-0.10.2\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Create Train and Validate functions"
      ],
      "metadata": {
        "id": "U-mCMBzyz_gY"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "from tqdm import tqdm\n",
        "from torchmetrics import Accuracy, AUROC, F1Score, Precision, Recall"
      ],
      "metadata": {
        "id": "kV8Yj6EZ9XhB"
      },
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# training function\n",
        "def train(model, dataloader, optimizer, criterion, train_data, device):\n",
        "    print('Training')\n",
        "    model.train()\n",
        "    counter = 0\n",
        "    train_running_loss = 0.0\n",
        "    # instantiate metrics\n",
        "    acc = Accuracy()\n",
        "    auc = AUROC()\n",
        "    f1_score = F1Score()\n",
        "    precision = Precision()\n",
        "    recall = Recall()\n",
        "    preds = []\n",
        "    labels = []\n",
        "    for i, data in tqdm(enumerate(dataloader), total=int(len(train_data)/dataloader.batch_size)):\n",
        "        counter += 1\n",
        "        data, target = data['image'].to(device), data['label'].to(device)\n",
        "        labels.append(target.cpu().numpy().argmax(axis=1))\n",
        "        optimizer.zero_grad()\n",
        "        outputs = model(data)\n",
        "        # apply sigmoid activation to get all the outputs between 0 and 1\n",
        "        outputs = torch.sigmoid(outputs)\n",
        "        loss = criterion(outputs, target)\n",
        "        train_running_loss += loss.item()\n",
        "        # backpropagation\n",
        "        loss.backward()\n",
        "        # update optimizer parameters\n",
        "        optimizer.step()\n",
        "        preds.append(outputs.detach().cpu().numpy().argmax(axis=1))\n",
        "        \n",
        "    train_loss = train_running_loss / counter\n",
        "    preds = torch.tensor(np.concatenate(preds))\n",
        "    labels = torch.tensor(np.concatenate(labels))\n",
        "    train_acc = acc(preds, labels).item()\n",
        "    \n",
        "    return train_loss, train_acc"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-10-28T04:24:16.962617Z",
          "iopub.execute_input": "2022-10-28T04:24:16.963764Z",
          "iopub.status.idle": "2022-10-28T04:24:16.976195Z",
          "shell.execute_reply.started": "2022-10-28T04:24:16.963715Z",
          "shell.execute_reply": "2022-10-28T04:24:16.975023Z"
        },
        "trusted": true,
        "id": "5y0w7dPrvJl9"
      },
      "execution_count": 20,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# validation function\n",
        "def validate(model, dataloader, criterion, val_data, device):\n",
        "    print('Validating')\n",
        "    model.eval()\n",
        "    counter = 0\n",
        "    val_running_loss = 0.0\n",
        "    # instantiate metrics\n",
        "    acc = Accuracy()\n",
        "    auc = AUROC()\n",
        "    f1_score = F1Score()\n",
        "    precision = Precision()\n",
        "    recall = Recall()\n",
        "    preds = []\n",
        "    labels = []\n",
        "    with torch.no_grad():\n",
        "        for i, data in tqdm(enumerate(dataloader), total=int(len(val_data)/dataloader.batch_size)):\n",
        "            counter += 1\n",
        "            data, target = data['image'].to(device), data['label'].to(device)\n",
        "            labels.append(target.cpu().numpy().argmax(axis=1))\n",
        "            # make predictions\n",
        "            outputs = model(data)\n",
        "            # apply sigmoid activation to get all the outputs between 0 and 1\n",
        "            outputs = torch.sigmoid(outputs)\n",
        "            loss = criterion(outputs, target)\n",
        "            val_running_loss += loss.item()\n",
        "            preds.append(outputs.detach().cpu().numpy().argmax(axis=1))\n",
        "        \n",
        "        val_loss = val_running_loss / counter\n",
        "        preds = torch.tensor(np.concatenate(preds))\n",
        "        labels = torch.tensor(np.concatenate(labels))\n",
        "        val_acc = acc(preds, labels).item()\n",
        "        return val_loss, val_acc"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-10-28T04:24:20.716225Z",
          "iopub.execute_input": "2022-10-28T04:24:20.716601Z",
          "iopub.status.idle": "2022-10-28T04:24:20.727583Z",
          "shell.execute_reply.started": "2022-10-28T04:24:20.716567Z",
          "shell.execute_reply": "2022-10-28T04:24:20.726191Z"
        },
        "trusted": true,
        "id": "F43wHu87vJl-"
      },
      "execution_count": 21,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import torch.optim as optim\n",
        "import matplotlib\n",
        "from torch.utils.data import DataLoader\n",
        "matplotlib.style.use('ggplot')\n",
        "# initialize the computation device\n",
        "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-10-28T04:24:24.617804Z",
          "iopub.execute_input": "2022-10-28T04:24:24.618166Z",
          "iopub.status.idle": "2022-10-28T04:24:24.623867Z",
          "shell.execute_reply.started": "2022-10-28T04:24:24.618134Z",
          "shell.execute_reply": "2022-10-28T04:24:24.622628Z"
        },
        "trusted": true,
        "id": "TWolPeqivJl-"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Set model parameters"
      ],
      "metadata": {
        "id": "2RkrFR2w0EfW"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#intialize the model\n",
        "from torch.optim.lr_scheduler import ReduceLROnPlateau\n",
        "\n",
        "ENet_model = model(pretrained=False, requires_grad=True).to(device)\n",
        "# learning parameters\n",
        "lr = 0.0001\n",
        "epochs = 25\n",
        "batch_size = 4\n",
        "optimizer = optim.Adam(ENet_model.parameters(), lr=lr)\n",
        "scheduler = ReduceLROnPlateau(optimizer, 'min')\n",
        "criterion = nn.BCELoss()"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-10-28T04:24:26.626925Z",
          "iopub.execute_input": "2022-10-28T04:24:26.630098Z",
          "iopub.status.idle": "2022-10-28T04:24:28.432430Z",
          "shell.execute_reply.started": "2022-10-28T04:24:26.630044Z",
          "shell.execute_reply": "2022-10-28T04:24:28.431444Z"
        },
        "trusted": true,
        "id": "siSktp4cvJmA",
        "outputId": "4559217e-7208-45c5-b3cb-5b2867c663cc",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/torchvision/models/_utils.py:209: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and will be removed in 0.15, please use 'weights' instead.\n",
            "  f\"The parameter '{pretrained_param}' is deprecated since 0.13 and will be removed in 0.15, \"\n",
            "/usr/local/lib/python3.7/dist-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=None`.\n",
            "  warnings.warn(msg)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "train_data = ImageDataset(\n",
        "    train_df, train=True, test=False\n",
        ")\n",
        "# validation dataset\n",
        "valid_data = ImageDataset(\n",
        "    train_df, train=False, test=False\n",
        ")\n",
        "# train data loader\n",
        "train_loader = DataLoader(\n",
        "    train_data, \n",
        "    batch_size=batch_size,\n",
        "    shuffle=True\n",
        ")\n",
        "# validation data loader\n",
        "valid_loader = DataLoader(\n",
        "    valid_data, \n",
        "    batch_size=batch_size,\n",
        "    shuffle=False\n",
        ")"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-10-28T04:24:32.210103Z",
          "iopub.execute_input": "2022-10-28T04:24:32.210655Z",
          "iopub.status.idle": "2022-10-28T04:24:32.243373Z",
          "shell.execute_reply.started": "2022-10-28T04:24:32.210615Z",
          "shell.execute_reply": "2022-10-28T04:24:32.242479Z"
        },
        "trusted": true,
        "id": "lUxQij0_vJmA",
        "outputId": "7f7cf372-d419-459d-baa5-8f6c5fa7fcac",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Number of training images: 25570\n",
            "Number of validation images: 4513\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Create save model class function"
      ],
      "metadata": {
        "id": "89DxeKW50Kez"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "class SaveBestModel:\n",
        "    \"\"\"\n",
        "    Class to save the best model while training. If the current epoch's \n",
        "    validation loss is less than the previous least less, then save the\n",
        "    model state.\n",
        "    \"\"\"\n",
        "    def __init__(\n",
        "        self, best_valid_loss=float('inf')\n",
        "    ):\n",
        "        self.best_valid_loss = best_valid_loss\n",
        "        \n",
        "    def __call__(\n",
        "        self, current_valid_loss, \n",
        "        epoch, model, optimizer, criterion\n",
        "    ):\n",
        "        if current_valid_loss < self.best_valid_loss:\n",
        "            self.best_valid_loss = current_valid_loss\n",
        "            print(f\"\\nBest validation loss: {self.best_valid_loss}\")\n",
        "            print(f\"\\nSaving best model for epoch: {epoch+1}\\n\")\n",
        "            file_path = f'drive/MyDrive/Colab Notebooks/Enet-ep{epoch+1}-val{current_valid_loss:.3f}.pth'\n",
        "            torch.save({\n",
        "                'epoch': epoch+1,\n",
        "                'model_state_dict': model.state_dict(),\n",
        "                'optimizer_state_dict': optimizer.state_dict(),\n",
        "                'loss': criterion,\n",
        "                }, file_path)"
      ],
      "metadata": {
        "id": "5TSUnoCtJQBS"
      },
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Save plots"
      ],
      "metadata": {
        "id": "ZyawvU8qDbYx"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def save_plots(train_acc, valid_acc, train_loss, valid_loss):\n",
        "    \"\"\"\n",
        "    Function to save the loss and accuracy plots to disk.\n",
        "    \"\"\"\n",
        "    # accuracy plots\n",
        "    plt.figure(figsize=(10, 7))\n",
        "    plt.plot(\n",
        "        train_acc, color='green', linestyle='-', \n",
        "        label='train accuracy'\n",
        "    )\n",
        "    plt.plot(\n",
        "        valid_acc, color='blue', linestyle='-', \n",
        "        label='validataion accuracy'\n",
        "    )\n",
        "    plt.xlabel('Epochs')\n",
        "    plt.ylabel('Accuracy')\n",
        "    plt.legend()\n",
        "    plt.savefig('drive/MyDrive/Colab Notebooks/Enet-acc.png')\n",
        "    \n",
        "    # loss plots\n",
        "    plt.figure(figsize=(10, 7))\n",
        "    plt.plot(\n",
        "        train_loss, color='orange', linestyle='-', \n",
        "        label='train loss'\n",
        "    )\n",
        "    plt.plot(\n",
        "        valid_loss, color='red', linestyle='-', \n",
        "        label='validataion loss'\n",
        "    )\n",
        "    plt.xlabel('Epochs')\n",
        "    plt.ylabel('Loss')\n",
        "    plt.legend()\n",
        "    plt.savefig('drive/MyDrive/Colab Notebooks/Enet-loss.png')"
      ],
      "metadata": {
        "id": "eH_PFKRVDeH1"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Create earlystopper class function"
      ],
      "metadata": {
        "id": "M-c9Qc4d0UhR"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "class EarlyStopper:\n",
        "    def __init__(self, patience=1, min_delta=0):\n",
        "        self.patience = patience\n",
        "        self.min_delta = min_delta\n",
        "        self.counter = 0\n",
        "        self.min_validation_loss = np.inf\n",
        "\n",
        "    def early_stop(self, validation_loss):\n",
        "        if validation_loss < self.min_validation_loss:\n",
        "            self.min_validation_loss = validation_loss\n",
        "            self.counter = 0\n",
        "        elif validation_loss > (self.min_validation_loss + self.min_delta):\n",
        "            self.counter += 1\n",
        "            if self.counter >= self.patience:\n",
        "                return True\n",
        "        return False"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-10-28T04:24:36.170818Z",
          "iopub.execute_input": "2022-10-28T04:24:36.171515Z",
          "iopub.status.idle": "2022-10-28T04:24:36.178005Z",
          "shell.execute_reply.started": "2022-10-28T04:24:36.171477Z",
          "shell.execute_reply": "2022-10-28T04:24:36.176852Z"
        },
        "trusted": true,
        "id": "LDFpjyp_vJmB"
      },
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Train model"
      ],
      "metadata": {
        "id": "bSIP5Yw40d7l"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# start the training and validation\n",
        "train_loss = []\n",
        "valid_loss = []\n",
        "train_acc = []\n",
        "val_acc = []\n",
        "early_stopper = EarlyStopper(patience=5, min_delta=0.001)\n",
        "save_best_model = SaveBestModel() # initialize SaveBestModel class\n",
        "for epoch in range(epochs):\n",
        "    print(f\"Epoch {epoch+1} of {epochs}\")\n",
        "    train_epoch_loss, train_epoch_acc = train(\n",
        "        ENet_model, train_loader, optimizer, criterion, train_data, device\n",
        "    )\n",
        "    valid_epoch_loss, val_epoch_acc = validate(\n",
        "        ENet_model, valid_loader, criterion, valid_data, device\n",
        "    )\n",
        "    if early_stopper.early_stop(valid_epoch_loss):             \n",
        "        break\n",
        "    train_loss.append(train_epoch_loss)\n",
        "    valid_loss.append(valid_epoch_loss)\n",
        "    train_acc.append(train_epoch_acc)\n",
        "    val_acc.append(val_epoch_acc)\n",
        "    print(f'Train Loss: {train_epoch_loss:.4f}; Val Loss: {valid_epoch_loss:.4f}; Train accuracy: {train_epoch_acc:.4f}; Val accuracy: {val_epoch_acc:.4f}')\n",
        "    # save the best model till now if we have the least loss in the current epoch\n",
        "    save_best_model(\n",
        "        valid_epoch_loss, epoch, ENet_model, optimizer, criterion\n",
        "    )\n",
        "    print('='*50) # gap\n",
        "\n",
        "save_plots(train_acc, val_acc, train_loss, valid_loss)\n",
        "print('PLOTS SAVED')"
      ],
      "metadata": {
        "execution": {
          "iopub.status.busy": "2022-10-28T04:24:39.710609Z",
          "iopub.execute_input": "2022-10-28T04:24:39.710988Z"
        },
        "trusted": true,
        "id": "umWa-LbavJmC",
        "outputId": "c81c677e-4b6a-4e4c-d8b0-c7c9018cff6b",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "execution_count": null,
      "outputs": [
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/torchmetrics/utilities/prints.py:36: UserWarning: Metric `AUROC` will save all targets and predictions in buffer. For large datasets this may lead to large memory footprint.\n",
            "  warnings.warn(*args, **kwargs)\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1 of 30\n",
            "Training\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "  0%|          | 0/6392 [00:00<?, ?it/s]/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:56: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n",
            "6393it [55:18,  1.93it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Validating\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "1126it [07:01,  2.67it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train Loss: 0.2887; Val Loss: 0.2701; Train accuracy: 0.4308; Val accuracy: 0.4373\n",
            "\n",
            "Best validation loss: 0.27008894395552985\n",
            "\n",
            "Saving best model for epoch: 1\n",
            "\n",
            "==================================================\n",
            "Epoch 2 of 30\n",
            "Training\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "6393it [53:15,  2.00it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Validating\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "1126it [07:03,  2.66it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train Loss: 0.2763; Val Loss: 0.2648; Train accuracy: 0.4333; Val accuracy: 0.4895\n",
            "\n",
            "Best validation loss: 0.26484752869891864\n",
            "\n",
            "Saving best model for epoch: 2\n",
            "\n",
            "==================================================\n",
            "Epoch 3 of 30\n",
            "Training\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "6393it [53:31,  1.99it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Validating\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "1126it [07:03,  2.66it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train Loss: 0.2719; Val Loss: 0.2589; Train accuracy: 0.4349; Val accuracy: 0.4317\n",
            "\n",
            "Best validation loss: 0.2588627720268222\n",
            "\n",
            "Saving best model for epoch: 3\n",
            "\n",
            "==================================================\n",
            "Epoch 4 of 30\n",
            "Training\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "6393it [53:18,  2.00it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Validating\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "1126it [07:02,  2.67it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train Loss: 0.2665; Val Loss: 0.2580; Train accuracy: 0.4372; Val accuracy: 0.4666\n",
            "\n",
            "Best validation loss: 0.257986861266209\n",
            "\n",
            "Saving best model for epoch: 4\n",
            "\n",
            "==================================================\n",
            "Epoch 5 of 30\n",
            "Training\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "6393it [53:13,  2.00it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Validating\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "1126it [07:05,  2.65it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train Loss: 0.2613; Val Loss: 0.2500; Train accuracy: 0.4460; Val accuracy: 0.4610\n",
            "\n",
            "Best validation loss: 0.25000611321461347\n",
            "\n",
            "Saving best model for epoch: 5\n",
            "\n",
            "==================================================\n",
            "Epoch 6 of 30\n",
            "Training\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "6393it [53:29,  1.99it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Validating\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "1126it [07:08,  2.63it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train Loss: 0.2523; Val Loss: 0.2343; Train accuracy: 0.4856; Val accuracy: 0.6040\n",
            "\n",
            "Best validation loss: 0.2342516876571873\n",
            "\n",
            "Saving best model for epoch: 6\n",
            "\n",
            "==================================================\n",
            "Epoch 7 of 30\n",
            "Training\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "6393it [53:23,  2.00it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Validating\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "1126it [07:05,  2.65it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train Loss: 0.2431; Val Loss: 0.2315; Train accuracy: 0.5421; Val accuracy: 0.6249\n",
            "\n",
            "Best validation loss: 0.23147968413907088\n",
            "\n",
            "Saving best model for epoch: 7\n",
            "\n",
            "==================================================\n",
            "Epoch 8 of 30\n",
            "Training\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "6393it [54:42,  1.95it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Validating\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "1126it [07:03,  2.66it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train Loss: 0.2374; Val Loss: 0.2245; Train accuracy: 0.5726; Val accuracy: 0.6251\n",
            "\n",
            "Best validation loss: 0.2244645284842799\n",
            "\n",
            "Saving best model for epoch: 8\n",
            "\n",
            "==================================================\n",
            "Epoch 9 of 30\n",
            "Training\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "6393it [53:26,  1.99it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Validating\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "1126it [07:01,  2.67it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train Loss: 0.2351; Val Loss: 0.2245; Train accuracy: 0.5784; Val accuracy: 0.6018\n",
            "==================================================\n",
            "Epoch 10 of 30\n",
            "Training\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "6393it [53:15,  2.00it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Validating\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "1126it [07:00,  2.68it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train Loss: 0.2323; Val Loss: 0.2268; Train accuracy: 0.5925; Val accuracy: 0.5412\n",
            "==================================================\n",
            "Epoch 11 of 30\n",
            "Training\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "6393it [53:03,  2.01it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Validating\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "1126it [07:02,  2.66it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train Loss: 0.2302; Val Loss: 0.2227; Train accuracy: 0.5969; Val accuracy: 0.6171\n",
            "\n",
            "Best validation loss: 0.2226692042768425\n",
            "\n",
            "Saving best model for epoch: 11\n",
            "\n",
            "==================================================\n",
            "Epoch 12 of 30\n",
            "Training\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "6393it [53:38,  1.99it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Validating\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "1126it [07:11,  2.61it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train Loss: 0.2288; Val Loss: 0.2232; Train accuracy: 0.6007; Val accuracy: 0.5976\n",
            "==================================================\n",
            "Epoch 13 of 30\n",
            "Training\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "6393it [53:04,  2.01it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Validating\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "1126it [07:08,  2.63it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train Loss: 0.2271; Val Loss: 0.2196; Train accuracy: 0.5988; Val accuracy: 0.6154\n",
            "\n",
            "Best validation loss: 0.2195812863749361\n",
            "\n",
            "Saving best model for epoch: 13\n",
            "\n",
            "==================================================\n",
            "Epoch 14 of 30\n",
            "Training\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "6393it [53:41,  1.98it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Validating\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "1126it [07:08,  2.63it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train Loss: 0.2255; Val Loss: 0.2187; Train accuracy: 0.6017; Val accuracy: 0.6258\n",
            "\n",
            "Best validation loss: 0.21869684002803866\n",
            "\n",
            "Saving best model for epoch: 14\n",
            "\n",
            "==================================================\n",
            "Epoch 15 of 30\n",
            "Training\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "6393it [53:47,  1.98it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Validating\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "1126it [07:12,  2.60it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train Loss: 0.2238; Val Loss: 0.2193; Train accuracy: 0.6014; Val accuracy: 0.6023\n",
            "==================================================\n",
            "Epoch 16 of 30\n",
            "Training\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "6393it [53:57,  1.97it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Validating\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "1126it [07:09,  2.62it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train Loss: 0.2225; Val Loss: 0.2148; Train accuracy: 0.5998; Val accuracy: 0.5952\n",
            "\n",
            "Best validation loss: 0.21480766724510578\n",
            "\n",
            "Saving best model for epoch: 16\n",
            "\n",
            "==================================================\n",
            "Epoch 17 of 30\n",
            "Training\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "6393it [54:16,  1.96it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Validating\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "1126it [07:11,  2.61it/s]\n"
          ]
        },
        {
          "metadata": {
            "tags": null
          },
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train Loss: 0.2210; Val Loss: 0.2136; Train accuracy: 0.6009; Val accuracy: 0.6060\n",
            "\n",
            "Best validation loss: 0.2136137347509605\n",
            "\n",
            "Saving best model for epoch: 17\n",
            "\n",
            "==================================================\n",
            "Epoch 18 of 30\n",
            "Training\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "6393it [53:49,  1.98it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Validating\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "1126it [07:06,  2.64it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Loss: 0.2186; Val Loss: 0.2118; Train accuracy: 0.6012; Val accuracy: 0.6040\n",
            "\n",
            "Best validation loss: 0.21180420065196115\n",
            "\n",
            "Saving best model for epoch: 18\n",
            "\n",
            "==================================================\n",
            "Epoch 19 of 30\n",
            "Training\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "6393it [54:20,  1.96it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Validating\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "1126it [07:10,  2.61it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Train Loss: 0.2173; Val Loss: 0.2099; Train accuracy: 0.6003; Val accuracy: 0.5958\n",
            "\n",
            "Best validation loss: 0.20991460148569321\n",
            "\n",
            "Saving best model for epoch: 19\n",
            "\n",
            "==================================================\n",
            "Epoch 20 of 30\n",
            "Training\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            " 57%|█████▋    | 3646/6392 [30:30<23:46,  1.92it/s]"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Evaluate model"
      ],
      "metadata": {
        "id": "Uaa5H3Sd4A3d"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model_path = 'drive/MyDrive/Colab Notebooks/Enet-ep19-val0.210.pth'"
      ],
      "metadata": {
        "id": "PEJ3zT6C5Aae"
      },
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# initialize the computation device\n",
        "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
        "#intialize the model\n",
        "test_model = model(pretrained=False, requires_grad=False).to(device)\n",
        "# load the model checkpoint\n",
        "checkpoint = torch.load(model_path, map_location=torch.device('cpu'))\n",
        "# load model weights state_dict\n",
        "test_model.load_state_dict(checkpoint['model_state_dict'])\n",
        "test_model.eval()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pLQXP0jCHHkv",
        "outputId": "1126c4e0-4deb-4dfb-9dc0-19c703e9004a"
      },
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/torchvision/models/_utils.py:209: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and will be removed in 0.15, please use 'weights' instead.\n",
            "  f\"The parameter '{pretrained_param}' is deprecated since 0.13 and will be removed in 0.15, \"\n",
            "/usr/local/lib/python3.7/dist-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=None`.\n",
            "  warnings.warn(msg)\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "EfficientNet(\n",
              "  (features): Sequential(\n",
              "    (0): Conv2dNormActivation(\n",
              "      (0): Conv2d(3, 24, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
              "      (1): BatchNorm2d(24, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (2): SiLU(inplace=True)\n",
              "    )\n",
              "    (1): Sequential(\n",
              "      (0): FusedMBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(24, 24, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(24, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.0, mode=row)\n",
              "      )\n",
              "      (1): FusedMBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(24, 24, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(24, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.005, mode=row)\n",
              "      )\n",
              "    )\n",
              "    (2): Sequential(\n",
              "      (0): FusedMBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(24, 96, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(96, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(48, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.01, mode=row)\n",
              "      )\n",
              "      (1): FusedMBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(48, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(192, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(48, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.015000000000000003, mode=row)\n",
              "      )\n",
              "      (2): FusedMBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(48, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(192, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(48, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.02, mode=row)\n",
              "      )\n",
              "      (3): FusedMBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(48, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(192, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(48, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.025, mode=row)\n",
              "      )\n",
              "    )\n",
              "    (3): Sequential(\n",
              "      (0): FusedMBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(48, 192, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(192, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(192, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.030000000000000006, mode=row)\n",
              "      )\n",
              "      (1): FusedMBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(64, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.035, mode=row)\n",
              "      )\n",
              "      (2): FusedMBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(64, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.04, mode=row)\n",
              "      )\n",
              "      (3): FusedMBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(64, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(64, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.045, mode=row)\n",
              "      )\n",
              "    )\n",
              "    (4): Sequential(\n",
              "      (0): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=256, bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(256, 16, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(16, 256, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.05, mode=row)\n",
              "      )\n",
              "      (1): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=512, bias=False)\n",
              "            (1): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(512, 32, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(32, 512, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.05500000000000001, mode=row)\n",
              "      )\n",
              "      (2): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=512, bias=False)\n",
              "            (1): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(512, 32, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(32, 512, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.06000000000000001, mode=row)\n",
              "      )\n",
              "      (3): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=512, bias=False)\n",
              "            (1): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(512, 32, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(32, 512, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.065, mode=row)\n",
              "      )\n",
              "      (4): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=512, bias=False)\n",
              "            (1): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(512, 32, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(32, 512, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.07, mode=row)\n",
              "      )\n",
              "      (5): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=512, bias=False)\n",
              "            (1): BatchNorm2d(512, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(512, 32, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(32, 512, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(128, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.075, mode=row)\n",
              "      )\n",
              "    )\n",
              "    (5): Sequential(\n",
              "      (0): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(128, 768, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(768, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(768, 768, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=768, bias=False)\n",
              "            (1): BatchNorm2d(768, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(768, 32, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(32, 768, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(768, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(160, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.08, mode=row)\n",
              "      )\n",
              "      (1): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(160, 960, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 960, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=960, bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(960, 40, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(40, 960, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(160, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.085, mode=row)\n",
              "      )\n",
              "      (2): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(160, 960, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 960, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=960, bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(960, 40, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(40, 960, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(160, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.09, mode=row)\n",
              "      )\n",
              "      (3): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(160, 960, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 960, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=960, bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(960, 40, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(40, 960, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(160, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.095, mode=row)\n",
              "      )\n",
              "      (4): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(160, 960, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 960, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=960, bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(960, 40, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(40, 960, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(160, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.1, mode=row)\n",
              "      )\n",
              "      (5): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(160, 960, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 960, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=960, bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(960, 40, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(40, 960, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(160, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.10500000000000001, mode=row)\n",
              "      )\n",
              "      (6): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(160, 960, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 960, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=960, bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(960, 40, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(40, 960, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(160, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.11000000000000001, mode=row)\n",
              "      )\n",
              "      (7): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(160, 960, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 960, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=960, bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(960, 40, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(40, 960, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(160, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.11500000000000002, mode=row)\n",
              "      )\n",
              "      (8): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(160, 960, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 960, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=960, bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(960, 40, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(40, 960, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 160, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(160, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.12000000000000002, mode=row)\n",
              "      )\n",
              "    )\n",
              "    (6): Sequential(\n",
              "      (0): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(160, 960, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 960, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), groups=960, bias=False)\n",
              "            (1): BatchNorm2d(960, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(960, 40, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(40, 960, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(960, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.125, mode=row)\n",
              "      )\n",
              "      (1): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 1536, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 1536, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=1536, bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(1536, 64, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(64, 1536, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.13, mode=row)\n",
              "      )\n",
              "      (2): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 1536, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 1536, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=1536, bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(1536, 64, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(64, 1536, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.135, mode=row)\n",
              "      )\n",
              "      (3): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 1536, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 1536, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=1536, bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(1536, 64, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(64, 1536, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.14, mode=row)\n",
              "      )\n",
              "      (4): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 1536, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 1536, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=1536, bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(1536, 64, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(64, 1536, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.14500000000000002, mode=row)\n",
              "      )\n",
              "      (5): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 1536, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 1536, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=1536, bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(1536, 64, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(64, 1536, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.15, mode=row)\n",
              "      )\n",
              "      (6): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 1536, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 1536, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=1536, bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(1536, 64, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(64, 1536, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.155, mode=row)\n",
              "      )\n",
              "      (7): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 1536, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 1536, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=1536, bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(1536, 64, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(64, 1536, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.16, mode=row)\n",
              "      )\n",
              "      (8): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 1536, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 1536, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=1536, bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(1536, 64, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(64, 1536, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.165, mode=row)\n",
              "      )\n",
              "      (9): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 1536, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 1536, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=1536, bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(1536, 64, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(64, 1536, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.17, mode=row)\n",
              "      )\n",
              "      (10): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 1536, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 1536, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=1536, bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(1536, 64, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(64, 1536, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.175, mode=row)\n",
              "      )\n",
              "      (11): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 1536, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 1536, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=1536, bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(1536, 64, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(64, 1536, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.18, mode=row)\n",
              "      )\n",
              "      (12): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 1536, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 1536, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=1536, bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(1536, 64, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(64, 1536, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.185, mode=row)\n",
              "      )\n",
              "      (13): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 1536, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 1536, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=1536, bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(1536, 64, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(64, 1536, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.19, mode=row)\n",
              "      )\n",
              "      (14): MBConv(\n",
              "        (block): Sequential(\n",
              "          (0): Conv2dNormActivation(\n",
              "            (0): Conv2d(256, 1536, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (1): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 1536, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), groups=1536, bias=False)\n",
              "            (1): BatchNorm2d(1536, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "            (2): SiLU(inplace=True)\n",
              "          )\n",
              "          (2): SqueezeExcitation(\n",
              "            (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "            (fc1): Conv2d(1536, 64, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (fc2): Conv2d(64, 1536, kernel_size=(1, 1), stride=(1, 1))\n",
              "            (activation): SiLU(inplace=True)\n",
              "            (scale_activation): Sigmoid()\n",
              "          )\n",
              "          (3): Conv2dNormActivation(\n",
              "            (0): Conv2d(1536, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "            (1): BatchNorm2d(256, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "          )\n",
              "        )\n",
              "        (stochastic_depth): StochasticDepth(p=0.195, mode=row)\n",
              "      )\n",
              "    )\n",
              "    (7): Conv2dNormActivation(\n",
              "      (0): Conv2d(256, 1280, kernel_size=(1, 1), stride=(1, 1), bias=False)\n",
              "      (1): BatchNorm2d(1280, eps=0.001, momentum=0.1, affine=True, track_running_stats=True)\n",
              "      (2): SiLU(inplace=True)\n",
              "    )\n",
              "  )\n",
              "  (avgpool): AdaptiveAvgPool2d(output_size=1)\n",
              "  (classifier): Sequential(\n",
              "    (0): Dropout(p=0.2, inplace=True)\n",
              "    (1): Linear(in_features=1280, out_features=11, bias=True)\n",
              "  )\n",
              ")"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# prepare the test dataset and dataloader\n",
        "test_data = ImageDataset(\n",
        "    train_df, train=False, test=True\n",
        ")\n",
        "test_loader = DataLoader(\n",
        "    test_data, \n",
        "    batch_size=1,\n",
        "    shuffle=False\n",
        ")"
      ],
      "metadata": {
        "id": "bUgrJlhhIfIT"
      },
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# get the list of label names from train_df\n",
        "tube_statuses = train_df.columns.values[1:12]"
      ],
      "metadata": {
        "id": "S6-Hkf5_JVxQ"
      },
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Run a Loop to Get the Predictions\n",
        "We will iterate over the test data loader and get the predictions for 1 batch (10 images)"
      ],
      "metadata": {
        "id": "kR9hnKmuKPWa"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def preds_preview():\n",
        "    for counter, data in enumerate(test_loader):\n",
        "        image, target = data['image'].to(device), data['label']\n",
        "        # get all the index positions where value == 1\n",
        "        target_indices = [i for i in range(len(target[0])) if target[0][i] == 1]\n",
        "        # get the predictions by passing the image through the model\n",
        "        outputs = test_model(image) # predictions\n",
        "        outputs = torch.sigmoid(outputs)\n",
        "        outputs = outputs.detach().cpu()\n",
        "        prob_indices = torch.flatten(np.argwhere(outputs[0]>0.4)) # adjust probability threshold here, need flatten to convert 2d to 1d \n",
        "        # print(prob_indices) # 1d list of indices e.g. 6 to correspond to label\n",
        "        string_predicted = ''\n",
        "        string_actual = ''\n",
        "        for i in range(len(prob_indices)):\n",
        "            string_predicted += f\"{tube_statuses[prob_indices[i]]}    \" #string concat\n",
        "        for i in range(len(target_indices)):\n",
        "            string_actual += f\"{tube_statuses[target_indices[i]]}    \"\n",
        "        image = image.squeeze(0)\n",
        "        image = image.detach().cpu().numpy()\n",
        "        image = np.transpose(image, (1, 2, 0))\n",
        "        plt.imshow(image)\n",
        "        plt.axis('off')\n",
        "        plt.title(f\"PREDICTED: {string_predicted}\\nACTUAL: {string_actual}\")\n",
        "        # plt.savefig(f\"drive/MyDrive/Colab Notebooks/inferences/inference_{counter}.jpg\")\n",
        "        plt.show()\n",
        "        \n",
        "preds_preview()"
      ],
      "metadata": {
        "outputId": "5af9103a-13aa-42c7-a869-f513ccaceb20",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "xyoD_ahut0gD"
      },
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:56: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbsAAAEJCAYAAAANa4lgAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOy9eZhcVdU9vKq6a+opc5okQiIkRgaJ+EhwAAKBV2aQVwWRUYMGUHFAAUHRgAEHZFR5ISCgQRFRQdDwE5kREBBxCIOihCmJTElI0umuHs73R759s+6qfaurSae6xbufp56quvfcM9+99tpnyoQQAlJJJZVUUknlDSzZoc5AKqmkkkoqqWxqScEulVRSSSWVN7ykYJdKKqmkksobXlKwSyWVVFJJ5Q0vKdilkkoqqaTyhpcU7FJJJZVUUnnDSwp2qaTyBpRjjjkGe+6551BnI5VBkEwmg4ULFw51Nv7j5b8a7I455hhkMhlkMhk0NjZi8uTJOO644/DKK69EYex+JpNBU1MTttlmG5x33nmxeJYsWRILx59zzz3XDdPc3Ixp06bhyCOPxH333efGd++998au33///Tj44IPR3t6OYrGIrbbaCkcccQQeeeQRXHXVVYl5sM/Xvva1Qc9rrbLbbrshk8ng+9//fuz6vffei0wmgyVLlsSu33LLLdh3330xbtw4FAoFvPnNb8YBBxyAG264AX19ff2WdcqUKa8rnyZ33nknMpkMJk+ejM7Ozti9PffcE8ccc8xGxf+fLtX64osvvoh8Pl/R1ibXXXcdstks/vGPf0TXfvOb32CvvfbCmDFjUCqVMH36dBx33HH4+9//vtF5tff85JNPjl1//vnnkclkcOedd250GqkMf/mvBjsA2GWXXbBs2TIsWbIEF110EX7+85/jqKOOioX57ne/i2XLlmHx4sX4zGc+g1NOOQWXXXZZRVw33ngjli1bFvuccMIJbpjFixfjkksuQQgBO++8cwWAqlx55ZXYZZddkM/ncc011+Dxxx/HT3/6U0yZMgWf+cxncOihh8bSPeSQQ6Ky2ecLX/hCXfKaJMViEfPmzcPq1aurhjvzzDOx3377YfLkybjuuuvw5JNP4uabb8ZBBx2EefPmYenSpbF8//znPwcAPPLII9G1hx566HXlUeXFF1/EBRdcMChxsZTL5UGPs17SX18cP348DjroICxYsMB9fsGCBdhtt90wbdo0AOvb+4ADDsDUqVPxi1/8Ak888QR+8IMfIJ/P48tf/vKg5LlYLOKiiy7CM888MyjxmYQQ0N3dPahxprKJJPwXy9FHHx322GOP2LWvf/3rIZvNho6OjhBCCADCj370o1iYd7zjHeF///d/o/9PP/10ABDuueeexLSqhfniF78YGhsbw1NPPeWGfeGFF0KhUAhz585143711VdrKtumyGutMmvWrHD44YeHzTffPJx22mnR9XvuuScACE8//XQIIYSHHnooAAjf+ta3EuPq6+uL/b/jjjsCgPDcc88NKE/VxOI89dRTw4gRI8JLL70U3dtjjz3C0UcfHf0vl8vhlFNOCRMnTgy5XC5svfXW4ZprronFByBceOGF4bDDDgttbW3hkEMOCVdeeWVoaGgIt99+e9huu+1CsVgMs2bNCi+88EK46667wtvf/vbQ1NQU9thjj/D8889Hcf3rX/8KBx98cJgwYUIolUphu+22Cz/84Q9j6SW1/8ZKrX3x1ltvDQDCQw89FLv/z3/+M2QymXDttdeGEEJ4+OGHA4BwzjnnVI1vY+Too48Os2fPDjNnzgwf+chHouvPPfdcABDuuOOO6NoTTzwR9t1339Dc3Byam5vD/vvvH/7xj39E97nN3v72t4dcLhd+85vfhFmzZoWPfexj4fTTTw/jxo0LI0aMCKeddlro7e0N8+bNC+PHjw9jx46N9f0QQrjmmmvCzJkzQ1tbWxgzZkzYd999w5NPPhkL4+mgVAYu//XMTqVUKqGvrw89PT0V90IIuO222/D4448jn88PWponn3wyent78ctf/tK9f91116GrqyvRyh01atSg5aU/8fI6ZcqUmtx6xWIR8+fPx/nnn4/nn3/eDbNw4UI0Nzfjs5/9bGI8mUxmwPl+vTJ37lxsttlmmDdvXmKY0047DQsWLMAFF1yAv/3tbzjiiCNwxBFH4LbbbouFmzdvHt7znvfgkUcewde//nUAQF9fH+bNm4fLL78cv//97/HCCy/g0EMPxRlnnIFLLrkEv//97/H888/j85//fBTPmjVrMHv2bCxatAh//etf8YlPfAIf/ehHcccdd2yaSiCptS/uscce2GqrrSrY3RVXXIGxY8fi4IMPBgD86Ec/QlNTU6x8XnwbK+am/8lPfoKHH37YDbNu3Tq8733vQ2dnJ+666y7cddddWLNmDfbee+8YE+/r68Mpp5yC8847D0888QTe+c53AgCuv/56dHd3495778V5552Hs88+G/vttx/WrFmDe+65B+eeey7OPvtsLFq0KIrL6vKRRx7BrbfeioaGBuy3337/0cx/2MpQo+1Qilq/ixcvDltuuWXYaaedomsAQqFQCM3NzaGxsTEACE1NTeGBBx6IwhgTKpVKkUVon/vuuy8WJolRtbe3h+OPP94Ne/zxx4e2traNKtumymsIIcyePTuceuqpVfMza9asMGfOnNDX1xfe8Y53hKOOOiqEUMns9tlnn7D99tvHnr3pppti+Vy4cGHs/qZkds8991y44YYbQi6XC3//+99DCHFmt3bt2pDP58P3vve92PPvf//7w+677x79BxA+9rGPxcJceeWVAUD405/+FF371re+FQCEhx9+OLp23nnnhTFjxlTN74EHHhiOPfbY6P+mYnYD6Yvf+MY3Qmtra1izZk0IIYSenp4wceLE8IUvfCEKs88++4S3ve1tg55PFq6L97///WHWrFkhhEpmd/nll4dSqRRj8cuXLw/FYjFcffXVIYQNbXb33XfH0pg1a1aYMWNG7No222wTtttuu9i17bffPpx00kmJeX3llVcCgHDvvfdG15Ayu0GR/3pmd+edd6KlpQWlUgnbbbcdttxyS/z4xz+OhZk/fz4effRR3HHHHXjve9+Ls88+GzvttFNFXFdeeSUeffTR2GeHHXaoKR8hhETGEjbBXt2DmdfbbrsN55xzTk3PZjIZfPvb38bChQvx6KOPJsbPsvvuu0d57Ozs3Kgxkn322QctLS3RpxY56KCD8O53vxunnHJKxb2nnnoK5XIZu+66a+z6rFmzsHjx4ti1mTNnVjyfyWTwtre9Lfq/2WabAQC233772LVXXnkFvb29AICOjg6ceuqp2HbbbTF69Gi0tLTgN7/5zaCPR3kykL740Y9+FJ2dnbj22msBAL/+9a+xbNkyfOITn3hd8bE8++yzsXY87rjjanrum9/8Jn7/+9/jV7/6VcW9xYsXY5tttsHYsWOja+3t7Zg+fXpFW+64444Vz8+YMSP2f7PNNou1o1178cUXo/+PPvooDj74YLz5zW9Ga2srtthiCwCoS1v+t0njUGdgqGWnnXbC1VdfjcbGRkycONF1T7a3t2Pq1KmYOnUqbrjhBrzlLW/BDjvsUKHgJk2ahKlTpw44Dy+99BJeeuklbLnllu796dOn47XXXsPzzz+PN73pTQOO35NNlddaZPbs2dhnn33wxS9+EV/96ldj997ylrfgrrvuQrlcjtqiubn5deXVk8svvxzr1q0b8HPnnnsudtppp4oZsgOR5ubmimvZbBYNDQ3RfzMicrlcxTUDhi9+8Yu48cYbcd5552H69Olobm7GSSedhFWrVr3uvNUqA+mLNlHlsssuw5w5cyomplh8d999d6y9a5GJEyfGjKW2traannvLW96CuXPn4pRTTom5EwciDQ0NKBaLFde5zYD17eZd6+vrA7DeaHnf+96HnXfeGVdeeSXa29sBANtuu23qxtwE8l/P7EqlEqZOnYopU6bU9LKNHTsWn/zkJ3HiiScOGuP69re/jYaGhmgcQ+VDH/oQCoVCNM6jsmLFikHJRy3SX15rlW9961u444478Jvf/CZ2/fDDD0dHR8frnvHZnxjI26dW2XHHHfHhD384NqMVAKZOnYpCoYC77747dv2uu+7CdtttNyh5Vrn77rtx+OGH45BDDsGMGTOw5ZZbDsoU/VpkoH1x7ty5ePDBB7Fo0SIsWrQIc+fOjd0/4ogjqrZ3Ut9ubGyMteP48eNrLsNXv/pVLF26tGJG9bbbbovHHnsML7/8cnTt3//+N5588slN0paPP/44XnrpJcyfPx+77bYbtt56a6xYsWKTeHJSSZnd65JPfepT+Pa3v41rr70Whx12WHT91VdfxfLly2Nhm5ub0draWhGmq6sLTz31FK666ipcc801+M53vpPIliZNmoTvfve7mDt3LlauXImPf/zj2GqrrfDqq6/ixhtvxB133FGhbPuTwczrHnvsgZkzZ9bsygSAbbbZBnPmzKmY1r/jjjvijDPOwOmnn46nn34aH/7wh/HmN78Zq1atwi233IK+vr4YE6qnnH322XjrW9+KbDaLQw45BADQ1NSEE088EV/5ylcwbtw4zJgxA9dffz1uvPFG3HrrrZskH9OnT8eNN96ID3zgA2hpacF5552HpUuXRsxgU8pA+6JNVDn88MMxevToCiPpne98Z9Tezz33HA499FBMnjwZS5cuxXXXXYcXXngB11133aCWYdy4cTj11FNx1llnxa5/5CMfwZlnnolDDz0U3/72txFCwBe+8AVMmjQJhx566KDmAQAmT56MQqGAiy++GCeddBKWLFmCU089ta4TsP6b5L+e2b0eaW9vx1FHHYUzzjgjNmvzoIMOwoQJE2KfL33pS7FnLczWW28dWbn33nsvPve5z1VN89hjj8Vdd92Fzs5OHHbYYZg+fTo++MEP4umnn8ZFF1004DIMZl7/+c9/YtmyZQPOw5lnnukC17x583DTTTfh2WefxYc+9CFMnToVe+65J+68804sXLgQhx9++IDTGgyZMmUKPv3pT1e4QefPn4+Pf/zj+OxnP4vtttsOCxcuxMKFC7HHHntsknycf/75mDx5MnbffXfssccemDRpEj74wQ9ukrQ8GUhfzGQy+PjHP44VK1bg6KOPdr0n8+bNww033IB//OMfeP/734/p06fjmGOOQVdXF84+++xNUobPfe5zsbE5YL2X57e//S0KhQJ23XVXzJo1C83NzbjlllsGdfa1ydixY7Fw4ULceuut2HbbbfGFL3wB5557LrLZVC1vCsmElDOnkkoqqaTyBpfUhEgllVRSSeUNLynYpZJKKqmk8oaXFOxSSSWVVFJ5w0sKdqmkkkoqqbzhJQW7VIaNTJkyJXH9ViqbXgbzDDxty9122w3HHnvsoMSdyn++JB1jtimlZrB74YUXUCgUMHHiRHeTZGBwzluzc8S8jYKTXsb+8jYYL9rXvva1xHy//PLLmDJlSr/l6++zsWekWR5tDRhLY2Mjrrrqqo2K/z9dFi5ciF133RUjRoxAc3MztttuO5x88sl44YUX8Mtf/hKZTAaPPfaY++wJJ5yAyZMnR7tf9PT04OKLL8bMmTPR2tqKtrY27LDDDpg/f/6gLPLX/mS7kTzxxBMbHfdQyC9+8YtNtlHAYEq1c/X+9Kc/IZPJVGyEYPKtb30Lzc3NsZ1sqvW5jZWBnhH53y41g90VV1yB/fffHyNHjsRNN91UcX+wz1sbiPSXt8GSKVOmVJwBt2zZMowZMwYPPfRQ9P/BBx8EED8z7plnnok9c9JJJ1XEd+GFF250HovFIq6//no88MADGx2Xyn/yuV1z5szBnDlzsOuuu2LRokV47LHHcNFFF2H58uX4zne+gwMOOAATJkxwz2Dr6OjAj3/8Y8yZMwfZbBbd3d3Yb7/9cPrpp+OQQw7B7bffjr/85S+YP38+HnjgAVx99dWDkudTTjkFy5Ytw9KlS3HzzTdj1apV2G+//TYqzr6+vmiPzU0hSdtcjR49uuYtvYZK+jtXb4cddsA73/nOxHP6Lr/8chx66KEYMWIEgP773GBIrWdEDlQ2dT8ZEqllt+je3t6wxRZbhF/96lfhG9/4Rth7771j9wfzvLVqO9h7z/SXtxA27Li/MfLVr341bLXVVjWFreXMuIHEV6tYnB/84AfDe9/73ti9hoaGcOWVV0b/ly5dGg499NAwYsSI6Bw1PnvM2uHmm28O733ve0OhUAjf//73oza46KKLwqRJk0Jzc3OYM2dOKJfL4ZJLLglbbLFFGDlyZPj4xz8eurq6ovh++9vfhlmzZoVRo0aFtra2sOuuu4Y//OEPsTxOnjw5nHXWWYNaJyGEcP311wcA4Sc/+Yl73/rn6aefHsaMGRM6Oztj9+0MMztT7txzzw2ZTCY6JSIpvo0Rry5+9atfBQCx+O+///6wyy67hGKxGEaOHBkOO+yw8O9//zu6b33i2muvDdOnTw8NDQ3hscceC6+88ko45JBDQlNTUxg/fnw4/fTTw1FHHVXxfl100UVh+vTpoVAohKlTp4avf/3robu7O5bP008/PRx//PFh9OjRYebMmW7+9R20/2eeeWZob28Po0aNCkceeWRYvXp1LP2f/OQnYcaMGaFQKITJkyeHz33uc9EpCoMptZ6rt2DBgtDY2BiWLVsWu2/vi52GUmuf2xip9YzIEF5/P5k8eXL48pe/HI477rgwYsSIMG7cuHDxxReHzs7O8KlPfSqMHDkyTJw4MVx88cWxvF1wwQVhxowZobm5ObS3t4dDDz00LF26NLpfi44cbKmJ2S1atAhdXV3YZ599cOSRR+K2226LUeShPG+tv7zVIscccwymTJmySfI3FPKNb3wDDz30UHSCt0oIAe9///vxxBNP4Oabb8aDDz6I9vZ2/M///E9sX0AAOOmkk3DKKafg8ccfxwEHHAAAePDBB/Hwww/j1ltvxU9+8hMsXLgQBx54IO677z7ccsstWLhwIX70ox/hiiuuiOJZs2YNTjjhBNx///247777MG3aNOy999545ZVXNl1F/P/yox/9CFOnTsWHP/xh9771z2OPPRYrVqzAL37xi9j9BQsWYL/99sOkSZOi+GbPno13v/vdVeMbTFm5ciV+/OMfY+utt47iX758Od73vvfhTW96Ex588EHcdNNN+Nvf/laxm8rSpUvx/e9/H1dffTUee+wxvOlNb8KcOXPwxz/+ETfddBNuv/12LFmypOI8xa997Ws499xzcc455+Dxxx/HhRdeiEsvvbTibL+LLroI48ePx/33348rr7yy5jJdf/31ePXVV3HnnXfi2muvxc0334xvfvOb0f2rrroKxx9/PE466SQ89thj+OEPf4jf/e53NZ9wMBCp9Vy9ww47DKVSqaKcCxYswPbbbx+dhlJrn9tYqeWMyI3pJwBw8cUXY9q0aXj44Ydx4okn4tOf/nR0UsNDDz2ET33qUzjxxBMrhgDOPfdc/PWvf8Uvf/lLPPvss4l1UTepBREPPPDA8PnPfz76v9dee4XTTz89+j+Y560NlNn1l7cQ+md2p556apg9e3bV/H71q18NmUym4gw4PXsthKFndiGE8NnPfjZMnTo1lMvlEEKc2f3ud78LAMLixYujZzs7O8Nmm20W5s2bF0LY0A7eCdjjxo2LsbZ99923ghEdeOCB4QMf+EBiXnt7e8PIkSNjZ9NtKma39dZbhwMOOKCmsHvvvXfsHLrHHnssYrgmpVIpfPrTnx70fLJMnjw55PP50NzcHJqamgKA8OY3vzk88cQTUZgvf/nLYdKkSbG2ePTRRwOAcNddd4UQNvTbZ555Jgrzj3/8IwAIv/3tb6NrXV1dYeLEidH7tXbt2lAqlcKiRYti+br66qvDiBEjYvn03p1amJ2+O8cdd1x417veFYvjkksuiYW56667KtjtYMhAztU77rjjwpZbbhn6+vpCCOtZWrFYDN/97nejMAPpc69Xaj0j8vX2kxDWt8FBBx0U/e/t7Q2tra1h//33j10bOXJkBbtjeeSRRwKAyDsyLJndCy+8gF//+texyRNHH300fvCDH0STQcIQ7ThWS95qkXPOOafiVGlPNt9884oz4DblGKEJn9u1zz771PTMV77yFbz88su45JJLKu4tXrwYY8aMwTbbbBNdKxQK2GmnnWo6g23rrbeO7RW42WabYfr06SgUCrFrfG7X008/jSOPPBJTp05FW1sb2trasGrVqmF3BtsnPvEJ3HnnnXjqqacArLfYt9hii1i9v97+fs0118Ta8pprrqka/pOf/CQeffRR/PnPf8Y999yDrbfeGvvvv380PrN48WK8613virXFjBkzMGLEiFg7tre3R+ekAYgs8Pe85z3RtXw+HzujbfHixVi3bl202bR95s6di1WrVuGll16Kwnp9pBbR898mTpyIf//73wDWHyX1zDPP4POf/7zb/619BksG2kf+9a9/4fbbbwewnsVls1kcccQRrys+lnvuuSdW3lr2Bu3vjMjX2084rEk2m8W4ceNi5/Rls1mMHz8+9r7feeed2GuvvbD55pujtbUVO++8M4ChPaev31MPrrjiCvT29lYc7Nnb24ubbroJBx988KCet2aDu6tWraqIa+XKlbFzpGrJ22BKLpcbtHPVBiLcgUulUk3PjB49GqeffjrOPPNMHH300a87be8MtoGe2wUA+++/P8aOHYvvfe972HzzzZHP57HzzjvX5dwu7/DNJDnggAOw2WabYcGCBTjrrLPwwx/+ECeeeGJsc97p06cnztqsJgceeGDs0N/+TikYPXp01N+mTp2KK664AhMmTMBPf/rTAc0u9tqwP7G2+9nPfoa3vOUtbt42Jn4AFZsrc5+x7wsvvBC77757xbODda6jyUDO1bOJKpdddhn22GOPiokpFl+tfY7lne98Z+x953quJtXOiKxVktpxoO/7s88+i3333RdHHnkkzjjjDIwdOxbPP/889txzzyE9p68qs+vr68MVV1yB0047rYLRHHbYYdF5UIN53tq0adOQzWajGY0mPT09+NOf/oS3vvWtA8rbG0H43C4bN6pFPv3pT6O1tRXz58+PXd92223xyiuvxBR2V1cX/vCHP2ySc7ssrVNPPRV77bUXttlmGxSLxZgluCnliCOOwFNPPRWdmK3C/bOxsREf+9jHcNVVV+G6667DypUrMWfOnIr4br/9dtx///39xsfS2toaa0s+TqkWsRMi7NSFbbfdFg888EBMgfz5z3/GqlWrqrajMfr77rsvulYul/HQQw9F/7fddlsUi0X861//iuXZPpv6mKX29nZsvvnmePLJJ930vcNTN0YGeq7e3LlzccMNN+Dmm2/GX//6V/ecvlr7HIudr2mfWsEOSD4j8vX2k9crDz30ENatW4cLLrgA733vezF9+vSIsQ+lVGV2ixYtwnPPPYe5c+dW0NtjjjkG++yzD5YsWYIpU6YM2nlrLS0tOP744/GlL30JpVIJO+64I1auXInzzz8fL7/8ctSpBpI3YP3ZbErx29rasOWWW+JLX/oSHnzwwX5dmb29vRVnwAHrj+pobBx+RwMWCgWcffbZ+OhHPxpjWbNnz8bMmTPxkY98BN/73vcwYsQInHXWWejs7MTxxx8/6PkYNWoUxo0bhwULFmCrrbbCK6+8gpNPPrlmlrqx8sEPfhBHHXUUjj76aCxevBj77rsvJk2ahKeffhpXXXUVRo0aFVNyxx57LM455xyceOKJsYkpJp/5zGfw//7f/8Nee+2FM844A7vtthvGjRuHxx9/HP/3f/+H3XffHZ/5zGc2Ot9r1qyJ+tu///1vnHXWWSiVSthrr70ArD9X8cILL8QxxxyD0047DStXrsQJJ5yAXXbZBbvssktivFOnTsWBBx6IT37yk7j00kvR3t6Ob3zjG7Hp6y0tLTjttNNw2mmnIZPJYM8990RPTw/++te/4k9/+lNsIsmmkvnz52POnDkYNWoUDjroIORyOTz++ONYtGgRLr300kFNa6Dn6h122GH4/Oc/j6OOOio2McVkoH1uMCTpjMjX209er0ybNg2ZTAbf+c53cPjhh+PPf/4zzjzzzEFPZ6BSldlddtll2GmnnVw/7uzZszF69GhcfvnlAAb3vLXzzjsPn/vc5zB//nxsv/32OOCAA7B69Wr84Q9/wLRp0wacNwD45S9/iR122CH2OeGEEwAAy5Ytwz//+c9+87VkyZKKM+AmTJjg+smHi3z4wx/GjBkzYmMImUwGN9xwA9761rdiv/32w4477ojly5fj1ltvrTjjazAkm83iZz/7Gf75z39i++23xzHHHIPPfvazmDBhwqCnlSRXX301LrvssmgsYeutt8YnP/lJtLe344tf/GIs7JQpU/C+970PK1aswCc+8YmKuHK5HBYtWoSzzjoL1157LWbNmoW3ve1t+NKXvoSZM2dulNuY5Zvf/GbUx2bPno0VK1bglltuidyK7e3t+O1vf4vnn38eO+64I/bff39st912uP766/uN+wc/+AHe/va3Y//998esWbMwadKkCrf/V77yFZx33nlYsGABZsyYgZ133hnnn39+3WYuH3nkkbjuuutw8803Y+bMmdhxxx3xta99bUDejYHIQM7Va25uxuGHH57YR4CB9bnBEu+MyI3pJ69Htt9+e1x88cW49NJLsc022+Dcc8+tAOChkPQ8u1RSSSWVVN7wku6NmUoqqaSSyhteUrBLJZVUUknlDS8p2KWSSiqppPKGlxTsUkkllVRSecNLCnappJJKKqm84WX4LQ57A8pVV12FkSNHAli/GP61117DAw88gNtvvx3Lly+vODrHzjAbDOlvsq3e98InxaHLGbLZLBoaGqLvxsZGNDQ0uJ9sNotcLheFy+fzyOfzKBaLaGpqQrFYjK43NjYim80im81Gv+3b0s1kMlG8fA6ciV233/at6yM5Dvvu6+tDCCG2S0QIAQ0NDeju7kYmk0FjY2MUPz+byWQQQkB3d3cUj33sXk9PD3p7e2P3LQ1OD1jff/j4lZ6enug5vW9x9vb2oqenB6tXr0ZHR0cUr6XFYS08f1u8Ftaet2t9fX2xtrVt46fVmBwAACAASURBVMrlspsPi8PqCVh/fFS5XEa5XEZ3d3esb9nv/t4Lu9fQ0IC2tjaEELBmzZpYPWaz2SjtXC6HMWPGoK2tDQ0NDbF66e3tjdonm82iVCph7dq1WLFiBfL5PHK5HF566aUob/YOW5k6OjqivlGr8FrYVAZfUrCrg4wcORKjR4+OXqZMJoN8Ph97ee0F9BR1NekPzDQeVSL2QvcXhxdGgY3BjQGJr/F/BrnGxkbkcjkUCgU0NzejWCyioaEBuVwO+Xw+FqfGx+Bka4xYkWqdslLM5XKxsjGwWhqmnFnpGtiVy+VYHrjO7NPX14eenp4YiJn09fVFQMgAUu3DYRUkGbz4ngGPKXWNx+Ky6wzAek/B0erN2jSXy0XGTHd3dwzkLH2rG2tLC1Mul7F69eroOU6jvz4KIAKmYrGIVatWVbS5heE64+2vuru70dHRgb6+vqgM1kYjRoxAR0cHenp60NLSgsbGRnR3d0ftb/kcTGM1lcGTFOyGQHp7e6OXWdmRib2EGyv9sbJqQOflTVkbgxwDBLMdZl78bKFQiJQjx2Ufjc8DOAUyVqT9icUDIKao7L8XDz9j4GvWvzI/rWMGP02zGrhxvjgeBnRmd6zglW2aUjdQZKXPTM0+PT09EXtMYo1W/kKhELUlt7sxNWW89t/q0PLU0NCAjo6OWBqcptc3rR4aGhpQLBYjZsVhPS+GAWyhUEA2m40xT2PPDQ0N6OzsRKFQQKlUik4iZ2avbZTK8JMU7OognmVeLpdjFutAX5RaXIt6zWNo1cBOFZKyN/1weGVcHEcul4uUon4s/iTF6AFdUv2xweCBhrWHgarHglnBeqyb25QZg8Zldc+gpGnVYnh4Lr0kt62JgRW7bQ3M1aXJYNLX14dcLhcDO3VFGgiFEGKGC4BYOzY0NMT6u5XD7lk7mPvQ8s1sUl2nDLYWn3kEzH2pRoPVg+Whp6cHnZ2dkduc02QPQE9PD9auXRv13b6+PuTz+Wi/SWOxA/XMpFI/ScGuDqJWvikMdc/0x+ZqZWFJ15N+s7B7SQFOGZyyLbWw+RrHY25JZYPmVmQ3IrM8D+CSPknl4npmRqX1b+Xg8SVgAxszhe6xIq1fZnQWTpkgx2XPsJuZDRX9reWzcSm+bmDErNKYixpi+t+YlzI67se9vb0R2PE4qLU5gxK7dS2f3NbGBru6uiryYuWzumMQzOfzGDFiBFasWFEBgixanwZ4pVIp1q46rmwst1QqRemZgdPY2FhRnyngDS9Jwa6OwtYlT1io9blar3v3ktylbL167E3H3RjglH1542ee25Ov628b/OdwnNckF91A6lCfY5bhAQ6HYSVm5eSzE716V6C03wq2Xn71vgIdg6GWT8XajsHe8mLp8EQNTt9Yngd4yu4KhUIsDwwUFpZBhQ0AYD1Lam5ujvqTuf25nXSSUkNDA9rb22Nsslwux9JkY4Prvre3F11dXZGhlc/nozzyGHAmk8G6detQKpVQKpVi7WLMjsdfUxlekoJdncWsWp1x5kk1JVhLeFXMJqYcvHEyZVv68dyJwAYFxICYNKnEm7DC15PcmEkA57ko+b9a8gBiEzUsrLLv/sCErysoer+T3J9J96rFx3mxPHqTJEzxWptw/OZaZJBX0LO88dicpWNtxuBnfSGXy0V5MLDr6uqKwIDLa89bfkulElpaWtDV1RW9LzZpxeKz8V5jgq2trWhvb8fy5ctRLBYxcuTIaFycP+yCDSFE8Wj/7e3tRT6fxy677ILXXnsNDz/8cNQvVq9ejVwuh9bWVqxatQqdnZ0V/SaV4Scp2NVB1Dq3QXGdoMLhPWXeXxpJ/5lpeRNLmL0xyCjAJbE3dn0qaKkVzh+dfMJgmzQBxcqj7MRTMEmuJHWNMXvkMEn3ent7Y6yDAcbuWTjOmwdoSe7DpPZMAuiktrfwWg6dHMNl9tIyQOSxSfs20OSPtan1E3NtKuAq4BuY5XI5lEqlCJyMVRnzstnMFufo0aPR3d0dTSSxmbI8tsfgZ3lqaWmJ4hs9ejRKpVIUz5gxY/CRj3wEK1aswNKlS7Fu3Tr09fVF7tUxY8Ygk8lg6dKlaGxsjPLLrDgFvuEjKdjVUezlNmbHCg6oPk7jxZV0XcGtGntjhWQKsZp7kgGK73nMzZuhqc9438wYaxFlV8rGPBcmAxnPvFTXIitiAzdW0uVyOZauMisGPG1Pdqv1N8bklYEBgtNOAkYtLz+vv5VZajwGGDpRhVmgtaelZwaPgrvFq16BfD4feUAaGxvR1dUVTQjhMjc1NaG5uRkvvvgiMplMbCmLDRnYmCIvlWhra8OYMWOQz+fR1NSEUaNGRQfqlstltLW1oVAooFgsYuzYsVi5ciVWrFiBcrmMrq4urFy5EhMnTowMVzupnCfypDJ8JAW7OgkrSPbt1+qW7I/dKcDZuFd/4MbK0gM7DuO5Hlk5eeBmCg5AYlzV3KLMilgRKxgquHlAx/eYWXjMi9PTsR5z8ZkC5rL19PRUzE7VZQFJ/xWkkgC/lr7g9Q11X3qAxmzZc6UmuUq53UJYv67PJnfkcjmUy+WoTiwPXFZmoDajsru7Gz09Pchms9FCdQNVezafz2PMmDHo7OxER0dHrC8bmywWixXeA2D9qeDNzc3o7OxEZ2cnVq5cGU026e7uRqlUispjgNjY2IhyuYx169bhtddeixbDr1y5Eg0NDZFLM5XhJynY1UHUZWMvSDVFXE1ZA5WzJg3cdEp/kstRgcz7KAgxsCkTqwaopjA0LgVUEwU4D4RUNCzXkxcOqHTlMdhrnfOYFI/nmfXOTM7iYXDk2ZrMGrms6sKsxa2pZdO6SKoXjp/z6gGtPq8grmNuXh+zdWy6zIFd+bwbDIOaCY8FWr5GjRqFpqYmLF++PMYmecMCc2t2d3dj7dq1WLt2LdatWxctT7AlBTbDsrm5OWKWllapVEIul8P48ePR0NCAlStXYu3atVi5cmW08cG6desilpeyuuEnKdjVQRTsbEskj9n15740MLH1Pmw9MxABcIHF+91fOL7G6+A8sPMA0suPB6gmrFz7Az6tp2psSMOzO8/+K+BZuazt7JqVh/Otz/J9BS5OU/OetBwhidHqfwYsnl3JbFUZmv03gPHylQSw1q42m9H6h7GkYrEIANEYms1+tPE+yyOv6eN8MhvN5/NR2oVCAaNGjYoYYKFQQKFQQFNTE5qampDNZtHZ2Ym1a9eio6MjcoPau8fvEufZylAsFqOyFYtFhBDQ0tKCnp6eCMDNNVoul9HU1BSxPq63VIaHpGBXB2EF1NvbW7H/n47HqGuN2Vt/DI6fY2XruSx1nI4VfBJDUzDj/7pjSrVPrXnicF69MthwXSoT8cIwS/EYo2cUeNctHv7NIM9Ao32C29z+MwPy+pAnXhnUC6Dhue6U2XnpJz3LfZH7n03fLxaLsQktvMi9t7c3tmWb1ycZ8Nh4GDFiBHK5HFavXo3W1lYUi8UorXXr1qGzszP6tvfN+ikbNgZklj8DbltCEcL6WZu9vb3RLioGnNlsFiNGjIjcttlsFsuWLXPbKJWhlRTs6iRJzI7vs3I3ULN1Z7z2TN2THtAx6FicqrQ9luWxNg+EeHadsjsFNAXvJOaorlbOt1dP/D8JDD32x/mopb00XVOGlmddTMx7QPJ1dVN7AJYEeJ67UZ/xyuiBPd/nb8u3Mt4kZtnQ0BC5JxlIOF88W9UYMU8e0XYoFAoxdscGBuelsbERY8eORSaTwZgxYyKAs6UA5mrmvslplcvlaDw5l8thwoQJ2HzzzaP1hJnMehepgTbvrGLltMX2xWIRo0ePxpo1a7DZZpthzZo10USaVIaPpGBXB1E3prldWNHZy2NjDLzPoC4NUDeZAon99hiWN7XfA84kV6fH6Kq5QlWSmJyXVwvvSRJL0jD9Xbd4tG75ty66ZobhMSplldxe3jidhbVr3rKAJOBK8g6oscHjhhyHMmN71gNqBT0DCjaQlB329PRESwY4XmNKFpZPjeBxUM6XhWlubsaoUaMwcuTIyEW5Zs2aiMFlMplodxNLj/fJDCHEjMhMJoPtt98e06ZNqxh7BhC5XC0Ptlm5gXJDQwMmTJiAF154AQ0NDdhss83Q0dGB1atXI5XhIynY1UFUUZj1aAPn2eyGtUM22K3jH2rhKjjwPY+tKRusBopJbkgPnHTSSzWg4ng4ba8syoI4Du+3gojF4eVD24brhvPJcXvjW0mgzgpfQdTiS2J0mjcui7K6JEbH+WSgAzZMrvHS0b7BedSJJZw3dWHqRso2Tqcu76ampph70eK1RegK9vl8HqVSCa2trWhqasLKlSuxcuXKyJ1YLBajrbx4P09jm3zNytvb24vOzk785S9/wdq1a5HNZqNxP5sVumTJEuTz+Shvvb29aGpqikC5VCph/PjxeO2117By5UpkMuvH81KwG16Sgl0dxCzCbDaLlpYWbL755pg2bRr6+vqwZs2aCPx0k2R171lcACoAw34zgFQDOA3PYKMAZul5gMd5UKWu41IeuCUxOn6G42PRnTg0bBJw8rflx8urgqbndvTymwRMuuRAgYPdplpmj3GpePlnFqZMU40BrhON38CIf2t+edswjZ9nWZqYK5Pri92Xdq2hoSHaVcXepbVr1+LVV1+NJq3wyQmWJ3Yl2vvDM0DN1RxCwHPPPYfXXnstmiU9cuTIaJKLvZ8hhGgcr7u7G01NTVF85XIZzc3NWLNmTXQMUCrDS1Kwq4PYold7UUeNGoVx48Zh+fLlkbWpa+N0xxGgUvkrWCS5Gz3m5YEYp+ExHWV1SYzMAwSPPaoLjQHKKyuHUTDjbwWGaiCYdJ2BgcvkKUqOQ38nuRl17R6wQSHr5tNeHWicnK4yNWWlXr0w0Hh1qAvRre2SJtJwepo/rsNMJhO59LXP2a4kBjoGXrbYnNeQGnvjfTSbmpoQwoaTDQyA2B3a2tqK5uZmtLe3I5/Po6enB2vWrInG66zu8vk8urq6IlDt6emJLYPI5XLYYost0N3djdWrV0f9Q9swlaGTFOzqIOai5MFtG5crFAoxZqdH6CS5K1lBKqtKYmf83R8IanrMLBUQPfaicXpjQ/ytaXLaeo+lGsPheC3/dp/D8Mw8C69Km4Vdgd6pCPpf02TGwkCq7jwdu2Mg47i1zGoMcB0w0Gp4rUOvzyXVDRswapDwQm4bU7NF4nysDqdjm0EXCgV0dXVFa+JsSy8zENlYtHzwbGXeIozdmADQ0tKCtrY2NDU1YZtttsGLL74YpbPFFlvgxRdfxLJly9DU1IStttoKf/zjH6O+0tLSgqampghcu7q6ovxyXaeAN3wkBbs6iMeE7IW0gXpdXmAgxodtKgBo3Dr1WxkbP6NxWBh1d1WbOKIvsYIc54/j91hXEijzt4GBpmdhPODTNFlhm+JLYjsWRhmfTTJiYLJ8qcvTi5Pzrm5NbSegkj15IMjlTXJPantqHrwycNm1nXQBOfcfHpuzNI0NMZDreJ+yOVuzNmLECDQ3N6OtrQ0hrB876+joiNLi2ZFsdAGI3qFMJhNbAG9G6Pjx46Mjfl566aVoCcPmm2+OlStXYurUqWhtbY3W9Znb0ibflMvlaB2fjfnpsEMqQy8p2A2BGNjZRBTexsgmqChjY2XD4w88AcUDuCQwqeWeBzqquL3xLqAS5Pg7qU6S0uUwnHYSg+H7ds1zdeqCa46bRV2zpjCz2WzkTuN7vOyAAUnzyeNVCpLcltlstup4GBBneVpmBVAFeK03jzF6dWltbOWwXUcUwKxMxsRsQ2Uz7NgtaG5FK3tbW1vs/ejrW78Rs71Dxgp59xVgw/IGqzuvHDYWOHr0aHR2dqK1tTXaKiyE9UsKpk6dihEjRsTO1rPjingdX1dXF7q6ujB69Gi0trbGPAmpDA9Jwa4O4iltUwzm3jQloee48TeDGFuOCnae8tdvD0j5uoKF981pMLvxWJ/naq02OcVLhyVp3EnZJZfHc2/qBBUtAz9rO/CzywxANHMQ2HDSAbt+PWNA3ZcKYlp39iy7RdVl7IG0V49cXmVvWmagciIQx81pZLPZaIstj22aK5ENITPubHamrduzMGZYWH5sO65qM1q1L/H+rAzMhUIBLS0tWLt2bbQofOTIkVixYgW6urrwyiuvYMKECXj22WfR2tqKNWvWRPtndnV1xcYBzUPT2NiI8ePHY8mSJbE6TWXoJQW7IRB2vdiMNHbFeDMdWUGwy9Lu8xiBZ7Xzs3rNYx0eaCa9tJYHVpZ2XZ9VkLM8KXNISk+VrI4F8iQaAwYGCI1L88pgpZNxOJ/8TNKsUK5fTpOVtdajMjUFSnU/MxNjlqr9wDNqPMDwDAMudxLAKNNm0OQxM25721zZTgjXPsAnKRh71lmdvIidy2j3Ozs7o/eDwdvq8dVXX8WKFSuQyWTw6quvRicaPPnkk+jo6MAzzzyD7u7u6DQEi4fX4VldGtMcO3ZsFD6V4SMp2NVBPKvdGJltXcQTUuzlZNDzAI6VMKelysbSU2Vp8XL+dKIGP5tUNgVHj61p+h77suseu1HRuLSO+Dr/9gDPXI9J5TRDhMvK6x+1nNyWBj66Tk3LwgYDA4dXbi4HA4FuVJBUB1wuHv/jOvFYkuY/iUlZvHw6OICK0+ptAwV7BzhublP7GIO2sWxm2AyQlqZNhtG42VDiOuDnOzo6ojV4mUwmWpTOY5AMyDy71o4P8uo8laGTFOzqKJ5CNEWlOzco8CnQKctQJaFMzq4lMTZlOTqRRIElaexIgS4J+DyQSAKnamBrogvDNc8M6joTkV1cXt40T+z65HtaJ8zg9CBXrlcPaLz69eqE48nn87FZh8z0FDxV4bOoi9Sra4/h8cJ1Zq7Wh81daWnaLife+kN91sDE3gUDd7vO25Lxu8CbSzOoZTKZigOUrQxmNOg7ZEDGdWv54zovl8vR2XapDB9Jwa4OwsrGPmad2gA9g5xdN6vXFLm+lPwCe25ETltBTUGqFlD0ysIuPG8MzotLFWwS+Hp50fwr400CRY+BWvzKqDSMMh1T2Oq+VJbgGQbMDDV//OE69srkAZ7ljafcewyR4+S6YyPAJnawaJ4U7BgAuE/wuJxds7Vu3Ie53hlIbNII17Glx8sWvL7suVk5jD1voGf50bA8Nmfvq+XTPAPG8srlcqxOU3Y3PCQFuzoKu2BKpRJGjhwZWYUGcMViMZpiXSqVYmMS9mLZbDA7F8/22rQJFEDlxAG7pmyK89bfdf5vioHHQphxemyuPzbB4frLmwGGt9xC47S6649J8scbp+I0rexm6Vu+1fXMeVGQY0bD5VfmxIraU5wKqplMJlogneTatLh0fFg9BAreWida15xHW09qM46tTnTxvOZd28n6ubqEObzVveXZJpHYczpGWs11zv89ALc2sjWyBnL8MXBOgW74SAp2dRB+gXO5HNra2jBp0iQ0Nzejo6MDwIap0mbxshWtbiWbyGJHk/ALZxY9AyA/qxMcOI/6W12X/JuVi11Xiz6pLjw3LC/UVhZXy/NaZwoAJp7iN2WmQM3P6LfnarTfzLbtW5mbKmwWHdPTfOlzHB/nRRdWa/2Y0maQ0LpWxqf3LH0FAj6pg1kkL/dIYrI83mZuQXUxW1rs0bAy2CJv6/vcBpp/bWtjndwW9hxvIJDL5aK1gHYorJ3GzkZnKsNHUrCrg/BLqdftxTKGYIuVeZIKT5fOZOILY5WBMPNTFsjniFn6rEBV6XoKIuk3A6lazMoyFawUOFWZc/xsDGh+VdlrPpPKwm4sZVFJgMr1CPjT81Wh238PvNQlrfnTcnnjijoBw/pMY2NjtAia86Vl4x1WTLgPVhvHszyZ+93G45h18dibGiPWp7u6uqJ0bVmO5Y/L7DE7e39sf8tqk7e4r/IQglemJCPI6rK3txf5fB6rVq2Kvb+pDC9Jwa4OYi+LWt+ea0snpZgC0vOxvPgARIvVLX4ewO/p6UFXV1dk6bNy1Be9FpDQ8nn31Wpm0OI8WlgPrKyOVIFqGlYnDBxJdeblld2s1ZS6x7yVNSkQGqCysmWA5RmLXp60ndkt6zEeft7qj4HAU94MeGqkcL0mGW22bpTXtQFxRsd9zfJhC7TZgDNWaCcacL2rkWTxmkdD3ymv7TjPtsOKlo+NM68sFtZcliNGjMCqVasq2GEqw0PSFqmDqBVqYxktLS2R64OtR3YrMVvT5QgmrJzMnQlsGB/hLcdsp3Yd70tSpOoS9MCsGtOyMJxvc8PazDedHcliioMXK3uAZfXHY2p61Ixu4aQAou44Zm/VXI+eFc8AyGmx4cPKnUHcc+tx/SiDrsYI+b+5NXksL8kdrIYB169OKuGdgLgcbKhwXVmZenp6onVtZghYPHx6OfctNpjYkON1fJ5oHOaG5D6VxBq53rWN+LlSqYSenh50dHRUAHoqQy8p2NVJWOnxQHYIoWJXCX3pvMF1C8fxq2Jkt5m9fKaQDHBZUbCrk19oE+93NaDzmAcvt2B2qZNoLG4+YFNdaVxnynTY5WuiCsjue4DHdc/CLmWP2Wl7e4YCn6LNM201Lp3Cr4qW24LzxWOF2ne4/b2y23PWbzSMxcHgZuXgtmDvBNdbX19fZGR1d3dXTIgBNuxl6Y3jJrE8Xnyu7E/7op2Lx4DJbL4/5u+Fs7B2qGvK7IafpC1SJ7EXnXdJN4sW8M+LU5Cwb0/ZmrDi9MaUkqxkBj+z/HV2p8fq7JuVC19TxspHsXiKjuNmoLN4vcXfCoKs9NSdpeOQPDNQ41HgYcVv9W9jNNa+DI46gcfCMEgwyHNfYePF2sTb49HrL7wTj4IEuzW5PrwJI5YOT8ywuuUlMdz2PNnH0uCTB3TiiPYhe87C63vBhiAzTWP+2set/ixfttE0GwScd+3PLFw/OgvYPBW2UN7290xZ3fCRFOzqIPyShBAi1435+i2MKVB2RbJSYquWlY/G71nr/J8VgoZl16e5mtjVWY3d8W8PWNmFx3nTfBjQ2X6hOvmDLeykcSRTlJY271+p7jtmRDz5x1OYlg67SHVqutaJlYHrl5WlhbeymHuPFz2zMmaGas8aYzZw6uuLH3vD4bTfcN/S+uQ2080O1BthYMJGmZXDAE77kfYPm5RiTI2ZqvYxBlY+W477iRlMNvbHy3mUySoL52+vr1pb2ni65bWpqSk2CzWV4SEp2NVZ7MXh3d+VAbELzV54W6+kFi4QV378Yib9NlHQY6veFIABHy9nsGe9+HRMkZWpt/MGP8vpMdBZPJwGg7ayNWDDqdnqXuN82bMcPx8HA8QtfzNOuM04T/yMTmRghqKswMLYOKqyHxYDWC6D5ckA3sDdxtIYyMxlzqCloMngDCCmuJNYuNcvrSwK1p5RYAaO7S9p9cgMUevZxAwzC2NMmPteX1+f6yngNvDE8s1GGue5oaEBxWIxYq8hBJRKpWjM0fOIpDI0koJdHUTdGWyJ24utoKTWtikNm2nJ8fHYhlrbLElAo2NGHvBZfnm80ZsWrunx89XqhZmBKWh2CyqrZbeiuuFYURtwGgBoHMy2WOkxkKoxwWCrblplIzrVntva7pt7z8axmP14zNzaXOuQFTMDFs+OtLBWF8wIWXTsSw0ibjM2CKx+rH+YS9La0wM6ZWdmJDAz57bgujHwtva18tpEHHvOdiPSd4PrzvvP32oY2MeMis7OzqisdjhtKsNHUrCrk6g1aL59VmjqPlEFacyCp+DbPc+Vx2mrsrPf6srx4mIGybPxGGwsH/wcMwizjFWJcLys1DSMKXdL11x8XD8ctylbmwRhRgKPA7G1bvlgha1uLR3nUdDjsDyT1lOcXBZjdDw2puVn40OZP7eR1rWdzaZjmMzoebo+gxwbKJp3BVkGN+4bnKdMJhMxVs03b53HhgPXtbYJu2utzSxuM5gM8HgfThMPeLm86s62cF4dGFCbUZqC3fCTFOzqKGxlNjc3Ry8jj4txWBNW5Oai4QF5ID5jU1kOxwPE3WzsckxSmJwHczWaJc4si8fF1F2nEyAAVOQhhBApDWaQnH8DMG+SA5ffxHautzLxNk48PsnPqRHB5dexNr6nAMDfHNYmm6jLTVkTAx/H3dvbG7mWGRDYNanPMbgpY9N8K5tNMpasfnTckmdZahrM2JV9e+G9umbhfDKLNoDncU5mp5yW/ub3xFuSoCBudWRjgbacyMt/KkMnKdjVQTxXiVnONigPbFDMHmPhF02nowMbWAlb1zyl2mMTZsnzTvSq9DzQ1LwBG8CNlamnvO0+j51wGcz655mgFs4YnW20q2NnXD9cN1ZeYy120rW1gY3n6GJoBQUdV7VvZiHsdmW2waDAU/PZCNC+ouBlcVq92AJxzkMSK+TyWJzsgmSlbKxMZ1tyW9j4H/cVXs6gXgNOg1m1GTjcd7T+qwGG1W8SG7ehAg/wOR0WjUN/VzMarA60DVIZeknBrk6iLy4rL/tvi6c7OzsjRQbEWZG9aLyAVl84VXQ6HZ6VlbEnG+dQpW7pe0pU3TycPlvvnH+dqKLPW544zxZXd3d3bFxE8+cpRbumi/Itfh7vBOLnlXngwQDJ7JXLwPVsbkKbwGBxsbJnULA6YiDisljZDVSUxStTZrBloOPfSWJx81pAblfLD7MoBkCPPXKf5QlXmi6XOal/VzNMrM0NhPk6P6OGgGegec9yHeo7zcsyPCMxlaGRFOzqLJ4LhZVBJpNBc3MzGhsboy2UGBTVpanAZMITJLyXjpWxMSm1/L21VDxpRHck0fyxBcxuLgU6XmuoLJYVLLvHeOahpeUZBZy+gYEdGaNlsvEtm91nwKZlUiOF25Q3XOZ4edKJKlUruzFPbjvOH7cNM0juRwp2Vl5dBK79xWOIFsYMIi4X59vyZe2jZdN+ygYP16WuDeS+ZsL5StoAQYGZZ6Tqu8TxeqJ9ksNzXSqg2zvMQxOpDK2kYFcHSWIcaomyRZX6zwAAIABJREFUoiuVSshkMujs7IwpGHupqw3cq3LWcSd283F4+/B4km4DxWlzHJwGl88Umt1TVsAAY6JjOqy4WKFxXjj/Wi5TrMbujIUoq7O0mT2ZsmMlrUaK5Ye/TbjMmg6Dtf1WBum5IL12ZrbDbNMmfvC1JONI3XYKWNaWPBYGxF3ollfOE//mfOfz+YjxcvrsXudn2dAw40HHpLk9PKDiOJJYm8foVfT9tX5m75s3+ziVoZUU7OokSZajAh6wQWEWi0X09fXFNsMFKtc9sSLjxc88wO6NJyW9yAZuFi9PFNH07BrHz/EmTds3kFNXEOeR68OUiM284+c8RcozRU3xscvW8qtsk/NvylSBiEHXyq//TTxFa+E5Hxqe88W/VcHqs7xFlwnXqRoqmpaF54/lgbf+4jWQfI3Hm5X1KpADiE00MrejlUvL4nkFksDfq39uAwZmzSfXDdc5h7d643RqdQ+nMjSSgt0QiVqPnrWdyWRi+1eapQ7Ex/GMgZlCYpccA4y6x0yp8DRxdl1yGGCDdc8LdnVHDQYKViJWPksP2LDGi8GMf2s+WdTtZsLM0dKw37bThdWjAhwrwGqWObcXK2B1Keo3K29vFxVPiXOde+zeY35J7MyumWgf5A8zX2WFlobOxmWjQsHZS5fP2eMxSHZzchw8+5SNAjVSDCgVyDhPbPBp3vhbGa8HdJaOMvFUhpekYDcEoqxEXyR++bLZ9bM1u7q6YmuFPCvc4uXZbd5EEXVr2viVAR0/p4u3TRho2ZrlsScWZlusQK3cDE4KWAoIdo/HMxlcVcnaf9uVhdkmMxG7pul77ee1lZcPBkEGdh3v8Vg315ml642XKfhq3XKaDKTKDnUrL86f9TUvXm4PDq91xP2TJ7Nw/Vk63BacDzUMPTau4gE+17sHUMqo1ajg32zIVJsslcrQSgp2dRDt+MxavBfDUw4AYi48i8dT8PoC87RufsYUbqFQiFnzukNK0ngc55cX+PKuFgzOyoSSRMe/FKy5XlkBqqWuLkhv3NFjBknl4zpTkGMGoQpZZ7ly23JZ7Fv7B4dRI4DbVsHMhFmkZzh4ylz3AlUGzUfxmHHDC8YVGO3DJ15UAwU1ONRl6LU5178HiurZ4Hwp4Gm+PCbH9WJtwRObUsAbXpKCXR2kFteGd4+ZgW2Qyy8ZK1yPjagSNmVjv+0F5k2HmQl6VjPHx/nmcQoGSWUdSa4jdUuxGxWIH/tiz1s6PH6kcSmw2TUDAF7QzWVUJWnx2vEtvAWWlVVdXuzm5bxpf0gCa4uP2Z2OgarLWftPkqHlgSkbRLYZtbWhApvVv536wGsks9kNW3exWF75nrJwBXfPOPKME01L3d7MuLzw6hL28qDgzAYdt0HK7IanpGBXR9GXVgHLRAHKrHdee+e9tPxCmzLSuJQB8to1HSfzQIm/uRy6oBzYsP+iZznrbwYwZWd2nV18ymr5OoOaKiRVTgxQHF/SZAMDPBsn5bJYeJ69ykfzKONRBe4pymqGghku6tq251gB2/IHnmikbM2u264uOiu2r2/9Hq2FQgH5fD7yAJh7nQHPJqswSNu1aq4+D8C0n+i3ZzxwecxgUMbLrn/tC/0ZqMo4te8UCoV0RuYwkxTshkA8FqHWvd3nbY5MTLmxwmClqDPlTOw+uyttkgArNVXCSZYtgBiz4DR4PIYnH+izut2TKjFVOlw2HTOye7xtmebPqytlbxxGJ98wI9J2YMZoYT2GYXlXkNaycrzKuq3eGEhVaXMedFcPK6u6ZfkZbW+7Z/0G2HDiABsHPAWfr+lkKC4rl1fbymN2npFo+fDE8p4EQApenI5nGCQxTavLYrEYM4ZSGXpJwa4OkvSS8H9WnnadlZ2yKrXceWKIB4QWN1vt/DvJVVTNHcOsh4GBFSuvZTMgYkVpB2l6LDLJytaxLwYQrhP78O4prDy5XjzAU2DjdO2/jUFZeAZFZZ2aL2YinL4aNxqGt4OzNHVBttU9H5fD7crxe4YRl1snxZj7l9fI2XV+zoDQwJ/7hYJHEnuzsDpRRfuJF4cakrx5AYvGrQycd5Dx3gdllNXemVSGTlKwq7MoCLHiU7ejutCUSai7zXtJgQ1Wre1wwYxOZyNyHlXJ2T375kW/DL4669P+M4vk8TadxeYpFe+6shQtdzW2qu5Etej5twfEmj8FMa5DblcGKB3nURDXOJllmuGgu6No/pmhevlm0XE8jz1Ze1lf6unpiXb6N8PFgIUPk2Wm73kCWJLyqHXL4b12tnvMXNX9rc9rHrkd1TD1hiG4vlMZXpKCXR2lmvtGWYwqQ0+U+bElDMRBh08m4G2dvPx47jQN57EcABUTPiw9VhjKdFgZekqfy8rKisvN7l62wrkuuGw6G9Bz72kdM/BwPEmiz7FxkVRGzoOJ5YVdpJx/BlVuD2bZrISZyTKrtt+8mYCyLDNw1K3a27v+TL5isYhMJhMzODj9WsTrb9pWyoC9vmrpK2B5rE/rXWeven1X6zPpfUpleEgKdnUQBRIFKRZ9+fh6khWpogqKx+fYbenlr7/8m5gyY6Djl95OKNCDSFnZqEIEUME2WemwK9FbTsELkdVNaEqay64GBY/psMtQFwuzgvMMFGUTIYRYvXMYbbMk5mJ1YflkpsKuSn6Wx8+UTXG61l4KHLwA3tK1majeRCbrY8Y0OZwtSVEGlsTqksCL66gWtq1pcf3pTFl1rapXhQ0Ai0fr3PNKpDI8JAW7Oom+jDo+p+H45fGAjsV7wVgZmrJVhaaAwHlQ5avXrQyeJW2nbnPeFYCV9Wi+WKGYItL1agwuml8P9BQkeAzJgI7BU93IzICYKXjtoYYGMwwe3+L24zIqkHrKnFkzH7rKefQYENeP1QXXg3oUrB5shmFXV1dsazDtq1zXXJZcLheN13JfU/Hcg5x3BcJaQE8/7E7mnXy0zTn/XHfqYdD+m8rwkxTs6iDeS2nXFahYoXIYzxLmF8u7by+ld5imvqSazyTG6Cllu97b24t169bFNulNiksVhzempGBnwuNtJp7bieuF42FGmLSbiqXNU9a1vFov2r4MhsyEmS3y/aRt2jgdPpZJ3YxaL5YW39OxUQvH46vcPtxePAHF1n1qHTDr5aOjeCINbyfG8WtcfD3pviee4eYBnZXXGKlNqikUCpEngOOw31yv3H9SGd6Sgt0QSpIVqEpbFZ8pLY1DQdVeZI+BVLNKqykSU3pq8Zrb0mOhHhjYXp5sXVuelaFo2az87BY069zyx8zJwiqrsHg9JpE084/j5GsMyHbN2A+DhOVPAVTbXIGSy2xtyu5j7Q+qiPszlrTeLY8MqNlsNrZLCte11T8bDTyJhscD7Z7mSRliNcDTNlRhUFMjxuvjlm97plQquUsHtE08phxCQKFQQLFYdPObytBICnZ1kGqApmM7nuvJs4A9JazKWIGnGpvT/0nsRRU1A50+75U1k8nEduJQJpGkjDhOdkFxWY2F8enrQHwszZQfK3UAMfZh+VG2yMqdN7Nm5c9tyu5Efl4PiLWwli9mxPy8uS0NtO2Z7u5uFAqFKA1L3wMUZtL2PINn0mkUvAE5t4WVhyeu6KQgdY3y2sYkENbvJMDzwFuNAq8/8m/t5z09Pejs7EQmk4ktLVEDxa5xXXmzbVMZHpKCXZ0kCfC8MMzkTCH1x7i8dFjZ8benYDzQ5TgZMBRUeXYnj6HZb7bsdXacp5D5t+eSYmBR8DZlxWNw6kK09V/sxtS0LD0ur5VBmaCCOeedZzwyMHhtz/Wmk3a6u7vR1dUVO3HC0iwUCjFDhIGFDQQWAwLdA5UB1PLMdcb5bWxsjHZP4b1DGXSsnS2v9tvcmZ4nQNvAawvvGWtT7Ve1AJ22h7k1Od9WH/oMGzdeXKkMD0nBro6SpLhNdOzEYz72m+Pz3Ci1KBAL29/LaS+8Wqq2ZMBbO6fKn91bvJuGlw9VkFx+dV1aXWl98PgVP5PL5RIZqrrnksYRk9hTkkHB9aFs0QNNBj4eU2Kg9ybpMLBrflSY4Vp7MFhxPHbf9spkowFARd6AyvFCu6aAXq3/JQGRXtN61/6fxBq1bbVNurq6YsaE9g9+nusvKc+pDK2kYFcHSQIhfZl5hpeGT3o5PUWW9MInuSZV+KXnl1tZoylMzqMpOZ1mrm49S0OXC3B9cZ3wzEnLFytRC+sBFLvpLI96AC6A2KQEr851AojHzvSegmZSG3KdM4vQtXkcdxJLsbJ4DCepHSwdBVMeu7P25jIpYHF6yjDZfWrlU8aZVP+19P1q/V7rSNsu6brNPLUxZs/7weOYXh5SGR6Sgl2dJKnzK7DYoD6/fJ4C53g9kEgCzFpfQlZYDHSq+CxeZiWaD2YDpgDZfcaKkhWuN5lBy23uNEvb1pzp5Ax15/E2V6y4vTIkGQSs6KspZl3AriCgYdldppNc1L3NgMR5tevMonmykubV6kOVPufBXLA2RsvAx7Nctd3VGLH7PDvTY7hefSYZGAyozIy9Z/pjXfxsX19fNH5nrlzu95p3NYhSGT6Sgl2dpJp7xV7Ocrkcbb3E40wKJvYsv9TV2KOm4/3WvHqMjsdmOB0GC6/MwIbTo01J89ZhwAaGp6eg69RujsOu8dZUzBwsPmZ0FheP2TEIWdk9N6YXtxoj9iy3S5JrS5Wq/bbp+lw2ngijLmMFKWVWDJTsKldG7s3ItAXiJrYxgbrc2a3s9UVdrM91w+5sj60mGXlcXmvTpHquFr9nAKoB0t3dHdW1Ap6CazoxZXhKCnZ1En1p9YVnoAPiExZ0LKyam0SZnH7rPU6Hw2UymYq1Rnx6AbsQ1aoGNuyYwazC7nV3d0ezN1nZWxqs1Bn4OQ3+WFidPMDxqAuKy8RAqDMrua7Y7cbsiz8MPNXan/PBbcAuVlWePOalu6Yw+Nh/Bm8uo/YrLpPVo41v2gQUmwlqSwm0XNbODNQWN0/Q8fqeTshRA4Kf0ee1X3gGofec9y5wGhqmXC6jsbERhUIhZjTZtxcX10MqQy8p2NVBktwa9lLbTDsWnilYbSBf2Z33grNS8YDSAzzdTYQVCoMvgw/HZ+yrs7OzgtXZDivKCFnJWz74WQVZnuRgLjVW6pZHr460DAooqvisLpnl9ceMvXtcLp2JyhNYtD25/Sx9zRcfzKtLO3i7NPsoA7F0bGmIzaK1w337+voqZpJyOW0SC+dFDRNuD7uv48G1MjMrD89WVgPCAzK+7xl/Gp7LZmDOfV4ZfAgBuVwOTU1NSGX4SAp2QyCskBkQ7B67yUxCCBFA8DgTx2nhWEFWc9UkKWJmBTzrkMHPnvfi9c7IMyXZ09MTuSp5LIonkSgDSfqw4rX4eJkDx6PswZ7h8jCAaF3qPa/eNG5P2XrGAYAYK7VwHlu0eI19cR4tjlwuVzFOaABkrlddlhFCiFgcK21z4XEbKhBwWQx8GGQ9wGPmzn1UXbK1MjoOM5C+nsQCNRyAaOYxuzM1XW2nVIaPpGBXB7GXXC35EEI00F/L82pNKovheC2MXeO4VNTCZXcZpw9sAAQdmLdwVh5T1jbVn5/j/xbOGBa7I7neqo2DeMpRJ0J49VkNtFXx6j3P5WWiY3S2fyVQud6NgcfiYsOH28MAgvsRnz4OILbA29zIBmCmqLUdOK/c5jaj0wMgr750xqwtScjlchV1zF4GNkZqATr7KKOrxg5V1KjxwInLauFto2tvMk+S8ZfK8JAU7Ooo/FIZc+OdRzwL1bM8WfnrGFN/1mV/7jUGIgUki9/bfsnGcxRgDfB4lim7P7064ue5HMxk2O3IwKR14TE3nRWoZWRhoPTYWhKDUwWoR/IoY2Mgs3SZwavxwUYCAynvIsNjcczWuWzcj7hNmDkWi8UKkNH2sji0D5bL5cg9q3Wvblmv/jzQqsbC+P2ya0lgpO+Htiu3Lb+zzO7U+Exl+EoKdnUSVYrGgvpbZ2RhWVghJrljagU/jVvj1IkoniuQ2amyT54Wz0yDFYrHtBR0PaBTZmCAYfXAIOC5KPW31qGXL85fEptg4GRXnbqftZ6VxXl1xHnWuuGZsnauHOfVY7v6nLW1jn9y2Y2FWV2zC9pjazYeXSgUKgCVy8F1y22ofZT7YRLDrsYQPXbNv6uBlgGesTtONwW84S0p2NVBVHFmMpnYWIiFGYjFyta1B3oeQ/JAzwM7TZetf1bMpsi88Ua12D13WlKeGxoaollvrPh1MgmXUXcOqaYw1dL3DABmhjqGqPHos8xcDDgsPOedlW6SyzRJvIk6DK69vb2xmZR8j8vC7M9jMh77YabMi9EtHs6j1Z9N3c/n87EwtoavFibHzNIzEpXR6TXOH7eTGqEcHz9n/7u7u6O69fqw9w6nMvSSgl0dxHsRbDyD75soKCnQ2TdPdqhFWVZjesoG9ZpOrLAZpKxAPaBT1qETK9RNx8rX7jNY8Meu6eQOS5vzlOQ29eolqR28cHzdY24KHLp0gdPxFKfHYDwwUibb3d0dm1HJxy6ZMDDwh40SbVevjnh3EZ20wfkyl6YaVDrRhp/lfHJeqhmG1eqUpZZ4TJi1cv0yYOq7kMrwkhTs6iCq/HivQ8/q9J5NelmBDSyv2gxNTt+u8e8ksOSxFovHxho9pczx8qw7dtOxG1NdaHzdc/kx4Hhl9ZigxwrsuipfdncyGGv+koBX20frjheBewyV688DJi6Hzc4tlUooFovRRtH2XHd3N5qamqLxO1vywVuQcf7YuFDDhcGV88R1VywWY6ceZDKZ2HrKEEI0wYPrS9kdt6OWW8cFvb7gGSRJjMsz+JLYHYfjzca5LnQGcirDR9LWqIMo2LBC4jBJQGSiLznf413nPfeKCU+ZTmKUnFdV1ux+ZSBTpaTAyuu8uIwWh+WNv3kcSMGLy8PxsuuQWQOHYaPDEwVOFY5f69BjYqz81TWbJMoE7ZrVN7vybEeTUqmEjo6OqC/wSebmGmZXcG/v+sN2eWatpavjq1YOD4C5vLlcLratlufm7O7ujia9cNz9saGkd8HuJRkd/Iz21SS2zmkqEwU2TAZiVsvxNTQ0pOvshpmkYFcH0ZeHXZh2zbOiPfGUAr+Atby8XhgNx1aqKThz33iMTsthx9mwq86z0j3mZN+elc27mPCzzODsmrpIeXzLrHJmNMrSPFeYN4FE60Dzza5YrWcvHvudtLzD6pfDd3Z2orm5GU1NTRHg2QJ+AzY+C88OJzWGruOuyn4tn6zYGQTsWZ5FanW9bt06qPT09CCfz1fUgwr3Fe7fnFetUy8ONRQ13/2F53ayb2OpyuD43fHes1SGRlKwq7PY1GUFO5YkIEpyr/ALXw3wlGGqeJYxp8nn1vFsPI7XFCPvqcggmOR64uuqRBkkGTiSmIGu10piVqy0k1ymXG+6+4eOY/ZnSHC5qxk0uiCf07P/6mq0A0fNpdnR0RE9s3bt2qjeisVixcJ7zmsSY1V2zOG0vrkubIG77n/JnggFE6/dk5hbEsPzWKCCZC0GllcHDHa6kUEt4JvK0EgKdnUQfmkNMOx/kvWnL3fSC6jPqEuwmksnycJVMAAqJ9TwPVOYuvaIZwZyeBUduwI2uGUNwHl3fd5GjWchMoAq61NFZv91yj8/x8DEwvGpxc9xA3CB2Ws3ZTAaXpU3g7gxbqunfD4fU8TmWrTy8mLvxsbG6Fmd2Wnp8n+7xmOqOlZleWTAs7R1vDZpg2iuFw/4+2NyXv1y+2s6tcTlAaox1KQZu6kMH0nBrg6SpMhqsQKTLGaOW9Pp7/wzBilvzZHmx2OjJjp+xWDLQGdxc11YXngmn7ImA1FmNTYGxPFofVkeqgGMKWp10yXVGzM5dUsmMdMk5qP3tV48Zqj1DsSPD8pkMpFrrVQqRYBmQNPV1YXOzs4YQ+bTIrgd+wN8LZc3nqcTc5Q1eiCkde0tMbBnk64lGXIeGHtxDYSVsZGhjNOrt1SGTlKwq4PwC5x0nAlLtRfZnuuPKei4FitbHoPjfLCitvR6e3sr9uhkhc+WujJVLjcDreXP6oKZGishmzqvbE1B26sTZh/e7MJaRAFKFZimqeCq7EGf05miScsMvPj0mhkFdr9QKER1YLMwzWjJ5XLROkatawUAb3IG1wUQZ9j2X8uo45b8HF+38cUkRqd5SQIqvZ8E4p4o+7drXh7s/TBXZsrshq+kYFdHUddcNWaXpJhrBUi1inVMKunlVYuXwZnH6cxS90CE421oaIgULSs+PcrHUyzsHmXlZ4rFm0ChZVCg5frxZmxqGM2fLk1goPfaQlmMPq/peu2r7eYZFOwCtE9XV1d0bFQ2m40Ajnfutz5pdWouTlXaVsfeZBxl2WYEKZPT+rP20aOZuK9wHVSrI72ufYnbyP5bP6wmamxoPkIIsUXm7KVIZXhJCnZ1EnX7eC+gheP7SVLNLaNxsXKs9rxa8axwmIUwW0oqK6drJ1tr+RjQWFF4oMvx2YJey5eyLf62MEClWzcJfAwIQ9hwmCxve8blTDIgkhiHKvGk+huIC4xBXo0AM0r4xAJuu56engrDRZmsVx6uT+13zKg9Y4YZpNaPxZdkjNVSb9Xyy3GrUeLVO5fNxhz1nk4oMmlpaama31TqKynY1UnU6tYXzmNmJhouyVL1lIG9oB5TZFCx30lLCezDp2ZzPllJ6FZoPAnEAx6OU5mit8yAFaiBnrcI3WOMdo8VtdYLu1V5EkV/E0cYrLmOLW/KtFX6c7GpQmZjQduQ20pdzWx0GRBa/njSiwdmXv6YKfFvi8PW3VlYS5fzo67spPJ79ZVUj5pnD9yAeN9hg85L0xM7tkpd7ry0IpWhlxTs6iD2QldjXkkK1HvZPPZXzc3DbhVvXImfYQaibITHXZKUiJ1CbmxIQdTiYWXC44LGpiyv9q1ApuN3PPPT2ItXVwpuCsRaPk67GttQ5c/Az99qYOjsR6/tNE7uP14fshmXPI6k+fFmiaoh4fUNrx7UeOA290DAAzx7VmdsenWt15Leq/7EM+YsHl3Cw2Xi+rDnzJXpnXWXyvCQFOzqIGxtJ1mi1URdS8r6qsWlLIlfXBbPRcdgZgDCwK2gbIuW+Tl2eerEBasXHsvksSG+x+Bp5dSxOmYP+owyy6T6tLIwi1Mw4DpTw0H/e4rU7mk+ktoxKQ1uBwXDvr6+yOjI5/PRFl3GoHXzZTXIrHzVGFkSMAOIxW87vPAYqfZDraP+AGOwAE7js29vHNg+akhZ3+et0jhvqQwPScGuDuK9XEkumGr3PBZXy3OsHE30pUwCSXu5M5kNC4E5TgDRmBwDkzE0XYtlzxrD0HFBziOHs9+8gFeZmLIVZRvqJmWwsv/Mgi1PGo7r2mNbGi6JDSlz4Ovclp5xw2VM6ivGdLW8yqSsDXWWrom5irkMOnbFxkZSH1HPgII4i04aSqoPlSRGmPTucX/gutXxORZmd1w+G/8ciAs0lfpJCnZ1EM+aTFKKdi1JeepvTafafbvHriVVOsyK7H8+n4+OidH8lsvlaANie46Vnr38+Xw+mhnIa6hUeaoi0QksamVzvj2WxHszctmYYSjw8RFESSyG02Q24DEEVbRe3nXyDLdXNfaiefQUubk1efzOyqltxqKuPG8MT9PnutT+zAfLKnDac17dKgPtjzH1Z0gmeSY8putNSLHr3GbG7mxdYxJQpjJ0koJdHURfPH5p+lNmQOUYTS0vu5ceKwEbY+gvHm+3Eouzq6srdnCrPcMTYhj4DBBV0fPCZsufzRAENixct//82xvjUQudFZcqNGVWPOanwMQKTBU6l0eZiAKxKUoeR/VcorUwA4+xeGxF289El48om+K4ueyavpVL01DFb/nwJvNoOlym/qTaO6H14NWR946o0ZcUL79T6rJNZfhICnZ1EH25vJeoGnvwftszfN1TDNVYoLkFkxiBKSl1QfFCZUtXFYa5xWw7JVN65XI5Cm9K3gMxBhMOZ9c4nLIinn4PbFBwvL6Mn/XcTspQFNh09p7XTgzgHlByeTQeBg/LTzXlyeNwbGSoC5b7nrptPYD16peF883t522ekGS0eYCtzC4JdGoxEI3N2XVtT6/9uB61TfgZTt/c+dbPdBF9KkMrKdjVQTxL0nPTDFT0ZasGdBqW3TlJLzBb6vai9/b2orOzM7arisWl8Vj4vr4+NDY2olgsRvHaR0FF3WE8o1DZlDdeZXmwiRg2td7yUG1WJ19nUPXAThU3P+eBljIgjc8zgJLCeB/PS8Dx8Fowvq6TgLSPVOtP1cSb3MJpWhiuLzVSFAyTwDGJ/XFfUsDkdvHqLykue16ZoBmP5XI56ucjR46sWkep1FdSsKuD8Muk1/X3QKxY7wX3QE1/M+MxQPBAw4uHQU7H3pQZ2r2uri60tLREx8qYO9PLl4GcLuJOAigA7iw4tsIZrDQcA6kaJEmMRhVof0xADQhPoSqoefXp/ebw1iZqTNm4pf3OZDa4jg0Evb1PGZisjyhAaT75Hq/H5D5j4o0BJvU7rXd+zmujJODituGw7FbV/HhjzPws/7bxa/NmpDJ8JAW7IRJWqqq8OEw1sOJw3u9qyoHve9c91sm7iVh8NnHFy6OOtZmSNXDSAX4GGVaquhBd85rkamN3od33xu24vF59WHmUaVlcniGjdeyxV24TZTWeQaAGkX4n5UPbkQGHD1jVtmD27c3E9QwIvW/tze3IIOyBPJeLDalq/dS7p3WgdaZ14wEkA56FScqvzcjkk01SGT6Sgl2dxF52VUj6kurUe/7W6/x8NWvXU44aH7+4nkXKywJMAdjMMz5lwZQkn2dnz5obMZ/PJ+aHr5syZHcqKy1LkxmBxsnhOO9JdcDxq1JXUPfYoIb1QL0aO7D/1RS4V2caD/cLXZJhkyi8tPUZ7VfMHJXpcZpJgKdl9I75Ufd5Utn7Y3ZeuTxw475j9cJuTb1mcXmzNW08O90qbPhJCnZ1EFVeDHzVrFML61nP1dhdEkjyNU9psHJW429cAAAgAElEQVRR5ZU0/mKsyRieDsyrYjYgtMXnbMEzMDCbS2K4HmB5wKQAqfFwPvnDLrUk9sK/q7FHrT9t7yRmqPf523s+ib3bb2bNvLbOAxB1+aohpvXqGRmWjvUNZZA6S5af9USZnwf6HL+Wn0XHUvWbwc0AuNpyBCsbrzlNZfhICnZ1EAa7JIbnuT1YeSQtcq3G1rz/3ovvPc/CJxQw07K47MBQVoisfO05BrRcLhfbHFqNAc6bXrN4zB1qzyaNuySxVa8u+bcyNK0/L15lIwoIqoS9vuEJx6P1w3nxjCM2KGyMVsul+dG8JpWZwYuBVOuSjRf1EvCYq8equRxee/XXtlo3HFdSu+rYK4O13WOwZLCzvWFTsBtekoJdHcXcHDozThW6KjR+Pomtea6tJGtZLWCNU4FF93XkF51nTFoYtYgtTnVn2vlqloZNfmFFrMzO0uS0OK+cXgghNvkmCfRYydfKOKr91nz11wach1rCeu2V5PrTuuNDVHlyknodPNbD5VOx55IOaDWx9mHm4/VHM2Y8BufVgZY3yXOhTM0rh87Q1bbg3X/snvVlm5Rjh+SmMnwkBbs6iSl6b1GrhkuyapPuef/tGe+3Wq5ePBY+aeakTT7xFAqDFKdvioKBhOtFXbu6BtADM53tqIo1CbyYyXh1rWXyjAdPmXkKUtPSNLV+PQZnYZlVKBvjMKywdUwqaYq/1ZnWqT2na9W0bCFs2J3H8qH5tzSsL7ARpO2RZIjpO+DVE9+rxjSTnvfi8jZYsP7IJ5YD612Z6Tq74SUp2NVBjLkwo2OwUUXKkzFYYZn057apRTxFwkCSzWajF1iVgc4sTAJOE2VMtvAW2ACm6v4yZajpmZLSMT5PPNDlPKmx4Vn1SRM5PGXJbaYuLk7Ti4/rk/Ov+VMQ4L5jbNkDKc4Xl1UNLwU6yz/Xv5c+tz+3mwfkSW5ArUvOn0p/xl41NszsLqkfqAGlfd6Ym9Utb6dnfbdYLCYaRanUX1Kwq4NY5zewYyt2IMCVZIGqJLECjUctdA7PIMcvdQgh5qqqVgYPEOwaL/BWl5bli/dyZOXKbjZzi3J+DOR4RiHnn1ml1oPWDbMiDuMxPa1PZq5JTNJjV7UoyCTGom5fZVfmttT6ZCPCRF3Q2hbK1rhu7MMsT8ukbNFj31wfSde9yVv9vQO13PfCZDKZaIKVjTmzu9XK3dfXF1uTmsrQSwp2dRAGO92FnpWvPpOkXD2w4udM+gNEjZuVCruZLJyyKS+PQJyZ8rZRnK6nMO2/ukiTFKwqYFX8SfFr3r37Jl64JCVs4b3na2Eo3CcUjJU9KkiaElYXJbctnxhh9eflm4FSy659NYkdWZzeOB6nwfnVMbRqrL1a+yaxuqR8cvm5Tb3+Yf2/WCyiXC6jXC6ju7s7MiT4Oe37qQytpGBXB1FQMItPz/fSE62rWbaqJJMUsieqPD1m4ykFAyFP2fKzNmuNN35WYOcJFbYLvpdvA0xmenadXW5cj165eC9NLafWIbPCJEDnZ6uJMgTNn1eXPAFH68IzKjQ/XrtZ/Wm+THkbAzNR74MyXAWbJONM+zK7VO26ego07+rW5zJoHuwZzzjgfpdkpDBTVAOQP8ViEd3d3VixYgW6urpQKBRi6SeBZSpDJynYDYGwq0PPY2PrXhWhKpkkFxhLNean4ZLYA4AIaBQMOU/exAYPfJiBqPtLnzPRCRWcjgdYyoa9euFxoiSWp4o0iaF4YKZxeXVi9eaFZbH+YuknzZj0FLSl4YGDNy5o4TkOPoBX82UfdaHafTbu+DoDrrZhUl9IAjwtF9eBFye/Y5YGu3LV2NE2yWazaGpqwooVKypmBOv2dakMD0nBrg7C4MYvtb0UDAQc1kRfYlVaScBXDdi8a9XC80w0BRUWnnygJ5sDiO2souurdFspPjSWWYzFbUrU4tdJBFonDBhJQOZN0EiqM73uKTiLU08Gt++ktmPxGIsyHo7PAxWPqVYD+CTGqczIc7Vy2tyeIYTYQbCcnpaV0/SMC82zllfL6bWnMjYuGzNBr34sb01NTejs7Iz1ZWvv1I05vCQFuzqJB3ZA8rjP/9fety030hxHF0gCIMj9DrYvfOO38Ps/gh/DshSytOvlAUcS/8VGDhOJrOrBShrwV3RFIADM9HmmKyurT5Xia3V4vpYxERfXKQQoSig2B7ywZhWkmL0BoKB09/v9mTJgQMX/akYf3KVQTlCk7rw7bj9tG1aKbvxU2YOyTAcgGgfhqgkYGs8pbhZmUgyCiKvKHBNR8HyUrTHAKdvL3iGeKMRh3BIb5DOfz89YNT9Xri+79bVdtCzufyUV6Lv7+m4cj8e4v7+P3377LdbrtTVqu3we6WA3gWiH5xmZyjIivHJtAZzed4rSKSyXhrqvtAyqPLFEAUoJi2vZvTWfzwcw2u/3ww4TnIcqf541id/MDJjJuVMSFNC0nbN2iDjdc9PFqdrPMbaWWyx7Rs5tpyCSARFv4cZlYwNEr49ZRuL+K4t3rG42+9g6DMcuaZshrHOJjn12aoi4+Fm7cvuyuHE4nAf5+PgYETGc7wjpSw8+l3Swm0jgpsPvylpVafn/KyDU7xbgcRidjKLpH4/H2O12ZwuCAYBQapgFyIDolB2+WQkjXYCZmyCDPJhR8rIFLjOUurYr4un/v4UpaJsr4Gra+k5wO0OYeSmwuDxUgfMzUoPBrWvTslYAFXE6uzLbexPuZ955h+Nm7cr1cff0mjMcxoRxAIW2gvDm5/P5PB4fH+Pl5eVkBvNisTgrW5frSQe7CQQKhBeVR5yfIg1x11habFDZnLOIVTnydQY7dF5VhAA0BnAVHJyKdBEeQLRYLE4AkOvF7kvM1sSsQa6PrqljF10GKgpkOvvRWfyujV3b454bX3N5abk0fceQ+Dp/VElzHRV0cI0nYyCOuhMzYygz0NC+YJAKyvxOaZs712z2O2s/Vy59hvwfz8SNsbm+pnnwOO/Ly8tJ+l0+j3Swm0CcC9ONk2icKj1VDhHtcTv97dJkxYn4qtze338cyAqXlCrY4/E4TESBsoOrE+ExgH93dxfz+TwiTsfqsvLp0gMAqtZZf7tvBlbXfq12bd3LZktmUoEdh+G2YCCczT6OW8J9TJbQiTnqatTrEJ0E48qXtUF2ll3Ex+bizp3q2pjfz6oNtZyZseLak9sA91y5tC6oA+6t1+uTheZdPod0sJtA1ALXiSoKRE6xOCaXWbl6PVMAroNDdL0fT8uG61LDs/sQcjgcTlydqtyg8AB6GRPhuKqMlCUATHUxs1rimdXOaTu2kClEbQ+Wap/EDEi4LJXoNHpm5rPZj/PreGKKxkUcZdnM4tUb4VyebAwhDR6r5PDqsmy1rwOdCkwylpf9Z6NH321nhLilC3d3d7Farc42j+jyOaSD3UQCZaLT6CNyi95J1om10zslwJ2vUqCqiHi8bbvdxuFwODmbDGGwfi7iw80JRevccBiz4fx0rI0VSgZ2+ltdf1z/MYyuSpvTwT3nOlTlnDEhBQFXDg6blRHvFodXUNfyOCar7kudtOIMFte2unSGDTn+z4e3urpVbZ09N/2trLAyBJ3o83czhLmtHx4e4vn5uQPdJ5MOdhMIK7TKvZEB3RgWwt8VU3AsLmOHLLe3t8PWSKy0MM2fQe54PA5HnmgeqIPuhgIFiXzd0TwMTu6/U4pIk8fMsvE8ZzBoGfW+MhkFXU5X29lNxsiArwIp1Atpcjvy0ovMa+AAGOvG2BhhgGoxMWU+vGZSnz0A1NWX09TrDvgcoLl33okzblw5kB6/Uwh3c3MTy+Wyn2f3CaWD3QTCYIfp4Hy9ilcp49Y1VkCZRaxKIlPUcM1oXGZzUIRgcwpIzs2ZKXEoQFWibqJH1kaONVXtrQquUubcXgzMri1dHq4c7nk5gM2MFWVyDEwZqwc4oj25zXnGoYIRPy8GGG0HzjPbQFwnyWRt5v47w9DFr5izMr8qLb7H74dOSFqtVmdjyV2uK/1pTCDM2OACbIWN8K407ZROIXN8pOmUbKaQ9T7KjbE1TCzhRd9wo/EBrMhXp5gzCGbtpGCp8VTB85ggp6HsQtvA1ZnLc4l1riDlACHLv1La7l4VHvXU437QJryuEXVk5qbMTMvuXJjKqJzhhLKpweKWQeC+m0Xq6l0xvgyYx7RnZUxVILxYLOL3338/qW+X60oHu4mElS+fdBxx+dhD1smclTpGUeK63kMaYGoRMUwk4XE83Q0FaQEcs7I6AcvgumZjeGpZ89ggK05V6MzEtC2ycbzWf3dd29QxMjZiVLLn1AJUHVPi627huM7YxDUttzI4bivHHLUcEae75PBkFzZG2H3aao+sTd1913YaNzNw3DOqvCWz2SxWq5UtU5frSAe7iYQVhk7sUMWtkt3X31XnqyxZ58KBsJuLt+ECi2OgY3DisTFn7bOCdEALtpZZ+TyRheMxUPIC9GxnEAUFduk5I6IyFrQ9lYVqnplBk6Vb5aVMhI0Gx6QQTg9adcDFk1a4jIjLz53LoKDDZdRJS+/v78Mp5w5w3LPQ++oZ4N/8TDKm5UAye1Yt74qODXe5vnSwm0hY6TpXJpQNOqMDCHUf6X381/tOUbl4GSuZzWbDsgAoqcPhMHw4X90wWhUgKxFlCNxWnCYMBHV1oUxYwxcRJ4vO2W2na80cO0L6ymTQ9lqmTBzb4jqraFitv5YxA119hs5VDLeaTqpx7I3TcZMxXNsgHhsNuoSFAZnrw2N3/JvfFceStV0c0OJd0HWD3Ne4Tvq7EmfEabt0ub50sJtA1Ird7/eDolYwY2sZ4SPqDlO5ZlpxHZCqopvP5ydADbemztTDOJ66GTNgyKxnsIS7u7tYLpfx9vY27DuIe/hmJewWRyvzUABgRZxNermkrbke/DsDMo03hrW4fFjBOlcwv384MUKfkTNOuO34m/PF+81rMRFPT2Dg3wx+DEjs+h5rXLgw2pb6jrP7lncC4vQyQ6z1HLt8PulgN7HMZh9H1rACUTcSwqoVzem4e04BtkBQFYpa6iizuiyPx6Pd/09BPLO2kQYmv6hbErujgLlxe0EhstJmxaRtyGCI+05xaVs6gBpjUChwaf4aJwM6DZuxG+Th4qD+jnFy+cDSdVan5q2sCCClDEyBUoEa6aqR0moXxz5Z1Hugv1l4z1XMOObZxCqOxbaeXZfPIR3sJhBnccJqZCXvJLMcKxDM0lFXmKYHxQR2BtchQE6VFS8lgMLjMmRKCr/BBqBkmEXOZrPhKBgGWbgpwU7Qljr9X9up1ZYZkFVMS8Fcw2fgqWnpb1eGLH9lr7oDCtpKJ+/oUgO0N9/n58NlqeqO56rXGfAYlNkTwHXNxk312pj2YWFQdm3M77Obvat1528Om733Xa4nHewmELZ8FVSycbiI07PJnHLJrGFORwFKLXqXJ8+y1K3BOG93bI0qElXCDG6Iy3HgxkXbbDab2O/3JwpyPp8PTBPKM7PcOQ/XRgwAGbBlbZv9dkqYnzvf07QVxMYAHV/ndlCAYLDBEhJmXlwudk1qmpXRoGDGY2I8nsrfOAED+eu6O8eKtb1ce7eMBmWibqmKrgvlcrby07hdrisd7CYQdf3c3t7GYrE4We/kLFgIT5pwVm11TcuRpa/TxDE2x2nxJBp0Yp34gTpyXsfj8YSFsRsX5YZyw+/lchnv7+/x+vo6ACM2lwZ4quJR5aNtr0Ck5c6MgIqNIS0389Ip9gxQnTjQqgwbrbfG5ecHRc4Mm/PlONnMRi2rsjJOk58hl5mNlez9ZSDN6t3yIrh43Kb6bjLg8exiPsXDvXf8+/7+PubzeazX67P8u0wvHewmEO4U6Lir1eps2YGLpywvou2ec8wPoq5GHStjUHZsSbcGc4d2Il1MaIESYYWiCpiV8HK5jNvb23h9fR1mruLMOmZ1ypa4zRzLzNqWDY7smjMqNB19NhWwueenxogqa342LOpGztJ39ce2bmzE6LsaEScsmtuEDQbNRxk3vAUKeAAVx2KzeiiYtwDYxc+YM4fhpTfwSjD4MSvGfV3LmPXvLtNKB7sJRC3A1WoVDw8PEVErP6eAK7dI1qmcAm0BrVPaCoyqLFRJzmazdLxH68RAh8W4b29vwwL2+Xwey+VymLASESdWtzI9bavMWFBlnQHfmDbX9hir7Fwa7oO68TcbHspiHWPiNnJlYMNFn40ey8P5K0g4IwDlxIxLZsTMIF37ZG3LbabXNL3M8FGW5/LmttH1nvpB2OVyafPrch3pYDeB8NjHbDaL3377bZh8EVFb4QpIClpOnMXKabGyUlcOp80KU616juum86sLywEQg8rNzY/DXB8fH2OxWMR6vT6xnAF4i8ViUIzu0FFWQJlhwIDCAO6AziluTacyGjQ80qzSg+gzYZcft1+rDApw6mnge85oiPiYSKJuPjwbGDbONcyCyTIMdJxOxsazfqDldICYscAsn6yNVNAePPaJD5bhdFb3eaSD3QTCYHd/fx+Pj49n7j8Wp7wqBujEdXSnyF0c/Gb3mAJCxLmbCnVVJqJsQ+sCgGLX0Gq1itlsFtvt9szqhhuMXaMKnFxGLYe2j86KzRSrm9rv6lK5L/W35lEpYAU2BjsHZMoKlcHomKeul2PWpe+DM4z4nXbtw+zucDicLFvBfbftnGszZzi4elaAp8+kMnjUEFFDQ1nffr+PzWaTssku00sHuwmEWQSYC3f8CN8JM8Bz9xxwsSLQGXes0DQdZmBQYpq3KnR2p+E+wAhKEHkxM+A0wdrW63Ucj8dYrVZxf38/nIq+3W5jNvsxRR4TBbDOjgGpWsqhbaltXbU3113bnZWfi6/PVpWvU9RZGpwf31Nw0/haf06LAYx/K9Ar4OM5Ix01LrQsnMfhcDg5sLcyJlxdFOiysBngqOuU01IjamyafC9zy3a5jnSwm0Bg8S2Xy3h4eDhZEK2irhPX0VQJuPscrlJAmgYrHmY6cNVw+VxaDBqopypysDdMNIHbBzMwX15eYr/fx3a7jYeHh3h4eIj39x8npGMpBIDRWfFcZ83fKUfH6Bi82O2L8Jqnew5jruvzZpDROM4ty16DLB7ATA0s3OOp/1w/ne6v7cDvE9zK/M4B0DhfB2zc7ryfqdahakM1Epzou8Lx3L3MaBhzv8vnkw52Ewg608PDw8mJAc6qh2TgpfcqgVswc2khjFMUOj6XrZ2CIuWlBRzegRGHQ3uwwry/v4/Z7Meel9++fYvVahWPj4+xWq1OFKgr+1ilo+y0+tawWpeWOAbG/929TLlX4asyZqyImbwCIgMoL+rn8TvHVpkFYmxavQnIQ92ZbEjA0HDl5zpn7cxtxIYLypL1PVzHUVbuGWQgybJYLPqZdp9I+pOYQGazH6631Wpl19ap2+NSdsD30QF1T0I3NthKHxY/KwlOE8LKDIozs8z1EFhcx7Tt+XweDw8PcX9/PywoPxwO8fz8HIvFYrg3m30cPaRjROpa029VUHqvZVxkIKkM+lILv8rPKeZL81Bg4/SwyJwn/nD6CI+Txt1z1vcE7yGWHCiDZeDhMw91VxVtn7FMrhXWgSOL2zVIw2s/4HcI7vYun0P6k5hAjsfjcHIxAxIzu8radyyJha/r9GedCeriOmHAZDBmQANAsdWPuAyAzBgQltcrMUNDnliGwC7f4/EYm80m7u/vT5Qz19XVQz/VNHfXLjxBxwFd1Y6451yVEacTaFicMuV8W8qb01ElrW3AgKBLC/hdxX9+drimDFN/43k5I0iBt2LTY+qdgaBrUy2/lkvB30nVr8YAcpdppIPdBALFrZ1C14c5xezcJVkeFZtzVrLe03y5jKx4uNy6G0rEh4Wrkx60PrwjBdLltsFC8ogYTj+AgPGhrm6huqapyg5lcGCgbaRKXxmia0v+ztb8oW2QhlO6CgSZ69IZUe45czl0zZhjvCijcw269PVoHgVLra/WW9s5q28Fcq7faB9qgRfcrDq+mEn2jnT5HNLBbgKBO4M7F493sEWtUrmqmKkw81CQy1wxVWdUNqkKicGFgS1jL+qi5RmVYK/M7rBwHOM5AD+kixmaTtlW1r8qbHahufYHcOoMT/dbDRRtZ6donaLXOmSMDfey3xxWlT+DMNqBz5JTsOGPgrwuQndxcU/LCdHZwQ7IXDuyOOOk1bYZ4OKenlDSMka6fF7pYDeBQLHz7EseeHedzblw+Do+vPCa10k5BuI6Pis/zY8ZGHdodh9yPM5b82cl1pr9qOXnNsERLNyWXK8qLteN28SN+XH5MmDTPLL2c8q55d5SxenAS/PicJmyd0ww4hTUdbwT4bC1mJZLT6DgMVmd5RkRg9HC4RzYaBn1ndXfrf+uD+izVtnv9yfueRdGn3VloHa5nnSwm0C0M6mriDuhhnNx8A1mxCChaaiC0zT4W8ur8aEIOT3eEQOSKUsoRj2lna1+ZgkcDuvssMic4yAPbiO1xDl9zdO1AaeLeO4g0gw8HAPS9r2EESCOGhNaHvc7A0Ztc1xzaasxxcYanhXc2gBNLb9+uwkuavxwmdUY0bpqfhpHjcoqLj4wrLS8OtbqytUyaLpMKx3sJhBmMhmL4okfEafKUq1PuPQgCjbOAubr+ptFy8fAgll1OuvSgTXniyUQnK4CjU4agULFjMvdbnd2QjrXgccWte3YfZa5inVNmTMEVBlXyqz1DCpl7fLMDBMtlyp4ncqv+ejzjTjfdQbtg3ZnA4BBj58BZm1yGRXcAZBaF35eWb0d43PPnuuq7Zq1OYc/Ho+x3W7Pxnad0cHtNcaA6TKtdLCbQKAIssW6ER8uOscMWGnz4ZIRp0yI83O/M5eQllV/M7A5RsXWvSsLj4c5kGNFeTgcTmZg8gnpTqnhvysz2oeXT2ibt9qA/ztF5pRaKz0tZwZ4zvBwcfSd4bBVfsxSnILW9mHGrctNOB9dfpAZbcjbTV7Kytsy0rTumbFXMW8F5sPhMIwhZ4zV9VcslenyOaSD3QTCYOAmVWiHU8s84mOCBivxyjp1ClLDZGXNrkGR8Z6ULn0FZ5RfGS6us9LEmByuYTZcy+XI7TKbfex4z0qIF/Ljnlsb5xSYA52q3ce06dhn6ACxYoaQ7NlrHq5dFFgAcNwO7NLk95QNE87HgambkJKBGj8LV3fH5rJwmqYTfld3u92Z65aBmj0N3O7Ok9DlOtLB7gqi7hbucDzzC6JT9HV3+QrAWspTFbsqnQyINYzGx84oiKNKjeNpHjrRxY0jKVsAmHG5lDHwuJtuOOwUZcZGXPu2jAgnqrz5mqZfpYH6ZgZO9R+A48CUDQa+xgxf09L3WRmbM5DYRVgZUFquMYCvkrE+F47zAbvDXp7KhtVAcP2py3Wlg91EwgrTdQJlHxCsNeP7CnSZwq2UgeuUyrq4XFCKcDHyRBVV1jhR3OXr2IwqONSTGYOzkjkeDIBseyeuA9LjT7W0gMO7NlSpQFHbtpVm9jw4rUzB6n295gCOwY/BDOOlHEdd08x2kBcMDJTBtb9OJtLJQBzeAUxWHzWm9Dq3hxPujxg3hrdCjTwVZ0B1ua50sJtQlM04xYTOBZBDp4drrwJL/nb33HUFOJ5kgHIxY3LgxooMa9ZUmWj9uFyYxefqpZMgHBvU8ro2RfmgxFnRQhmr1c/KtdW+lQXv4jom6cI7Rc3f2W9lWCotJa/1wntRTeTBPbg81T3q3h1dOsMGHd59TWMMiLj7Lg39qHC5DodD7HY7u0GEA+/O6j6XdLCbQLKO5DoZOgncJcfj8Wy6v/tuMQKVCgAV8NQdyLPoGDwAGrwrDKerbJBZnC6qd+XT9nJlUGCDQHny2Ke2qS6I5m/Oo2rLSsE5AMpAVAE7Y2Gctssvk6x9Mwa6WCzi7e0tdrvdSXh8eOJUBeoR+VE+/F6xu7Rl3GUGAYetwF/b0TFi3MMxU/f397YOWZ27XF862E0g2dIA/OZ7vEUWGE+EP4YG1yPGzbRURdQKjzgcHmHu7u6GzX15IooCkgNPZWW6nEDrzG4uBWGE47iHw+HslHM3C7Sy5t2MQm7DDNRUUbr/rk2dO4yXUnD6lzCGTHFz/pxuxQR5Y3Ce5Vq9l7zmEoZbRAweAGXPzKSyurbYnbanjj0qw0MYzjd7jsfjMV5fX8+MoxZL73J96WA3gWRWOnc2AAbGA3BuGyxel6a6dyAKfJWlmrljcN9NUsA3z7Tk8ABATo/DQ5kw0LlNsvWjYMf5RHwwUIAbAI/DYP2XmyWnoO/GCB3oZUq5pbCd0aOg6FiK/s4ULQOYC6PXWumgTXm/SH0/8Dy1jZVl8/gqt4V71vqut8TVu6qrvo9qZGj+7+/v8fz8fOKyR3y3/rPL55AOdhMLKzF0Hp5teTweY7/fx8vLS9zf35+wJic/Yzlmyo6VPLuTuOwcFmVjAGJWpEpM00FeCsysoNgYqMaBHPg4o4KVWgVCY1gvx3HGQBZODQptk4jzU64rZV+xthZjg2RMSZ8BPm6RuSp9ZeRcJiwr4ZPKcZ89Ac5o0+faahsHbBFhmZyKc8kj/PPz88n7q+/y3d1dfPnyJW3zLtNKB7sJRC1GCJgJpuljATV29HfrlbKOmSnX7B6uZ1azKhe1uCNOzx3DtzsOBffYBcaz2rRuWke28pVBIE9lXFwHncGKI2p0V5dL2y9jzFn8semMAbWxLMc924xturzUIHKTL5jlqVeA4xyPx4EV4n3QGbRVHbUdM6Ab0zZq/DBQ63uoH5R9vV4PfXc+nw/p6TvY5XNIB7sJJLMeeVkB7xSCOLvd7gQQMiXlXFut8JVCcACrCsaNFeqsTVxjoGPF7sZS3NIKpAPjAO5eZZEMyspIucxunLDVBupm4+suLJenEk2rMmYUeLjt3DPjb40zhum5vJnNszAIZhOqjsfj2eHF+rwRXsdMuVxjmap+u7huDFLbiPsvezze39+H8xXn8/lZuDHt3GU66WA3gbgXnpcVYE53XhYAACAASURBVJNj7Wzb7TaWy6Vdd1Tl42a7aZgxylxBjsPxjhoKVg58OS1WljwZRc/GQxxeAK6H07JiUZeSqycm/ThlrCCeLWbPAE7DsIwFmIxhX8r2MnB0eWZxFKyZvcMNqfEAUm7sFve13d/e3s7WR/JzyETfvVb9+H/WpvxOsQvTgR/W3rGBmoFql+tLB7sJRDsWXB/H43GYzq2dFsof4xoRXomyOAXccgXxdWYjChxw/fE9tb5xjBFb7Dz2gqNd+DrSVUXB7QNGwGNEqpSVNerEFQZmbGbNblhs5Yb4UOiqrLP2zJ6HGgkVO2Pm4PJQBttSpi79S8rprjOI3dzcDOs/uZ34XDw+BSF7v1BvPA9Oi8vuWHDWDsruxwi/R/rRcBA+9qiK0+X60sFuYuEZXPv9fgA6ZXVQFLxzRUuxZsqAryF9vZa53lSp6HoqVUiLxeLkrLmI82NjoBz4AFiE02UH6tJkppOxF5SPFRGUMJQp2thNkIn4GBN0TJMlu+YkY9VjxTHRShFzvIwxalpjATTi/HQE/OaxVJ6Y4lgx8uXxPVxnw6bVZlUdXRvxR8fpWuN2uvyFz3fUNumg93mkg90EghcejA7uD5zNpmG5U0FhZx1njOKsXFoZIKpycjMstUzKBDkdndWWlUcBi5cKKFgrGPJ4EtYAOsanIK1pVuCVtVclVdszqP+MYrxEyWsZMrZa5aWs+e7u7uxAVw4PrwQbFo5p4lnr+8flzOqL+O55ahgWt7YO/S6bUMYGGr/P7DmoGGeX60kHuwnk5uYmFovFoNT4ENKIfDA8IgZgnM/nsVgsbPqqGDIlkSlmd91Z4eye0jVUbqwG9YXlq7NL3dR0XlTvLHpmuaxMAHDuJO2IHywSLlOkXS2rcG3o0lW3n4apWN4lz0PLyWlwPg7oHDvSOC4fBw4Ij3bkmcScBxsV2kbs/kZYB4AcT+Noe7TApdXXFOQYBHkrO37HefIV3i0GvdlsFsvlMi1Tl2mlg90EslwuBwt4u90OrkvXAfGff8OdCcCMaI/fQTK3EX6z8Po6KCUeU9OZoZoGszdWTm7TaFZ8vJSBJ59kO1RwfFVAmN3qFI+mo7MyKwWteY9leJqvtptrB77v2GxWLidjw2UGT/W8dcyW3dez2elmCOqS1PeGPQAZE8b/FrhnbNelPZbdMgAyILLXQo3A2ex8W7Eu15MOdhMI74qCQyAhqoy5Q0V8MBCwlmyqfMYs3P9MgXB5ADY8bpUpBgegHEbX46mlDpDjLaRUqbGwYgHzVbbHDEKBU9OCYs7aR+tWsTpX/woIlVFmkoGeSw8KmMvP9/RdcWkq0HF7alm4bZlZ83PW2ZVaRlcHLV8WlkWNSH1GzOIqlqfPxRlrWheN69hxl+tJB7sJBECHLcCUuanyYHbFzA7r7vhsO00DMgYQK+WN//jocgYGEXZVouOrxY0yIyyvk5vP58NOGihDi9XxeiescVJmxErIsTYNq/cyUNNvB5J8jZ+nimtzZa1VeJeexlflm4FExpoc+OPZ6bvqgNa1TcUY3aSWlmQg6dopAzxNj8vCv90OL27sb2zZu0wjHewmkO12e+JWc5MRVFmrwgC7WywWJywE9yNqZXopy8PHHbPCogP5qCOnzR1/sVjEzc3HXpi6fyWzBc6X8warOx5/uC2xI78qHzcGw+u5ALRc3qw9nPvrEkWsvyuLPwMGJy3m4NLKjC1XRvcMWdGr2w7vauaS5YX9uKbt7wy5qt3UAFGwzgBbAU8NAweGzhDUdtZ0unwO6WA3gfAZWGpZu2/uiDwpBC67bANdFtfBNVz2m9PWiSgKwmCd3PnVjQQ2ikNdeasuzU8Zhq6rU6VW1Z8BWK3tMQYCp+XCZIrOhdW8M2NFlekYBufydnXMGGPVFi49tyTAMcZs2QjS0Ly5PTMW2gK+LA4PC2gfg1ei6ovuv5ajZUR0ua50sJtAKoDT3+iY6Hw8MQXAAoan4FlZ5BG1MoMoAGVKCTPUWAGwqxJANZ/Ph91ikB67FF2+fI2VpFNkEAVlVTyZcaEKS9uqKpe779o9S0efjSroLK4DRzUGNI8M0DLgyJR1xpI4Pd4STJ8Fr6fjQ165LFkbch5uvM/Vp+p3CnpaDi0311c/Gk/r3eX60sFuAoH1iE6lnY87GcLyeh5snot1d2B3OumCLXeI3le3kxONj2Nd8J9BOeJ8IS0UGT6cJ7ssOa9LGEoF2hxW2xpp8XUFbJR/DGurwmT3nBJ0LE3z4Pj6O2sDLUcFIhVb4jhqYGVGA9qQ33v3rrJbU9fiOaaUsbnsf3YvAzy+l8XTYQidpTmmj3WZXjrYTSDOzVGN22ELJu2IvDPIfr8/meWoebUUZ6bs8FsVMJYhcDgHGszmHLBlO5aMZUiuTspkIWgfxwq13KgzM0+XD7fPJaJAxW2s5eE4Gdhp2i4tLnurrbSe7tuFUYbl4vE7rQwcbe4AGu89PAb6XurkGI3L+eg9x+qyumh5Oe/j8Xgyw1rL8ssvv5y1cZfrSAe7K4oqMrA6bBEWcb6xLsLu9/thQe8l+eh17sAVQHI5IuKkjLjPywf4g3RcHpXrylnyl4K41hPjoAyADHIKdFxml8+Y8iFcxUA0XguUxuSZhc2A1aUHgHJKP+J0bSbfYyDDf95VR8sElyavzVM2xe9SJpVBkLE6XTenbB+igIj7mGXt0tYNrrtcT/qTmEAYrJRRaIdWtoSp9QjDoLjf70/Gw5zircCBlT1f1/hqgfOAPjo0ZlkywOl0/0qhc57KgLJyt1ifxomIk9mll7aLMjIO41hYBsKO4SFNN6W9qtsYIBzTXq1n4+qJ37zZAI/XskGh47lIk40NLa9OThpr/GQMV4FIPy6Mtj9fQ944g3I+n9tlN10+h3Swm0gqN5NzabISYGWATgSww1IExOG4ml5VnjGgwWvqkMfd3V0sl0u7A0lWngwsXFkrUGkpY1f/zADI8td4rNTVTavpsNJXRpUZGC5/VzYGEgWD6tk7Vpe1wZj3QtPj+uryg5ubm5PNuREf7YH1khAGDlcuV/eW6PNxIOeAX/su1wcf14+7fB7pYDeBOGbHoOVcJc6y5q20mN1hrRgrmTEunwyE8DtT5Ap0eiCnKvIKnNz/FvBkdVFg1Hq6+2OAIgNqxG+5khUIWnVWEGul2wJ6zcOBR5W+liuLq22pzCwiTg4r1nT5Obk+4d5pZWacP6eFPuMAia+zMaNr8Fyd8d8d04V0unwO6WA3gaBzqvtDB9jRwXSBtDIwBjXdRkwVEitjBSNXRg6X1QXgxietqyKq0tD8HKutGFqmhCv2pyDsGI7Gd3EU1LOy4ppbp1iBiz5zrY8r6yVSAZh757K42n4VU2cA4TBuDM89c8e6qvczKz9Ex78V0Ny9rHz4jwk4Y+rT5TrSwW4CyWZe4pstT0xQYfclT8/GNQ3vDnjNLH6nzCGOuag1ywDnGF3E+WnpTrK89F5LiaiCc5NMtN4qylq1rTTPDAzYjZftCqPM3TGbLA/Nz5WT/ztF7dpWr7vn32q3qmyc/vH4ce4hNiTQeAqW1bPT8jjjScXtUKRs0I2h6w4wuL7b7YY1sTrRpsvnkA52E4izHB34HY/HwULkDsrLDtyOIvv9fthdPQO1llTKimeaRXxs3Mzl1CUGqkirfDKgydJT5ZgxuSpflrHsSBWxtjWeD7eFbouVGRhj8q3KXQGbhtXwCnyt58HxszZxaTDDi/g4HojBjOMxADnXflYuV19cy/phiz3yO4jnzMz0eDzGZrMZJmopi+1yfelgN7G4DseLUhnsWKEw4Ol9LEIfO8256oAMXup2jYizvSw13iUdPHMPKQOpwnCeDgRdGTNFmDEdvq8bVDsGBGFWoIzPsbusbC3Acgpa2ZRLo2KGWRla+SMtnVzCz0fbQsetM9Fxu6yulejz4e3C0M/cM+VhAn12+N5sNvH4+HjSx798+dKsV5dppIPdBFJZ5ug4AKz9fj+AHndsAA/PvES6cGXqaQjIo7LAXTkdYCENnYyiaWaAxOlkbVOVKRNnhVcMUN1jY5gWlB3HYTeV7g6ia654f1OXtpbHAZIDI63n2PbmfBwYtkAvA0y3swiny655Bve7uzvrMtS8+Pm590YnmWi8jMlhJiW3hfYFfgf0XcK9/X4f2+12mDA2xpXfZTrpYDeB8PZf+OgWQwjDu6c4378DGkxUWS6XpeKqXDQsFbPUTuwYlkqmHLNrlUJtxeVr3A5OCbo2cIoy4vyUblZ+qrxZ9Jy8FvtwRgSuV+A2hhm5eimYVFI9B7cDCafJBpuO9XLaypxadVR2qW3WMmR4Qhinp8/KuSW1bofDIZ6enmKxWJyk1eVzSAe7CYQVATozfuOb2R2zJ9x3CpOVB6xTBscWK8iUF7uZ+JpTUpyPy9fdZ9eeXsvSdPfG5jMWOF2buPC6cwy7JTkeA2P2DLP6tcri2HomFVt0UrEnzVfBqbXlFsdjl7BO5uG1nArI2bIaZ9zpfbfEwLVDZrw4Y4Tr/Pr6Gl++fBnq1pcefB7pYDehZJ2K3SkKBOig3MGgCJhRVK5MxHOTSDKFyvfZqnXWuJOK4VVKvEozC6PxnZXeSicLp2mxEtR20bTdGCvnlQFexu6qsmnYqs4VyLp0sv8ttsxpuWfC1/R9VgOB09N4Tlph+Hk6w0Gfkdug2pXl/f09np+fY7FYxN3dXWd2n0g62E0gOq05UzhszUacKlR0NkzV1k2Oj8fj4MqsFEDG9jKFkH1cnJbbRvNWttOSjGlkdWgpdQ6TAV3WRvqMWBlGnM9OdellZcH1inlVCryS6v3LypSxQuehcHXhuBmIc51ubn4c7ssTR5T5aTk1X1dHBlQGMlcW3QAAZYn4YHI4DUT7xcvLS3z58uVsR5gu15UOdhOIsyAh3Ol4cTi7KNldBFcldyIohGxhK+fVElVGDLRjrOQsP1VSXO+K4bhyjWWCl9Q3U45OXFl5+zAFvYwJZcyyUthj61WVt2KEThz4qfvPtZ3mU4ETz7Tk943f6axs+l66tkRYXjKAPucAlSesoN/pe8fgfDweT3Y42m638euvv/aNoD+R9CcxgSgw4Zq6xDBQDtBjRTmbnR6Y6lxomKjixtyQDyueSgk51xLf03BVnTmOsroWs6uYkF7PgH4sCLfCtNioYwRatirPrBwOnJyhMAa0LmUamr6Ln9VX3xO9pv8VdGDU8dFSajTq+5eBnBpk+ux0JqhOSHHPAM8b6wVxtNV8Po/b29u4v7+Ph4eHn2r3Ln9/6WA3sSjIRfzoaNiBIcKzPV3zxruo4D9vDu2USkvZQFxYrYMC7VgWpKyuiqv3WmWqpALJVn2dMnUgz8/LlcsBR8XAMxboyuDYjHPXaRn0WWheWl71UlTgp4BcMXI1CB1bqlhuVQe+ps8NrI3Bze3pyUx9NpudnNmI8Tmkt9/v4+vXr/H09HS23rDL9aSD3USiLI7ZGDoIu3LA9JjhwZLE/YhzFyLPNquAwgGNUxYVq3FAl4V3Ut1z9ysrW/+zgtGTB1r5ZnVy191aKgcEYxhSqyyZ0nTgmEkLEDRMBXZZu2dl5zAOnHFdWRXOuuPNo7lMP/uOYbkA9xkGV46PccTlchmr1WoYj4OR+vLyEpvNZkhvt9tFxI/xuw50n0M62E0o3KnwwXID3TUdnR4KVfcQhID18eL0w+FglyAgbVzLFBP/HwMMLo0KXDNQrdxRFRNsKdjs2phyV3Gy+4696xFNfE9lTJ21zBzegcklgKjlcCynAl29NoY1KqAz64w4HbvWe4jHHpCx9cOYnY6Nc7rz+TxWq1Xc398Pi8efnp5iu93GZrM52aMWAiB9eXk5q2OX60gHuwmEFY26a97e3mK73Q6WIMBN2Z8qMVUU6KToZOzKzDqaKhl3Ta9r+MqNlOXrFG/m4qpcX5qGK0clY1xfY/NtXdc9MjMw4vz+FgWpdavaxN1zgKxA4OLru5MZOVU5ZrOPMTNdU+d2oamMpFZZlaHiP8bgHh8f4/7+Pt7f3+P19TXW6/UJG2QDFoCJ4YTO6D6XdLCbQHQrpIgP4APYbbfbs1PH1a2pHZ5BUZmi7iOo4hiAAz7IGNDJ3IouTgWEmcJyynMsC8vYTRbftUtWN8eW1Rhx4V0Zqjo54MhYYGXguLbIrnGcDBSrNufyOwCu2kbdmY4dV8aAllfjqRGK/3d3d/H4+BgPDw9xPB7j5eUl1uv1Sd9CeXR2J67PZrOTg5W7XF862E0gDHToKOg4b29vQ0eazWYnExzYOmbwarmX2JXZUgaO3Tj3mVMSLFU+LWU+Jh5fc4q3ZdG3wFqlWidXgVV2zdWblX3lFszuZcwtMxJaLEsBz71nDqz0etZeY5k533Os1zFATTPrI5wOv8tY0rNareKXX36Jm5ubAeR4CEEBzfUTHKi8XC7P8u1yPelgN4FUrG6/38dmsxnAbr/fR8SHslVr1n2Y6eE/LFC1LCuXE36rErsU3MbcH5tGBQKtNMdY/1nYMXV27XaJVO2u4Vp1yfLW552BHP92YOEMq6rOFfPMyqbhszCurSr222LWET/Y3K+//hoPDw+x3W7j6elpGFpwBiD6G4OtGhWbzWboz12uLx3sJhB1zQCQDodDbDab2Gw2EREnLhFs+4WwCmrs4nHgx67M1u7rFduqFEsmlcJWMKmUvEtrDMvJ4masTMvlXIIcvqW4x+RZySUANQbk+JpjTi6Oq9+YOjMbq57dWAMkew4tw4LLUBkwx+OPGZi///57rFareH5+jufn55PJJloGZoMs6JtYknB7exv/8i//cvHEmS7/GOlgN4EwCDGj2+12Jy5MBjC4M5WhYfxP13LxuCADHvbnU2XeYm4sFWi0LGYXz6Xn0nbt6MpTpe3KxWEdAFe/XTlb5XeA2SoX389AZizgZiDhlHbF8i6RDLDZOKsMgTHtMCbPCugATNja69u3b7HdbuN4PD0kGf2JN2jXduF8MHtztVrF9+/fRwF7l3+8dLCbQBjoIj42bT4cDvH6+jqAEu+xCMBzC2x5obnussJApxNVWtJidxzmZ92UGXBWrroxgJyVLWOlCv56v2J2FROs3HYVkLswMIBYLgU0vpYZJqrYHbiMKbcL556vu1+Vk1laVQ7Ec+5FsC4F019++SWOx2N8/fo1drtd2hYcR9uHy3d7exsPDw/x8PAQT09P8b//+78/ZSx0+ftLB7sJhDsNAAhuzO12O4Cd7ovJ4AYBcDHzy1xGyCcid+Fo589YSgU2XOZWuIwNteJVddC0FYw0T2fxZwzDhXFAOUapa3n0fgWUY8SxIXd/DIhlbjfHZqq8mSXxvUves6y9XPu7NDUOyrpYLOLt7S3+9Kc/xWazSU8aycqCfo148/k8fv3117i/v4+//vWv8T//8z/x/fv30cZal3+sdLCbSNgdgo+ur8Omt7e3t2ebPTM7ROe5vb09mZSii2P52B+2aseUtWJlrDRbTE/DVWHdvbGMolUG939smcaGcayxAjYH0I5ZXQJ8Grb1H9cyo6GKl+Wt7aQTqBD2kvbPQJalxco5HPrOH//4x3h6ejrpS1X98a2zde/v74eNn//v//4v/vjHP8bz83Mfq/tE0sFuAmGgw+/D4RDr9Tr2+33MZrMBkCJyJoH/ADYc9Jqt49NxPKeMnbSUUBVfy+pcPa241fVM6WYMTu9p+R27dflkirzFVJyMBa/ZbHa2RZam4dhMlr6+B2PSyso9hvW34rvwYxg/55sxOBeOXY3z+TzW63W8vLwMbayne/Cm7frN7PDLly+xWq3i/f09/vrXv8bXr19ju93aNulyPelgN4Ew0GEfzO12G8/Pz2eTSNxpBxEf1jFPVMEYQ6aYGPBarkBWlC2AqwDH/W6l1QIqDXuJtADelbliNA4cs3I5t2FWPy0TP4sWU6nKyr/5k8XlurgNkbN2GFO/6vmNDeeejQNFdw+/F4tFRMQwCxrhYTxyefhZHI8fLsu7u7thbC4i4unpKdbrdex2u2GSS5fPJR3sJhDMvsQ4HdbWbTabE8bDnVx3S3fsbjabDUf6sNuTO6djfZweRJkXX9M4lTWdWfwtFtC65/JtsQsHcBk4jQHwyq3l0tL8K9aV5c111PhVeVxYvsa/xzA9J+6+ew/4m42zyvDI8muxT01L7+M4ntfX15PlBQjLm6+7smOmJY7ueX19HT6YZHY4HIYJYj9joHX5x0gHuwlkv98PE1MwCxM7M/DicZ58cjweTxgfxuN4XI7BDmEzwHNgNkZaTC9TeMoSK2bprv9MuJ8BsCpcFa/Fsqr4jvFl5WDAG5ufA7nsd4sd6XvjQLIFmvyfZxw7lox3/W95RtVzv7u7i91uN+xdyR4XHlvkda44xuf+/j5Wq1Xc3d3FZrOJp6en4aSD4/E4nFyOvTPf39+HdLpcXzrYTSC8Mexut4vNZnNy9AdAEICFa+h8fHglK4nb29vhxAQcGglR5aQgiXstEMK9n3VjZgzSxXH3NYzezxRvVQbNb6y4smsbZuk5gBvDVDleBnqOiWVMzl3n78og0nbPWG3W7g7YnFSTqTR9bZtM0F9ms4/Tz8Hksq34cGYdTjyYzWbD7iqvr6/D5DL2wrj1rq7sXaaXDnYTCGZF8ikHcGGic/ERIbAutQMzEPJsMCxQXy6XqQLiTqez4zK3kpNMsfwsM3P3W6xR/yuLbIUdoxhZHHPJ2Azn49J0AMdpVv+r9CtQ43s66YLjIO3qjD7XBi5sVkeXn4YZA7iub7SeL4xD3auW+8PNzc1wZt3j4+NgQG42m3h+fh629uOJQ2xIMtjx0p8u15cOdhMJrMi3t7fB9aFjA2Bw6ubJLHq2KNfr9dBJVQlwB0R6Y8DJAUmLnWRpueuZYmpZ/q10LwHELL2W8syU8xiWN+Z+Vpax8ZlZZCxOf+M/r93kfDJWrs+Lw3G7Z2ll31ymrL0ZNKv2wLg2Zj9zXnBTYixuuVzG3d3dcB7der0elglp3ZCWMjoYr9ls2i7TSwe7CYSXCmC87nj8MSZ3d/fjEfC0Z46nCkAteNx7e3uL19fXE2uUrdexlrZe1/CZ4nXSYoyubhlQjRVVilqHStGPTXds/CyOA8QMTB1QZHkrm8PYr3u+VTsjfHZEVMY+M7aZ/XcblWdMT9PBNZ65XNWNmVfED5bHMyrv7+9jPp8Pp5DAW4KJZWowIi1MDOP9aPlEkw52n0c62E0gACN0pJeXlxMXBzo8OhIWuGrH144NixLKabPZxHq9jvv7+yFMpuxwD+mwctN7Whf9XbmzMqX1M0BRKb2M9bSYXAamY1kl5+3GRJ27Tg0GDZeV5xIjhdNuudKq56dlcM9UQdgxacfkItqA59pwDEvXsmK3IbA4jI//8ssvsdlshjE4GKToD8yOeZcglIHH/xgU1YV5iWHV5R8jHewmEGZ26FR3d3d2cJw7PyssXv8DS1XdJ2CNv/3225lyudS1WFnWem1s2u5ei3FeUvafLdMYpam/3T0nLdYxJi1V+Nl+mcrsFPS0PFl7OaPEgXRWD8RjRlq1D4cdwyazumt4fnfQf7BFGFjbn//852FhOfoZysGszLUxAxr3UUw40/J0wLuudLCbQLA84HA4xPPz8/Cf3YuODbBwON7aiMf3MB6oYy5Vurh3CZMZI2OVVeYSq+Jk6YxVyE6BZ2GqslSMkJW4/ub7mbLmMJy2pqNGg35nbaJjuO6w2rHi2pvfPcfotOzud1WWzJjj8AycAKfD4TDMiH5+fo6vX78O/TEiTpYLuLZlkOPJLmx4AgB58ksHuutLB7sJBJNRdrtdvL6+DsDEbhJmc6oEeScL7oisSPEBe8SiV51SzaLKIgvjGFYGEmOujckrE8cItS6t8C0Wc4lk4DMmrF5XyZjumLIqeLj2YuWN+45ZZcZBBlaZcDpZ+q4e7tlpGjqDlF2PAKL9fj9MGnGAxRurw4gE4+Oy8rgc2lAPWcbmES3vRJfppIPdBIJZXJjVBZcHD2hHfCgNdFBVytnYBluXADt2dVVgBqncXC1FdmlHHuMGc+mqy6sFMGOBocovS98xK2cYVExTjZqx7eieZ/a7eu541yLixB2HmYtZfR3AO8CrDAEFTYCIY5iuL7j6sdHH7n31okB4EglPEIPRyOkgffx2SxgAlAx2cGde8ny7/GOkg90EAgtvs9mcuDAxGM4MDx0PwmN1zPDQ6fRYIOSDgXPEVYVTuYe0Y16qiCvF5/JrsaHs2hgGWN27hB1p+LGuqQqUsnIy63LlcMwe9y9lzDrpiZmennLvyt5i0tmz0veQ6+QmZ3EZXHsy0GVldOnyBC98ePxOr3E/Y6CL+ABPZpE6dtfletLBbgLh3VPY+ouIoSNFfLhi0CkzFyS7Z6CUuPNhM9rHx8cS3Co3lHO/ZO4wLZuL45SdWveaTpankxaDaQFOxj5UqjLztUtdfJqHKyMDE28tV8VB/vqOVGw14hQEdOwrY3ecb/Z8Hbi5jysTA1LF5rVMXH7uV1p33stSN1BHm7BbVHdOwUYRfDgzG51drisd7CYQgB0OagWzizjtQPjgGisoXMe6PBYoJN6T7/X1dThfK1M4ThFXriYnY1ifKr+MgWT5jWVpLcDLylvFbTEUvqbjNg7kMoDI8s3KoqDhypQ9Xy1PlqayPA7P+epvJ9l7pQDsjCO+3mqrlpHk2DC7MbmsejCyuioZHLEeD2tcMRlGx/y6XE862E0gu91usPzgtmTLOeJ8yyG2YnVcJeJcsULw/+Xlxe7qrlIp/xaIOWGF4cCtUlhjQPNvVRxj2NiYvDWsYwsu/YoRZaypVQc2lKr2qfJyaarArakMTOuuz71i1bpbEG92ruVw9cvYJjPg6r0HKCFvfLsJL/hGn+R+CrA7Ho8Do+Mxuy7Xlw52Ewg6wmazOZnJdTgchp0c2LLkadDorMwIOSw6HndyBlYFzojxLkQOWynRDDA1LSfOkq/yjPyP+AAACpRJREFU0bhZ/q38xuSTKVaX5pgyatifkQoA+b3RZ13NyuU0tC313eGF1fhWdx/iuXfHsTj3n9/9FhC3GHL2fLBBO86FxJIe7mPcp5wRqutcI+JkvA5enS7Xlw52E8hms4nFYhHb7faM0UXEmSLR/3xenYbRjgQw3Ww2di8/Fyfib5/QcQngaLyxkgFjxdY0j8q1NbZcY1lvpsi5fFm+CgTOYNEyRXywr4xVZfkpoOm75tg58nJsT8vmZgdrPfRdbXkYlNFl7ZEZa4jLk8T4aB91X7IRAQOU7+E3wuz3+1iv12f17HId6WA3gTw/P8fj42Nst9uh4/PMLR4HUBeMKkpmhLiOdXrc6Q6Hw8AkVeFVSomFFc6lkikhxxovTdfFVYVaxWUlqdICQ9cWWRtd2nb8bDJGws+uAlPHQPhaZmDxb2cc8XuK9423t9PJLBrPuSUrVjjG0PoZ4bbT9lE3JoO/frOXhpcAYQMJrKv92X7U5e8nHewmkO/fv8fx+MOvD1CCNQnQiogz9sbAyL+Pxx9nbfHU6Ig4sSqxD6dao0i3EpTBTYbJxAGoAxf91rhj2WOLiTkWlf1XduPAw8XlMBqvxUo0DQcEXD43Jqj5cR0c2Ol9rX9VZy6fsi89wcPVI2sHfUe4TJewZy3/GEMG7cfnRVZgx//ZqIQ7FG7M9/f3+P79+7Dhe5fPIR3sJhCAjk5b5rV23Nlub2/tjEy2KhEW4ix37PnHYVnRZYomk7EKXNPMlJAqt7+n9evcYVo+LZvey9LUMmt6nG4GzplbswqjebPxw/V0ZXLr6VriGK0zmJjpRZwvnRmTD6fl2mwM06sYqfYNTtd5VnQcHKInGwDseIxut9vFy8vL0BZdPod0sJtA3t7e4tu3b7FarYYBcAY7sD0IK0xeNA7FxlsZRcRZfMh2uz1Z55Mp9RYjU1GGpjLGQs8s+Ra7qK45V15WV1VurXSyOlSAxZNCKrBw5XXpcxjH5iLOJ5A40XauGGXmSs3KzZubMyPVPFpMj3+7clRGgP53oBfxccwPp8+uSLSRzsBEHflA5v1+H5vNJr5//z4AJU+w6S7M60sHuwkEsyNns9lw2gEAisfZ0Cmc9Qqg4/PvWJkgPDM/dEAek2DJOqAyhFZ4F7/678S5PC/J5xJ2qr8z9qJlw++MHWcAruslK6DT9LK6OZDlazohRNNRj8HPGBiclssf43lstGWg13p2Y40ajeMMAgjOs5vP50P59vv9mVtTAY8ZHYBuvV7H8/Nz7Ha7uL29PRlmQDod8K4rHewmED6hHKcg66C4LjaPOF1wDnGdnN2jzPBw5I+buFG59VRaDG6sO/RnwJLBJQOdFmsaoxwrRjmmnFkc10ZVWaoysFeAw45lORWb4jTd86wA110HkHG6PHtTw1Zpjnk3K+OoYnYYMkA6WCaE/LhPsmGAsuz3+2EiCk/WWS6XcXNzM0xKy9qwy3TSwW4CgSsRFiAsSXSkw+EwMD6IghcvOGcX5vF4PJuogu+3t7fhoNhMKheWu16lU7Gbn5HK6q8Yx6XM8lKAztx/yly4PRyzGFOusUCWiWOhmkdEfuI33jl33RkZyl45jJ4SoJ6JVp2zcFUb6dibaw/uSwx2mEDmxthxTBBONUc/RLoAUN5lZewz6/KPkQ52E8h+vx9edrA7npXJioFBTcfpIs73+eO4qoCQ35j9+TKlnoFVC9S0PC7M3yJj3FiteE4ylsJxnYHQUrwIewnA4flWrrAK9Pk/P8uKkTtgzlhW9X5U7yWfNBARJdPTMjqW7oDatTPKy+ANw5P7HPoLPDDsrtzv97HdbmO9XsdutzspqzJANmrdmHqXaaWD3QTCYwDv7+8n7E6VBndcnW3HpyWoElF3DD44aeFnrEqnwMdaqC7MP8qFk6VbKUCnpMf8vwTcNG4GVprmJdKKq8CVPRdlHy23rgKn1oVF14ty2+lWXZpmVufKIND6633+zOfzs3KxRwUsDpur86Svqu3f3t5OwK6zuutLB7sJRF/0zWYzDGCjo2Mj54g4mSHG7hFYnbPZbFjbpJYlJr9A0EHdBBWUbYy70LG2MczNubjGSAUKWRlbddDfrbyzuiiAZOUc42pzLIZ/uzauyu2YHpe5CqPpjGGRVVn4N384HXYR4j6vP3UGigMwLSd/+IQI/sxms/iP//iP+NOf/hR//vOfY7PZxNvbW+x2uzgcDvHy8jKAHCauqDGpZdBnxa7NDnjXlQ52E4hTGq+vr4NVqR0kU6bs9sG5eFAOET8UBzol7h0OhwHsOH0tmyoRB2KuQ6u0GF3L7ZWlOdYd2ipjS0E7F5/7rvLm+FU5tKyZYYF7lxgK7pllCrdlLIzNKwNJZWxcJ50AEnG6fILTytyVCuj4IB1dOsCuyX//93+P//zP/4z/+q//iv/+7/8e3JQYk9ONnFEmeFJcu+E3b8zQge760sHuSoLthNiFE3GqVNEx7+7uTjaLZkXGCmA2mw2Ax3v88bZhGctRpahyKTtqSYsFslTA1gK9FhPi3648jsFqXAULTisDtKpeWVmziUYKahmDczIGRDOWnRk2zoBy7aLhkB67NLnO6BN6T8HsePzYygxh4I5cr9ex3W5js9nEy8tLPDw8xGaziT/84Q/x9evXs8NXuf7oq3xqSfZeoHxZf+oyvXSwm0i0Q0f8OPrn9fXVjmnM5/Mzlscz2OAe4Q4HaxO7OGDRLGaM6W4qEHVxZSyKFZdLJ6u3yhgXWUsqhlgBnQO4zFU4hsm6tB3b0fKMMR4urcslrDozdBxwV/c1Dw2XgZvG47PjwJzYYwHhBevs/jwePw5H5lmTmOm83W7j27dv8fLyMkwweX5+jn/913+N2WwW3759O9u0QevEpyK49tffCs5drisd7CaQzIWEySroROzjPx6Pw9gd0sB1zO7COiFYkRjTA7BhScLr6+swqO5cpKrwx7jqMiv9ElHw1DJVcS4Rp4haYKtg0GKQlSLXNNS4cGlqOhUIabhLDQy9V7WXuu5a7K51DyDC5VAGpe5LNSa4Pgx0mD2JnU7wG+2PfBeLRdpHuYxuVmXVVrPZ7CS/DnjXlQ52E0imuOBuAbvTew6UeOEqpkZDOby9vQ2D6bytEQbaeb/NrIw/U7efdWteAnJ6P8tTQYTb/mfZZAbsFUi0yqwskH+3gLUqkyujli2rQ2Z8cFn1yJwsbOuaut9hmDHYcX7KcrX+SJO9HxwGhiFmSAKEDodD/Nu//Vv8/vvv8Yc//OEE9PAbQwIAuqqOeq9a49plWulgN4GgM7oZkXC/PD09xfv7+7AGD5/dbjeM2e33+4j4WOyqh77OZrPBlQOFdHNzE3/5y1/iL3/5S8xms1iv10PnRRykqdO/2cJWpez2fVQZq/jGAIZrz9b1sSyuSkdBopKWu8oZMFm6zg0K40XTYsaeXc/KUNWpBaAufPYM+Ld7Hgp+bnw6A1kYcWzk8cGs8JKs1+vYbDZDOjAEF4tFrNfrkzE6dmny2J8zpLhNxoB/l+vI7PizZnmXLl26dOny/4n08ye6dOnSpcs/vXSw69KlS5cu//TSwa5Lly5duvzTSwe7Ll26dOnyTy8d7Lp06dKlyz+9dLDr0qVLly7/9PL/AKeTXdLv7Tr6AAAAAElFTkSuQmCC\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Should output 10 images for inference using test dataset. \n",
        "\n",
        "Looks like the model consistently output 3 labels under ETT, NGT and CVC even though not all 3 categories are labelled for every image. "
      ],
      "metadata": {
        "id": "JZs7sA4iMPZL"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Construct Kaggle submission file\n",
        "\n",
        "__Reference:__\n",
        "1. [Kaggle notebook](https://www.kaggle.com/code/ammarali32/resnet200d-inference-single-model-lb-96-5)"
      ],
      "metadata": {
        "id": "ugCYbg3YRset"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "test2_models = [test_model.to(device)] # make model become an iterable"
      ],
      "metadata": {
        "id": "X3U-Nult1PJ4"
      },
      "execution_count": 27,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "TEST_PATH = 'test' # define for following class function"
      ],
      "metadata": {
        "id": "xhp2xzTK0wHJ"
      },
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "class TestDataset(Dataset):\n",
        "    def __init__(self, df, transform=None):\n",
        "        self.df = df\n",
        "        self.file_names = df['StudyInstanceUID'].values\n",
        "        self.transform = transform\n",
        "        \n",
        "    def __len__(self):\n",
        "        return len(self.df)\n",
        "\n",
        "    def __getitem__(self, idx):\n",
        "        file_name = self.file_names[idx]\n",
        "        file_path = f'{TEST_PATH}/{file_name}.jpg'\n",
        "        image = cv2.imread(file_path)\n",
        "        image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)\n",
        "        if self.transform:\n",
        "            augmented = self.transform(image=image)\n",
        "            image = augmented['image']\n",
        "        return image"
      ],
      "metadata": {
        "id": "lrGNuL_7yWsc"
      },
      "execution_count": 29,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import albumentations\n",
        "from albumentations import *\n",
        "from albumentations.pytorch import ToTensorV2"
      ],
      "metadata": {
        "id": "0wVJEGmA0gt4"
      },
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def get_transforms():\n",
        "        return Compose([\n",
        "            Resize(400, 400),\n",
        "            Normalize(\n",
        "            ),\n",
        "            ToTensorV2(),\n",
        "        ])"
      ],
      "metadata": {
        "id": "2-j_tpYlybg_"
      },
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def inference(models, test_loader, device):\n",
        "    tk0 = tqdm(enumerate(test_loader), total=len(test_loader))\n",
        "    probs = []\n",
        "    for i, (images) in tk0:\n",
        "        images = images.to(device)\n",
        "        avg_preds = []\n",
        "        for model in models:\n",
        "            with torch.no_grad():\n",
        "                y_preds1 = model(images)\n",
        "                #y_preds2 = model(images.flip(-1)) # vertical flip \n",
        "            #y_preds = (y_preds1.sigmoid().to('cpu').numpy() + y_preds2.sigmoid().to('cpu').numpy()) / 2\n",
        "            y_preds = y_preds1.sigmoid().to('cpu').numpy()\n",
        "            avg_preds.append(y_preds)\n",
        "        avg_preds = np.mean(avg_preds, axis=0) # 2d nested prob arrays for each batch of 4 images\n",
        "        # best to convert preds using 0.4 threshold conversion to binary classes before appending using if-else statment\n",
        "        new_avg_preds = []\n",
        "        for prob_arr in avg_preds:\n",
        "            label_arr = [1 if prob >= 0.4 else 0 for prob in prob_arr] # threshold 0.4 probability\n",
        "            new_avg_preds.append(label_arr)\n",
        "        probs.append(new_avg_preds) \n",
        "    probs = np.concatenate(probs)\n",
        "    return probs"
      ],
      "metadata": {
        "id": "NrOVEkvKym-F"
      },
      "execution_count": 48,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test = pd.read_csv('sample_submission.csv')"
      ],
      "metadata": {
        "id": "KMA72uioy4Ak"
      },
      "execution_count": 49,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "test_dataset = TestDataset(test, transform=get_transforms())\n",
        "test_loader = DataLoader(test_dataset, batch_size=4, shuffle=False, \n",
        "                         num_workers=4 , pin_memory=True)\n",
        "predictions = inference(test2_models, test_loader, device)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tWiWZMFAy96P",
        "outputId": "3e92e5e7-c22e-4137-a230-979e9114dcb9"
      },
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 896/896 [11:43<00:00,  1.27it/s]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "target_cols = test.iloc[:, 1:12].columns.tolist()\n",
        "test[target_cols] = predictions\n",
        "test[['StudyInstanceUID'] + target_cols].to_csv('drive/MyDrive/Colab Notebooks/kaggle_submission.csv', index=False)\n",
        "test.head()"
      ],
      "metadata": {
        "id": "o8qXw_jr_t91",
        "outputId": "a6b3025c-ecf7-41a6-a468-97ca5dbae670",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 322
        }
      },
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                    StudyInstanceUID  ETT - Abnormal  \\\n",
              "0  1.2.826.0.1.3680043.8.498.46923145579096002617...               0   \n",
              "1  1.2.826.0.1.3680043.8.498.84006870182611080091...               0   \n",
              "2  1.2.826.0.1.3680043.8.498.12219033294413119947...               0   \n",
              "3  1.2.826.0.1.3680043.8.498.84994474380235968109...               0   \n",
              "4  1.2.826.0.1.3680043.8.498.35798987793805669662...               0   \n",
              "\n",
              "   ETT - Borderline  ETT - Normal  NGT - Abnormal  NGT - Borderline  \\\n",
              "0                 0             1               0                 0   \n",
              "1                 0             1               0                 0   \n",
              "2                 0             1               0                 0   \n",
              "3                 0             0               0                 0   \n",
              "4                 0             1               0                 0   \n",
              "\n",
              "   NGT - Incompletely Imaged  NGT - Normal  CVC - Abnormal  CVC - Borderline  \\\n",
              "0                          0             0               0                 0   \n",
              "1                          0             0               0                 1   \n",
              "2                          1             0               0                 1   \n",
              "3                          0             0               0                 1   \n",
              "4                          1             0               0                 1   \n",
              "\n",
              "   CVC - Normal  Swan Ganz Catheter Present  \n",
              "0             1                           0  \n",
              "1             0                           0  \n",
              "2             1                           0  \n",
              "3             0                           0  \n",
              "4             0                           0  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-a4332e0b-8f0d-4351-9146-3e4227f7a028\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>StudyInstanceUID</th>\n",
              "      <th>ETT - Abnormal</th>\n",
              "      <th>ETT - Borderline</th>\n",
              "      <th>ETT - Normal</th>\n",
              "      <th>NGT - Abnormal</th>\n",
              "      <th>NGT - Borderline</th>\n",
              "      <th>NGT - Incompletely Imaged</th>\n",
              "      <th>NGT - Normal</th>\n",
              "      <th>CVC - Abnormal</th>\n",
              "      <th>CVC - Borderline</th>\n",
              "      <th>CVC - Normal</th>\n",
              "      <th>Swan Ganz Catheter Present</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1.2.826.0.1.3680043.8.498.46923145579096002617...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1.2.826.0.1.3680043.8.498.84006870182611080091...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1.2.826.0.1.3680043.8.498.12219033294413119947...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1.2.826.0.1.3680043.8.498.84994474380235968109...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1.2.826.0.1.3680043.8.498.35798987793805669662...</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a4332e0b-8f0d-4351-9146-3e4227f7a028')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-a4332e0b-8f0d-4351-9146-3e4227f7a028 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-a4332e0b-8f0d-4351-9146-3e4227f7a028');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 52
        }
      ]
    }
  ]
}