flair_cadec_v2.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "xONJpNLKtGie"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "import os\n",
        "import pickle"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nw6RM8uotGih",
        "outputId": "d850bf2d-7bc6-44ca-c8b7-799c8670b279"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        }
      ],
      "source": [
        "from google.colab import drive\n",
        "drive.mount(\"/content/drive\")\n",
        "\n",
        "os.chdir('/content/drive/MyDrive/SI 630 project')\n",
        "# os.listdir('.')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "P9LwpBadtGih",
        "outputId": "704fb914-4baf-4ca0-a285-7619157624d9"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "['sct',\n",
              " 'text',\n",
              " 'original - unedited version of edited files',\n",
              " 'meddra - unedited version of edited files',\n",
              " 'sct - unedited version of edited files',\n",
              " 'original',\n",
              " 'meddra']"
            ]
          },
          "execution_count": 4,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "os.listdir('datasets/CADEC/CADEC.v2/cadec')\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "VNZtmxiPb5uV"
      },
      "outputs": [],
      "source": [
        "suf = '_v2'\n",
        "\n",
        "data_save_dir = 'data'+suf"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "BIUl2-W9tGii",
        "outputId": "1d08f361-ad55-48bd-ea8b-43980f3feabf"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-ac7b53be-a150-443d-bdb7-3eaa5103789a\">\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>ner</th>\n",
              "      <th>meddra</th>\n",
              "      <th>counts_exploded</th>\n",
              "      <th>counts</th>\n",
              "      <th>counts_exploded_rank</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Drug</td>\n",
              "      <td>Lipitor</td>\n",
              "      <td>1081</td>\n",
              "      <td>1073</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ADR</td>\n",
              "      <td>CONCEPT_LESS</td>\n",
              "      <td>1016</td>\n",
              "      <td>266</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ADR</td>\n",
              "      <td>Severe_pain</td>\n",
              "      <td>722</td>\n",
              "      <td>247</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ADR</td>\n",
              "      <td>Myalgia</td>\n",
              "      <td>619</td>\n",
              "      <td>305</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ADR</td>\n",
              "      <td>Pain</td>\n",
              "      <td>292</td>\n",
              "      <td>260</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1294</th>\n",
              "      <td>Finding</td>\n",
              "      <td>Senility</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1233</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1295</th>\n",
              "      <td>Drug</td>\n",
              "      <td>Seroquel</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1233</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1296</th>\n",
              "      <td>Drug</td>\n",
              "      <td>Solaraze</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1233</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1297</th>\n",
              "      <td>Drug</td>\n",
              "      <td>Tagamet</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1233</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1298</th>\n",
              "      <td>Drug</td>\n",
              "      <td>testosterone</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1233</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>1299 rows × 5 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ac7b53be-a150-443d-bdb7-3eaa5103789a')\"\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-ac7b53be-a150-443d-bdb7-3eaa5103789a 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-ac7b53be-a150-443d-bdb7-3eaa5103789a');\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",
              "  "
            ],
            "text/plain": [
              "          ner        meddra  counts_exploded  counts  counts_exploded_rank\n",
              "0        Drug       Lipitor             1081    1073                     1\n",
              "1         ADR  CONCEPT_LESS             1016     266                     2\n",
              "2         ADR   Severe_pain              722     247                     3\n",
              "3         ADR       Myalgia              619     305                     4\n",
              "4         ADR          Pain              292     260                     5\n",
              "...       ...           ...              ...     ...                   ...\n",
              "1294  Finding      Senility                1       1                  1233\n",
              "1295     Drug      Seroquel                1       1                  1233\n",
              "1296     Drug      Solaraze                1       1                  1233\n",
              "1297     Drug       Tagamet                1       1                  1233\n",
              "1298     Drug  testosterone                1       1                  1233\n",
              "\n",
              "[1299 rows x 5 columns]"
            ]
          },
          "execution_count": 6,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "meddra_counts = pd.read_csv(data_save_dir+'/meddra_counts.csv')\n",
        "meddra_counts"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "xzc4Zo0PtGii",
        "outputId": "ea99a2d4-6e17-47be-eae7-4dab0706a837"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-2f0fef57-5422-4d06-8869-ca5c141095a9\">\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>ner</th>\n",
              "      <th>meddra</th>\n",
              "      <th>counts_exploded</th>\n",
              "      <th>counts</th>\n",
              "      <th>counts_exploded_rank</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ADR</td>\n",
              "      <td>CONCEPT_LESS</td>\n",
              "      <td>1016</td>\n",
              "      <td>266</td>\n",
              "      <td>2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>ADR</td>\n",
              "      <td>Severe_pain</td>\n",
              "      <td>722</td>\n",
              "      <td>247</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>ADR</td>\n",
              "      <td>Myalgia</td>\n",
              "      <td>619</td>\n",
              "      <td>305</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>ADR</td>\n",
              "      <td>Pain</td>\n",
              "      <td>292</td>\n",
              "      <td>260</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>ADR</td>\n",
              "      <td>Arthralgia</td>\n",
              "      <td>280</td>\n",
              "      <td>163</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1268</th>\n",
              "      <td>ADR</td>\n",
              "      <td>Labyrinthitis</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1233</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1269</th>\n",
              "      <td>ADR</td>\n",
              "      <td>Dysphagia</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1233</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1275</th>\n",
              "      <td>ADR</td>\n",
              "      <td>Lupus_erythematosus</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1233</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1278</th>\n",
              "      <td>ADR</td>\n",
              "      <td>Cataract</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1233</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1283</th>\n",
              "      <td>ADR</td>\n",
              "      <td>Crying_associated_with_mood</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1233</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>800 rows × 5 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-2f0fef57-5422-4d06-8869-ca5c141095a9')\"\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-2f0fef57-5422-4d06-8869-ca5c141095a9 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-2f0fef57-5422-4d06-8869-ca5c141095a9');\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",
              "  "
            ],
            "text/plain": [
              "      ner                       meddra  counts_exploded  counts  \\\n",
              "1     ADR                 CONCEPT_LESS             1016     266   \n",
              "2     ADR                  Severe_pain              722     247   \n",
              "3     ADR                      Myalgia              619     305   \n",
              "4     ADR                         Pain              292     260   \n",
              "5     ADR                   Arthralgia              280     163   \n",
              "...   ...                          ...              ...     ...   \n",
              "1268  ADR                Labyrinthitis                1       1   \n",
              "1269  ADR                    Dysphagia                1       1   \n",
              "1275  ADR          Lupus_erythematosus                1       1   \n",
              "1278  ADR                     Cataract                1       1   \n",
              "1283  ADR  Crying_associated_with_mood                1       1   \n",
              "\n",
              "      counts_exploded_rank  \n",
              "1                        2  \n",
              "2                        3  \n",
              "3                        4  \n",
              "4                        5  \n",
              "5                        6  \n",
              "...                    ...  \n",
              "1268                  1233  \n",
              "1269                  1233  \n",
              "1275                  1233  \n",
              "1278                  1233  \n",
              "1283                  1233  \n",
              "\n",
              "[800 rows x 5 columns]"
            ]
          },
          "execution_count": 7,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "meddra_counts.loc[meddra_counts.ner == 'ADR',:]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hAGPini7tGii",
        "outputId": "21347cac-9bc2-4df8-c92e-e94640f1b5b9"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "[Sentence[117]: \"i feel a bit drowsy & have a little blurred vision , so far no gastric problems . ive been on arthrotec 50 for over 10 years on and off , only taking it when i needed it . due to my arthritis getting progressively worse , to the point where i am in tears with the agony , gp ' s started me on 75 twice a day and i have to take it . every day for the next month to see how i get on , here goes . so far its been very good , pains almost gone , but i feel a bit weird , didnt have that when on 50 .\" → [\"i\"/O/O, \"feel\"/O/O, \"a\"/O/O, \"bit\"/B-ADR/B-Drowsy, \"drowsy\"/E-ADR/E-Drowsy, \"&\"/O/O, \"have\"/O/O, \"a\"/O/O, \"little\"/B-ADR/B-Blurred_vision_-_hazy, \"blurred\"/I-ADR/I-Blurred_vision_-_hazy, \"vision\"/E-ADR/E-Blurred_vision_-_hazy, \",\"/O/O, \"so\"/O/O, \"far\"/O/O, \"no\"/O/O, \"gastric\"/B-ADR/B-Excessive_upper_gastrointestinal_gas, \"problems\"/E-ADR/E-Excessive_upper_gastrointestinal_gas, \".\"/O/O, \"ive\"/O/O, \"been\"/O/O, \"on\"/O/O, \"arthrotec\"/S-Drug/S-Arthrotec, \"50\"/O/O, \"for\"/O/O, \"over\"/O/O, \"10\"/O/O, \"years\"/O/O, \"on\"/O/O, \"and\"/O/O, \"off\"/O/O, \",\"/O/O, \"only\"/O/O, \"taking\"/O/O, \"it\"/O/O, \"when\"/O/O, \"i\"/O/O, \"needed\"/O/O, \"it\"/O/O, \".\"/O/O, \"due\"/O/O, \"to\"/O/O, \"my\"/O/O, \"arthritis\"/S-Disease/S-Arthritis, \"getting\"/O/O, \"progressively\"/O/O, \"worse\"/O/O, \",\"/O/O, \"to\"/O/O, \"the\"/O/O, \"point\"/O/O, \"where\"/O/O, \"i\"/O/O, \"am\"/O/O, \"in\"/O/O, \"tears\"/O/O, \"with\"/O/O, \"the\"/O/O, \"agony\"/S-Symptom/S-Agony, \",\"/O/O, \"gp\"/O/O, \"'\"/O/O, \"s\"/O/O, \"started\"/O/O, \"me\"/O/O, \"on\"/O/O, \"75\"/O/O, \"twice\"/O/O, \"a\"/O/O, \"day\"/O/O, \"and\"/O/O, \"i\"/O/O, \"have\"/O/O, \"to\"/O/O, \"take\"/O/O, \"it\"/O/O, \".\"/O/O, \"every\"/O/O, \"day\"/O/O, \"for\"/O/O, \"the\"/O/O, \"next\"/O/O, \"month\"/O/O, \"to\"/O/O, \"see\"/O/O, \"how\"/O/O, \"i\"/O/O, \"get\"/O/O, \"on\"/O/O, \",\"/O/O, \"here\"/O/O, \"goes\"/O/O, \".\"/O/O, \"so\"/O/O, \"far\"/O/O, \"its\"/O/O, \"been\"/O/O, \"very\"/O/O, \"good\"/O/O, \",\"/O/O, \"pains\"/S-Symptom/S-Pain, \"almost\"/O/O, \"gone\"/O/O, \",\"/O/O, \"but\"/O/O, \"i\"/O/O, \"feel\"/B-ADR/B-Malaise, \"a\"/I-ADR/I-Malaise, \"bit\"/I-ADR/I-Malaise, \"weird\"/E-ADR/E-Malaise, \",\"/O/O, \"didnt\"/O/O, \"have\"/O/O, \"that\"/O/O, \"when\"/O/O, \"on\"/O/O, \"50\"/O/O, \".\"/O/O],\n",
              " Sentence[39]: \"hunger pangs . brilliant , i have a new lease of life , i walk up & down steps properly , no longer sideways like a toddler , hip pain as gone other than if i jar it .\" → [\"hunger\"/B-ADR/B-Hunger_pain, \"pangs\"/E-ADR/E-Hunger_pain, \".\"/O/O, \"brilliant\"/O/O, \",\"/O/O, \"i\"/O/O, \"have\"/O/O, \"a\"/O/O, \"new\"/O/O, \"lease\"/O/O, \"of\"/O/O, \"life\"/O/O, \",\"/O/O, \"i\"/O/O, \"walk\"/B-Symptom/B-CONCEPT_LESS, \"up\"/I-Symptom/I-CONCEPT_LESS, \"&\"/I-Symptom/I-CONCEPT_LESS, \"down\"/I-Symptom/I-CONCEPT_LESS, \"steps\"/I-Symptom/I-CONCEPT_LESS, \"properly\"/O/O, \",\"/O/O, \"no\"/O/O, \"longer\"/O/O, \"sideways\"/E-Symptom/E-CONCEPT_LESS, \"like\"/O/O, \"a\"/O/O, \"toddler\"/O/O, \",\"/O/O, \"hip\"/B-Symptom/B-Hip_pain, \"pain\"/E-Symptom/E-Hip_pain, \"as\"/O/O, \"gone\"/O/O, \"other\"/O/O, \"than\"/O/O, \"if\"/O/O, \"i\"/O/O, \"jar\"/O/O, \"it\"/O/O, \".\"/O/O],\n",
              " Sentence[51]: \"no side effects for the first two months . then vaginal bleeding 2 wks after menstral cycle . stomach pain . canker sores in my mouth . headache . been off for 1 week still have bleeding . helped my pain alot . too scared to take this drug again .\" → [\"no\"/O/O, \"side\"/O/O, \"effects\"/O/O, \"for\"/O/O, \"the\"/O/O, \"first\"/O/O, \"two\"/O/O, \"months\"/O/O, \".\"/O/O, \"then\"/O/O, \"vaginal\"/B-ADR/B-Vaginal_bleeding, \"bleeding\"/E-ADR/E-Vaginal_bleeding, \"2\"/O/O, \"wks\"/O/O, \"after\"/O/O, \"menstral\"/O/O, \"cycle\"/O/O, \".\"/O/O, \"stomach\"/B-ADR/B-Stomach_ache, \"pain\"/E-ADR/E-Stomach_ache, \".\"/O/O, \"canker\"/B-ADR/B-Aphthous_ulcer, \"sores\"/E-ADR/E-Aphthous_ulcer, \"in\"/O/O, \"my\"/O/O, \"mouth\"/O/O, \".\"/O/O, \"headache\"/S-ADR/S-Headache, \".\"/O/O, \"been\"/O/O, \"off\"/O/O, \"for\"/O/O, \"1\"/O/O, \"week\"/O/O, \"still\"/O/O, \"have\"/O/O, \"bleeding\"/O/O, \".\"/O/O, \"helped\"/O/O, \"my\"/O/O, \"pain\"/S-Symptom/S-Pain, \"alot\"/O/O, \".\"/O/O, \"too\"/O/O, \"scared\"/O/O, \"to\"/O/O, \"take\"/O/O, \"this\"/O/O, \"drug\"/O/O, \"again\"/O/O, \".\"/O/O],\n",
              " Sentence[68]: \"1st pill taken with food , a few hours after i experienced shortness of breath , a sense of depression , cramping , upset stomache will stop taking immediately . honestly can not recommend this drug , i am still experiencing side effects just after 1 pill and im supposed to take twice / day 75mg . i just cant keep going have to try something else .\" → [\"1st\"/O/O, \"pill\"/O/O, \"taken\"/O/O, \"with\"/O/O, \"food\"/O/O, \",\"/O/O, \"a\"/O/O, \"few\"/O/O, \"hours\"/O/O, \"after\"/O/O, \"i\"/O/O, \"experienced\"/O/O, \"shortness\"/B-ADR/B-Dyspnoea, \"of\"/I-ADR/I-Dyspnoea, \"breath\"/E-ADR/E-Dyspnoea, \",\"/O/O, \"a\"/O/O, \"sense\"/O/O, \"of\"/O/O, \"depression\"/S-ADR/S-Depression, \",\"/O/O, \"cramping\"/S-ADR/S-Muscle_cramp, \",\"/O/O, \"upset\"/B-ADR/B-Upset_stomach, \"stomache\"/E-ADR/E-Upset_stomach, \"will\"/O/O, \"stop\"/O/O, \"taking\"/O/O, \"immediately\"/O/O, \".\"/O/O, \"honestly\"/O/O, \"can\"/O/O, \"not\"/O/O, \"recommend\"/O/O, \"this\"/O/O, \"drug\"/O/O, \",\"/O/O, \"i\"/O/O, \"am\"/O/O, \"still\"/O/O, \"experiencing\"/O/O, \"side\"/O/O, \"effects\"/O/O, \"just\"/O/O, \"after\"/O/O, \"1\"/O/O, \"pill\"/O/O, \"and\"/O/O, \"im\"/O/O, \"supposed\"/O/O, \"to\"/O/O, \"take\"/O/O, \"twice\"/O/O, \"/\"/O/O, \"day\"/O/O, \"75mg\"/O/O, \".\"/O/O, \"i\"/O/O, \"just\"/O/O, \"cant\"/O/O, \"keep\"/O/O, \"going\"/O/O, \"have\"/O/O, \"to\"/O/O, \"try\"/O/O, \"something\"/O/O, \"else\"/O/O, \".\"/O/O],\n",
              " Sentence[50]: \"i have had no side effects been taking arthrotec a little over a year , have not noticed any side effects . it does help alot i noticed that when there are times when i forget to take it i cant stand or walk for any lengths of time .\" → [\"i\"/O/O, \"have\"/O/O, \"had\"/O/O, \"no\"/O/O, \"side\"/O/O, \"effects\"/O/O, \"been\"/O/O, \"taking\"/O/O, \"arthrotec\"/S-Drug/S-Arthrotec, \"a\"/O/O, \"little\"/O/O, \"over\"/O/O, \"a\"/O/O, \"year\"/O/O, \",\"/O/O, \"have\"/O/O, \"not\"/O/O, \"noticed\"/O/O, \"any\"/O/O, \"side\"/O/O, \"effects\"/O/O, \".\"/O/O, \"it\"/O/O, \"does\"/O/O, \"help\"/O/O, \"alot\"/O/O, \"i\"/O/O, \"noticed\"/O/O, \"that\"/O/O, \"when\"/O/O, \"there\"/O/O, \"are\"/O/O, \"times\"/O/O, \"when\"/O/O, \"i\"/O/O, \"forget\"/O/O, \"to\"/O/O, \"take\"/O/O, \"it\"/O/O, \"i\"/B-Symptom/B-Reduced_mobility, \"cant\"/I-Symptom/I-Reduced_mobility, \"stand\"/I-Symptom/I-Reduced_mobility, \"or\"/I-Symptom/I-Reduced_mobility, \"walk\"/I-Symptom/I-Reduced_mobility, \"for\"/I-Symptom/I-Reduced_mobility, \"any\"/I-Symptom/I-Reduced_mobility, \"lengths\"/I-Symptom/I-Reduced_mobility, \"of\"/I-Symptom/I-Reduced_mobility, \"time\"/E-Symptom/E-Reduced_mobility, \".\"/O/O],\n",
              " Sentence[89]: \"it seems as if bleeding forever is a side effect but doctors tell you that its not the drug . i have been on arthotec for three months i still have pain and stiffness . at first week into this drug i was in so much pain i sat on the floor at work and thought i was gaving a baby . i always feel weak and almost fainted a bunch of times . consider somthing that works i am switching today as this isnt in my head .\" → [\"it\"/O/O, \"seems\"/O/O, \"as\"/O/O, \"if\"/O/O, \"bleeding\"/S-ADR/S-Bleeding, \"forever\"/O/O, \"is\"/O/O, \"a\"/O/O, \"side\"/O/O, \"effect\"/O/O, \"but\"/O/O, \"doctors\"/O/O, \"tell\"/O/O, \"you\"/O/O, \"that\"/O/O, \"its\"/O/O, \"not\"/O/O, \"the\"/O/O, \"drug\"/O/O, \".\"/O/O, \"i\"/O/O, \"have\"/O/O, \"been\"/O/O, \"on\"/O/O, \"arthotec\"/S-Drug/S-Arthrotec, \"for\"/O/O, \"three\"/O/O, \"months\"/O/O, \"i\"/O/O, \"still\"/O/O, \"have\"/O/O, \"pain\"/B-Symptom/B-Pain, \"and\"/I-Symptom/I-Pain, \"stiffness\"/E-Symptom/E-Pain, \".\"/O/O, \"at\"/O/O, \"first\"/O/O, \"week\"/O/O, \"into\"/O/O, \"this\"/O/O, \"drug\"/O/O, \"i\"/O/O, \"was\"/O/O, \"in\"/O/O, \"so\"/B-ADR/B-Pain, \"much\"/I-ADR/I-Pain, \"pain\"/E-ADR/E-Pain, \"i\"/O/O, \"sat\"/O/O, \"on\"/O/O, \"the\"/O/O, \"floor\"/O/O, \"at\"/O/O, \"work\"/O/O, \"and\"/O/O, \"thought\"/O/O, \"i\"/O/O, \"was\"/O/O, \"gaving\"/O/O, \"a\"/O/O, \"baby\"/O/O, \".\"/O/O, \"i\"/O/O, \"always\"/O/O, \"feel\"/B-ADR/B-Malaise_and_fatigue, \"weak\"/I-ADR/I-Malaise_and_fatigue, \"and\"/I-ADR/I-Malaise_and_fatigue, \"almost\"/I-ADR/I-Malaise_and_fatigue, \"fainted\"/E-ADR/E-Malaise_and_fatigue, \"a\"/O/O, \"bunch\"/O/O, \"of\"/O/O, \"times\"/O/O, \".\"/O/O, \"consider\"/O/O, \"somthing\"/O/O, \"that\"/O/O, \"works\"/O/O, \"i\"/O/O, \"am\"/O/O, \"switching\"/O/O, \"today\"/O/O, \"as\"/O/O, \"this\"/O/O, \"isnt\"/O/O, \"in\"/O/O, \"my\"/O/O, \"head\"/O/O, \".\"/O/O],\n",
              " Sentence[56]: \"some diarrhea , some constipation , fatigue . it works very well for the heel ; i couldnt walk on it very well before taking it . i take 75mg . twice a day . also have severe osteoarthritis in the knees and hands , but it doesnt seem to relieve that pain at all .\" → [\"some\"/O/O, \"diarrhea\"/S-ADR/S-Diarrhoea, \",\"/O/O, \"some\"/O/O, \"constipation\"/S-ADR/S-Constipation, \",\"/O/O, \"fatigue\"/S-ADR/S-Fatigue, \".\"/O/O, \"it\"/O/O, \"works\"/O/O, \"very\"/O/O, \"well\"/O/O, \"for\"/O/O, \"the\"/B-Symptom/B-Walking_disability, \"heel\"/I-Symptom/I-Walking_disability, \";\"/O/O, \"i\"/I-Symptom/I-Walking_disability, \"couldnt\"/I-Symptom/I-Walking_disability, \"walk\"/I-Symptom/I-Walking_disability, \"on\"/I-Symptom/I-Walking_disability, \"it\"/O/O, \"very\"/I-Symptom/I-Walking_disability, \"well\"/E-Symptom/E-Walking_disability, \"before\"/O/O, \"taking\"/O/O, \"it\"/O/O, \".\"/O/O, \"i\"/O/O, \"take\"/O/O, \"75mg\"/O/O, \".\"/O/O, \"twice\"/O/O, \"a\"/O/O, \"day\"/O/O, \".\"/O/O, \"also\"/O/O, \"have\"/O/O, \"severe\"/B-Disease/B-Osteoarthritis_of_knee/B-Osteoarthrosis_of_hand, \"osteoarthritis\"/I-Disease/I-Osteoarthritis_of_knee/I-Osteoarthrosis_of_hand, \"in\"/I-Disease/I-Osteoarthritis_of_knee/I-Osteoarthrosis_of_hand, \"the\"/I-Disease/I-Osteoarthritis_of_knee/I-Osteoarthrosis_of_hand, \"knees\"/E-Disease/E-Osteoarthritis_of_knee, \"and\"/O/O, \"hands\"/E-Disease/E-Osteoarthrosis_of_hand, \",\"/O/O, \"but\"/O/O, \"it\"/O/O, \"doesnt\"/O/O, \"seem\"/O/O, \"to\"/O/O, \"relieve\"/O/O, \"that\"/O/O, \"pain\"/S-Symptom/S-Pain, \"at\"/O/O, \"all\"/O/O, \".\"/O/O],\n",
              " Sentence[84]: \"ive been on 75mg twice daily for 13 days , with recent body swelling , face , wrists , abdomen , thighs - clothes fit tight . insomina is much worse . i have forgetfulnes and confussion - really bad - almost dangerous . have felt \" empty stomach \" feeling , but not bothersome . have had more loose stools , but tolerable . im looking for a more natural therapy , checking into ha ( hyaluronic acid ) to replace arthrotec .\" → [\"ive\"/O/O, \"been\"/O/O, \"on\"/O/O, \"75mg\"/O/O, \"twice\"/O/O, \"daily\"/O/O, \"for\"/O/O, \"13\"/O/O, \"days\"/O/O, \",\"/O/O, \"with\"/O/O, \"recent\"/O/O, \"body\"/B-ADR/B-Swelling_of_body_region, \"swelling\"/B-ADR/E-ADR/B-Abdominal_swelling/B-Facial_swelling/E-Swelling_of_body_region/B-Swelling_of_lower_limb/B-Swelling_of_wrist_joint, \",\"/O/O, \"face\"/E-ADR/E-Facial_swelling, \",\"/O/O, \"wrists\"/E-ADR/E-Swelling_of_wrist_joint, \",\"/O/O, \"abdomen\"/E-ADR/E-Abdominal_swelling, \",\"/O/O, \"thighs\"/E-ADR/E-Swelling_of_lower_limb, \"-\"/O/O, \"clothes\"/O/O, \"fit\"/O/O, \"tight\"/O/O, \".\"/O/O, \"insomina\"/S-ADR/S-Insomnia, \"is\"/O/O, \"much\"/O/O, \"worse\"/O/O, \".\"/O/O, \"i\"/O/O, \"have\"/O/O, \"forgetfulnes\"/B-ADR/B-Forgetful, \"and\"/I-ADR/I-Forgetful, \"confussion\"/E-ADR/E-Forgetful, \"-\"/O/O, \"really\"/O/O, \"bad\"/O/O, \"-\"/O/O, \"almost\"/O/O, \"dangerous\"/O/O, \".\"/O/O, \"have\"/O/O, \"felt\"/O/O, \"\"\"/B-ADR/B-Feeling_empty, \"empty\"/I-ADR/I-Feeling_empty, \"stomach\"/I-ADR/I-Feeling_empty, \"\"\"/I-ADR/I-Feeling_empty, \"feeling\"/E-ADR/E-Feeling_empty, \",\"/O/O, \"but\"/O/O, \"not\"/O/O, \"bothersome\"/O/O, \".\"/O/O, \"have\"/O/O, \"had\"/O/O, \"more\"/O/O, \"loose\"/B-ADR/B-Loose_stool, \"stools\"/E-ADR/E-Loose_stool, \",\"/O/O, \"but\"/O/O, \"tolerable\"/O/O, \".\"/O/O, \"im\"/O/O, \"looking\"/O/O, \"for\"/O/O, \"a\"/O/O, \"more\"/O/O, \"natural\"/O/O, \"therapy\"/O/O, \",\"/O/O, \"checking\"/O/O, \"into\"/O/O, \"ha\"/B-Drug/B-CONCEPT_LESS, \"(\"/I-Drug/I-CONCEPT_LESS, \"hyaluronic\"/I-Drug/I-CONCEPT_LESS, \"acid\"/I-Drug/I-CONCEPT_LESS, \")\"/E-Drug/E-CONCEPT_LESS, \"to\"/O/O, \"replace\"/O/O, \"arthrotec\"/S-Drug/S-Arthrotec, \".\"/O/O],\n",
              " Sentence[68]: \"none that i can think of . i recently had bilateral knee replacement . prior to that , arthrotec was the only drug that helped . i tried every drug on the market , but , to no avail . i took 50 mg . 3 times a day of arthrotec and was almost pain free . i highly recommend it to anyone that suffers from arthritis .\" → [\"none\"/O/O, \"that\"/O/O, \"i\"/O/O, \"can\"/O/O, \"think\"/O/O, \"of\"/O/O, \".\"/O/O, \"i\"/O/O, \"recently\"/O/O, \"had\"/O/O, \"bilateral\"/O/O, \"knee\"/O/O, \"replacement\"/O/O, \".\"/O/O, \"prior\"/O/O, \"to\"/O/O, \"that\"/O/O, \",\"/O/O, \"arthrotec\"/S-Drug/S-Arthrotec, \"was\"/O/O, \"the\"/O/O, \"only\"/O/O, \"drug\"/O/O, \"that\"/O/O, \"helped\"/O/O, \".\"/O/O, \"i\"/O/O, \"tried\"/O/O, \"every\"/O/O, \"drug\"/O/O, \"on\"/O/O, \"the\"/O/O, \"market\"/O/O, \",\"/O/O, \"but\"/O/O, \",\"/O/O, \"to\"/O/O, \"no\"/O/O, \"avail\"/O/O, \".\"/O/O, \"i\"/O/O, \"took\"/O/O, \"50\"/O/O, \"mg\"/O/O, \".\"/O/O, \"3\"/O/O, \"times\"/O/O, \"a\"/O/O, \"day\"/O/O, \"of\"/O/O, \"arthrotec\"/S-Drug/S-Arthrotec, \"and\"/O/O, \"was\"/O/O, \"almost\"/O/O, \"pain\"/S-Symptom/S-Pain, \"free\"/O/O, \".\"/O/O, \"i\"/O/O, \"highly\"/O/O, \"recommend\"/O/O, \"it\"/O/O, \"to\"/O/O, \"anyone\"/O/O, \"that\"/O/O, \"suffers\"/O/O, \"from\"/O/O, \"arthritis\"/S-Finding/S-Arthritis, \".\"/O/O],\n",
              " Sentence[28]: \"hospitalization due extreme rectal bleed that required blood transfusion . i was extremely sick and initially felt i might not survive . should be removed from market .\" → [\"hospitalization\"/O/O, \"due\"/O/O, \"extreme\"/O/O, \"rectal\"/B-ADR/B-Rectal_haemorrhage, \"bleed\"/E-ADR/E-Rectal_haemorrhage, \"that\"/O/O, \"required\"/O/O, \"blood\"/O/O, \"transfusion\"/O/O, \".\"/O/O, \"i\"/O/O, \"was\"/O/O, \"extremely\"/B-ADR/B-Generally_unwell, \"sick\"/E-ADR/E-Generally_unwell, \"and\"/O/O, \"initially\"/O/O, \"felt\"/B-ADR/B-CONCEPT_LESS, \"i\"/I-ADR/I-CONCEPT_LESS, \"might\"/I-ADR/I-CONCEPT_LESS, \"not\"/I-ADR/I-CONCEPT_LESS, \"survive\"/E-ADR/E-CONCEPT_LESS, \".\"/O/O, \"should\"/O/O, \"be\"/O/O, \"removed\"/O/O, \"from\"/O/O, \"market\"/O/O, \".\"/O/O]]"
            ]
          },
          "execution_count": 8,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "with open(data_save_dir+'/cadec_flair_objects.pkl', 'rb') as f:\n",
        "    docs = pickle.load(f)\n",
        "docs[:10]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sW10pPAPtGij",
        "outputId": "afc060cd-941d-45b2-8ab4-06d97858cad7"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "1186"
            ]
          },
          "execution_count": 9,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "len(docs)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EMT5TsNStGij",
        "outputId": "51bc4d93-a94e-4701-a95f-8b9827592e49"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "Sentence[117]: \"i feel a bit drowsy & have a little blurred vision , so far no gastric problems . ive been on arthrotec 50 for over 10 years on and off , only taking it when i needed it . due to my arthritis getting progressively worse , to the point where i am in tears with the agony , gp ' s started me on 75 twice a day and i have to take it . every day for the next month to see how i get on , here goes . so far its been very good , pains almost gone , but i feel a bit weird , didnt have that when on 50 .\" → [\"i\"/O/O, \"feel\"/O/O, \"a\"/O/O, \"bit\"/B-ADR/B-Drowsy, \"drowsy\"/E-ADR/E-Drowsy, \"&\"/O/O, \"have\"/O/O, \"a\"/O/O, \"little\"/B-ADR/B-Blurred_vision_-_hazy, \"blurred\"/I-ADR/I-Blurred_vision_-_hazy, \"vision\"/E-ADR/E-Blurred_vision_-_hazy, \",\"/O/O, \"so\"/O/O, \"far\"/O/O, \"no\"/O/O, \"gastric\"/B-ADR/B-Excessive_upper_gastrointestinal_gas, \"problems\"/E-ADR/E-Excessive_upper_gastrointestinal_gas, \".\"/O/O, \"ive\"/O/O, \"been\"/O/O, \"on\"/O/O, \"arthrotec\"/S-Drug/S-Arthrotec, \"50\"/O/O, \"for\"/O/O, \"over\"/O/O, \"10\"/O/O, \"years\"/O/O, \"on\"/O/O, \"and\"/O/O, \"off\"/O/O, \",\"/O/O, \"only\"/O/O, \"taking\"/O/O, \"it\"/O/O, \"when\"/O/O, \"i\"/O/O, \"needed\"/O/O, \"it\"/O/O, \".\"/O/O, \"due\"/O/O, \"to\"/O/O, \"my\"/O/O, \"arthritis\"/S-Disease/S-Arthritis, \"getting\"/O/O, \"progressively\"/O/O, \"worse\"/O/O, \",\"/O/O, \"to\"/O/O, \"the\"/O/O, \"point\"/O/O, \"where\"/O/O, \"i\"/O/O, \"am\"/O/O, \"in\"/O/O, \"tears\"/O/O, \"with\"/O/O, \"the\"/O/O, \"agony\"/S-Symptom/S-Agony, \",\"/O/O, \"gp\"/O/O, \"'\"/O/O, \"s\"/O/O, \"started\"/O/O, \"me\"/O/O, \"on\"/O/O, \"75\"/O/O, \"twice\"/O/O, \"a\"/O/O, \"day\"/O/O, \"and\"/O/O, \"i\"/O/O, \"have\"/O/O, \"to\"/O/O, \"take\"/O/O, \"it\"/O/O, \".\"/O/O, \"every\"/O/O, \"day\"/O/O, \"for\"/O/O, \"the\"/O/O, \"next\"/O/O, \"month\"/O/O, \"to\"/O/O, \"see\"/O/O, \"how\"/O/O, \"i\"/O/O, \"get\"/O/O, \"on\"/O/O, \",\"/O/O, \"here\"/O/O, \"goes\"/O/O, \".\"/O/O, \"so\"/O/O, \"far\"/O/O, \"its\"/O/O, \"been\"/O/O, \"very\"/O/O, \"good\"/O/O, \",\"/O/O, \"pains\"/S-Symptom/S-Pain, \"almost\"/O/O, \"gone\"/O/O, \",\"/O/O, \"but\"/O/O, \"i\"/O/O, \"feel\"/B-ADR/B-Malaise, \"a\"/I-ADR/I-Malaise, \"bit\"/I-ADR/I-Malaise, \"weird\"/E-ADR/E-Malaise, \",\"/O/O, \"didnt\"/O/O, \"have\"/O/O, \"that\"/O/O, \"when\"/O/O, \"on\"/O/O, \"50\"/O/O, \".\"/O/O]"
            ]
          },
          "execution_count": 10,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "temp = docs[0]\n",
        "temp"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2EeTdfHstGij",
        "outputId": "8ad27be9-2190-4f25-a568-b7c9438c4361"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "{'ner': ['Token[0]: \"i\"'/'O' (1.0),\n",
              "  'Token[1]: \"feel\"'/'O' (1.0),\n",
              "  'Token[2]: \"a\"'/'O' (1.0),\n",
              "  'Token[3]: \"bit\"'/'B-ADR' (1.0),\n",
              "  'Token[4]: \"drowsy\"'/'E-ADR' (1.0),\n",
              "  'Token[5]: \"&\"'/'O' (1.0),\n",
              "  'Token[6]: \"have\"'/'O' (1.0),\n",
              "  'Token[7]: \"a\"'/'O' (1.0),\n",
              "  'Token[8]: \"little\"'/'B-ADR' (1.0),\n",
              "  'Token[9]: \"blurred\"'/'I-ADR' (1.0),\n",
              "  'Token[10]: \"vision\"'/'E-ADR' (1.0),\n",
              "  'Token[11]: \",\"'/'O' (1.0),\n",
              "  'Token[12]: \"so\"'/'O' (1.0),\n",
              "  'Token[13]: \"far\"'/'O' (1.0),\n",
              "  'Token[14]: \"no\"'/'O' (1.0),\n",
              "  'Token[15]: \"gastric\"'/'B-ADR' (1.0),\n",
              "  'Token[16]: \"problems\"'/'E-ADR' (1.0),\n",
              "  'Token[17]: \".\"'/'O' (1.0),\n",
              "  'Token[18]: \"ive\"'/'O' (1.0),\n",
              "  'Token[19]: \"been\"'/'O' (1.0),\n",
              "  'Token[20]: \"on\"'/'O' (1.0),\n",
              "  'Token[21]: \"arthrotec\"'/'S-Drug' (1.0),\n",
              "  'Token[22]: \"50\"'/'O' (1.0),\n",
              "  'Token[23]: \"for\"'/'O' (1.0),\n",
              "  'Token[24]: \"over\"'/'O' (1.0),\n",
              "  'Token[25]: \"10\"'/'O' (1.0),\n",
              "  'Token[26]: \"years\"'/'O' (1.0),\n",
              "  'Token[27]: \"on\"'/'O' (1.0),\n",
              "  'Token[28]: \"and\"'/'O' (1.0),\n",
              "  'Token[29]: \"off\"'/'O' (1.0),\n",
              "  'Token[30]: \",\"'/'O' (1.0),\n",
              "  'Token[31]: \"only\"'/'O' (1.0),\n",
              "  'Token[32]: \"taking\"'/'O' (1.0),\n",
              "  'Token[33]: \"it\"'/'O' (1.0),\n",
              "  'Token[34]: \"when\"'/'O' (1.0),\n",
              "  'Token[35]: \"i\"'/'O' (1.0),\n",
              "  'Token[36]: \"needed\"'/'O' (1.0),\n",
              "  'Token[37]: \"it\"'/'O' (1.0),\n",
              "  'Token[38]: \".\"'/'O' (1.0),\n",
              "  'Token[39]: \"due\"'/'O' (1.0),\n",
              "  'Token[40]: \"to\"'/'O' (1.0),\n",
              "  'Token[41]: \"my\"'/'O' (1.0),\n",
              "  'Token[42]: \"arthritis\"'/'S-Disease' (1.0),\n",
              "  'Token[43]: \"getting\"'/'O' (1.0),\n",
              "  'Token[44]: \"progressively\"'/'O' (1.0),\n",
              "  'Token[45]: \"worse\"'/'O' (1.0),\n",
              "  'Token[46]: \",\"'/'O' (1.0),\n",
              "  'Token[47]: \"to\"'/'O' (1.0),\n",
              "  'Token[48]: \"the\"'/'O' (1.0),\n",
              "  'Token[49]: \"point\"'/'O' (1.0),\n",
              "  'Token[50]: \"where\"'/'O' (1.0),\n",
              "  'Token[51]: \"i\"'/'O' (1.0),\n",
              "  'Token[52]: \"am\"'/'O' (1.0),\n",
              "  'Token[53]: \"in\"'/'O' (1.0),\n",
              "  'Token[54]: \"tears\"'/'O' (1.0),\n",
              "  'Token[55]: \"with\"'/'O' (1.0),\n",
              "  'Token[56]: \"the\"'/'O' (1.0),\n",
              "  'Token[57]: \"agony\"'/'S-Symptom' (1.0),\n",
              "  'Token[58]: \",\"'/'O' (1.0),\n",
              "  'Token[59]: \"gp\"'/'O' (1.0),\n",
              "  'Token[60]: \"'\"'/'O' (1.0),\n",
              "  'Token[61]: \"s\"'/'O' (1.0),\n",
              "  'Token[62]: \"started\"'/'O' (1.0),\n",
              "  'Token[63]: \"me\"'/'O' (1.0),\n",
              "  'Token[64]: \"on\"'/'O' (1.0),\n",
              "  'Token[65]: \"75\"'/'O' (1.0),\n",
              "  'Token[66]: \"twice\"'/'O' (1.0),\n",
              "  'Token[67]: \"a\"'/'O' (1.0),\n",
              "  'Token[68]: \"day\"'/'O' (1.0),\n",
              "  'Token[69]: \"and\"'/'O' (1.0),\n",
              "  'Token[70]: \"i\"'/'O' (1.0),\n",
              "  'Token[71]: \"have\"'/'O' (1.0),\n",
              "  'Token[72]: \"to\"'/'O' (1.0),\n",
              "  'Token[73]: \"take\"'/'O' (1.0),\n",
              "  'Token[74]: \"it\"'/'O' (1.0),\n",
              "  'Token[75]: \".\"'/'O' (1.0),\n",
              "  'Token[76]: \"every\"'/'O' (1.0),\n",
              "  'Token[77]: \"day\"'/'O' (1.0),\n",
              "  'Token[78]: \"for\"'/'O' (1.0),\n",
              "  'Token[79]: \"the\"'/'O' (1.0),\n",
              "  'Token[80]: \"next\"'/'O' (1.0),\n",
              "  'Token[81]: \"month\"'/'O' (1.0),\n",
              "  'Token[82]: \"to\"'/'O' (1.0),\n",
              "  'Token[83]: \"see\"'/'O' (1.0),\n",
              "  'Token[84]: \"how\"'/'O' (1.0),\n",
              "  'Token[85]: \"i\"'/'O' (1.0),\n",
              "  'Token[86]: \"get\"'/'O' (1.0),\n",
              "  'Token[87]: \"on\"'/'O' (1.0),\n",
              "  'Token[88]: \",\"'/'O' (1.0),\n",
              "  'Token[89]: \"here\"'/'O' (1.0),\n",
              "  'Token[90]: \"goes\"'/'O' (1.0),\n",
              "  'Token[91]: \".\"'/'O' (1.0),\n",
              "  'Token[92]: \"so\"'/'O' (1.0),\n",
              "  'Token[93]: \"far\"'/'O' (1.0),\n",
              "  'Token[94]: \"its\"'/'O' (1.0),\n",
              "  'Token[95]: \"been\"'/'O' (1.0),\n",
              "  'Token[96]: \"very\"'/'O' (1.0),\n",
              "  'Token[97]: \"good\"'/'O' (1.0),\n",
              "  'Token[98]: \",\"'/'O' (1.0),\n",
              "  'Token[99]: \"pains\"'/'S-Symptom' (1.0),\n",
              "  'Token[100]: \"almost\"'/'O' (1.0),\n",
              "  'Token[101]: \"gone\"'/'O' (1.0),\n",
              "  'Token[102]: \",\"'/'O' (1.0),\n",
              "  'Token[103]: \"but\"'/'O' (1.0),\n",
              "  'Token[104]: \"i\"'/'O' (1.0),\n",
              "  'Token[105]: \"feel\"'/'B-ADR' (1.0),\n",
              "  'Token[106]: \"a\"'/'I-ADR' (1.0),\n",
              "  'Token[107]: \"bit\"'/'I-ADR' (1.0),\n",
              "  'Token[108]: \"weird\"'/'E-ADR' (1.0),\n",
              "  'Token[109]: \",\"'/'O' (1.0),\n",
              "  'Token[110]: \"didnt\"'/'O' (1.0),\n",
              "  'Token[111]: \"have\"'/'O' (1.0),\n",
              "  'Token[112]: \"that\"'/'O' (1.0),\n",
              "  'Token[113]: \"when\"'/'O' (1.0),\n",
              "  'Token[114]: \"on\"'/'O' (1.0),\n",
              "  'Token[115]: \"50\"'/'O' (1.0),\n",
              "  'Token[116]: \".\"'/'O' (1.0)],\n",
              " 'meddra': ['Token[0]: \"i\"'/'O' (1.0),\n",
              "  'Token[1]: \"feel\"'/'O' (1.0),\n",
              "  'Token[2]: \"a\"'/'O' (1.0),\n",
              "  'Token[3]: \"bit\"'/'B-Drowsy' (1.0),\n",
              "  'Token[4]: \"drowsy\"'/'E-Drowsy' (1.0),\n",
              "  'Token[5]: \"&\"'/'O' (1.0),\n",
              "  'Token[6]: \"have\"'/'O' (1.0),\n",
              "  'Token[7]: \"a\"'/'O' (1.0),\n",
              "  'Token[8]: \"little\"'/'B-Blurred_vision_-_hazy' (1.0),\n",
              "  'Token[9]: \"blurred\"'/'I-Blurred_vision_-_hazy' (1.0),\n",
              "  'Token[10]: \"vision\"'/'E-Blurred_vision_-_hazy' (1.0),\n",
              "  'Token[11]: \",\"'/'O' (1.0),\n",
              "  'Token[12]: \"so\"'/'O' (1.0),\n",
              "  'Token[13]: \"far\"'/'O' (1.0),\n",
              "  'Token[14]: \"no\"'/'O' (1.0),\n",
              "  'Token[15]: \"gastric\"'/'B-Excessive_upper_gastrointestinal_gas' (1.0),\n",
              "  'Token[16]: \"problems\"'/'E-Excessive_upper_gastrointestinal_gas' (1.0),\n",
              "  'Token[17]: \".\"'/'O' (1.0),\n",
              "  'Token[18]: \"ive\"'/'O' (1.0),\n",
              "  'Token[19]: \"been\"'/'O' (1.0),\n",
              "  'Token[20]: \"on\"'/'O' (1.0),\n",
              "  'Token[21]: \"arthrotec\"'/'S-Arthrotec' (1.0),\n",
              "  'Token[22]: \"50\"'/'O' (1.0),\n",
              "  'Token[23]: \"for\"'/'O' (1.0),\n",
              "  'Token[24]: \"over\"'/'O' (1.0),\n",
              "  'Token[25]: \"10\"'/'O' (1.0),\n",
              "  'Token[26]: \"years\"'/'O' (1.0),\n",
              "  'Token[27]: \"on\"'/'O' (1.0),\n",
              "  'Token[28]: \"and\"'/'O' (1.0),\n",
              "  'Token[29]: \"off\"'/'O' (1.0),\n",
              "  'Token[30]: \",\"'/'O' (1.0),\n",
              "  'Token[31]: \"only\"'/'O' (1.0),\n",
              "  'Token[32]: \"taking\"'/'O' (1.0),\n",
              "  'Token[33]: \"it\"'/'O' (1.0),\n",
              "  'Token[34]: \"when\"'/'O' (1.0),\n",
              "  'Token[35]: \"i\"'/'O' (1.0),\n",
              "  'Token[36]: \"needed\"'/'O' (1.0),\n",
              "  'Token[37]: \"it\"'/'O' (1.0),\n",
              "  'Token[38]: \".\"'/'O' (1.0),\n",
              "  'Token[39]: \"due\"'/'O' (1.0),\n",
              "  'Token[40]: \"to\"'/'O' (1.0),\n",
              "  'Token[41]: \"my\"'/'O' (1.0),\n",
              "  'Token[42]: \"arthritis\"'/'S-Arthritis' (1.0),\n",
              "  'Token[43]: \"getting\"'/'O' (1.0),\n",
              "  'Token[44]: \"progressively\"'/'O' (1.0),\n",
              "  'Token[45]: \"worse\"'/'O' (1.0),\n",
              "  'Token[46]: \",\"'/'O' (1.0),\n",
              "  'Token[47]: \"to\"'/'O' (1.0),\n",
              "  'Token[48]: \"the\"'/'O' (1.0),\n",
              "  'Token[49]: \"point\"'/'O' (1.0),\n",
              "  'Token[50]: \"where\"'/'O' (1.0),\n",
              "  'Token[51]: \"i\"'/'O' (1.0),\n",
              "  'Token[52]: \"am\"'/'O' (1.0),\n",
              "  'Token[53]: \"in\"'/'O' (1.0),\n",
              "  'Token[54]: \"tears\"'/'O' (1.0),\n",
              "  'Token[55]: \"with\"'/'O' (1.0),\n",
              "  'Token[56]: \"the\"'/'O' (1.0),\n",
              "  'Token[57]: \"agony\"'/'S-Agony' (1.0),\n",
              "  'Token[58]: \",\"'/'O' (1.0),\n",
              "  'Token[59]: \"gp\"'/'O' (1.0),\n",
              "  'Token[60]: \"'\"'/'O' (1.0),\n",
              "  'Token[61]: \"s\"'/'O' (1.0),\n",
              "  'Token[62]: \"started\"'/'O' (1.0),\n",
              "  'Token[63]: \"me\"'/'O' (1.0),\n",
              "  'Token[64]: \"on\"'/'O' (1.0),\n",
              "  'Token[65]: \"75\"'/'O' (1.0),\n",
              "  'Token[66]: \"twice\"'/'O' (1.0),\n",
              "  'Token[67]: \"a\"'/'O' (1.0),\n",
              "  'Token[68]: \"day\"'/'O' (1.0),\n",
              "  'Token[69]: \"and\"'/'O' (1.0),\n",
              "  'Token[70]: \"i\"'/'O' (1.0),\n",
              "  'Token[71]: \"have\"'/'O' (1.0),\n",
              "  'Token[72]: \"to\"'/'O' (1.0),\n",
              "  'Token[73]: \"take\"'/'O' (1.0),\n",
              "  'Token[74]: \"it\"'/'O' (1.0),\n",
              "  'Token[75]: \".\"'/'O' (1.0),\n",
              "  'Token[76]: \"every\"'/'O' (1.0),\n",
              "  'Token[77]: \"day\"'/'O' (1.0),\n",
              "  'Token[78]: \"for\"'/'O' (1.0),\n",
              "  'Token[79]: \"the\"'/'O' (1.0),\n",
              "  'Token[80]: \"next\"'/'O' (1.0),\n",
              "  'Token[81]: \"month\"'/'O' (1.0),\n",
              "  'Token[82]: \"to\"'/'O' (1.0),\n",
              "  'Token[83]: \"see\"'/'O' (1.0),\n",
              "  'Token[84]: \"how\"'/'O' (1.0),\n",
              "  'Token[85]: \"i\"'/'O' (1.0),\n",
              "  'Token[86]: \"get\"'/'O' (1.0),\n",
              "  'Token[87]: \"on\"'/'O' (1.0),\n",
              "  'Token[88]: \",\"'/'O' (1.0),\n",
              "  'Token[89]: \"here\"'/'O' (1.0),\n",
              "  'Token[90]: \"goes\"'/'O' (1.0),\n",
              "  'Token[91]: \".\"'/'O' (1.0),\n",
              "  'Token[92]: \"so\"'/'O' (1.0),\n",
              "  'Token[93]: \"far\"'/'O' (1.0),\n",
              "  'Token[94]: \"its\"'/'O' (1.0),\n",
              "  'Token[95]: \"been\"'/'O' (1.0),\n",
              "  'Token[96]: \"very\"'/'O' (1.0),\n",
              "  'Token[97]: \"good\"'/'O' (1.0),\n",
              "  'Token[98]: \",\"'/'O' (1.0),\n",
              "  'Token[99]: \"pains\"'/'S-Pain' (1.0),\n",
              "  'Token[100]: \"almost\"'/'O' (1.0),\n",
              "  'Token[101]: \"gone\"'/'O' (1.0),\n",
              "  'Token[102]: \",\"'/'O' (1.0),\n",
              "  'Token[103]: \"but\"'/'O' (1.0),\n",
              "  'Token[104]: \"i\"'/'O' (1.0),\n",
              "  'Token[105]: \"feel\"'/'B-Malaise' (1.0),\n",
              "  'Token[106]: \"a\"'/'I-Malaise' (1.0),\n",
              "  'Token[107]: \"bit\"'/'I-Malaise' (1.0),\n",
              "  'Token[108]: \"weird\"'/'E-Malaise' (1.0),\n",
              "  'Token[109]: \",\"'/'O' (1.0),\n",
              "  'Token[110]: \"didnt\"'/'O' (1.0),\n",
              "  'Token[111]: \"have\"'/'O' (1.0),\n",
              "  'Token[112]: \"that\"'/'O' (1.0),\n",
              "  'Token[113]: \"when\"'/'O' (1.0),\n",
              "  'Token[114]: \"on\"'/'O' (1.0),\n",
              "  'Token[115]: \"50\"'/'O' (1.0),\n",
              "  'Token[116]: \".\"'/'O' (1.0)]}"
            ]
          },
          "execution_count": 11,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "temp.annotation_layers"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "id": "4CsGKKB6tGik"
      },
      "outputs": [],
      "source": [
        "\n",
        "# define your NER tags\n",
        "label_dict = {\n",
        "    'ADR': 'ADR',\n",
        "    'Drug': 'Drug',\n",
        "    'Finding': 'Finding',\n",
        "    'Symptom': 'Symptom',\n",
        "    'Disease': 'Disease'\n",
        "}"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QmNsy8AltGik",
        "outputId": "cfbc0c84-cadf-44b9-e3d1-4bf93f1dec86"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:37:07,977 Filtering empty sentences\n",
            "2023-04-21 01:37:07,998 Corpus: 948 train + 119 dev + 119 test sentences\n",
            "2023-04-21 01:37:08,000 Computing label dictionary. Progress:\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "948it [00:00, 6077.09it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:37:08,161 Dictionary created for label 'ner' with 22 values: O (seen 79057 times), I-ADR (seen 4449 times), E-ADR (seen 3506 times), B-ADR (seen 3349 times), S-ADR (seen 1363 times), S-Drug (seen 1286 times), B-Finding (seen 200 times), E-Finding (seen 200 times), I-Finding (seen 149 times), B-Disease (seen 119 times), E-Disease (seen 119 times), E-Symptom (seen 114 times), S-Disease (seen 112 times), S-Symptom (seen 112 times), E-Drug (seen 111 times), B-Symptom (seen 110 times), B-Drug (seen 110 times), S-Finding (seen 109 times), I-Symptom (seen 104 times), I-Drug (seen 50 times)\n",
            "False\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:37:16,713 SequenceTagger predicts: Dictionary with 22 tags: <unk>, O, I-ADR, E-ADR, B-ADR, S-ADR, S-Drug, B-Finding, E-Finding, I-Finding, B-Disease, E-Disease, E-Symptom, S-Disease, S-Symptom, E-Drug, B-Symptom, B-Drug, S-Finding, I-Symptom, I-Drug, I-Disease\n",
            "True\n",
            "False\n"
          ]
        }
      ],
      "source": [
        "from flair.data import Corpus\n",
        "\n",
        "from flair.embeddings import WordEmbeddings, FlairEmbeddings, StackedEmbeddings, TransformerWordEmbeddings\n",
        "from flair.models import SequenceTagger\n",
        "from flair.trainers import ModelTrainer\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "from torch.optim import AdamW, SGD\n",
        "\n",
        "from flair.visual.training_curves import Plotter\n",
        "plotter = Plotter()\n",
        "\n",
        "\n",
        "\n",
        "train_docs, test_docs = train_test_split(docs, test_size=0.2, random_state=321)\n",
        "dev_docs, test_docs = train_test_split(test_docs, test_size=0.5, random_state=123)\n",
        "\n",
        "corpus = Corpus(train=train_docs, dev=dev_docs, test=test_docs) # Corpus: 948 train + 119 dev + 119 test sentences\n",
        "corpus.filter_empty_sentences()\n",
        "# # create a tag dictionary\n",
        "labels = ['ADR', 'Drug', 'Finding', 'Symptom', 'Disease']\n",
        "# tag_dictionary = Dictionary(add_unk=True)\n",
        "# for label in labels:\n",
        "#     tag_dictionary.add_item(label)\n",
        "tag_dictionary = corpus.make_label_dictionary(label_type='ner')\n",
        "print(tag_dictionary.span_labels)\n",
        "\n",
        "# create a stacked embedding using GloVe and Flair embeddings\n",
        "stacked_embeddings = StackedEmbeddings([\n",
        "    WordEmbeddings('glove',\n",
        "                   force_cpu=False,\n",
        "                #    embedding_length=1024,\n",
        "                  #  fine_tune=True,\n",
        "                   ),\n",
        "    # TransformerWordEmbeddings('emilyalsentzer/Bio_ClinicalBERT', layers=\"-1\",\n",
        "    #                                    subtoken_pooling=\"first\",\n",
        "    #                                    is_document_embedding=True,\n",
        "    #                                    fine_tune=True,\n",
        "    #                                    use_context=True,\n",
        "    #                                    max_seq_length=1024,\n",
        "    #                                    ),\n",
        "    FlairEmbeddings('news-forward',\n",
        "                    # fine_tune=True\n",
        "                    ),\n",
        "    FlairEmbeddings('news-backward',\n",
        "                    # fine_tune=True\n",
        "                    ),\n",
        "])\n",
        "# https://github.com/flairNLP/flair/blob/master/resources/docs/embeddings/FLAIR_EMBEDDINGS.md\n",
        "\n",
        "# create a SequenceTagger object and specify the tag dictionary and embeddings to use\n",
        "tagger = SequenceTagger(hidden_size=256,\n",
        "                        embeddings=stacked_embeddings,\n",
        "                        tag_dictionary=tag_dictionary,\n",
        "                        tag_type='ner',\n",
        "                        # use_crf=False,\n",
        "                        # use_rnn=False,\n",
        "                        # reproject_embeddings=False,\n",
        "                        )\n",
        "tagger.predict_spans = False\n",
        "\n",
        "print(tagger._determine_if_span_prediction_problem(tagger.label_dictionary))\n",
        "print(tagger.predict_spans)\n",
        "\n",
        "# https://github.com/flairNLP/flair/blob/master/flair/trainers/trainer.py\n",
        "# create a ModelTrainer object and train the tagger\n",
        "trainer = ModelTrainer(tagger, corpus)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "id": "fh0zU2jHAA4J"
      },
      "outputs": [],
      "source": [
        "# !tensorboard --logdir=runs\n",
        "# http://localhost:6006/"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UKTMzxV8zF9o",
        "outputId": "fa32220d-f998-4fea-b33b-55cfae7fbc0e"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:37:16,972 tensorboard logging path is runs/tb_logs_v2\n",
            "2023-04-21 01:37:16,983 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:37:16,984 Model: \"SequenceTagger(\n",
            "  (embeddings): StackedEmbeddings(\n",
            "    (list_embedding_0): WordEmbeddings(\n",
            "      'glove'\n",
            "      (embedding): Embedding(400001, 100)\n",
            "    )\n",
            "    (list_embedding_1): FlairEmbeddings(\n",
            "      (lm): LanguageModel(\n",
            "        (drop): Dropout(p=0.05, inplace=False)\n",
            "        (encoder): Embedding(300, 100)\n",
            "        (rnn): LSTM(100, 2048)\n",
            "      )\n",
            "    )\n",
            "    (list_embedding_2): FlairEmbeddings(\n",
            "      (lm): LanguageModel(\n",
            "        (drop): Dropout(p=0.05, inplace=False)\n",
            "        (encoder): Embedding(300, 100)\n",
            "        (rnn): LSTM(100, 2048)\n",
            "      )\n",
            "    )\n",
            "  )\n",
            "  (word_dropout): WordDropout(p=0.05)\n",
            "  (locked_dropout): LockedDropout(p=0.5)\n",
            "  (embedding2nn): Linear(in_features=4196, out_features=4196, bias=True)\n",
            "  (rnn): LSTM(4196, 256, batch_first=True, bidirectional=True)\n",
            "  (linear): Linear(in_features=512, out_features=24, bias=True)\n",
            "  (loss_function): ViterbiLoss()\n",
            "  (crf): CRF()\n",
            ")\"\n",
            "2023-04-21 01:37:16,985 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:37:16,987 Corpus: \"Corpus: 948 train + 119 dev + 119 test sentences\"\n",
            "2023-04-21 01:37:16,988 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:37:16,990 Parameters:\n",
            "2023-04-21 01:37:16,991  - learning_rate: \"0.100000\"\n",
            "2023-04-21 01:37:16,992  - mini_batch_size: \"12\"\n",
            "2023-04-21 01:37:16,993  - patience: \"1\"\n",
            "2023-04-21 01:37:16,995  - anneal_factor: \"0.5\"\n",
            "2023-04-21 01:37:16,996  - max_epochs: \"100\"\n",
            "2023-04-21 01:37:16,997  - shuffle: \"True\"\n",
            "2023-04-21 01:37:16,999  - train_with_dev: \"False\"\n",
            "2023-04-21 01:37:17,000  - batch_growth_annealing: \"False\"\n",
            "2023-04-21 01:37:17,001 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:37:17,002 Model training base path: \"flair_ner_model_v2\"\n",
            "2023-04-21 01:37:17,004 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:37:17,005 Device: cuda:0\n",
            "2023-04-21 01:37:17,006 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:37:17,008 Embeddings storage mode: gpu\n",
            "2023-04-21 01:37:17,009 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:37:20,067 epoch 1 - iter 7/79 - loss 4.98871697 - time (sec): 3.06 - samples/sec: 2991.09 - lr: 0.100000\n",
            "2023-04-21 01:37:22,361 epoch 1 - iter 14/79 - loss 6.30638063 - time (sec): 5.35 - samples/sec: 3056.26 - lr: 0.100000\n",
            "2023-04-21 01:37:25,128 epoch 1 - iter 21/79 - loss 5.49719748 - time (sec): 8.12 - samples/sec: 3010.07 - lr: 0.100000\n",
            "2023-04-21 01:37:27,818 epoch 1 - iter 28/79 - loss 4.91810763 - time (sec): 10.81 - samples/sec: 3016.80 - lr: 0.100000\n",
            "2023-04-21 01:37:30,558 epoch 1 - iter 35/79 - loss 4.49004769 - time (sec): 13.55 - samples/sec: 3047.24 - lr: 0.100000\n",
            "2023-04-21 01:37:33,296 epoch 1 - iter 42/79 - loss 4.41306718 - time (sec): 16.28 - samples/sec: 3038.24 - lr: 0.100000\n",
            "2023-04-21 01:37:37,066 epoch 1 - iter 49/79 - loss 4.19839069 - time (sec): 20.06 - samples/sec: 2886.06 - lr: 0.100000\n",
            "2023-04-21 01:37:40,944 epoch 1 - iter 56/79 - loss 4.02508830 - time (sec): 23.93 - samples/sec: 2780.58 - lr: 0.100000\n",
            "2023-04-21 01:37:43,316 epoch 1 - iter 63/79 - loss 3.92796592 - time (sec): 26.30 - samples/sec: 2847.01 - lr: 0.100000\n",
            "2023-04-21 01:37:46,345 epoch 1 - iter 70/79 - loss 3.87202191 - time (sec): 29.33 - samples/sec: 2835.43 - lr: 0.100000\n",
            "2023-04-21 01:37:49,193 epoch 1 - iter 77/79 - loss 3.78429805 - time (sec): 32.18 - samples/sec: 2856.27 - lr: 0.100000\n",
            "2023-04-21 01:37:50,761 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:37:50,763 EPOCH 1 done: loss 3.7356 - lr 0.100000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.71it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:38:12,193 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:38:14,931 TRAIN : loss 1.893505573272705 - f1-score (weighted avg)  0.1212\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:05<00:00,  1.81it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:38:21,460 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:38:21,526 DEV : loss 1.9621220827102661 - f1-score (weighted avg)  0.1224\n",
            "2023-04-21 01:38:21,535 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 01:38:21,539 saving best model\n",
            "2023-04-21 01:38:23,549 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:38:25,553 epoch 2 - iter 7/79 - loss 2.67268115 - time (sec): 2.00 - samples/sec: 4214.37 - lr: 0.100000\n",
            "2023-04-21 01:38:27,020 epoch 2 - iter 14/79 - loss 3.12651882 - time (sec): 3.47 - samples/sec: 4731.92 - lr: 0.100000\n",
            "2023-04-21 01:38:28,721 epoch 2 - iter 21/79 - loss 3.32248311 - time (sec): 5.17 - samples/sec: 4792.40 - lr: 0.100000\n",
            "2023-04-21 01:38:30,257 epoch 2 - iter 28/79 - loss 3.57550108 - time (sec): 6.71 - samples/sec: 4797.07 - lr: 0.100000\n",
            "2023-04-21 01:38:31,945 epoch 2 - iter 35/79 - loss 3.60043622 - time (sec): 8.39 - samples/sec: 4806.55 - lr: 0.100000\n",
            "2023-04-21 01:38:34,374 epoch 2 - iter 42/79 - loss 3.62271722 - time (sec): 10.82 - samples/sec: 4594.34 - lr: 0.100000\n",
            "2023-04-21 01:38:36,148 epoch 2 - iter 49/79 - loss 3.52473111 - time (sec): 12.60 - samples/sec: 4636.84 - lr: 0.100000\n",
            "2023-04-21 01:38:37,941 epoch 2 - iter 56/79 - loss 3.84916292 - time (sec): 14.39 - samples/sec: 4678.23 - lr: 0.100000\n",
            "2023-04-21 01:38:39,567 epoch 2 - iter 63/79 - loss 3.79394267 - time (sec): 16.02 - samples/sec: 4699.46 - lr: 0.100000\n",
            "2023-04-21 01:38:41,539 epoch 2 - iter 70/79 - loss 3.78839245 - time (sec): 17.99 - samples/sec: 4683.52 - lr: 0.100000\n",
            "2023-04-21 01:38:43,252 epoch 2 - iter 77/79 - loss 3.76371094 - time (sec): 19.70 - samples/sec: 4696.75 - lr: 0.100000\n",
            "2023-04-21 01:38:43,635 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:38:43,637 EPOCH 2 done: loss 3.7644 - lr 0.100000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.67it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:39:05,315 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:39:08,079 TRAIN : loss 2.6930272579193115 - f1-score (weighted avg)  0.0699\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.55it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:39:10,974 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:39:11,041 DEV : loss 2.6982486248016357 - f1-score (weighted avg)  0.0729\n",
            "2023-04-21 01:39:11,050 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 01:39:11,054 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:39:12,923 epoch 3 - iter 7/79 - loss 3.42127798 - time (sec): 1.87 - samples/sec: 4492.46 - lr: 0.100000\n",
            "2023-04-21 01:39:14,490 epoch 3 - iter 14/79 - loss 3.30537078 - time (sec): 3.43 - samples/sec: 4782.15 - lr: 0.100000\n",
            "2023-04-21 01:39:16,650 epoch 3 - iter 21/79 - loss 3.32627054 - time (sec): 5.59 - samples/sec: 4597.17 - lr: 0.100000\n",
            "2023-04-21 01:39:19,011 epoch 3 - iter 28/79 - loss 3.24062048 - time (sec): 7.95 - samples/sec: 4516.48 - lr: 0.100000\n",
            "2023-04-21 01:39:20,664 epoch 3 - iter 35/79 - loss 3.35287275 - time (sec): 9.61 - samples/sec: 4581.96 - lr: 0.100000\n",
            "2023-04-21 01:39:22,335 epoch 3 - iter 42/79 - loss 3.27128971 - time (sec): 11.28 - samples/sec: 4592.75 - lr: 0.100000\n",
            "2023-04-21 01:39:23,831 epoch 3 - iter 49/79 - loss 3.35850528 - time (sec): 12.77 - samples/sec: 4652.19 - lr: 0.100000\n",
            "2023-04-21 01:39:25,462 epoch 3 - iter 56/79 - loss 3.29298585 - time (sec): 14.41 - samples/sec: 4725.46 - lr: 0.100000\n",
            "2023-04-21 01:39:27,081 epoch 3 - iter 63/79 - loss 3.35434792 - time (sec): 16.03 - samples/sec: 4776.01 - lr: 0.100000\n",
            "2023-04-21 01:39:29,181 epoch 3 - iter 70/79 - loss 3.38246214 - time (sec): 18.12 - samples/sec: 4736.59 - lr: 0.100000\n",
            "2023-04-21 01:39:30,536 epoch 3 - iter 77/79 - loss 3.46518252 - time (sec): 19.48 - samples/sec: 4769.63 - lr: 0.100000\n",
            "2023-04-21 01:39:30,955 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:39:30,956 EPOCH 3 done: loss 3.4829 - lr 0.100000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.69it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:39:52,512 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:39:56,451 TRAIN : loss 2.871452808380127 - f1-score (weighted avg)  0.0673\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.64it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:39:59,280 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:39:59,345 DEV : loss 3.0032033920288086 - f1-score (weighted avg)  0.0632\n",
            "2023-04-21 01:39:59,354 Epoch     3: reducing learning rate of group 0 to 5.0000e-02.\n",
            "2023-04-21 01:39:59,356 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 01:39:59,359 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:40:01,127 epoch 4 - iter 7/79 - loss 3.98643937 - time (sec): 1.77 - samples/sec: 4683.52 - lr: 0.050000\n",
            "2023-04-21 01:40:02,613 epoch 4 - iter 14/79 - loss 3.40115906 - time (sec): 3.25 - samples/sec: 5088.96 - lr: 0.050000\n",
            "2023-04-21 01:40:04,212 epoch 4 - iter 21/79 - loss 3.22085479 - time (sec): 4.85 - samples/sec: 4908.81 - lr: 0.050000\n",
            "2023-04-21 01:40:06,829 epoch 4 - iter 28/79 - loss 2.89564295 - time (sec): 7.47 - samples/sec: 4496.81 - lr: 0.050000\n",
            "2023-04-21 01:40:08,342 epoch 4 - iter 35/79 - loss 2.82942186 - time (sec): 8.98 - samples/sec: 4719.92 - lr: 0.050000\n",
            "2023-04-21 01:40:09,865 epoch 4 - iter 42/79 - loss 2.82470698 - time (sec): 10.50 - samples/sec: 4740.00 - lr: 0.050000\n",
            "2023-04-21 01:40:11,573 epoch 4 - iter 49/79 - loss 2.78574573 - time (sec): 12.21 - samples/sec: 4805.04 - lr: 0.050000\n",
            "2023-04-21 01:40:13,307 epoch 4 - iter 56/79 - loss 2.72692259 - time (sec): 13.95 - samples/sec: 4828.88 - lr: 0.050000\n",
            "2023-04-21 01:40:14,849 epoch 4 - iter 63/79 - loss 2.68016492 - time (sec): 15.49 - samples/sec: 4874.87 - lr: 0.050000\n",
            "2023-04-21 01:40:16,867 epoch 4 - iter 70/79 - loss 2.59727744 - time (sec): 17.51 - samples/sec: 4817.73 - lr: 0.050000\n",
            "2023-04-21 01:40:18,447 epoch 4 - iter 77/79 - loss 2.57393949 - time (sec): 19.09 - samples/sec: 4837.85 - lr: 0.050000\n",
            "2023-04-21 01:40:18,897 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:40:18,898 EPOCH 4 done: loss 2.5599 - lr 0.050000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.69it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:40:40,420 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:40:43,174 TRAIN : loss 1.9468613862991333 - f1-score (weighted avg)  0.067\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.64it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:40:46,003 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:40:46,068 DEV : loss 2.089040517807007 - f1-score (weighted avg)  0.0689\n",
            "2023-04-21 01:40:46,077 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 01:40:46,080 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:40:47,674 epoch 5 - iter 7/79 - loss 2.36757401 - time (sec): 1.59 - samples/sec: 5244.58 - lr: 0.050000\n",
            "2023-04-21 01:40:49,343 epoch 5 - iter 14/79 - loss 2.43022098 - time (sec): 3.26 - samples/sec: 4912.78 - lr: 0.050000\n",
            "2023-04-21 01:40:51,000 epoch 5 - iter 21/79 - loss 2.31296725 - time (sec): 4.92 - samples/sec: 4970.18 - lr: 0.050000\n",
            "2023-04-21 01:40:52,835 epoch 5 - iter 28/79 - loss 2.18476798 - time (sec): 6.75 - samples/sec: 4931.55 - lr: 0.050000\n",
            "2023-04-21 01:40:54,485 epoch 5 - iter 35/79 - loss 2.15867041 - time (sec): 8.40 - samples/sec: 4818.37 - lr: 0.050000\n",
            "2023-04-21 01:40:57,004 epoch 5 - iter 42/79 - loss 2.15258018 - time (sec): 10.92 - samples/sec: 4664.99 - lr: 0.050000\n",
            "2023-04-21 01:40:58,584 epoch 5 - iter 49/79 - loss 2.13998522 - time (sec): 12.50 - samples/sec: 4719.49 - lr: 0.050000\n",
            "2023-04-21 01:41:00,384 epoch 5 - iter 56/79 - loss 2.08896828 - time (sec): 14.30 - samples/sec: 4731.88 - lr: 0.050000\n",
            "2023-04-21 01:41:02,139 epoch 5 - iter 63/79 - loss 2.09025623 - time (sec): 16.06 - samples/sec: 4805.48 - lr: 0.050000\n",
            "2023-04-21 01:41:03,805 epoch 5 - iter 70/79 - loss 2.08469431 - time (sec): 17.72 - samples/sec: 4812.90 - lr: 0.050000\n",
            "2023-04-21 01:41:05,400 epoch 5 - iter 77/79 - loss 2.06537077 - time (sec): 19.32 - samples/sec: 4791.02 - lr: 0.050000\n",
            "2023-04-21 01:41:05,768 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:41:05,770 EPOCH 5 done: loss 2.0561 - lr 0.050000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.68it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:41:28,576 Evaluating as a multi-label problem: True\n",
            "2023-04-21 01:41:31,325 TRAIN : loss 1.2105978727340698 - f1-score (weighted avg)  0.1036\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.67it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:41:34,128 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:41:34,193 DEV : loss 1.304074764251709 - f1-score (weighted avg)  0.1122\n",
            "2023-04-21 01:41:34,202 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 01:41:34,206 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:41:36,280 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:41:38,179 epoch 6 - iter 7/79 - loss 1.77905556 - time (sec): 1.90 - samples/sec: 4699.21 - lr: 0.050000\n",
            "2023-04-21 01:41:40,405 epoch 6 - iter 14/79 - loss 1.84236833 - time (sec): 4.12 - samples/sec: 4166.61 - lr: 0.050000\n",
            "2023-04-21 01:41:41,938 epoch 6 - iter 21/79 - loss 1.81905389 - time (sec): 5.66 - samples/sec: 4369.05 - lr: 0.050000\n",
            "2023-04-21 01:41:43,762 epoch 6 - iter 28/79 - loss 1.82260448 - time (sec): 7.48 - samples/sec: 4293.97 - lr: 0.050000\n",
            "2023-04-21 01:41:45,533 epoch 6 - iter 35/79 - loss 1.80871179 - time (sec): 9.25 - samples/sec: 4393.17 - lr: 0.050000\n",
            "2023-04-21 01:41:46,986 epoch 6 - iter 42/79 - loss 1.79340064 - time (sec): 10.70 - samples/sec: 4583.20 - lr: 0.050000\n",
            "2023-04-21 01:41:48,901 epoch 6 - iter 49/79 - loss 1.84211908 - time (sec): 12.62 - samples/sec: 4656.36 - lr: 0.050000\n",
            "2023-04-21 01:41:50,350 epoch 6 - iter 56/79 - loss 1.88961002 - time (sec): 14.07 - samples/sec: 4700.09 - lr: 0.050000\n",
            "2023-04-21 01:41:52,601 epoch 6 - iter 63/79 - loss 1.93627826 - time (sec): 16.32 - samples/sec: 4629.90 - lr: 0.050000\n",
            "2023-04-21 01:41:54,453 epoch 6 - iter 70/79 - loss 1.96495414 - time (sec): 18.17 - samples/sec: 4632.77 - lr: 0.050000\n",
            "2023-04-21 01:41:56,015 epoch 6 - iter 77/79 - loss 1.96412205 - time (sec): 19.73 - samples/sec: 4661.69 - lr: 0.050000\n",
            "2023-04-21 01:41:56,555 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:41:56,557 EPOCH 6 done: loss 1.9521 - lr 0.050000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.65it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:42:18,347 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:42:21,107 TRAIN : loss 1.5016216039657593 - f1-score (weighted avg)  0.0817\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.62it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:42:25,081 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:42:25,150 DEV : loss 1.6057008504867554 - f1-score (weighted avg)  0.0831\n",
            "2023-04-21 01:42:25,161 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 01:42:25,165 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:42:26,883 epoch 7 - iter 7/79 - loss 1.45956417 - time (sec): 1.72 - samples/sec: 4609.93 - lr: 0.050000\n",
            "2023-04-21 01:42:29,027 epoch 7 - iter 14/79 - loss 1.70038587 - time (sec): 3.86 - samples/sec: 4555.96 - lr: 0.050000\n",
            "2023-04-21 01:42:30,524 epoch 7 - iter 21/79 - loss 1.75884807 - time (sec): 5.36 - samples/sec: 4633.20 - lr: 0.050000\n",
            "2023-04-21 01:42:32,173 epoch 7 - iter 28/79 - loss 1.77012083 - time (sec): 7.01 - samples/sec: 4750.12 - lr: 0.050000\n",
            "2023-04-21 01:42:33,759 epoch 7 - iter 35/79 - loss 1.76443774 - time (sec): 8.59 - samples/sec: 4777.51 - lr: 0.050000\n",
            "2023-04-21 01:42:36,097 epoch 7 - iter 42/79 - loss 1.72697787 - time (sec): 10.93 - samples/sec: 4592.72 - lr: 0.050000\n",
            "2023-04-21 01:42:37,875 epoch 7 - iter 49/79 - loss 1.69631983 - time (sec): 12.71 - samples/sec: 4620.53 - lr: 0.050000\n",
            "2023-04-21 01:42:39,491 epoch 7 - iter 56/79 - loss 1.67999092 - time (sec): 14.32 - samples/sec: 4680.85 - lr: 0.050000\n",
            "2023-04-21 01:42:40,969 epoch 7 - iter 63/79 - loss 1.63910039 - time (sec): 15.80 - samples/sec: 4703.23 - lr: 0.050000\n",
            "2023-04-21 01:42:42,768 epoch 7 - iter 70/79 - loss 1.64095856 - time (sec): 17.60 - samples/sec: 4741.71 - lr: 0.050000\n",
            "2023-04-21 01:42:44,537 epoch 7 - iter 77/79 - loss 1.65503941 - time (sec): 19.37 - samples/sec: 4770.89 - lr: 0.050000\n",
            "2023-04-21 01:42:45,013 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:42:45,015 EPOCH 7 done: loss 1.6521 - lr 0.050000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.68it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:43:06,632 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:43:09,370 TRAIN : loss 1.2262790203094482 - f1-score (weighted avg)  0.1321\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.66it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:43:12,181 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:43:12,246 DEV : loss 1.3367215394973755 - f1-score (weighted avg)  0.115\n",
            "2023-04-21 01:43:12,256 Epoch     7: reducing learning rate of group 0 to 2.5000e-02.\n",
            "2023-04-21 01:43:12,258 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 01:43:12,262 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:43:14,099 epoch 8 - iter 7/79 - loss 1.66657451 - time (sec): 1.84 - samples/sec: 4856.85 - lr: 0.025000\n",
            "2023-04-21 01:43:16,343 epoch 8 - iter 14/79 - loss 1.42175850 - time (sec): 4.08 - samples/sec: 4545.02 - lr: 0.025000\n",
            "2023-04-21 01:43:17,973 epoch 8 - iter 21/79 - loss 1.42912292 - time (sec): 5.71 - samples/sec: 4853.13 - lr: 0.025000\n",
            "2023-04-21 01:43:19,466 epoch 8 - iter 28/79 - loss 1.44099904 - time (sec): 7.20 - samples/sec: 4871.65 - lr: 0.025000\n",
            "2023-04-21 01:43:21,201 epoch 8 - iter 35/79 - loss 1.41867851 - time (sec): 8.94 - samples/sec: 4835.46 - lr: 0.025000\n",
            "2023-04-21 01:43:23,075 epoch 8 - iter 42/79 - loss 1.41407292 - time (sec): 10.81 - samples/sec: 4785.61 - lr: 0.025000\n",
            "2023-04-21 01:43:24,679 epoch 8 - iter 49/79 - loss 1.38706206 - time (sec): 12.42 - samples/sec: 4863.42 - lr: 0.025000\n",
            "2023-04-21 01:43:26,400 epoch 8 - iter 56/79 - loss 1.37244910 - time (sec): 14.14 - samples/sec: 4798.17 - lr: 0.025000\n",
            "2023-04-21 01:43:28,072 epoch 8 - iter 63/79 - loss 1.34076677 - time (sec): 15.81 - samples/sec: 4824.56 - lr: 0.025000\n",
            "2023-04-21 01:43:30,019 epoch 8 - iter 70/79 - loss 1.32186478 - time (sec): 17.76 - samples/sec: 4793.04 - lr: 0.025000\n",
            "2023-04-21 01:43:31,749 epoch 8 - iter 77/79 - loss 1.30603801 - time (sec): 19.49 - samples/sec: 4767.29 - lr: 0.025000\n",
            "2023-04-21 01:43:32,124 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:43:32,126 EPOCH 8 done: loss 1.3127 - lr 0.025000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.68it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:43:53,728 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:43:57,737 TRAIN : loss 0.808732807636261 - f1-score (weighted avg)  0.2764\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.58it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:44:00,612 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:44:00,681 DEV : loss 0.883874773979187 - f1-score (weighted avg)  0.252\n",
            "2023-04-21 01:44:00,691 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 01:44:00,695 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:44:02,791 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:44:04,417 epoch 9 - iter 7/79 - loss 1.03268109 - time (sec): 1.62 - samples/sec: 5343.89 - lr: 0.025000\n",
            "2023-04-21 01:44:06,294 epoch 9 - iter 14/79 - loss 1.08023398 - time (sec): 3.50 - samples/sec: 5014.98 - lr: 0.025000\n",
            "2023-04-21 01:44:08,298 epoch 9 - iter 21/79 - loss 1.10937868 - time (sec): 5.50 - samples/sec: 4822.12 - lr: 0.025000\n",
            "2023-04-21 01:44:10,284 epoch 9 - iter 28/79 - loss 1.08558738 - time (sec): 7.49 - samples/sec: 4668.97 - lr: 0.025000\n",
            "2023-04-21 01:44:11,952 epoch 9 - iter 35/79 - loss 1.09322276 - time (sec): 9.16 - samples/sec: 4672.47 - lr: 0.025000\n",
            "2023-04-21 01:44:13,584 epoch 9 - iter 42/79 - loss 1.06952405 - time (sec): 10.79 - samples/sec: 4746.42 - lr: 0.025000\n",
            "2023-04-21 01:44:15,324 epoch 9 - iter 49/79 - loss 1.06721320 - time (sec): 12.53 - samples/sec: 4765.20 - lr: 0.025000\n",
            "2023-04-21 01:44:16,874 epoch 9 - iter 56/79 - loss 1.06387028 - time (sec): 14.08 - samples/sec: 4763.18 - lr: 0.025000\n",
            "2023-04-21 01:44:18,634 epoch 9 - iter 63/79 - loss 1.04943458 - time (sec): 15.84 - samples/sec: 4756.02 - lr: 0.025000\n",
            "2023-04-21 01:44:20,259 epoch 9 - iter 70/79 - loss 1.03118733 - time (sec): 17.47 - samples/sec: 4774.91 - lr: 0.025000\n",
            "2023-04-21 01:44:22,379 epoch 9 - iter 77/79 - loss 1.02698917 - time (sec): 19.59 - samples/sec: 4711.92 - lr: 0.025000\n",
            "2023-04-21 01:44:22,834 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:44:22,836 EPOCH 9 done: loss 1.0253 - lr 0.025000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.68it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:44:44,438 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:44:47,180 TRAIN : loss 0.7327685952186584 - f1-score (weighted avg)  0.3009\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.63it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:44:50,017 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:44:50,088 DEV : loss 0.8121908903121948 - f1-score (weighted avg)  0.2791\n",
            "2023-04-21 01:44:50,097 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 01:44:50,102 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:44:52,188 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:44:54,254 epoch 10 - iter 7/79 - loss 1.00773167 - time (sec): 2.06 - samples/sec: 4123.83 - lr: 0.025000\n",
            "2023-04-21 01:44:55,657 epoch 10 - iter 14/79 - loss 1.07813924 - time (sec): 3.47 - samples/sec: 4433.14 - lr: 0.025000\n",
            "2023-04-21 01:44:57,559 epoch 10 - iter 21/79 - loss 1.07165152 - time (sec): 5.37 - samples/sec: 4288.15 - lr: 0.025000\n",
            "2023-04-21 01:44:59,248 epoch 10 - iter 28/79 - loss 1.03211974 - time (sec): 7.06 - samples/sec: 4422.48 - lr: 0.025000\n",
            "2023-04-21 01:45:01,136 epoch 10 - iter 35/79 - loss 1.03955071 - time (sec): 8.95 - samples/sec: 4460.74 - lr: 0.025000\n",
            "2023-04-21 01:45:02,823 epoch 10 - iter 42/79 - loss 1.01745259 - time (sec): 10.63 - samples/sec: 4542.09 - lr: 0.025000\n",
            "2023-04-21 01:45:04,466 epoch 10 - iter 49/79 - loss 1.02959463 - time (sec): 12.28 - samples/sec: 4628.10 - lr: 0.025000\n",
            "2023-04-21 01:45:06,242 epoch 10 - iter 56/79 - loss 1.01411734 - time (sec): 14.05 - samples/sec: 4745.80 - lr: 0.025000\n",
            "2023-04-21 01:45:08,142 epoch 10 - iter 63/79 - loss 1.02773842 - time (sec): 15.95 - samples/sec: 4730.10 - lr: 0.025000\n",
            "2023-04-21 01:45:09,875 epoch 10 - iter 70/79 - loss 1.05473014 - time (sec): 17.68 - samples/sec: 4776.68 - lr: 0.025000\n",
            "2023-04-21 01:45:11,542 epoch 10 - iter 77/79 - loss 1.04887994 - time (sec): 19.35 - samples/sec: 4788.23 - lr: 0.025000\n",
            "2023-04-21 01:45:11,973 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:45:11,975 EPOCH 10 done: loss 1.0426 - lr 0.025000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.50it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:45:34,726 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:45:37,505 TRAIN : loss 0.6637446284294128 - f1-score (weighted avg)  0.313\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.63it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:45:40,343 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:45:40,412 DEV : loss 0.7185861468315125 - f1-score (weighted avg)  0.2974\n",
            "2023-04-21 01:45:40,423 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 01:45:40,427 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:45:42,496 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:45:44,216 epoch 11 - iter 7/79 - loss 0.87527988 - time (sec): 1.72 - samples/sec: 4922.83 - lr: 0.025000\n",
            "2023-04-21 01:45:45,989 epoch 11 - iter 14/79 - loss 0.93830027 - time (sec): 3.49 - samples/sec: 5000.35 - lr: 0.025000\n",
            "2023-04-21 01:45:47,818 epoch 11 - iter 21/79 - loss 0.98154244 - time (sec): 5.32 - samples/sec: 4905.76 - lr: 0.025000\n",
            "2023-04-21 01:45:49,653 epoch 11 - iter 28/79 - loss 1.03033627 - time (sec): 7.16 - samples/sec: 4823.98 - lr: 0.025000\n",
            "2023-04-21 01:45:51,390 epoch 11 - iter 35/79 - loss 1.01001989 - time (sec): 8.89 - samples/sec: 4822.91 - lr: 0.025000\n",
            "2023-04-21 01:45:52,966 epoch 11 - iter 42/79 - loss 1.01137811 - time (sec): 10.47 - samples/sec: 4810.30 - lr: 0.025000\n",
            "2023-04-21 01:45:54,665 epoch 11 - iter 49/79 - loss 1.00765217 - time (sec): 12.17 - samples/sec: 4766.05 - lr: 0.025000\n",
            "2023-04-21 01:45:57,335 epoch 11 - iter 56/79 - loss 1.02441373 - time (sec): 14.84 - samples/sec: 4565.26 - lr: 0.025000\n",
            "2023-04-21 01:45:58,632 epoch 11 - iter 63/79 - loss 1.02308587 - time (sec): 16.13 - samples/sec: 4668.95 - lr: 0.025000\n",
            "2023-04-21 01:46:00,334 epoch 11 - iter 70/79 - loss 1.02989488 - time (sec): 17.84 - samples/sec: 4685.69 - lr: 0.025000\n",
            "2023-04-21 01:46:02,109 epoch 11 - iter 77/79 - loss 1.01061885 - time (sec): 19.61 - samples/sec: 4737.55 - lr: 0.025000\n",
            "2023-04-21 01:46:02,480 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:46:02,482 EPOCH 11 done: loss 1.0041 - lr 0.025000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.68it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:46:24,103 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:46:28,134 TRAIN : loss 0.6288954019546509 - f1-score (weighted avg)  0.298\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.59it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:46:31,005 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:46:31,075 DEV : loss 0.6958194971084595 - f1-score (weighted avg)  0.2829\n",
            "2023-04-21 01:46:31,084 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 01:46:31,088 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:46:33,180 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:46:35,007 epoch 12 - iter 7/79 - loss 0.87704300 - time (sec): 1.82 - samples/sec: 4586.81 - lr: 0.025000\n",
            "2023-04-21 01:46:36,522 epoch 12 - iter 14/79 - loss 0.91744455 - time (sec): 3.34 - samples/sec: 4840.62 - lr: 0.025000\n",
            "2023-04-21 01:46:38,679 epoch 12 - iter 21/79 - loss 0.86429671 - time (sec): 5.50 - samples/sec: 4637.20 - lr: 0.025000\n",
            "2023-04-21 01:46:40,427 epoch 12 - iter 28/79 - loss 0.85178572 - time (sec): 7.24 - samples/sec: 4749.12 - lr: 0.025000\n",
            "2023-04-21 01:46:42,469 epoch 12 - iter 35/79 - loss 0.80498164 - time (sec): 9.29 - samples/sec: 4686.46 - lr: 0.025000\n",
            "2023-04-21 01:46:44,174 epoch 12 - iter 42/79 - loss 0.81087818 - time (sec): 10.99 - samples/sec: 4654.07 - lr: 0.025000\n",
            "2023-04-21 01:46:45,908 epoch 12 - iter 49/79 - loss 0.81576899 - time (sec): 12.73 - samples/sec: 4684.06 - lr: 0.025000\n",
            "2023-04-21 01:46:47,220 epoch 12 - iter 56/79 - loss 0.80881496 - time (sec): 14.04 - samples/sec: 4790.56 - lr: 0.025000\n",
            "2023-04-21 01:46:48,830 epoch 12 - iter 63/79 - loss 0.79595620 - time (sec): 15.65 - samples/sec: 4800.66 - lr: 0.025000\n",
            "2023-04-21 01:46:51,187 epoch 12 - iter 70/79 - loss 0.80368826 - time (sec): 18.00 - samples/sec: 4678.00 - lr: 0.025000\n",
            "2023-04-21 01:46:53,023 epoch 12 - iter 77/79 - loss 0.81776739 - time (sec): 19.84 - samples/sec: 4666.72 - lr: 0.025000\n",
            "2023-04-21 01:46:53,465 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:46:53,467 EPOCH 12 done: loss 0.8193 - lr 0.025000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.68it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:47:15,076 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:47:17,824 TRAIN : loss 0.8327720761299133 - f1-score (weighted avg)  0.115\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.64it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:47:20,651 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:47:20,717 DEV : loss 0.8827935457229614 - f1-score (weighted avg)  0.1004\n",
            "2023-04-21 01:47:20,726 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 01:47:20,730 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:47:22,222 epoch 13 - iter 7/79 - loss 0.95275090 - time (sec): 1.49 - samples/sec: 5408.21 - lr: 0.025000\n",
            "2023-04-21 01:47:23,703 epoch 13 - iter 14/79 - loss 0.87642921 - time (sec): 2.97 - samples/sec: 5265.98 - lr: 0.025000\n",
            "2023-04-21 01:47:25,552 epoch 13 - iter 21/79 - loss 0.84002383 - time (sec): 4.82 - samples/sec: 4905.09 - lr: 0.025000\n",
            "2023-04-21 01:47:27,025 epoch 13 - iter 28/79 - loss 0.87107405 - time (sec): 6.29 - samples/sec: 4976.86 - lr: 0.025000\n",
            "2023-04-21 01:47:28,899 epoch 13 - iter 35/79 - loss 0.84525313 - time (sec): 8.17 - samples/sec: 4773.79 - lr: 0.025000\n",
            "2023-04-21 01:47:30,753 epoch 13 - iter 42/79 - loss 0.81872112 - time (sec): 10.02 - samples/sec: 4778.35 - lr: 0.025000\n",
            "2023-04-21 01:47:32,708 epoch 13 - iter 49/79 - loss 0.81481005 - time (sec): 11.98 - samples/sec: 4828.92 - lr: 0.025000\n",
            "2023-04-21 01:47:34,875 epoch 13 - iter 56/79 - loss 0.80216630 - time (sec): 14.14 - samples/sec: 4717.93 - lr: 0.025000\n",
            "2023-04-21 01:47:36,619 epoch 13 - iter 63/79 - loss 0.78924204 - time (sec): 15.89 - samples/sec: 4707.98 - lr: 0.025000\n",
            "2023-04-21 01:47:38,280 epoch 13 - iter 70/79 - loss 0.78937895 - time (sec): 17.55 - samples/sec: 4756.53 - lr: 0.025000\n",
            "2023-04-21 01:47:39,905 epoch 13 - iter 77/79 - loss 0.77103393 - time (sec): 19.17 - samples/sec: 4810.27 - lr: 0.025000\n",
            "2023-04-21 01:47:40,681 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:47:40,683 EPOCH 13 done: loss 0.7723 - lr 0.025000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.48it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:48:03,507 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:48:06,301 TRAIN : loss 0.6623326539993286 - f1-score (weighted avg)  0.2994\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.57it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:48:09,181 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:48:09,255 DEV : loss 0.7103167772293091 - f1-score (weighted avg)  0.2812\n",
            "2023-04-21 01:48:09,265 Epoch    13: reducing learning rate of group 0 to 1.2500e-02.\n",
            "2023-04-21 01:48:09,267 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 01:48:09,271 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:48:10,948 epoch 14 - iter 7/79 - loss 0.73308079 - time (sec): 1.67 - samples/sec: 5136.93 - lr: 0.012500\n",
            "2023-04-21 01:48:12,541 epoch 14 - iter 14/79 - loss 0.74331531 - time (sec): 3.27 - samples/sec: 5178.15 - lr: 0.012500\n",
            "2023-04-21 01:48:14,435 epoch 14 - iter 21/79 - loss 0.73362449 - time (sec): 5.16 - samples/sec: 4987.08 - lr: 0.012500\n",
            "2023-04-21 01:48:16,299 epoch 14 - iter 28/79 - loss 0.74510234 - time (sec): 7.03 - samples/sec: 4866.59 - lr: 0.012500\n",
            "2023-04-21 01:48:18,212 epoch 14 - iter 35/79 - loss 0.71622259 - time (sec): 8.94 - samples/sec: 4827.92 - lr: 0.012500\n",
            "2023-04-21 01:48:19,858 epoch 14 - iter 42/79 - loss 0.72823520 - time (sec): 10.58 - samples/sec: 4755.40 - lr: 0.012500\n",
            "2023-04-21 01:48:22,035 epoch 14 - iter 49/79 - loss 0.73247046 - time (sec): 12.76 - samples/sec: 4637.01 - lr: 0.012500\n",
            "2023-04-21 01:48:23,693 epoch 14 - iter 56/79 - loss 0.74670624 - time (sec): 14.42 - samples/sec: 4708.84 - lr: 0.012500\n",
            "2023-04-21 01:48:25,260 epoch 14 - iter 63/79 - loss 0.73631281 - time (sec): 15.99 - samples/sec: 4755.81 - lr: 0.012500\n",
            "2023-04-21 01:48:26,766 epoch 14 - iter 70/79 - loss 0.72330187 - time (sec): 17.49 - samples/sec: 4826.81 - lr: 0.012500\n",
            "2023-04-21 01:48:28,162 epoch 14 - iter 77/79 - loss 0.71671941 - time (sec): 18.89 - samples/sec: 4882.84 - lr: 0.012500\n",
            "2023-04-21 01:48:28,871 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:48:28,873 EPOCH 14 done: loss 0.7192 - lr 0.012500\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.67it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:48:50,551 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:48:54,608 TRAIN : loss 0.4756223261356354 - f1-score (weighted avg)  0.339\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.58it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:48:57,485 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:48:57,558 DEV : loss 0.5125617980957031 - f1-score (weighted avg)  0.3341\n",
            "2023-04-21 01:48:57,568 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 01:48:57,573 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:48:59,720 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:49:01,432 epoch 15 - iter 7/79 - loss 0.60629326 - time (sec): 1.71 - samples/sec: 5216.67 - lr: 0.012500\n",
            "2023-04-21 01:49:02,998 epoch 15 - iter 14/79 - loss 0.62981133 - time (sec): 3.28 - samples/sec: 5063.52 - lr: 0.012500\n",
            "2023-04-21 01:49:04,473 epoch 15 - iter 21/79 - loss 0.63317849 - time (sec): 4.75 - samples/sec: 5106.10 - lr: 0.012500\n",
            "2023-04-21 01:49:06,092 epoch 15 - iter 28/79 - loss 0.62871413 - time (sec): 6.37 - samples/sec: 5164.33 - lr: 0.012500\n",
            "2023-04-21 01:49:08,411 epoch 15 - iter 35/79 - loss 0.63528031 - time (sec): 8.69 - samples/sec: 4759.73 - lr: 0.012500\n",
            "2023-04-21 01:49:10,683 epoch 15 - iter 42/79 - loss 0.63053944 - time (sec): 10.96 - samples/sec: 4558.40 - lr: 0.012500\n",
            "2023-04-21 01:49:12,588 epoch 15 - iter 49/79 - loss 0.63191560 - time (sec): 12.87 - samples/sec: 4575.63 - lr: 0.012500\n",
            "2023-04-21 01:49:14,522 epoch 15 - iter 56/79 - loss 0.63204363 - time (sec): 14.80 - samples/sec: 4583.23 - lr: 0.012500\n",
            "2023-04-21 01:49:16,067 epoch 15 - iter 63/79 - loss 0.63493171 - time (sec): 16.35 - samples/sec: 4577.30 - lr: 0.012500\n",
            "2023-04-21 01:49:18,138 epoch 15 - iter 70/79 - loss 0.61694466 - time (sec): 18.42 - samples/sec: 4522.01 - lr: 0.012500\n",
            "2023-04-21 01:49:20,196 epoch 15 - iter 77/79 - loss 0.61524402 - time (sec): 20.47 - samples/sec: 4519.20 - lr: 0.012500\n",
            "2023-04-21 01:49:20,793 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:49:20,795 EPOCH 15 done: loss 0.6144 - lr 0.012500\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.65it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:49:42,556 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:49:45,343 TRAIN : loss 0.45660433173179626 - f1-score (weighted avg)  0.3353\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.59it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:49:48,209 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:49:48,277 DEV : loss 0.49260765314102173 - f1-score (weighted avg)  0.332\n",
            "2023-04-21 01:49:48,286 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 01:49:48,290 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:49:50,359 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:49:52,388 epoch 16 - iter 7/79 - loss 0.77625754 - time (sec): 2.03 - samples/sec: 4710.64 - lr: 0.012500\n",
            "2023-04-21 01:49:54,081 epoch 16 - iter 14/79 - loss 0.64910287 - time (sec): 3.72 - samples/sec: 5203.75 - lr: 0.012500\n",
            "2023-04-21 01:49:55,663 epoch 16 - iter 21/79 - loss 0.64576249 - time (sec): 5.30 - samples/sec: 5017.00 - lr: 0.012500\n",
            "2023-04-21 01:49:57,209 epoch 16 - iter 28/79 - loss 0.66749797 - time (sec): 6.85 - samples/sec: 4990.09 - lr: 0.012500\n",
            "2023-04-21 01:49:59,024 epoch 16 - iter 35/79 - loss 0.63563116 - time (sec): 8.66 - samples/sec: 4903.20 - lr: 0.012500\n",
            "2023-04-21 01:50:01,078 epoch 16 - iter 42/79 - loss 0.61072401 - time (sec): 10.72 - samples/sec: 4722.87 - lr: 0.012500\n",
            "2023-04-21 01:50:03,155 epoch 16 - iter 49/79 - loss 0.61736805 - time (sec): 12.79 - samples/sec: 4633.90 - lr: 0.012500\n",
            "2023-04-21 01:50:05,014 epoch 16 - iter 56/79 - loss 0.60347539 - time (sec): 14.65 - samples/sec: 4648.60 - lr: 0.012500\n",
            "2023-04-21 01:50:06,787 epoch 16 - iter 63/79 - loss 0.61221721 - time (sec): 16.43 - samples/sec: 4649.41 - lr: 0.012500\n",
            "2023-04-21 01:50:08,393 epoch 16 - iter 70/79 - loss 0.60181825 - time (sec): 18.03 - samples/sec: 4641.74 - lr: 0.012500\n",
            "2023-04-21 01:50:10,033 epoch 16 - iter 77/79 - loss 0.60264049 - time (sec): 19.67 - samples/sec: 4704.77 - lr: 0.012500\n",
            "2023-04-21 01:50:10,588 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:50:10,590 EPOCH 16 done: loss 0.5996 - lr 0.012500\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.48it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:50:33,463 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:50:36,246 TRAIN : loss 0.409477174282074 - f1-score (weighted avg)  0.3364\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.62it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:50:39,095 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:50:39,174 DEV : loss 0.4551801085472107 - f1-score (weighted avg)  0.3189\n",
            "2023-04-21 01:50:39,185 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 01:50:39,189 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:50:41,256 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:50:42,760 epoch 17 - iter 7/79 - loss 0.62622443 - time (sec): 1.50 - samples/sec: 4937.74 - lr: 0.012500\n",
            "2023-04-21 01:50:44,631 epoch 17 - iter 14/79 - loss 0.56117783 - time (sec): 3.37 - samples/sec: 4787.00 - lr: 0.012500\n",
            "2023-04-21 01:50:46,506 epoch 17 - iter 21/79 - loss 0.56396270 - time (sec): 5.25 - samples/sec: 4736.86 - lr: 0.012500\n",
            "2023-04-21 01:50:48,666 epoch 17 - iter 28/79 - loss 0.57039429 - time (sec): 7.41 - samples/sec: 4584.64 - lr: 0.012500\n",
            "2023-04-21 01:50:51,119 epoch 17 - iter 35/79 - loss 0.54977418 - time (sec): 9.86 - samples/sec: 4299.03 - lr: 0.012500\n",
            "2023-04-21 01:50:52,904 epoch 17 - iter 42/79 - loss 0.55727802 - time (sec): 11.64 - samples/sec: 4341.77 - lr: 0.012500\n",
            "2023-04-21 01:50:54,533 epoch 17 - iter 49/79 - loss 0.57228176 - time (sec): 13.27 - samples/sec: 4468.09 - lr: 0.012500\n",
            "2023-04-21 01:50:56,070 epoch 17 - iter 56/79 - loss 0.57657723 - time (sec): 14.81 - samples/sec: 4468.25 - lr: 0.012500\n",
            "2023-04-21 01:50:57,993 epoch 17 - iter 63/79 - loss 0.58581275 - time (sec): 16.73 - samples/sec: 4537.59 - lr: 0.012500\n",
            "2023-04-21 01:50:59,551 epoch 17 - iter 70/79 - loss 0.59286774 - time (sec): 18.29 - samples/sec: 4601.01 - lr: 0.012500\n",
            "2023-04-21 01:51:01,052 epoch 17 - iter 77/79 - loss 0.58771115 - time (sec): 19.79 - samples/sec: 4667.97 - lr: 0.012500\n",
            "2023-04-21 01:51:01,468 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:51:01,469 EPOCH 17 done: loss 0.5862 - lr 0.012500\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.68it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:51:23,073 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:51:27,122 TRAIN : loss 0.4442141056060791 - f1-score (weighted avg)  0.3112\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.61it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:51:29,971 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:51:30,039 DEV : loss 0.4759809374809265 - f1-score (weighted avg)  0.3045\n",
            "2023-04-21 01:51:30,049 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 01:51:30,052 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:51:32,066 epoch 18 - iter 7/79 - loss 0.50164460 - time (sec): 2.01 - samples/sec: 4407.62 - lr: 0.012500\n",
            "2023-04-21 01:51:33,757 epoch 18 - iter 14/79 - loss 0.48586134 - time (sec): 3.70 - samples/sec: 4476.11 - lr: 0.012500\n",
            "2023-04-21 01:51:35,859 epoch 18 - iter 21/79 - loss 0.49829763 - time (sec): 5.81 - samples/sec: 4444.88 - lr: 0.012500\n",
            "2023-04-21 01:51:37,587 epoch 18 - iter 28/79 - loss 0.50613623 - time (sec): 7.53 - samples/sec: 4648.69 - lr: 0.012500\n",
            "2023-04-21 01:51:39,580 epoch 18 - iter 35/79 - loss 0.51040486 - time (sec): 9.53 - samples/sec: 4585.15 - lr: 0.012500\n",
            "2023-04-21 01:51:41,323 epoch 18 - iter 42/79 - loss 0.52427579 - time (sec): 11.27 - samples/sec: 4597.93 - lr: 0.012500\n",
            "2023-04-21 01:51:42,921 epoch 18 - iter 49/79 - loss 0.52933713 - time (sec): 12.87 - samples/sec: 4652.95 - lr: 0.012500\n",
            "2023-04-21 01:51:44,443 epoch 18 - iter 56/79 - loss 0.53569826 - time (sec): 14.39 - samples/sec: 4696.18 - lr: 0.012500\n",
            "2023-04-21 01:51:45,969 epoch 18 - iter 63/79 - loss 0.54345589 - time (sec): 15.92 - samples/sec: 4710.33 - lr: 0.012500\n",
            "2023-04-21 01:51:47,977 epoch 18 - iter 70/79 - loss 0.54175574 - time (sec): 17.92 - samples/sec: 4733.54 - lr: 0.012500\n",
            "2023-04-21 01:51:49,585 epoch 18 - iter 77/79 - loss 0.54911519 - time (sec): 19.53 - samples/sec: 4735.13 - lr: 0.012500\n",
            "2023-04-21 01:51:50,071 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:51:50,073 EPOCH 18 done: loss 0.5498 - lr 0.012500\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.65it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:52:11,889 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:52:14,662 TRAIN : loss 0.3911835849285126 - f1-score (weighted avg)  0.3512\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.56it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:52:17,553 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:52:17,625 DEV : loss 0.4246269762516022 - f1-score (weighted avg)  0.348\n",
            "2023-04-21 01:52:17,635 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 01:52:17,640 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:52:19,730 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:52:21,080 epoch 19 - iter 7/79 - loss 0.55096970 - time (sec): 1.35 - samples/sec: 5550.49 - lr: 0.012500\n",
            "2023-04-21 01:52:22,635 epoch 19 - iter 14/79 - loss 0.52813232 - time (sec): 2.90 - samples/sec: 5517.42 - lr: 0.012500\n",
            "2023-04-21 01:52:24,461 epoch 19 - iter 21/79 - loss 0.55591060 - time (sec): 4.73 - samples/sec: 5239.14 - lr: 0.012500\n",
            "2023-04-21 01:52:26,110 epoch 19 - iter 28/79 - loss 0.56068690 - time (sec): 6.38 - samples/sec: 5005.80 - lr: 0.012500\n",
            "2023-04-21 01:52:28,377 epoch 19 - iter 35/79 - loss 0.56305258 - time (sec): 8.65 - samples/sec: 4731.96 - lr: 0.012500\n",
            "2023-04-21 01:52:29,808 epoch 19 - iter 42/79 - loss 0.58916537 - time (sec): 10.08 - samples/sec: 4884.73 - lr: 0.012500\n",
            "2023-04-21 01:52:32,057 epoch 19 - iter 49/79 - loss 0.59284782 - time (sec): 12.33 - samples/sec: 4723.79 - lr: 0.012500\n",
            "2023-04-21 01:52:33,865 epoch 19 - iter 56/79 - loss 0.57658648 - time (sec): 14.13 - samples/sec: 4780.36 - lr: 0.012500\n",
            "2023-04-21 01:52:35,700 epoch 19 - iter 63/79 - loss 0.57513873 - time (sec): 15.97 - samples/sec: 4788.01 - lr: 0.012500\n",
            "2023-04-21 01:52:37,337 epoch 19 - iter 70/79 - loss 0.57876202 - time (sec): 17.61 - samples/sec: 4768.41 - lr: 0.012500\n",
            "2023-04-21 01:52:39,251 epoch 19 - iter 77/79 - loss 0.57175940 - time (sec): 19.52 - samples/sec: 4743.01 - lr: 0.012500\n",
            "2023-04-21 01:52:39,725 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:52:39,727 EPOCH 19 done: loss 0.5717 - lr 0.012500\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.48it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:53:02,542 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:53:05,310 TRAIN : loss 0.42172932624816895 - f1-score (weighted avg)  0.3557\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.62it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:53:08,150 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:53:08,219 DEV : loss 0.4609644114971161 - f1-score (weighted avg)  0.3603\n",
            "2023-04-21 01:53:08,229 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 01:53:08,233 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:53:10,186 epoch 20 - iter 7/79 - loss 0.53090659 - time (sec): 1.95 - samples/sec: 4681.20 - lr: 0.012500\n",
            "2023-04-21 01:53:12,204 epoch 20 - iter 14/79 - loss 0.54853996 - time (sec): 3.97 - samples/sec: 4350.89 - lr: 0.012500\n",
            "2023-04-21 01:53:14,181 epoch 20 - iter 21/79 - loss 0.50363431 - time (sec): 5.95 - samples/sec: 4490.76 - lr: 0.012500\n",
            "2023-04-21 01:53:15,936 epoch 20 - iter 28/79 - loss 0.50602708 - time (sec): 7.70 - samples/sec: 4635.02 - lr: 0.012500\n",
            "2023-04-21 01:53:17,736 epoch 20 - iter 35/79 - loss 0.50777336 - time (sec): 9.50 - samples/sec: 4653.59 - lr: 0.012500\n",
            "2023-04-21 01:53:19,408 epoch 20 - iter 42/79 - loss 0.51625424 - time (sec): 11.17 - samples/sec: 4658.06 - lr: 0.012500\n",
            "2023-04-21 01:53:21,083 epoch 20 - iter 49/79 - loss 0.52241924 - time (sec): 12.85 - samples/sec: 4724.76 - lr: 0.012500\n",
            "2023-04-21 01:53:22,689 epoch 20 - iter 56/79 - loss 0.52675318 - time (sec): 14.45 - samples/sec: 4767.81 - lr: 0.012500\n",
            "2023-04-21 01:53:24,360 epoch 20 - iter 63/79 - loss 0.52649882 - time (sec): 16.13 - samples/sec: 4769.95 - lr: 0.012500\n",
            "2023-04-21 01:53:26,190 epoch 20 - iter 70/79 - loss 0.53601259 - time (sec): 17.96 - samples/sec: 4742.79 - lr: 0.012500\n",
            "2023-04-21 01:53:27,551 epoch 20 - iter 77/79 - loss 0.53541956 - time (sec): 19.32 - samples/sec: 4779.35 - lr: 0.012500\n",
            "2023-04-21 01:53:28,018 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:53:28,020 EPOCH 20 done: loss 0.5367 - lr 0.012500\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.66it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:53:49,787 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:53:53,858 TRAIN : loss 0.4091442823410034 - f1-score (weighted avg)  0.2937\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.62it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:53:56,698 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:53:56,774 DEV : loss 0.4408794045448303 - f1-score (weighted avg)  0.2844\n",
            "2023-04-21 01:53:56,783 Epoch    20: reducing learning rate of group 0 to 6.2500e-03.\n",
            "2023-04-21 01:53:56,785 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 01:53:56,789 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:53:58,351 epoch 21 - iter 7/79 - loss 0.58469070 - time (sec): 1.56 - samples/sec: 4756.12 - lr: 0.006250\n",
            "2023-04-21 01:53:59,958 epoch 21 - iter 14/79 - loss 0.55565969 - time (sec): 3.17 - samples/sec: 4936.91 - lr: 0.006250\n",
            "2023-04-21 01:54:01,457 epoch 21 - iter 21/79 - loss 0.58022816 - time (sec): 4.67 - samples/sec: 4993.79 - lr: 0.006250\n",
            "2023-04-21 01:54:03,843 epoch 21 - iter 28/79 - loss 0.54199227 - time (sec): 7.05 - samples/sec: 4771.03 - lr: 0.006250\n",
            "2023-04-21 01:54:05,582 epoch 21 - iter 35/79 - loss 0.54242224 - time (sec): 8.79 - samples/sec: 4727.26 - lr: 0.006250\n",
            "2023-04-21 01:54:07,074 epoch 21 - iter 42/79 - loss 0.53008127 - time (sec): 10.28 - samples/sec: 4896.88 - lr: 0.006250\n",
            "2023-04-21 01:54:08,614 epoch 21 - iter 49/79 - loss 0.51750482 - time (sec): 11.82 - samples/sec: 4896.68 - lr: 0.006250\n",
            "2023-04-21 01:54:10,202 epoch 21 - iter 56/79 - loss 0.50875919 - time (sec): 13.41 - samples/sec: 4934.40 - lr: 0.006250\n",
            "2023-04-21 01:54:11,926 epoch 21 - iter 63/79 - loss 0.50161678 - time (sec): 15.13 - samples/sec: 4941.95 - lr: 0.006250\n",
            "2023-04-21 01:54:13,804 epoch 21 - iter 70/79 - loss 0.49706652 - time (sec): 17.01 - samples/sec: 4876.09 - lr: 0.006250\n",
            "2023-04-21 01:54:15,895 epoch 21 - iter 77/79 - loss 0.49946174 - time (sec): 19.10 - samples/sec: 4819.61 - lr: 0.006250\n",
            "2023-04-21 01:54:16,365 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:54:16,367 EPOCH 21 done: loss 0.5032 - lr 0.006250\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.68it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:54:37,943 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:54:40,731 TRAIN : loss 0.36594632267951965 - f1-score (weighted avg)  0.3453\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.61it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:54:43,580 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:54:43,649 DEV : loss 0.3956149220466614 - f1-score (weighted avg)  0.3266\n",
            "2023-04-21 01:54:43,658 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 01:54:43,662 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:54:45,768 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:54:47,820 epoch 22 - iter 7/79 - loss 0.45124281 - time (sec): 2.05 - samples/sec: 4365.36 - lr: 0.006250\n",
            "2023-04-21 01:54:49,353 epoch 22 - iter 14/79 - loss 0.49555582 - time (sec): 3.58 - samples/sec: 4493.69 - lr: 0.006250\n",
            "2023-04-21 01:54:51,253 epoch 22 - iter 21/79 - loss 0.47635277 - time (sec): 5.48 - samples/sec: 4596.30 - lr: 0.006250\n",
            "2023-04-21 01:54:52,853 epoch 22 - iter 28/79 - loss 0.47924326 - time (sec): 7.08 - samples/sec: 4634.55 - lr: 0.006250\n",
            "2023-04-21 01:54:54,538 epoch 22 - iter 35/79 - loss 0.46591205 - time (sec): 8.77 - samples/sec: 4736.42 - lr: 0.006250\n",
            "2023-04-21 01:54:56,125 epoch 22 - iter 42/79 - loss 0.47196160 - time (sec): 10.36 - samples/sec: 4832.96 - lr: 0.006250\n",
            "2023-04-21 01:54:57,918 epoch 22 - iter 49/79 - loss 0.48072255 - time (sec): 12.15 - samples/sec: 4806.86 - lr: 0.006250\n",
            "2023-04-21 01:54:59,509 epoch 22 - iter 56/79 - loss 0.48340797 - time (sec): 13.74 - samples/sec: 4846.55 - lr: 0.006250\n",
            "2023-04-21 01:55:01,323 epoch 22 - iter 63/79 - loss 0.47804700 - time (sec): 15.55 - samples/sec: 4838.65 - lr: 0.006250\n",
            "2023-04-21 01:55:03,387 epoch 22 - iter 70/79 - loss 0.46925927 - time (sec): 17.62 - samples/sec: 4768.83 - lr: 0.006250\n",
            "2023-04-21 01:55:04,998 epoch 22 - iter 77/79 - loss 0.46278082 - time (sec): 19.23 - samples/sec: 4787.93 - lr: 0.006250\n",
            "2023-04-21 01:55:05,464 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:55:05,466 EPOCH 22 done: loss 0.4600 - lr 0.006250\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.49it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:55:28,239 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:55:31,003 TRAIN : loss 0.35216522216796875 - f1-score (weighted avg)  0.3338\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.67it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:55:33,807 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:55:33,875 DEV : loss 0.38587358593940735 - f1-score (weighted avg)  0.3118\n",
            "2023-04-21 01:55:33,884 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 01:55:33,888 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:55:35,984 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:55:37,627 epoch 23 - iter 7/79 - loss 0.38650867 - time (sec): 1.64 - samples/sec: 4884.52 - lr: 0.006250\n",
            "2023-04-21 01:55:39,320 epoch 23 - iter 14/79 - loss 0.39284820 - time (sec): 3.33 - samples/sec: 5107.85 - lr: 0.006250\n",
            "2023-04-21 01:55:40,840 epoch 23 - iter 21/79 - loss 0.39981571 - time (sec): 4.85 - samples/sec: 5112.16 - lr: 0.006250\n",
            "2023-04-21 01:55:42,470 epoch 23 - iter 28/79 - loss 0.41451596 - time (sec): 6.48 - samples/sec: 5035.06 - lr: 0.006250\n",
            "2023-04-21 01:55:44,118 epoch 23 - iter 35/79 - loss 0.46072364 - time (sec): 8.13 - samples/sec: 5011.00 - lr: 0.006250\n",
            "2023-04-21 01:55:46,204 epoch 23 - iter 42/79 - loss 0.46250196 - time (sec): 10.22 - samples/sec: 4955.41 - lr: 0.006250\n",
            "2023-04-21 01:55:47,860 epoch 23 - iter 49/79 - loss 0.46219280 - time (sec): 11.87 - samples/sec: 4919.45 - lr: 0.006250\n",
            "2023-04-21 01:55:49,524 epoch 23 - iter 56/79 - loss 0.47025891 - time (sec): 13.54 - samples/sec: 4912.77 - lr: 0.006250\n",
            "2023-04-21 01:55:51,812 epoch 23 - iter 63/79 - loss 0.46724405 - time (sec): 15.83 - samples/sec: 4760.07 - lr: 0.006250\n",
            "2023-04-21 01:55:53,930 epoch 23 - iter 70/79 - loss 0.46071880 - time (sec): 17.94 - samples/sec: 4746.14 - lr: 0.006250\n",
            "2023-04-21 01:55:55,422 epoch 23 - iter 77/79 - loss 0.46427754 - time (sec): 19.44 - samples/sec: 4769.61 - lr: 0.006250\n",
            "2023-04-21 01:55:55,919 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:55:55,921 EPOCH 23 done: loss 0.4617 - lr 0.006250\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.67it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:56:17,597 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:56:21,638 TRAIN : loss 0.356751024723053 - f1-score (weighted avg)  0.3228\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.63it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:56:24,472 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:56:24,540 DEV : loss 0.3822615146636963 - f1-score (weighted avg)  0.3033\n",
            "2023-04-21 01:56:24,549 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 01:56:24,553 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:56:26,635 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:56:28,205 epoch 24 - iter 7/79 - loss 0.42153774 - time (sec): 1.57 - samples/sec: 5288.87 - lr: 0.006250\n",
            "2023-04-21 01:56:30,365 epoch 24 - iter 14/79 - loss 0.42961325 - time (sec): 3.73 - samples/sec: 4359.98 - lr: 0.006250\n",
            "2023-04-21 01:56:32,086 epoch 24 - iter 21/79 - loss 0.43951591 - time (sec): 5.45 - samples/sec: 4584.48 - lr: 0.006250\n",
            "2023-04-21 01:56:33,742 epoch 24 - iter 28/79 - loss 0.45264062 - time (sec): 7.11 - samples/sec: 4654.51 - lr: 0.006250\n",
            "2023-04-21 01:56:35,490 epoch 24 - iter 35/79 - loss 0.44932533 - time (sec): 8.85 - samples/sec: 4640.62 - lr: 0.006250\n",
            "2023-04-21 01:56:37,636 epoch 24 - iter 42/79 - loss 0.44470367 - time (sec): 11.00 - samples/sec: 4571.37 - lr: 0.006250\n",
            "2023-04-21 01:56:39,594 epoch 24 - iter 49/79 - loss 0.44495754 - time (sec): 12.96 - samples/sec: 4550.51 - lr: 0.006250\n",
            "2023-04-21 01:56:41,130 epoch 24 - iter 56/79 - loss 0.44274022 - time (sec): 14.49 - samples/sec: 4580.39 - lr: 0.006250\n",
            "2023-04-21 01:56:42,981 epoch 24 - iter 63/79 - loss 0.44256608 - time (sec): 16.34 - samples/sec: 4648.17 - lr: 0.006250\n",
            "2023-04-21 01:56:44,714 epoch 24 - iter 70/79 - loss 0.44571531 - time (sec): 18.08 - samples/sec: 4610.71 - lr: 0.006250\n",
            "2023-04-21 01:56:46,452 epoch 24 - iter 77/79 - loss 0.44900436 - time (sec): 19.82 - samples/sec: 4640.82 - lr: 0.006250\n",
            "2023-04-21 01:56:46,939 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:56:46,940 EPOCH 24 done: loss 0.4467 - lr 0.006250\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.65it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:57:08,707 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:57:11,471 TRAIN : loss 0.3347572386264801 - f1-score (weighted avg)  0.3298\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.54it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:57:14,375 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:57:14,444 DEV : loss 0.36759448051452637 - f1-score (weighted avg)  0.306\n",
            "2023-04-21 01:57:14,454 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 01:57:14,457 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:57:16,551 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:57:18,182 epoch 25 - iter 7/79 - loss 0.41551849 - time (sec): 1.63 - samples/sec: 5451.07 - lr: 0.006250\n",
            "2023-04-21 01:57:20,323 epoch 25 - iter 14/79 - loss 0.43681681 - time (sec): 3.77 - samples/sec: 4706.19 - lr: 0.006250\n",
            "2023-04-21 01:57:21,842 epoch 25 - iter 21/79 - loss 0.44478019 - time (sec): 5.29 - samples/sec: 4994.71 - lr: 0.006250\n",
            "2023-04-21 01:57:23,815 epoch 25 - iter 28/79 - loss 0.42779878 - time (sec): 7.26 - samples/sec: 4843.47 - lr: 0.006250\n",
            "2023-04-21 01:57:25,485 epoch 25 - iter 35/79 - loss 0.43508164 - time (sec): 8.93 - samples/sec: 4821.04 - lr: 0.006250\n",
            "2023-04-21 01:57:27,384 epoch 25 - iter 42/79 - loss 0.43491818 - time (sec): 10.83 - samples/sec: 4721.35 - lr: 0.006250\n",
            "2023-04-21 01:57:29,539 epoch 25 - iter 49/79 - loss 0.43594853 - time (sec): 12.99 - samples/sec: 4571.84 - lr: 0.006250\n",
            "2023-04-21 01:57:31,448 epoch 25 - iter 56/79 - loss 0.43186061 - time (sec): 14.90 - samples/sec: 4594.98 - lr: 0.006250\n",
            "2023-04-21 01:57:33,045 epoch 25 - iter 63/79 - loss 0.43460139 - time (sec): 16.49 - samples/sec: 4604.72 - lr: 0.006250\n",
            "2023-04-21 01:57:35,048 epoch 25 - iter 70/79 - loss 0.43644816 - time (sec): 18.50 - samples/sec: 4552.95 - lr: 0.006250\n",
            "2023-04-21 01:57:36,555 epoch 25 - iter 77/79 - loss 0.44035077 - time (sec): 20.00 - samples/sec: 4613.77 - lr: 0.006250\n",
            "2023-04-21 01:57:36,988 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:57:36,990 EPOCH 25 done: loss 0.4381 - lr 0.006250\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.47it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:57:59,912 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:58:02,691 TRAIN : loss 0.3291448950767517 - f1-score (weighted avg)  0.3047\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.63it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:58:05,524 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:58:05,612 DEV : loss 0.35930559039115906 - f1-score (weighted avg)  0.2821\n",
            "2023-04-21 01:58:05,622 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 01:58:05,626 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:58:07,734 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:58:09,542 epoch 26 - iter 7/79 - loss 0.40627068 - time (sec): 1.81 - samples/sec: 4603.02 - lr: 0.006250\n",
            "2023-04-21 01:58:11,517 epoch 26 - iter 14/79 - loss 0.40855086 - time (sec): 3.78 - samples/sec: 4620.51 - lr: 0.006250\n",
            "2023-04-21 01:58:12,935 epoch 26 - iter 21/79 - loss 0.40511692 - time (sec): 5.20 - samples/sec: 4807.82 - lr: 0.006250\n",
            "2023-04-21 01:58:15,150 epoch 26 - iter 28/79 - loss 0.39726056 - time (sec): 7.41 - samples/sec: 4673.60 - lr: 0.006250\n",
            "2023-04-21 01:58:16,659 epoch 26 - iter 35/79 - loss 0.39752622 - time (sec): 8.92 - samples/sec: 4675.81 - lr: 0.006250\n",
            "2023-04-21 01:58:18,923 epoch 26 - iter 42/79 - loss 0.39929690 - time (sec): 11.19 - samples/sec: 4530.11 - lr: 0.006250\n",
            "2023-04-21 01:58:20,963 epoch 26 - iter 49/79 - loss 0.41873473 - time (sec): 13.23 - samples/sec: 4504.98 - lr: 0.006250\n",
            "2023-04-21 01:58:22,689 epoch 26 - iter 56/79 - loss 0.43675350 - time (sec): 14.95 - samples/sec: 4575.71 - lr: 0.006250\n",
            "2023-04-21 01:58:24,306 epoch 26 - iter 63/79 - loss 0.43463331 - time (sec): 16.57 - samples/sec: 4616.33 - lr: 0.006250\n",
            "2023-04-21 01:58:25,930 epoch 26 - iter 70/79 - loss 0.43080356 - time (sec): 18.19 - samples/sec: 4621.37 - lr: 0.006250\n",
            "2023-04-21 01:58:27,568 epoch 26 - iter 77/79 - loss 0.42655136 - time (sec): 19.83 - samples/sec: 4678.97 - lr: 0.006250\n",
            "2023-04-21 01:58:27,963 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:58:27,965 EPOCH 26 done: loss 0.4239 - lr 0.006250\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.67it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:58:49,619 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:58:53,707 TRAIN : loss 0.32526880502700806 - f1-score (weighted avg)  0.3395\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.60it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:58:56,566 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:58:56,635 DEV : loss 0.3525862395763397 - f1-score (weighted avg)  0.3214\n",
            "2023-04-21 01:58:56,644 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 01:58:56,649 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:58:58,732 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:59:00,591 epoch 27 - iter 7/79 - loss 0.43382909 - time (sec): 1.86 - samples/sec: 3940.85 - lr: 0.006250\n",
            "2023-04-21 01:59:02,416 epoch 27 - iter 14/79 - loss 0.44328547 - time (sec): 3.68 - samples/sec: 4125.86 - lr: 0.006250\n",
            "2023-04-21 01:59:03,926 epoch 27 - iter 21/79 - loss 0.41795915 - time (sec): 5.19 - samples/sec: 4545.70 - lr: 0.006250\n",
            "2023-04-21 01:59:05,329 epoch 27 - iter 28/79 - loss 0.42492073 - time (sec): 6.60 - samples/sec: 4762.29 - lr: 0.006250\n",
            "2023-04-21 01:59:07,465 epoch 27 - iter 35/79 - loss 0.42124862 - time (sec): 8.73 - samples/sec: 4553.90 - lr: 0.006250\n",
            "2023-04-21 01:59:09,408 epoch 27 - iter 42/79 - loss 0.42622929 - time (sec): 10.67 - samples/sec: 4487.55 - lr: 0.006250\n",
            "2023-04-21 01:59:11,140 epoch 27 - iter 49/79 - loss 0.44199139 - time (sec): 12.41 - samples/sec: 4603.95 - lr: 0.006250\n",
            "2023-04-21 01:59:13,209 epoch 27 - iter 56/79 - loss 0.44756827 - time (sec): 14.47 - samples/sec: 4567.21 - lr: 0.006250\n",
            "2023-04-21 01:59:14,928 epoch 27 - iter 63/79 - loss 0.43982823 - time (sec): 16.19 - samples/sec: 4595.84 - lr: 0.006250\n",
            "2023-04-21 01:59:16,581 epoch 27 - iter 70/79 - loss 0.43710414 - time (sec): 17.85 - samples/sec: 4651.66 - lr: 0.006250\n",
            "2023-04-21 01:59:18,526 epoch 27 - iter 77/79 - loss 0.43673430 - time (sec): 19.79 - samples/sec: 4673.84 - lr: 0.006250\n",
            "2023-04-21 01:59:19,114 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:59:19,115 EPOCH 27 done: loss 0.4341 - lr 0.006250\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.66it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:59:40,815 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:59:43,589 TRAIN : loss 0.315780371427536 - f1-score (weighted avg)  0.3569\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.60it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:59:46,450 Evaluating as a multi-label problem: False\n",
            "2023-04-21 01:59:46,520 DEV : loss 0.3392994701862335 - f1-score (weighted avg)  0.3416\n",
            "2023-04-21 01:59:46,530 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 01:59:46,534 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 01:59:48,626 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 01:59:50,270 epoch 28 - iter 7/79 - loss 0.44658552 - time (sec): 1.64 - samples/sec: 5130.38 - lr: 0.006250\n",
            "2023-04-21 01:59:51,769 epoch 28 - iter 14/79 - loss 0.39492496 - time (sec): 3.14 - samples/sec: 5083.31 - lr: 0.006250\n",
            "2023-04-21 01:59:53,689 epoch 28 - iter 21/79 - loss 0.40176547 - time (sec): 5.06 - samples/sec: 4825.04 - lr: 0.006250\n",
            "2023-04-21 01:59:55,168 epoch 28 - iter 28/79 - loss 0.41003146 - time (sec): 6.54 - samples/sec: 4963.73 - lr: 0.006250\n",
            "2023-04-21 01:59:57,017 epoch 28 - iter 35/79 - loss 0.40682817 - time (sec): 8.39 - samples/sec: 4939.54 - lr: 0.006250\n",
            "2023-04-21 01:59:58,970 epoch 28 - iter 42/79 - loss 0.40789101 - time (sec): 10.34 - samples/sec: 4759.11 - lr: 0.006250\n",
            "2023-04-21 02:00:00,703 epoch 28 - iter 49/79 - loss 0.40454252 - time (sec): 12.07 - samples/sec: 4727.34 - lr: 0.006250\n",
            "2023-04-21 02:00:02,531 epoch 28 - iter 56/79 - loss 0.40842339 - time (sec): 13.90 - samples/sec: 4729.02 - lr: 0.006250\n",
            "2023-04-21 02:00:04,471 epoch 28 - iter 63/79 - loss 0.41409830 - time (sec): 15.84 - samples/sec: 4710.87 - lr: 0.006250\n",
            "2023-04-21 02:00:06,444 epoch 28 - iter 70/79 - loss 0.41198804 - time (sec): 17.82 - samples/sec: 4697.10 - lr: 0.006250\n",
            "2023-04-21 02:00:07,916 epoch 28 - iter 77/79 - loss 0.41373644 - time (sec): 19.29 - samples/sec: 4734.94 - lr: 0.006250\n",
            "2023-04-21 02:00:08,549 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:00:08,551 EPOCH 28 done: loss 0.4132 - lr 0.006250\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.49it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:00:31,324 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:00:34,073 TRAIN : loss 0.31246891617774963 - f1-score (weighted avg)  0.3607\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.57it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:00:36,952 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:00:37,028 DEV : loss 0.3461717367172241 - f1-score (weighted avg)  0.3415\n",
            "2023-04-21 02:00:37,038 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:00:37,043 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:00:38,748 epoch 29 - iter 7/79 - loss 0.43269769 - time (sec): 1.70 - samples/sec: 4522.98 - lr: 0.006250\n",
            "2023-04-21 02:00:40,380 epoch 29 - iter 14/79 - loss 0.43561438 - time (sec): 3.34 - samples/sec: 4553.35 - lr: 0.006250\n",
            "2023-04-21 02:00:42,068 epoch 29 - iter 21/79 - loss 0.42182164 - time (sec): 5.02 - samples/sec: 4619.37 - lr: 0.006250\n",
            "2023-04-21 02:00:43,742 epoch 29 - iter 28/79 - loss 0.42265483 - time (sec): 6.70 - samples/sec: 4750.12 - lr: 0.006250\n",
            "2023-04-21 02:00:45,521 epoch 29 - iter 35/79 - loss 0.41569343 - time (sec): 8.48 - samples/sec: 4810.07 - lr: 0.006250\n",
            "2023-04-21 02:00:47,273 epoch 29 - iter 42/79 - loss 0.41233593 - time (sec): 10.23 - samples/sec: 4824.21 - lr: 0.006250\n",
            "2023-04-21 02:00:48,966 epoch 29 - iter 49/79 - loss 0.41374633 - time (sec): 11.92 - samples/sec: 4825.29 - lr: 0.006250\n",
            "2023-04-21 02:00:51,084 epoch 29 - iter 56/79 - loss 0.40965942 - time (sec): 14.04 - samples/sec: 4726.62 - lr: 0.006250\n",
            "2023-04-21 02:00:53,131 epoch 29 - iter 63/79 - loss 0.41760609 - time (sec): 16.09 - samples/sec: 4741.96 - lr: 0.006250\n",
            "2023-04-21 02:00:54,943 epoch 29 - iter 70/79 - loss 0.41902554 - time (sec): 17.90 - samples/sec: 4731.96 - lr: 0.006250\n",
            "2023-04-21 02:00:56,748 epoch 29 - iter 77/79 - loss 0.42154867 - time (sec): 19.70 - samples/sec: 4694.41 - lr: 0.006250\n",
            "2023-04-21 02:00:57,175 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:00:57,177 EPOCH 29 done: loss 0.4208 - lr 0.006250\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.64it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:01:19,031 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:01:23,147 TRAIN : loss 0.3048126995563507 - f1-score (weighted avg)  0.3433\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.51it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:01:26,072 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:01:26,142 DEV : loss 0.33108487725257874 - f1-score (weighted avg)  0.3263\n",
            "2023-04-21 02:01:26,152 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:01:26,156 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:01:28,236 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:01:30,135 epoch 30 - iter 7/79 - loss 0.39362351 - time (sec): 1.90 - samples/sec: 4248.83 - lr: 0.006250\n",
            "2023-04-21 02:01:32,146 epoch 30 - iter 14/79 - loss 0.40134311 - time (sec): 3.91 - samples/sec: 4358.99 - lr: 0.006250\n",
            "2023-04-21 02:01:33,729 epoch 30 - iter 21/79 - loss 0.40698763 - time (sec): 5.49 - samples/sec: 4739.33 - lr: 0.006250\n",
            "2023-04-21 02:01:35,443 epoch 30 - iter 28/79 - loss 0.40869932 - time (sec): 7.20 - samples/sec: 4747.01 - lr: 0.006250\n",
            "2023-04-21 02:01:37,264 epoch 30 - iter 35/79 - loss 0.40518116 - time (sec): 9.03 - samples/sec: 4694.16 - lr: 0.006250\n",
            "2023-04-21 02:01:38,893 epoch 30 - iter 42/79 - loss 0.40348079 - time (sec): 10.65 - samples/sec: 4752.39 - lr: 0.006250\n",
            "2023-04-21 02:01:40,844 epoch 30 - iter 49/79 - loss 0.40795252 - time (sec): 12.61 - samples/sec: 4670.50 - lr: 0.006250\n",
            "2023-04-21 02:01:42,869 epoch 30 - iter 56/79 - loss 0.40471424 - time (sec): 14.63 - samples/sec: 4586.44 - lr: 0.006250\n",
            "2023-04-21 02:01:45,072 epoch 30 - iter 63/79 - loss 0.40020874 - time (sec): 16.83 - samples/sec: 4512.00 - lr: 0.006250\n",
            "2023-04-21 02:01:46,492 epoch 30 - iter 70/79 - loss 0.39639930 - time (sec): 18.25 - samples/sec: 4615.77 - lr: 0.006250\n",
            "2023-04-21 02:01:48,139 epoch 30 - iter 77/79 - loss 0.39981249 - time (sec): 19.90 - samples/sec: 4636.38 - lr: 0.006250\n",
            "2023-04-21 02:01:48,655 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:01:48,657 EPOCH 30 done: loss 0.3991 - lr 0.006250\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.65it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:02:10,446 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:02:13,237 TRAIN : loss 0.31649377942085266 - f1-score (weighted avg)  0.3433\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.56it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:02:16,124 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:02:16,192 DEV : loss 0.3373541533946991 - f1-score (weighted avg)  0.335\n",
            "2023-04-21 02:02:16,202 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:02:16,206 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:02:17,866 epoch 31 - iter 7/79 - loss 0.43654980 - time (sec): 1.66 - samples/sec: 4601.41 - lr: 0.006250\n",
            "2023-04-21 02:02:19,353 epoch 31 - iter 14/79 - loss 0.42965311 - time (sec): 3.15 - samples/sec: 4947.34 - lr: 0.006250\n",
            "2023-04-21 02:02:21,547 epoch 31 - iter 21/79 - loss 0.40736255 - time (sec): 5.34 - samples/sec: 4664.16 - lr: 0.006250\n",
            "2023-04-21 02:02:23,306 epoch 31 - iter 28/79 - loss 0.40639293 - time (sec): 7.10 - samples/sec: 4697.02 - lr: 0.006250\n",
            "2023-04-21 02:02:25,053 epoch 31 - iter 35/79 - loss 0.40861369 - time (sec): 8.84 - samples/sec: 4707.72 - lr: 0.006250\n",
            "2023-04-21 02:02:26,630 epoch 31 - iter 42/79 - loss 0.40407902 - time (sec): 10.42 - samples/sec: 4825.58 - lr: 0.006250\n",
            "2023-04-21 02:02:28,228 epoch 31 - iter 49/79 - loss 0.41536849 - time (sec): 12.02 - samples/sec: 4875.82 - lr: 0.006250\n",
            "2023-04-21 02:02:29,823 epoch 31 - iter 56/79 - loss 0.41697822 - time (sec): 13.62 - samples/sec: 4895.03 - lr: 0.006250\n",
            "2023-04-21 02:02:31,768 epoch 31 - iter 63/79 - loss 0.41303782 - time (sec): 15.56 - samples/sec: 4800.20 - lr: 0.006250\n",
            "2023-04-21 02:02:33,363 epoch 31 - iter 70/79 - loss 0.42075773 - time (sec): 17.15 - samples/sec: 4853.60 - lr: 0.006250\n",
            "2023-04-21 02:02:35,192 epoch 31 - iter 77/79 - loss 0.42234769 - time (sec): 18.98 - samples/sec: 4852.90 - lr: 0.006250\n",
            "2023-04-21 02:02:35,770 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:02:35,772 EPOCH 31 done: loss 0.4218 - lr 0.006250\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.64it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:02:58,841 Evaluating as a multi-label problem: True\n",
            "2023-04-21 02:03:01,623 TRAIN : loss 0.2990100085735321 - f1-score (weighted avg)  0.3693\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.61it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:03:04,470 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:03:04,540 DEV : loss 0.3215554654598236 - f1-score (weighted avg)  0.3648\n",
            "2023-04-21 02:03:04,550 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:03:04,553 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:03:06,637 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:03:08,260 epoch 32 - iter 7/79 - loss 0.40458333 - time (sec): 1.62 - samples/sec: 6013.47 - lr: 0.006250\n",
            "2023-04-21 02:03:10,369 epoch 32 - iter 14/79 - loss 0.39036225 - time (sec): 3.73 - samples/sec: 4882.67 - lr: 0.006250\n",
            "2023-04-21 02:03:11,859 epoch 32 - iter 21/79 - loss 0.39760320 - time (sec): 5.22 - samples/sec: 5012.70 - lr: 0.006250\n",
            "2023-04-21 02:03:13,363 epoch 32 - iter 28/79 - loss 0.40748424 - time (sec): 6.72 - samples/sec: 5071.76 - lr: 0.006250\n",
            "2023-04-21 02:03:14,929 epoch 32 - iter 35/79 - loss 0.39508722 - time (sec): 8.29 - samples/sec: 5092.25 - lr: 0.006250\n",
            "2023-04-21 02:03:17,033 epoch 32 - iter 42/79 - loss 0.40558726 - time (sec): 10.39 - samples/sec: 4847.21 - lr: 0.006250\n",
            "2023-04-21 02:03:18,789 epoch 32 - iter 49/79 - loss 0.40409614 - time (sec): 12.15 - samples/sec: 4826.29 - lr: 0.006250\n",
            "2023-04-21 02:03:21,254 epoch 32 - iter 56/79 - loss 0.39156002 - time (sec): 14.62 - samples/sec: 4731.55 - lr: 0.006250\n",
            "2023-04-21 02:03:22,749 epoch 32 - iter 63/79 - loss 0.39723596 - time (sec): 16.11 - samples/sec: 4745.02 - lr: 0.006250\n",
            "2023-04-21 02:03:24,818 epoch 32 - iter 70/79 - loss 0.38890788 - time (sec): 18.18 - samples/sec: 4727.38 - lr: 0.006250\n",
            "2023-04-21 02:03:26,292 epoch 32 - iter 77/79 - loss 0.39054831 - time (sec): 19.65 - samples/sec: 4714.13 - lr: 0.006250\n",
            "2023-04-21 02:03:26,794 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:03:26,796 EPOCH 32 done: loss 0.3906 - lr 0.006250\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.66it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:03:48,501 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:03:51,277 TRAIN : loss 0.3419773280620575 - f1-score (weighted avg)  0.3189\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.61it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:03:54,130 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:03:54,196 DEV : loss 0.35737770795822144 - f1-score (weighted avg)  0.3183\n",
            "2023-04-21 02:03:54,206 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:03:54,210 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:03:55,772 epoch 33 - iter 7/79 - loss 0.38023578 - time (sec): 1.56 - samples/sec: 4985.03 - lr: 0.006250\n",
            "2023-04-21 02:03:57,623 epoch 33 - iter 14/79 - loss 0.40063690 - time (sec): 3.41 - samples/sec: 4951.62 - lr: 0.006250\n",
            "2023-04-21 02:03:59,042 epoch 33 - iter 21/79 - loss 0.41917389 - time (sec): 4.83 - samples/sec: 5005.04 - lr: 0.006250\n",
            "2023-04-21 02:04:00,650 epoch 33 - iter 28/79 - loss 0.40753831 - time (sec): 6.44 - samples/sec: 5051.10 - lr: 0.006250\n",
            "2023-04-21 02:04:02,212 epoch 33 - iter 35/79 - loss 0.40098344 - time (sec): 8.00 - samples/sec: 5182.37 - lr: 0.006250\n",
            "2023-04-21 02:04:04,006 epoch 33 - iter 42/79 - loss 0.40258796 - time (sec): 9.79 - samples/sec: 5082.47 - lr: 0.006250\n",
            "2023-04-21 02:04:05,930 epoch 33 - iter 49/79 - loss 0.40032758 - time (sec): 11.72 - samples/sec: 5036.26 - lr: 0.006250\n",
            "2023-04-21 02:04:08,150 epoch 33 - iter 56/79 - loss 0.39777044 - time (sec): 13.94 - samples/sec: 4796.64 - lr: 0.006250\n",
            "2023-04-21 02:04:09,894 epoch 33 - iter 63/79 - loss 0.39511045 - time (sec): 15.68 - samples/sec: 4816.53 - lr: 0.006250\n",
            "2023-04-21 02:04:12,057 epoch 33 - iter 70/79 - loss 0.39205873 - time (sec): 17.84 - samples/sec: 4759.88 - lr: 0.006250\n",
            "2023-04-21 02:04:13,770 epoch 33 - iter 77/79 - loss 0.38631256 - time (sec): 19.56 - samples/sec: 4750.36 - lr: 0.006250\n",
            "2023-04-21 02:04:14,176 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:04:14,177 EPOCH 33 done: loss 0.3850 - lr 0.006250\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.48it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:04:37,024 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:04:39,817 TRAIN : loss 0.2769707143306732 - f1-score (weighted avg)  0.3561\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.55it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:04:42,719 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:04:42,794 DEV : loss 0.3003895580768585 - f1-score (weighted avg)  0.3316\n",
            "2023-04-21 02:04:42,803 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:04:42,808 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:04:44,904 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:04:46,298 epoch 34 - iter 7/79 - loss 0.36700913 - time (sec): 1.39 - samples/sec: 5196.72 - lr: 0.006250\n",
            "2023-04-21 02:04:47,998 epoch 34 - iter 14/79 - loss 0.37567686 - time (sec): 3.09 - samples/sec: 5065.98 - lr: 0.006250\n",
            "2023-04-21 02:04:49,827 epoch 34 - iter 21/79 - loss 0.36794053 - time (sec): 4.92 - samples/sec: 5074.41 - lr: 0.006250\n",
            "2023-04-21 02:04:51,419 epoch 34 - iter 28/79 - loss 0.37572487 - time (sec): 6.51 - samples/sec: 5164.49 - lr: 0.006250\n",
            "2023-04-21 02:04:53,055 epoch 34 - iter 35/79 - loss 0.38640994 - time (sec): 8.15 - samples/sec: 5251.95 - lr: 0.006250\n",
            "2023-04-21 02:04:55,232 epoch 34 - iter 42/79 - loss 0.39771952 - time (sec): 10.32 - samples/sec: 4978.68 - lr: 0.006250\n",
            "2023-04-21 02:04:56,952 epoch 34 - iter 49/79 - loss 0.40311196 - time (sec): 12.04 - samples/sec: 4941.05 - lr: 0.006250\n",
            "2023-04-21 02:04:59,215 epoch 34 - iter 56/79 - loss 0.40472433 - time (sec): 14.31 - samples/sec: 4784.99 - lr: 0.006250\n",
            "2023-04-21 02:05:01,123 epoch 34 - iter 63/79 - loss 0.40401744 - time (sec): 16.22 - samples/sec: 4737.38 - lr: 0.006250\n",
            "2023-04-21 02:05:02,955 epoch 34 - iter 70/79 - loss 0.40072869 - time (sec): 18.05 - samples/sec: 4747.79 - lr: 0.006250\n",
            "2023-04-21 02:05:04,319 epoch 34 - iter 77/79 - loss 0.40200521 - time (sec): 19.41 - samples/sec: 4745.21 - lr: 0.006250\n",
            "2023-04-21 02:05:04,856 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:05:04,857 EPOCH 34 done: loss 0.3993 - lr 0.006250\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.64it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:05:27,937 Evaluating as a multi-label problem: True\n",
            "2023-04-21 02:05:30,717 TRAIN : loss 0.27349668741226196 - f1-score (weighted avg)  0.3692\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.60it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:05:33,572 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:05:33,644 DEV : loss 0.291950523853302 - f1-score (weighted avg)  0.36\n",
            "2023-04-21 02:05:33,653 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:05:33,657 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:05:35,740 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:05:37,940 epoch 35 - iter 7/79 - loss 0.35230429 - time (sec): 2.20 - samples/sec: 4270.37 - lr: 0.006250\n",
            "2023-04-21 02:05:39,459 epoch 35 - iter 14/79 - loss 0.36154758 - time (sec): 3.72 - samples/sec: 4645.40 - lr: 0.006250\n",
            "2023-04-21 02:05:41,164 epoch 35 - iter 21/79 - loss 0.35803076 - time (sec): 5.42 - samples/sec: 4858.58 - lr: 0.006250\n",
            "2023-04-21 02:05:42,900 epoch 35 - iter 28/79 - loss 0.36272072 - time (sec): 7.16 - samples/sec: 4746.91 - lr: 0.006250\n",
            "2023-04-21 02:05:44,966 epoch 35 - iter 35/79 - loss 0.36395621 - time (sec): 9.22 - samples/sec: 4774.91 - lr: 0.006250\n",
            "2023-04-21 02:05:46,569 epoch 35 - iter 42/79 - loss 0.36627326 - time (sec): 10.83 - samples/sec: 4787.52 - lr: 0.006250\n",
            "2023-04-21 02:05:49,140 epoch 35 - iter 49/79 - loss 0.36586465 - time (sec): 13.40 - samples/sec: 4554.32 - lr: 0.006250\n",
            "2023-04-21 02:05:50,776 epoch 35 - iter 56/79 - loss 0.36718023 - time (sec): 15.03 - samples/sec: 4607.65 - lr: 0.006250\n",
            "2023-04-21 02:05:52,238 epoch 35 - iter 63/79 - loss 0.36856516 - time (sec): 16.50 - samples/sec: 4598.39 - lr: 0.006250\n",
            "2023-04-21 02:05:53,904 epoch 35 - iter 70/79 - loss 0.36721688 - time (sec): 18.16 - samples/sec: 4590.63 - lr: 0.006250\n",
            "2023-04-21 02:05:55,637 epoch 35 - iter 77/79 - loss 0.37626730 - time (sec): 19.89 - samples/sec: 4628.80 - lr: 0.006250\n",
            "2023-04-21 02:05:56,139 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:05:56,141 EPOCH 35 done: loss 0.3765 - lr 0.006250\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.64it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:06:17,959 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:06:20,744 TRAIN : loss 0.28911060094833374 - f1-score (weighted avg)  0.3611\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.58it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:06:23,615 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:06:23,683 DEV : loss 0.30994680523872375 - f1-score (weighted avg)  0.3725\n",
            "2023-04-21 02:06:23,693 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:06:23,697 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:06:25,220 epoch 36 - iter 7/79 - loss 0.44402137 - time (sec): 1.52 - samples/sec: 5941.01 - lr: 0.006250\n",
            "2023-04-21 02:06:26,542 epoch 36 - iter 14/79 - loss 0.41859720 - time (sec): 2.84 - samples/sec: 5747.09 - lr: 0.006250\n",
            "2023-04-21 02:06:28,185 epoch 36 - iter 21/79 - loss 0.39908062 - time (sec): 4.49 - samples/sec: 5548.47 - lr: 0.006250\n",
            "2023-04-21 02:06:29,846 epoch 36 - iter 28/79 - loss 0.38270123 - time (sec): 6.15 - samples/sec: 5345.65 - lr: 0.006250\n",
            "2023-04-21 02:06:31,883 epoch 36 - iter 35/79 - loss 0.37159876 - time (sec): 8.18 - samples/sec: 4997.00 - lr: 0.006250\n",
            "2023-04-21 02:06:33,566 epoch 36 - iter 42/79 - loss 0.37324318 - time (sec): 9.87 - samples/sec: 4954.19 - lr: 0.006250\n",
            "2023-04-21 02:06:35,379 epoch 36 - iter 49/79 - loss 0.36754002 - time (sec): 11.68 - samples/sec: 4907.59 - lr: 0.006250\n",
            "2023-04-21 02:06:37,294 epoch 36 - iter 56/79 - loss 0.36420995 - time (sec): 13.60 - samples/sec: 4863.81 - lr: 0.006250\n",
            "2023-04-21 02:06:39,603 epoch 36 - iter 63/79 - loss 0.36661371 - time (sec): 15.90 - samples/sec: 4762.49 - lr: 0.006250\n",
            "2023-04-21 02:06:41,377 epoch 36 - iter 70/79 - loss 0.36886969 - time (sec): 17.68 - samples/sec: 4782.92 - lr: 0.006250\n",
            "2023-04-21 02:06:43,097 epoch 36 - iter 77/79 - loss 0.37483844 - time (sec): 19.40 - samples/sec: 4794.88 - lr: 0.006250\n",
            "2023-04-21 02:06:43,491 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:06:43,492 EPOCH 36 done: loss 0.3758 - lr 0.006250\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.48it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:07:06,320 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:07:09,098 TRAIN : loss 0.3068969249725342 - f1-score (weighted avg)  0.3431\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.57it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:07:11,979 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:07:12,052 DEV : loss 0.32240429520606995 - f1-score (weighted avg)  0.3387\n",
            "2023-04-21 02:07:12,064 Epoch    36: reducing learning rate of group 0 to 3.1250e-03.\n",
            "2023-04-21 02:07:12,066 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:07:12,071 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:07:14,653 epoch 37 - iter 7/79 - loss 0.33983581 - time (sec): 2.58 - samples/sec: 4005.01 - lr: 0.003125\n",
            "2023-04-21 02:07:16,328 epoch 37 - iter 14/79 - loss 0.38127071 - time (sec): 4.26 - samples/sec: 4140.34 - lr: 0.003125\n",
            "2023-04-21 02:07:18,165 epoch 37 - iter 21/79 - loss 0.38211842 - time (sec): 6.09 - samples/sec: 4156.11 - lr: 0.003125\n",
            "2023-04-21 02:07:19,832 epoch 37 - iter 28/79 - loss 0.38708522 - time (sec): 7.76 - samples/sec: 4331.25 - lr: 0.003125\n",
            "2023-04-21 02:07:21,518 epoch 37 - iter 35/79 - loss 0.38509154 - time (sec): 9.45 - samples/sec: 4456.81 - lr: 0.003125\n",
            "2023-04-21 02:07:23,062 epoch 37 - iter 42/79 - loss 0.39964958 - time (sec): 10.99 - samples/sec: 4603.89 - lr: 0.003125\n",
            "2023-04-21 02:07:24,806 epoch 37 - iter 49/79 - loss 0.39896836 - time (sec): 12.73 - samples/sec: 4640.37 - lr: 0.003125\n",
            "2023-04-21 02:07:26,571 epoch 37 - iter 56/79 - loss 0.39451635 - time (sec): 14.50 - samples/sec: 4610.06 - lr: 0.003125\n",
            "2023-04-21 02:07:28,506 epoch 37 - iter 63/79 - loss 0.39107779 - time (sec): 16.43 - samples/sec: 4614.97 - lr: 0.003125\n",
            "2023-04-21 02:07:30,166 epoch 37 - iter 70/79 - loss 0.38483387 - time (sec): 18.09 - samples/sec: 4705.50 - lr: 0.003125\n",
            "2023-04-21 02:07:31,573 epoch 37 - iter 77/79 - loss 0.37982077 - time (sec): 19.50 - samples/sec: 4748.88 - lr: 0.003125\n",
            "2023-04-21 02:07:31,905 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:07:31,907 EPOCH 37 done: loss 0.3776 - lr 0.003125\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.64it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:07:53,735 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:07:57,775 TRAIN : loss 0.2786063849925995 - f1-score (weighted avg)  0.3525\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.60it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:08:00,636 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:08:00,705 DEV : loss 0.2996746599674225 - f1-score (weighted avg)  0.335\n",
            "2023-04-21 02:08:00,715 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:08:00,719 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:08:03,005 epoch 38 - iter 7/79 - loss 0.29215182 - time (sec): 2.28 - samples/sec: 4217.37 - lr: 0.003125\n",
            "2023-04-21 02:08:04,503 epoch 38 - iter 14/79 - loss 0.35601054 - time (sec): 3.78 - samples/sec: 4569.32 - lr: 0.003125\n",
            "2023-04-21 02:08:06,297 epoch 38 - iter 21/79 - loss 0.35486360 - time (sec): 5.58 - samples/sec: 4630.07 - lr: 0.003125\n",
            "2023-04-21 02:08:08,446 epoch 38 - iter 28/79 - loss 0.34428114 - time (sec): 7.73 - samples/sec: 4622.53 - lr: 0.003125\n",
            "2023-04-21 02:08:10,123 epoch 38 - iter 35/79 - loss 0.34553870 - time (sec): 9.40 - samples/sec: 4795.55 - lr: 0.003125\n",
            "2023-04-21 02:08:11,864 epoch 38 - iter 42/79 - loss 0.35064337 - time (sec): 11.14 - samples/sec: 4816.78 - lr: 0.003125\n",
            "2023-04-21 02:08:13,169 epoch 38 - iter 49/79 - loss 0.34614444 - time (sec): 12.45 - samples/sec: 4861.82 - lr: 0.003125\n",
            "2023-04-21 02:08:14,957 epoch 38 - iter 56/79 - loss 0.34749093 - time (sec): 14.24 - samples/sec: 4881.15 - lr: 0.003125\n",
            "2023-04-21 02:08:16,816 epoch 38 - iter 63/79 - loss 0.34626014 - time (sec): 16.10 - samples/sec: 4846.14 - lr: 0.003125\n",
            "2023-04-21 02:08:18,262 epoch 38 - iter 70/79 - loss 0.34577035 - time (sec): 17.54 - samples/sec: 4823.14 - lr: 0.003125\n",
            "2023-04-21 02:08:20,010 epoch 38 - iter 77/79 - loss 0.34648674 - time (sec): 19.29 - samples/sec: 4812.90 - lr: 0.003125\n",
            "2023-04-21 02:08:20,453 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:08:20,454 EPOCH 38 done: loss 0.3455 - lr 0.003125\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.65it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:08:42,266 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:08:45,049 TRAIN : loss 0.2833389639854431 - f1-score (weighted avg)  0.3505\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.62it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:08:47,887 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:08:47,956 DEV : loss 0.3018530011177063 - f1-score (weighted avg)  0.3502\n",
            "2023-04-21 02:08:47,966 Epoch    38: reducing learning rate of group 0 to 1.5625e-03.\n",
            "2023-04-21 02:08:47,968 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:08:47,971 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:08:49,650 epoch 39 - iter 7/79 - loss 0.34289651 - time (sec): 1.68 - samples/sec: 5133.52 - lr: 0.001563\n",
            "2023-04-21 02:08:51,875 epoch 39 - iter 14/79 - loss 0.33378136 - time (sec): 3.90 - samples/sec: 4489.81 - lr: 0.001563\n",
            "2023-04-21 02:08:53,476 epoch 39 - iter 21/79 - loss 0.33982872 - time (sec): 5.50 - samples/sec: 4582.29 - lr: 0.001563\n",
            "2023-04-21 02:08:55,025 epoch 39 - iter 28/79 - loss 0.34950435 - time (sec): 7.05 - samples/sec: 4672.51 - lr: 0.001563\n",
            "2023-04-21 02:08:56,938 epoch 39 - iter 35/79 - loss 0.34796158 - time (sec): 8.96 - samples/sec: 4643.64 - lr: 0.001563\n",
            "2023-04-21 02:08:58,539 epoch 39 - iter 42/79 - loss 0.35334955 - time (sec): 10.57 - samples/sec: 4743.61 - lr: 0.001563\n",
            "2023-04-21 02:09:00,458 epoch 39 - iter 49/79 - loss 0.35959626 - time (sec): 12.48 - samples/sec: 4744.18 - lr: 0.001563\n",
            "2023-04-21 02:09:02,172 epoch 39 - iter 56/79 - loss 0.35034793 - time (sec): 14.20 - samples/sec: 4748.96 - lr: 0.001563\n",
            "2023-04-21 02:09:03,701 epoch 39 - iter 63/79 - loss 0.34612400 - time (sec): 15.73 - samples/sec: 4779.75 - lr: 0.001563\n",
            "2023-04-21 02:09:06,057 epoch 39 - iter 70/79 - loss 0.34127320 - time (sec): 18.08 - samples/sec: 4652.30 - lr: 0.001563\n",
            "2023-04-21 02:09:07,811 epoch 39 - iter 77/79 - loss 0.34113675 - time (sec): 19.84 - samples/sec: 4686.40 - lr: 0.001563\n",
            "2023-04-21 02:09:08,178 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:09:08,180 EPOCH 39 done: loss 0.3411 - lr 0.001563\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.49it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:09:30,976 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:09:33,786 TRAIN : loss 0.2742469608783722 - f1-score (weighted avg)  0.3583\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.56it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:09:36,672 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:09:36,747 DEV : loss 0.2937588691711426 - f1-score (weighted avg)  0.3462\n",
            "2023-04-21 02:09:36,757 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:09:36,761 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:09:38,925 epoch 40 - iter 7/79 - loss 0.30451652 - time (sec): 2.16 - samples/sec: 3931.65 - lr: 0.001563\n",
            "2023-04-21 02:09:40,966 epoch 40 - iter 14/79 - loss 0.30543714 - time (sec): 4.20 - samples/sec: 4068.89 - lr: 0.001563\n",
            "2023-04-21 02:09:42,950 epoch 40 - iter 21/79 - loss 0.32522896 - time (sec): 6.19 - samples/sec: 4179.86 - lr: 0.001563\n",
            "2023-04-21 02:09:44,879 epoch 40 - iter 28/79 - loss 0.31407870 - time (sec): 8.12 - samples/sec: 4325.58 - lr: 0.001563\n",
            "2023-04-21 02:09:46,296 epoch 40 - iter 35/79 - loss 0.32395986 - time (sec): 9.53 - samples/sec: 4499.80 - lr: 0.001563\n",
            "2023-04-21 02:09:47,885 epoch 40 - iter 42/79 - loss 0.33314061 - time (sec): 11.12 - samples/sec: 4547.54 - lr: 0.001563\n",
            "2023-04-21 02:09:49,547 epoch 40 - iter 49/79 - loss 0.33564189 - time (sec): 12.78 - samples/sec: 4657.50 - lr: 0.001563\n",
            "2023-04-21 02:09:51,023 epoch 40 - iter 56/79 - loss 0.33635987 - time (sec): 14.26 - samples/sec: 4709.72 - lr: 0.001563\n",
            "2023-04-21 02:09:52,542 epoch 40 - iter 63/79 - loss 0.33607193 - time (sec): 15.78 - samples/sec: 4765.87 - lr: 0.001563\n",
            "2023-04-21 02:09:54,270 epoch 40 - iter 70/79 - loss 0.33318574 - time (sec): 17.51 - samples/sec: 4792.32 - lr: 0.001563\n",
            "2023-04-21 02:09:55,908 epoch 40 - iter 77/79 - loss 0.33596327 - time (sec): 19.14 - samples/sec: 4838.40 - lr: 0.001563\n",
            "2023-04-21 02:09:56,456 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:09:56,458 EPOCH 40 done: loss 0.3346 - lr 0.001563\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.66it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:10:18,188 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:10:22,241 TRAIN : loss 0.2766505181789398 - f1-score (weighted avg)  0.3438\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.58it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:10:25,112 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:10:25,184 DEV : loss 0.2953406274318695 - f1-score (weighted avg)  0.3306\n",
            "2023-04-21 02:10:25,193 Epoch    40: reducing learning rate of group 0 to 7.8125e-04.\n",
            "2023-04-21 02:10:25,195 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:10:25,198 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:10:27,152 epoch 41 - iter 7/79 - loss 0.30680235 - time (sec): 1.95 - samples/sec: 4770.31 - lr: 0.000781\n",
            "2023-04-21 02:10:28,705 epoch 41 - iter 14/79 - loss 0.31378448 - time (sec): 3.51 - samples/sec: 4705.05 - lr: 0.000781\n",
            "2023-04-21 02:10:30,998 epoch 41 - iter 21/79 - loss 0.35647853 - time (sec): 5.80 - samples/sec: 4355.48 - lr: 0.000781\n",
            "2023-04-21 02:10:32,841 epoch 41 - iter 28/79 - loss 0.35348881 - time (sec): 7.64 - samples/sec: 4393.96 - lr: 0.000781\n",
            "2023-04-21 02:10:34,993 epoch 41 - iter 35/79 - loss 0.35569531 - time (sec): 9.79 - samples/sec: 4267.84 - lr: 0.000781\n",
            "2023-04-21 02:10:36,858 epoch 41 - iter 42/79 - loss 0.35337161 - time (sec): 11.66 - samples/sec: 4315.81 - lr: 0.000781\n",
            "2023-04-21 02:10:38,668 epoch 41 - iter 49/79 - loss 0.34321458 - time (sec): 13.47 - samples/sec: 4399.47 - lr: 0.000781\n",
            "2023-04-21 02:10:40,376 epoch 41 - iter 56/79 - loss 0.33844169 - time (sec): 15.18 - samples/sec: 4489.44 - lr: 0.000781\n",
            "2023-04-21 02:10:41,783 epoch 41 - iter 63/79 - loss 0.34184331 - time (sec): 16.58 - samples/sec: 4565.33 - lr: 0.000781\n",
            "2023-04-21 02:10:43,294 epoch 41 - iter 70/79 - loss 0.33684717 - time (sec): 18.09 - samples/sec: 4617.54 - lr: 0.000781\n",
            "2023-04-21 02:10:44,845 epoch 41 - iter 77/79 - loss 0.33500488 - time (sec): 19.64 - samples/sec: 4718.45 - lr: 0.000781\n",
            "2023-04-21 02:10:45,254 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:10:45,256 EPOCH 41 done: loss 0.3356 - lr 0.000781\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.65it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:11:07,047 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:11:09,801 TRAIN : loss 0.2685922384262085 - f1-score (weighted avg)  0.3629\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.63it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:11:12,636 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:11:12,704 DEV : loss 0.28802013397216797 - f1-score (weighted avg)  0.3522\n",
            "2023-04-21 02:11:12,713 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:11:12,717 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:11:14,784 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:11:16,622 epoch 42 - iter 7/79 - loss 0.30611757 - time (sec): 1.84 - samples/sec: 4588.51 - lr: 0.000781\n",
            "2023-04-21 02:11:18,127 epoch 42 - iter 14/79 - loss 0.33218639 - time (sec): 3.34 - samples/sec: 5102.13 - lr: 0.000781\n",
            "2023-04-21 02:11:19,706 epoch 42 - iter 21/79 - loss 0.33648170 - time (sec): 4.92 - samples/sec: 5124.80 - lr: 0.000781\n",
            "2023-04-21 02:11:21,531 epoch 42 - iter 28/79 - loss 0.33530879 - time (sec): 6.74 - samples/sec: 5040.81 - lr: 0.000781\n",
            "2023-04-21 02:11:23,455 epoch 42 - iter 35/79 - loss 0.33614360 - time (sec): 8.67 - samples/sec: 4803.39 - lr: 0.000781\n",
            "2023-04-21 02:11:24,969 epoch 42 - iter 42/79 - loss 0.32497384 - time (sec): 10.18 - samples/sec: 4865.22 - lr: 0.000781\n",
            "2023-04-21 02:11:26,793 epoch 42 - iter 49/79 - loss 0.32468364 - time (sec): 12.01 - samples/sec: 4846.31 - lr: 0.000781\n",
            "2023-04-21 02:11:28,235 epoch 42 - iter 56/79 - loss 0.32786086 - time (sec): 13.45 - samples/sec: 4885.70 - lr: 0.000781\n",
            "2023-04-21 02:11:30,008 epoch 42 - iter 63/79 - loss 0.32579837 - time (sec): 15.22 - samples/sec: 4920.54 - lr: 0.000781\n",
            "2023-04-21 02:11:31,740 epoch 42 - iter 70/79 - loss 0.32892615 - time (sec): 16.95 - samples/sec: 4938.25 - lr: 0.000781\n",
            "2023-04-21 02:11:33,478 epoch 42 - iter 77/79 - loss 0.32777562 - time (sec): 18.69 - samples/sec: 4913.83 - lr: 0.000781\n",
            "2023-04-21 02:11:34,403 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:11:34,404 EPOCH 42 done: loss 0.3280 - lr 0.000781\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:23<00:00,  3.42it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:11:57,618 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:12:00,385 TRAIN : loss 0.26532042026519775 - f1-score (weighted avg)  0.3683\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.58it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:12:03,254 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:12:03,323 DEV : loss 0.28656190633773804 - f1-score (weighted avg)  0.3539\n",
            "2023-04-21 02:12:03,333 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:12:03,337 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:12:05,428 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:12:06,834 epoch 43 - iter 7/79 - loss 0.34081339 - time (sec): 1.40 - samples/sec: 5934.99 - lr: 0.000781\n",
            "2023-04-21 02:12:08,691 epoch 43 - iter 14/79 - loss 0.32263383 - time (sec): 3.26 - samples/sec: 5270.68 - lr: 0.000781\n",
            "2023-04-21 02:12:10,191 epoch 43 - iter 21/79 - loss 0.35030633 - time (sec): 4.76 - samples/sec: 5308.46 - lr: 0.000781\n",
            "2023-04-21 02:12:11,876 epoch 43 - iter 28/79 - loss 0.33897584 - time (sec): 6.45 - samples/sec: 5182.21 - lr: 0.000781\n",
            "2023-04-21 02:12:13,693 epoch 43 - iter 35/79 - loss 0.33550026 - time (sec): 8.26 - samples/sec: 5153.83 - lr: 0.000781\n",
            "2023-04-21 02:12:15,365 epoch 43 - iter 42/79 - loss 0.33089047 - time (sec): 9.93 - samples/sec: 5070.49 - lr: 0.000781\n",
            "2023-04-21 02:12:17,152 epoch 43 - iter 49/79 - loss 0.32978074 - time (sec): 11.72 - samples/sec: 5022.29 - lr: 0.000781\n",
            "2023-04-21 02:12:19,212 epoch 43 - iter 56/79 - loss 0.33243046 - time (sec): 13.78 - samples/sec: 4941.88 - lr: 0.000781\n",
            "2023-04-21 02:12:20,738 epoch 43 - iter 63/79 - loss 0.33099059 - time (sec): 15.31 - samples/sec: 4974.84 - lr: 0.000781\n",
            "2023-04-21 02:12:22,493 epoch 43 - iter 70/79 - loss 0.33295057 - time (sec): 17.06 - samples/sec: 4944.88 - lr: 0.000781\n",
            "2023-04-21 02:12:24,397 epoch 43 - iter 77/79 - loss 0.32904672 - time (sec): 18.97 - samples/sec: 4890.74 - lr: 0.000781\n",
            "2023-04-21 02:12:24,881 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:12:24,883 EPOCH 43 done: loss 0.3284 - lr 0.000781\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.69it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:12:46,461 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:12:50,587 TRAIN : loss 0.2665311396121979 - f1-score (weighted avg)  0.3586\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.54it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:12:53,492 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:12:53,561 DEV : loss 0.28629618883132935 - f1-score (weighted avg)  0.345\n",
            "2023-04-21 02:12:53,572 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:12:53,576 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:12:55,677 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:12:58,099 epoch 44 - iter 7/79 - loss 0.31015971 - time (sec): 2.42 - samples/sec: 3823.28 - lr: 0.000781\n",
            "2023-04-21 02:12:59,677 epoch 44 - iter 14/79 - loss 0.30535381 - time (sec): 4.00 - samples/sec: 4447.82 - lr: 0.000781\n",
            "2023-04-21 02:13:01,262 epoch 44 - iter 21/79 - loss 0.30872523 - time (sec): 5.58 - samples/sec: 4571.43 - lr: 0.000781\n",
            "2023-04-21 02:13:02,815 epoch 44 - iter 28/79 - loss 0.30663056 - time (sec): 7.14 - samples/sec: 4677.06 - lr: 0.000781\n",
            "2023-04-21 02:13:04,671 epoch 44 - iter 35/79 - loss 0.31231948 - time (sec): 8.99 - samples/sec: 4671.40 - lr: 0.000781\n",
            "2023-04-21 02:13:06,369 epoch 44 - iter 42/79 - loss 0.31580246 - time (sec): 10.69 - samples/sec: 4784.78 - lr: 0.000781\n",
            "2023-04-21 02:13:08,767 epoch 44 - iter 49/79 - loss 0.31330978 - time (sec): 13.09 - samples/sec: 4619.76 - lr: 0.000781\n",
            "2023-04-21 02:13:10,463 epoch 44 - iter 56/79 - loss 0.31327966 - time (sec): 14.78 - samples/sec: 4614.49 - lr: 0.000781\n",
            "2023-04-21 02:13:12,175 epoch 44 - iter 63/79 - loss 0.31870305 - time (sec): 16.50 - samples/sec: 4658.50 - lr: 0.000781\n",
            "2023-04-21 02:13:13,690 epoch 44 - iter 70/79 - loss 0.31977901 - time (sec): 18.01 - samples/sec: 4681.16 - lr: 0.000781\n",
            "2023-04-21 02:13:15,176 epoch 44 - iter 77/79 - loss 0.32240983 - time (sec): 19.50 - samples/sec: 4734.47 - lr: 0.000781\n",
            "2023-04-21 02:13:15,670 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:13:15,671 EPOCH 44 done: loss 0.3210 - lr 0.000781\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.66it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:13:37,388 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:13:40,156 TRAIN : loss 0.2599697709083557 - f1-score (weighted avg)  0.3671\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.56it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:13:43,047 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:13:43,115 DEV : loss 0.27935054898262024 - f1-score (weighted avg)  0.352\n",
            "2023-04-21 02:13:43,124 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:13:43,128 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:13:45,234 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:13:46,933 epoch 45 - iter 7/79 - loss 0.29894404 - time (sec): 1.70 - samples/sec: 4875.32 - lr: 0.000781\n",
            "2023-04-21 02:13:48,704 epoch 45 - iter 14/79 - loss 0.30657576 - time (sec): 3.47 - samples/sec: 5052.71 - lr: 0.000781\n",
            "2023-04-21 02:13:50,290 epoch 45 - iter 21/79 - loss 0.31067768 - time (sec): 5.05 - samples/sec: 4886.80 - lr: 0.000781\n",
            "2023-04-21 02:13:51,971 epoch 45 - iter 28/79 - loss 0.30754346 - time (sec): 6.74 - samples/sec: 4802.59 - lr: 0.000781\n",
            "2023-04-21 02:13:53,471 epoch 45 - iter 35/79 - loss 0.31482018 - time (sec): 8.24 - samples/sec: 4817.70 - lr: 0.000781\n",
            "2023-04-21 02:13:55,415 epoch 45 - iter 42/79 - loss 0.31764290 - time (sec): 10.18 - samples/sec: 4621.71 - lr: 0.000781\n",
            "2023-04-21 02:13:57,089 epoch 45 - iter 49/79 - loss 0.31958868 - time (sec): 11.85 - samples/sec: 4691.49 - lr: 0.000781\n",
            "2023-04-21 02:13:58,996 epoch 45 - iter 56/79 - loss 0.32957677 - time (sec): 13.76 - samples/sec: 4727.62 - lr: 0.000781\n",
            "2023-04-21 02:14:00,598 epoch 45 - iter 63/79 - loss 0.32697197 - time (sec): 15.36 - samples/sec: 4824.02 - lr: 0.000781\n",
            "2023-04-21 02:14:02,708 epoch 45 - iter 70/79 - loss 0.32398237 - time (sec): 17.47 - samples/sec: 4771.64 - lr: 0.000781\n",
            "2023-04-21 02:14:04,997 epoch 45 - iter 77/79 - loss 0.32257343 - time (sec): 19.76 - samples/sec: 4664.67 - lr: 0.000781\n",
            "2023-04-21 02:14:05,472 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:14:05,473 EPOCH 45 done: loss 0.3239 - lr 0.000781\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.47it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:14:28,403 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:14:31,176 TRAIN : loss 0.2602604031562805 - f1-score (weighted avg)  0.3743\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.61it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:14:34,029 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:14:34,099 DEV : loss 0.27898821234703064 - f1-score (weighted avg)  0.3521\n",
            "2023-04-21 02:14:34,108 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:14:34,112 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:14:36,173 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:14:37,778 epoch 46 - iter 7/79 - loss 0.33683449 - time (sec): 1.59 - samples/sec: 4886.02 - lr: 0.000781\n",
            "2023-04-21 02:14:39,640 epoch 46 - iter 14/79 - loss 0.31677008 - time (sec): 3.45 - samples/sec: 4475.43 - lr: 0.000781\n",
            "2023-04-21 02:14:41,315 epoch 46 - iter 21/79 - loss 0.33058767 - time (sec): 5.12 - samples/sec: 4647.23 - lr: 0.000781\n",
            "2023-04-21 02:14:42,667 epoch 46 - iter 28/79 - loss 0.33733451 - time (sec): 6.48 - samples/sec: 4894.20 - lr: 0.000781\n",
            "2023-04-21 02:14:44,343 epoch 46 - iter 35/79 - loss 0.33472935 - time (sec): 8.15 - samples/sec: 4915.77 - lr: 0.000781\n",
            "2023-04-21 02:14:46,102 epoch 46 - iter 42/79 - loss 0.33101815 - time (sec): 9.91 - samples/sec: 4893.68 - lr: 0.000781\n",
            "2023-04-21 02:14:47,974 epoch 46 - iter 49/79 - loss 0.32788668 - time (sec): 11.78 - samples/sec: 4857.25 - lr: 0.000781\n",
            "2023-04-21 02:14:49,818 epoch 46 - iter 56/79 - loss 0.32123002 - time (sec): 13.63 - samples/sec: 4898.80 - lr: 0.000781\n",
            "2023-04-21 02:14:51,743 epoch 46 - iter 63/79 - loss 0.31988377 - time (sec): 15.55 - samples/sec: 4821.42 - lr: 0.000781\n",
            "2023-04-21 02:14:53,401 epoch 46 - iter 70/79 - loss 0.31981036 - time (sec): 17.21 - samples/sec: 4832.26 - lr: 0.000781\n",
            "2023-04-21 02:14:55,570 epoch 46 - iter 77/79 - loss 0.32097277 - time (sec): 19.38 - samples/sec: 4765.21 - lr: 0.000781\n",
            "2023-04-21 02:14:56,013 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:14:56,015 EPOCH 46 done: loss 0.3235 - lr 0.000781\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.69it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:15:17,559 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:15:21,690 TRAIN : loss 0.2622503638267517 - f1-score (weighted avg)  0.3633\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.52it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:15:24,614 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:15:24,683 DEV : loss 0.2819312810897827 - f1-score (weighted avg)  0.3534\n",
            "2023-04-21 02:15:24,692 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:15:24,696 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:15:26,545 epoch 47 - iter 7/79 - loss 0.32388497 - time (sec): 1.85 - samples/sec: 4587.28 - lr: 0.000781\n",
            "2023-04-21 02:15:28,155 epoch 47 - iter 14/79 - loss 0.32631379 - time (sec): 3.46 - samples/sec: 4637.11 - lr: 0.000781\n",
            "2023-04-21 02:15:29,964 epoch 47 - iter 21/79 - loss 0.32976181 - time (sec): 5.27 - samples/sec: 4705.72 - lr: 0.000781\n",
            "2023-04-21 02:15:32,007 epoch 47 - iter 28/79 - loss 0.32133190 - time (sec): 7.31 - samples/sec: 4628.74 - lr: 0.000781\n",
            "2023-04-21 02:15:33,540 epoch 47 - iter 35/79 - loss 0.31758374 - time (sec): 8.84 - samples/sec: 4732.66 - lr: 0.000781\n",
            "2023-04-21 02:15:35,322 epoch 47 - iter 42/79 - loss 0.31651020 - time (sec): 10.62 - samples/sec: 4686.85 - lr: 0.000781\n",
            "2023-04-21 02:15:37,064 epoch 47 - iter 49/79 - loss 0.31655827 - time (sec): 12.37 - samples/sec: 4700.49 - lr: 0.000781\n",
            "2023-04-21 02:15:38,683 epoch 47 - iter 56/79 - loss 0.31954873 - time (sec): 13.98 - samples/sec: 4786.12 - lr: 0.000781\n",
            "2023-04-21 02:15:40,679 epoch 47 - iter 63/79 - loss 0.31755041 - time (sec): 15.98 - samples/sec: 4735.54 - lr: 0.000781\n",
            "2023-04-21 02:15:42,302 epoch 47 - iter 70/79 - loss 0.32160185 - time (sec): 17.60 - samples/sec: 4746.48 - lr: 0.000781\n",
            "2023-04-21 02:15:44,430 epoch 47 - iter 77/79 - loss 0.32194993 - time (sec): 19.73 - samples/sec: 4694.62 - lr: 0.000781\n",
            "2023-04-21 02:15:44,891 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:15:44,893 EPOCH 47 done: loss 0.3220 - lr 0.000781\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.69it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:16:06,459 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:16:09,227 TRAIN : loss 0.2607969045639038 - f1-score (weighted avg)  0.359\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.62it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:16:12,072 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:16:12,141 DEV : loss 0.27991682291030884 - f1-score (weighted avg)  0.3472\n",
            "2023-04-21 02:16:12,150 Epoch    47: reducing learning rate of group 0 to 3.9063e-04.\n",
            "2023-04-21 02:16:12,152 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:16:12,156 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:16:13,675 epoch 48 - iter 7/79 - loss 0.30574300 - time (sec): 1.52 - samples/sec: 4949.59 - lr: 0.000391\n",
            "2023-04-21 02:16:15,389 epoch 48 - iter 14/79 - loss 0.31229376 - time (sec): 3.23 - samples/sec: 4818.27 - lr: 0.000391\n",
            "2023-04-21 02:16:17,267 epoch 48 - iter 21/79 - loss 0.31802616 - time (sec): 5.11 - samples/sec: 4844.67 - lr: 0.000391\n",
            "2023-04-21 02:16:19,445 epoch 48 - iter 28/79 - loss 0.32642620 - time (sec): 7.29 - samples/sec: 4621.15 - lr: 0.000391\n",
            "2023-04-21 02:16:21,206 epoch 48 - iter 35/79 - loss 0.32230092 - time (sec): 9.05 - samples/sec: 4666.35 - lr: 0.000391\n",
            "2023-04-21 02:16:22,843 epoch 48 - iter 42/79 - loss 0.32621725 - time (sec): 10.69 - samples/sec: 4789.71 - lr: 0.000391\n",
            "2023-04-21 02:16:24,358 epoch 48 - iter 49/79 - loss 0.32142634 - time (sec): 12.20 - samples/sec: 4851.97 - lr: 0.000391\n",
            "2023-04-21 02:16:25,957 epoch 48 - iter 56/79 - loss 0.32030252 - time (sec): 13.80 - samples/sec: 4907.94 - lr: 0.000391\n",
            "2023-04-21 02:16:27,491 epoch 48 - iter 63/79 - loss 0.32234050 - time (sec): 15.33 - samples/sec: 4923.16 - lr: 0.000391\n",
            "2023-04-21 02:16:29,178 epoch 48 - iter 70/79 - loss 0.32161984 - time (sec): 17.02 - samples/sec: 4906.88 - lr: 0.000391\n",
            "2023-04-21 02:16:31,665 epoch 48 - iter 77/79 - loss 0.31786962 - time (sec): 19.51 - samples/sec: 4745.57 - lr: 0.000391\n",
            "2023-04-21 02:16:32,125 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:16:32,127 EPOCH 48 done: loss 0.3181 - lr 0.000391\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.68it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:16:53,723 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:16:57,744 TRAIN : loss 0.2598779499530792 - f1-score (weighted avg)  0.3593\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.60it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:17:00,602 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:17:00,672 DEV : loss 0.27880963683128357 - f1-score (weighted avg)  0.3396\n",
            "2023-04-21 02:17:00,682 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:17:00,685 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:17:02,768 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:17:05,014 epoch 49 - iter 7/79 - loss 0.31941550 - time (sec): 2.24 - samples/sec: 4011.01 - lr: 0.000391\n",
            "2023-04-21 02:17:06,369 epoch 49 - iter 14/79 - loss 0.30837480 - time (sec): 3.60 - samples/sec: 4710.08 - lr: 0.000391\n",
            "2023-04-21 02:17:08,191 epoch 49 - iter 21/79 - loss 0.31806694 - time (sec): 5.42 - samples/sec: 4696.06 - lr: 0.000391\n",
            "2023-04-21 02:17:09,878 epoch 49 - iter 28/79 - loss 0.30588275 - time (sec): 7.11 - samples/sec: 4880.57 - lr: 0.000391\n",
            "2023-04-21 02:17:11,643 epoch 49 - iter 35/79 - loss 0.30995451 - time (sec): 8.87 - samples/sec: 4934.25 - lr: 0.000391\n",
            "2023-04-21 02:17:13,335 epoch 49 - iter 42/79 - loss 0.31183409 - time (sec): 10.56 - samples/sec: 4944.97 - lr: 0.000391\n",
            "2023-04-21 02:17:15,680 epoch 49 - iter 49/79 - loss 0.31356645 - time (sec): 12.91 - samples/sec: 4707.35 - lr: 0.000391\n",
            "2023-04-21 02:17:17,336 epoch 49 - iter 56/79 - loss 0.31092666 - time (sec): 14.57 - samples/sec: 4725.56 - lr: 0.000391\n",
            "2023-04-21 02:17:19,525 epoch 49 - iter 63/79 - loss 0.31960112 - time (sec): 16.75 - samples/sec: 4628.70 - lr: 0.000391\n",
            "2023-04-21 02:17:21,222 epoch 49 - iter 70/79 - loss 0.31871476 - time (sec): 18.45 - samples/sec: 4630.44 - lr: 0.000391\n",
            "2023-04-21 02:17:22,783 epoch 49 - iter 77/79 - loss 0.32063847 - time (sec): 20.01 - samples/sec: 4646.72 - lr: 0.000391\n",
            "2023-04-21 02:17:23,197 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:17:23,199 EPOCH 49 done: loss 0.3193 - lr 0.000391\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.66it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:17:44,921 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:17:47,691 TRAIN : loss 0.260163813829422 - f1-score (weighted avg)  0.359\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.61it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:17:50,541 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:17:50,609 DEV : loss 0.27869558334350586 - f1-score (weighted avg)  0.3438\n",
            "2023-04-21 02:17:50,618 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:17:50,622 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:17:52,703 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:17:54,812 epoch 50 - iter 7/79 - loss 0.30427150 - time (sec): 2.11 - samples/sec: 3909.72 - lr: 0.000391\n",
            "2023-04-21 02:17:56,309 epoch 50 - iter 14/79 - loss 0.30310270 - time (sec): 3.60 - samples/sec: 4307.91 - lr: 0.000391\n",
            "2023-04-21 02:17:58,031 epoch 50 - iter 21/79 - loss 0.30633412 - time (sec): 5.33 - samples/sec: 4581.63 - lr: 0.000391\n",
            "2023-04-21 02:17:59,479 epoch 50 - iter 28/79 - loss 0.30179891 - time (sec): 6.77 - samples/sec: 4701.31 - lr: 0.000391\n",
            "2023-04-21 02:18:01,119 epoch 50 - iter 35/79 - loss 0.30780554 - time (sec): 8.41 - samples/sec: 4815.74 - lr: 0.000391\n",
            "2023-04-21 02:18:02,993 epoch 50 - iter 42/79 - loss 0.30780745 - time (sec): 10.29 - samples/sec: 4815.54 - lr: 0.000391\n",
            "2023-04-21 02:18:04,629 epoch 50 - iter 49/79 - loss 0.30837770 - time (sec): 11.92 - samples/sec: 4787.17 - lr: 0.000391\n",
            "2023-04-21 02:18:07,069 epoch 50 - iter 56/79 - loss 0.30866634 - time (sec): 14.36 - samples/sec: 4649.06 - lr: 0.000391\n",
            "2023-04-21 02:18:08,746 epoch 50 - iter 63/79 - loss 0.31502708 - time (sec): 16.04 - samples/sec: 4708.45 - lr: 0.000391\n",
            "2023-04-21 02:18:10,710 epoch 50 - iter 70/79 - loss 0.31537637 - time (sec): 18.00 - samples/sec: 4647.42 - lr: 0.000391\n",
            "2023-04-21 02:18:12,529 epoch 50 - iter 77/79 - loss 0.31719892 - time (sec): 19.82 - samples/sec: 4650.63 - lr: 0.000391\n",
            "2023-04-21 02:18:13,019 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:18:13,021 EPOCH 50 done: loss 0.3162 - lr 0.000391\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.48it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:18:35,846 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:18:38,615 TRAIN : loss 0.25912484526634216 - f1-score (weighted avg)  0.3608\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.64it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:18:41,443 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:18:41,513 DEV : loss 0.2773917615413666 - f1-score (weighted avg)  0.3466\n",
            "2023-04-21 02:18:41,524 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:18:41,527 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:18:43,594 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:18:45,388 epoch 51 - iter 7/79 - loss 0.34499692 - time (sec): 1.79 - samples/sec: 5495.44 - lr: 0.000391\n",
            "2023-04-21 02:18:47,117 epoch 51 - iter 14/79 - loss 0.34569072 - time (sec): 3.52 - samples/sec: 5189.33 - lr: 0.000391\n",
            "2023-04-21 02:18:48,668 epoch 51 - iter 21/79 - loss 0.32944144 - time (sec): 5.07 - samples/sec: 5132.67 - lr: 0.000391\n",
            "2023-04-21 02:18:50,553 epoch 51 - iter 28/79 - loss 0.32806635 - time (sec): 6.96 - samples/sec: 4930.88 - lr: 0.000391\n",
            "2023-04-21 02:18:52,367 epoch 51 - iter 35/79 - loss 0.32864007 - time (sec): 8.77 - samples/sec: 4906.47 - lr: 0.000391\n",
            "2023-04-21 02:18:54,106 epoch 51 - iter 42/79 - loss 0.32813924 - time (sec): 10.51 - samples/sec: 4856.68 - lr: 0.000391\n",
            "2023-04-21 02:18:55,711 epoch 51 - iter 49/79 - loss 0.32189813 - time (sec): 12.11 - samples/sec: 4818.77 - lr: 0.000391\n",
            "2023-04-21 02:18:57,531 epoch 51 - iter 56/79 - loss 0.32263836 - time (sec): 13.93 - samples/sec: 4778.95 - lr: 0.000391\n",
            "2023-04-21 02:18:59,493 epoch 51 - iter 63/79 - loss 0.31940172 - time (sec): 15.90 - samples/sec: 4675.68 - lr: 0.000391\n",
            "2023-04-21 02:19:02,006 epoch 51 - iter 70/79 - loss 0.31333066 - time (sec): 18.41 - samples/sec: 4607.35 - lr: 0.000391\n",
            "2023-04-21 02:19:03,668 epoch 51 - iter 77/79 - loss 0.31381988 - time (sec): 20.07 - samples/sec: 4602.92 - lr: 0.000391\n",
            "2023-04-21 02:19:04,075 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:19:04,077 EPOCH 51 done: loss 0.3151 - lr 0.000391\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.68it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:19:25,704 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:19:29,735 TRAIN : loss 0.25914886593818665 - f1-score (weighted avg)  0.3728\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.64it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:19:32,569 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:19:32,640 DEV : loss 0.27701565623283386 - f1-score (weighted avg)  0.3621\n",
            "2023-04-21 02:19:32,650 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:19:32,654 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:19:34,739 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:19:36,351 epoch 52 - iter 7/79 - loss 0.32215200 - time (sec): 1.61 - samples/sec: 5093.28 - lr: 0.000391\n",
            "2023-04-21 02:19:38,265 epoch 52 - iter 14/79 - loss 0.32328690 - time (sec): 3.52 - samples/sec: 4951.57 - lr: 0.000391\n",
            "2023-04-21 02:19:40,167 epoch 52 - iter 21/79 - loss 0.31457048 - time (sec): 5.43 - samples/sec: 4886.09 - lr: 0.000391\n",
            "2023-04-21 02:19:42,409 epoch 52 - iter 28/79 - loss 0.30512796 - time (sec): 7.67 - samples/sec: 4590.25 - lr: 0.000391\n",
            "2023-04-21 02:19:43,877 epoch 52 - iter 35/79 - loss 0.30464432 - time (sec): 9.14 - samples/sec: 4713.29 - lr: 0.000391\n",
            "2023-04-21 02:19:45,722 epoch 52 - iter 42/79 - loss 0.31345076 - time (sec): 10.98 - samples/sec: 4704.77 - lr: 0.000391\n",
            "2023-04-21 02:19:47,256 epoch 52 - iter 49/79 - loss 0.31336428 - time (sec): 12.52 - samples/sec: 4772.85 - lr: 0.000391\n",
            "2023-04-21 02:19:49,150 epoch 52 - iter 56/79 - loss 0.31117066 - time (sec): 14.41 - samples/sec: 4716.00 - lr: 0.000391\n",
            "2023-04-21 02:19:50,930 epoch 52 - iter 63/79 - loss 0.31510798 - time (sec): 16.19 - samples/sec: 4737.66 - lr: 0.000391\n",
            "2023-04-21 02:19:52,520 epoch 52 - iter 70/79 - loss 0.31824707 - time (sec): 17.78 - samples/sec: 4744.24 - lr: 0.000391\n",
            "2023-04-21 02:19:54,567 epoch 52 - iter 77/79 - loss 0.31606799 - time (sec): 19.83 - samples/sec: 4677.96 - lr: 0.000391\n",
            "2023-04-21 02:19:55,024 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:19:55,025 EPOCH 52 done: loss 0.3153 - lr 0.000391\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.67it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:20:16,720 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:20:19,483 TRAIN : loss 0.2593397796154022 - f1-score (weighted avg)  0.3623\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.54it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:20:22,394 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:20:22,464 DEV : loss 0.2768321931362152 - f1-score (weighted avg)  0.3523\n",
            "2023-04-21 02:20:22,474 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:20:22,478 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:20:24,552 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:20:26,128 epoch 53 - iter 7/79 - loss 0.29554426 - time (sec): 1.57 - samples/sec: 5429.31 - lr: 0.000391\n",
            "2023-04-21 02:20:27,950 epoch 53 - iter 14/79 - loss 0.31591092 - time (sec): 3.40 - samples/sec: 5142.10 - lr: 0.000391\n",
            "2023-04-21 02:20:29,731 epoch 53 - iter 21/79 - loss 0.31758533 - time (sec): 5.18 - samples/sec: 4943.21 - lr: 0.000391\n",
            "2023-04-21 02:20:31,704 epoch 53 - iter 28/79 - loss 0.32018689 - time (sec): 7.15 - samples/sec: 4788.87 - lr: 0.000391\n",
            "2023-04-21 02:20:33,088 epoch 53 - iter 35/79 - loss 0.31999363 - time (sec): 8.53 - samples/sec: 4854.77 - lr: 0.000391\n",
            "2023-04-21 02:20:35,280 epoch 53 - iter 42/79 - loss 0.31689266 - time (sec): 10.73 - samples/sec: 4604.68 - lr: 0.000391\n",
            "2023-04-21 02:20:36,826 epoch 53 - iter 49/79 - loss 0.32251613 - time (sec): 12.27 - samples/sec: 4652.50 - lr: 0.000391\n",
            "2023-04-21 02:20:38,727 epoch 53 - iter 56/79 - loss 0.31704725 - time (sec): 14.17 - samples/sec: 4643.10 - lr: 0.000391\n",
            "2023-04-21 02:20:40,627 epoch 53 - iter 63/79 - loss 0.31782312 - time (sec): 16.07 - samples/sec: 4620.20 - lr: 0.000391\n",
            "2023-04-21 02:20:42,489 epoch 53 - iter 70/79 - loss 0.31886264 - time (sec): 17.93 - samples/sec: 4656.97 - lr: 0.000391\n",
            "2023-04-21 02:20:44,302 epoch 53 - iter 77/79 - loss 0.31954216 - time (sec): 19.75 - samples/sec: 4684.86 - lr: 0.000391\n",
            "2023-04-21 02:20:44,884 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:20:44,886 EPOCH 53 done: loss 0.3198 - lr 0.000391\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.49it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:21:07,655 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:21:10,432 TRAIN : loss 0.26275917887687683 - f1-score (weighted avg)  0.3624\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.59it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:21:13,298 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:21:13,367 DEV : loss 0.28018224239349365 - f1-score (weighted avg)  0.3509\n",
            "2023-04-21 02:21:13,377 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:21:13,382 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:21:15,306 epoch 54 - iter 7/79 - loss 0.32503583 - time (sec): 1.92 - samples/sec: 4755.04 - lr: 0.000391\n",
            "2023-04-21 02:21:17,089 epoch 54 - iter 14/79 - loss 0.31712427 - time (sec): 3.70 - samples/sec: 4698.80 - lr: 0.000391\n",
            "2023-04-21 02:21:18,892 epoch 54 - iter 21/79 - loss 0.32437588 - time (sec): 5.51 - samples/sec: 4744.21 - lr: 0.000391\n",
            "2023-04-21 02:21:20,423 epoch 54 - iter 28/79 - loss 0.32838558 - time (sec): 7.04 - samples/sec: 4794.08 - lr: 0.000391\n",
            "2023-04-21 02:21:22,100 epoch 54 - iter 35/79 - loss 0.32349424 - time (sec): 8.72 - samples/sec: 4786.62 - lr: 0.000391\n",
            "2023-04-21 02:21:23,938 epoch 54 - iter 42/79 - loss 0.32007812 - time (sec): 10.55 - samples/sec: 4805.42 - lr: 0.000391\n",
            "2023-04-21 02:21:25,509 epoch 54 - iter 49/79 - loss 0.32100744 - time (sec): 12.12 - samples/sec: 4786.25 - lr: 0.000391\n",
            "2023-04-21 02:21:27,089 epoch 54 - iter 56/79 - loss 0.32253589 - time (sec): 13.70 - samples/sec: 4831.32 - lr: 0.000391\n",
            "2023-04-21 02:21:28,978 epoch 54 - iter 63/79 - loss 0.31998812 - time (sec): 15.59 - samples/sec: 4814.84 - lr: 0.000391\n",
            "2023-04-21 02:21:31,107 epoch 54 - iter 70/79 - loss 0.31783374 - time (sec): 17.72 - samples/sec: 4749.38 - lr: 0.000391\n",
            "2023-04-21 02:21:32,609 epoch 54 - iter 77/79 - loss 0.31807563 - time (sec): 19.22 - samples/sec: 4805.44 - lr: 0.000391\n",
            "2023-04-21 02:21:33,116 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:21:33,117 EPOCH 54 done: loss 0.3186 - lr 0.000391\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.67it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:21:54,805 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:21:58,871 TRAIN : loss 0.2592394948005676 - f1-score (weighted avg)  0.3614\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.62it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:22:01,715 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:22:01,784 DEV : loss 0.2774837911128998 - f1-score (weighted avg)  0.3479\n",
            "2023-04-21 02:22:01,794 Epoch    54: reducing learning rate of group 0 to 1.9531e-04.\n",
            "2023-04-21 02:22:01,796 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:22:01,807 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:22:03,365 epoch 55 - iter 7/79 - loss 0.31978848 - time (sec): 1.56 - samples/sec: 5145.38 - lr: 0.000195\n",
            "2023-04-21 02:22:04,869 epoch 55 - iter 14/79 - loss 0.32465061 - time (sec): 3.06 - samples/sec: 5235.87 - lr: 0.000195\n",
            "2023-04-21 02:22:06,321 epoch 55 - iter 21/79 - loss 0.32321479 - time (sec): 4.51 - samples/sec: 5119.43 - lr: 0.000195\n",
            "2023-04-21 02:22:07,821 epoch 55 - iter 28/79 - loss 0.31581071 - time (sec): 6.01 - samples/sec: 5041.59 - lr: 0.000195\n",
            "2023-04-21 02:22:09,494 epoch 55 - iter 35/79 - loss 0.30951357 - time (sec): 7.69 - samples/sec: 5072.20 - lr: 0.000195\n",
            "2023-04-21 02:22:11,563 epoch 55 - iter 42/79 - loss 0.31183598 - time (sec): 9.75 - samples/sec: 4913.29 - lr: 0.000195\n",
            "2023-04-21 02:22:14,157 epoch 55 - iter 49/79 - loss 0.30460236 - time (sec): 12.35 - samples/sec: 4659.43 - lr: 0.000195\n",
            "2023-04-21 02:22:15,785 epoch 55 - iter 56/79 - loss 0.30101334 - time (sec): 13.98 - samples/sec: 4770.01 - lr: 0.000195\n",
            "2023-04-21 02:22:17,781 epoch 55 - iter 63/79 - loss 0.30959103 - time (sec): 15.97 - samples/sec: 4725.32 - lr: 0.000195\n",
            "2023-04-21 02:22:19,035 epoch 55 - iter 70/79 - loss 0.31206030 - time (sec): 17.23 - samples/sec: 4824.68 - lr: 0.000195\n",
            "2023-04-21 02:22:20,788 epoch 55 - iter 77/79 - loss 0.31289450 - time (sec): 18.98 - samples/sec: 4844.71 - lr: 0.000195\n",
            "2023-04-21 02:22:21,463 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:22:21,464 EPOCH 55 done: loss 0.3150 - lr 0.000195\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.70it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:22:42,949 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:22:45,726 TRAIN : loss 0.2578314244747162 - f1-score (weighted avg)  0.3674\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.56it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:22:48,618 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:22:48,688 DEV : loss 0.2761448323726654 - f1-score (weighted avg)  0.3507\n",
            "2023-04-21 02:22:48,698 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:22:48,703 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:22:50,805 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:22:52,364 epoch 56 - iter 7/79 - loss 0.33083771 - time (sec): 1.56 - samples/sec: 5367.02 - lr: 0.000195\n",
            "2023-04-21 02:22:53,867 epoch 56 - iter 14/79 - loss 0.31818353 - time (sec): 3.06 - samples/sec: 5425.40 - lr: 0.000195\n",
            "2023-04-21 02:22:55,412 epoch 56 - iter 21/79 - loss 0.31644906 - time (sec): 4.60 - samples/sec: 5390.50 - lr: 0.000195\n",
            "2023-04-21 02:22:57,500 epoch 56 - iter 28/79 - loss 0.30572596 - time (sec): 6.69 - samples/sec: 4945.61 - lr: 0.000195\n",
            "2023-04-21 02:22:59,559 epoch 56 - iter 35/79 - loss 0.30808709 - time (sec): 8.75 - samples/sec: 4659.91 - lr: 0.000195\n",
            "2023-04-21 02:23:01,236 epoch 56 - iter 42/79 - loss 0.30722353 - time (sec): 10.43 - samples/sec: 4725.99 - lr: 0.000195\n",
            "2023-04-21 02:23:02,969 epoch 56 - iter 49/79 - loss 0.30782755 - time (sec): 12.16 - samples/sec: 4696.30 - lr: 0.000195\n",
            "2023-04-21 02:23:04,605 epoch 56 - iter 56/79 - loss 0.30706183 - time (sec): 13.80 - samples/sec: 4753.02 - lr: 0.000195\n",
            "2023-04-21 02:23:06,629 epoch 56 - iter 63/79 - loss 0.30939546 - time (sec): 15.82 - samples/sec: 4749.90 - lr: 0.000195\n",
            "2023-04-21 02:23:08,287 epoch 56 - iter 70/79 - loss 0.31129150 - time (sec): 17.48 - samples/sec: 4757.35 - lr: 0.000195\n",
            "2023-04-21 02:23:10,309 epoch 56 - iter 77/79 - loss 0.31236256 - time (sec): 19.50 - samples/sec: 4745.42 - lr: 0.000195\n",
            "2023-04-21 02:23:10,746 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:23:10,748 EPOCH 56 done: loss 0.3151 - lr 0.000195\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.47it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:23:33,673 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:23:36,455 TRAIN : loss 0.2572723925113678 - f1-score (weighted avg)  0.3681\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.58it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:23:39,329 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:23:39,398 DEV : loss 0.27588093280792236 - f1-score (weighted avg)  0.3505\n",
            "2023-04-21 02:23:39,408 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:23:39,412 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:23:41,508 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:23:42,911 epoch 57 - iter 7/79 - loss 0.33211424 - time (sec): 1.40 - samples/sec: 5041.97 - lr: 0.000195\n",
            "2023-04-21 02:23:44,672 epoch 57 - iter 14/79 - loss 0.30523148 - time (sec): 3.16 - samples/sec: 4946.66 - lr: 0.000195\n",
            "2023-04-21 02:23:46,231 epoch 57 - iter 21/79 - loss 0.31254194 - time (sec): 4.72 - samples/sec: 4833.88 - lr: 0.000195\n",
            "2023-04-21 02:23:47,912 epoch 57 - iter 28/79 - loss 0.31035077 - time (sec): 6.40 - samples/sec: 4872.98 - lr: 0.000195\n",
            "2023-04-21 02:23:49,590 epoch 57 - iter 35/79 - loss 0.31220224 - time (sec): 8.08 - samples/sec: 4829.10 - lr: 0.000195\n",
            "2023-04-21 02:23:51,593 epoch 57 - iter 42/79 - loss 0.30702980 - time (sec): 10.08 - samples/sec: 4711.60 - lr: 0.000195\n",
            "2023-04-21 02:23:53,479 epoch 57 - iter 49/79 - loss 0.30986638 - time (sec): 11.97 - samples/sec: 4724.07 - lr: 0.000195\n",
            "2023-04-21 02:23:55,427 epoch 57 - iter 56/79 - loss 0.31449391 - time (sec): 13.92 - samples/sec: 4735.00 - lr: 0.000195\n",
            "2023-04-21 02:23:57,396 epoch 57 - iter 63/79 - loss 0.31117041 - time (sec): 15.89 - samples/sec: 4699.74 - lr: 0.000195\n",
            "2023-04-21 02:23:59,041 epoch 57 - iter 70/79 - loss 0.31129723 - time (sec): 17.53 - samples/sec: 4707.68 - lr: 0.000195\n",
            "2023-04-21 02:24:01,302 epoch 57 - iter 77/79 - loss 0.31010053 - time (sec): 19.79 - samples/sec: 4700.50 - lr: 0.000195\n",
            "2023-04-21 02:24:01,662 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:24:01,664 EPOCH 57 done: loss 0.3110 - lr 0.000195\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.67it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:24:23,319 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:24:27,427 TRAIN : loss 0.2570406198501587 - f1-score (weighted avg)  0.3656\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.62it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:24:30,268 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:24:30,339 DEV : loss 0.2756916582584381 - f1-score (weighted avg)  0.3476\n",
            "2023-04-21 02:24:30,348 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:24:30,352 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:24:32,422 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:24:34,484 epoch 58 - iter 7/79 - loss 0.25049958 - time (sec): 2.06 - samples/sec: 4175.54 - lr: 0.000195\n",
            "2023-04-21 02:24:36,180 epoch 58 - iter 14/79 - loss 0.30163328 - time (sec): 3.75 - samples/sec: 4421.04 - lr: 0.000195\n",
            "2023-04-21 02:24:37,854 epoch 58 - iter 21/79 - loss 0.30531726 - time (sec): 5.43 - samples/sec: 4581.58 - lr: 0.000195\n",
            "2023-04-21 02:24:39,448 epoch 58 - iter 28/79 - loss 0.30850213 - time (sec): 7.02 - samples/sec: 4662.08 - lr: 0.000195\n",
            "2023-04-21 02:24:40,898 epoch 58 - iter 35/79 - loss 0.31401583 - time (sec): 8.47 - samples/sec: 4753.67 - lr: 0.000195\n",
            "2023-04-21 02:24:42,636 epoch 58 - iter 42/79 - loss 0.31207729 - time (sec): 10.21 - samples/sec: 4699.61 - lr: 0.000195\n",
            "2023-04-21 02:24:44,663 epoch 58 - iter 49/79 - loss 0.31026747 - time (sec): 12.24 - samples/sec: 4674.99 - lr: 0.000195\n",
            "2023-04-21 02:24:46,709 epoch 58 - iter 56/79 - loss 0.30583030 - time (sec): 14.28 - samples/sec: 4607.36 - lr: 0.000195\n",
            "2023-04-21 02:24:48,510 epoch 58 - iter 63/79 - loss 0.30756223 - time (sec): 16.09 - samples/sec: 4594.55 - lr: 0.000195\n",
            "2023-04-21 02:24:50,583 epoch 58 - iter 70/79 - loss 0.30976307 - time (sec): 18.16 - samples/sec: 4596.26 - lr: 0.000195\n",
            "2023-04-21 02:24:52,381 epoch 58 - iter 77/79 - loss 0.31461935 - time (sec): 19.96 - samples/sec: 4591.29 - lr: 0.000195\n",
            "2023-04-21 02:24:53,043 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:24:53,044 EPOCH 58 done: loss 0.3144 - lr 0.000195\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.63it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:25:14,957 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:25:17,728 TRAIN : loss 0.2568184435367584 - f1-score (weighted avg)  0.3662\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.63it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:25:20,565 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:25:20,632 DEV : loss 0.2748049199581146 - f1-score (weighted avg)  0.3497\n",
            "2023-04-21 02:25:20,642 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:25:20,646 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:25:22,731 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:25:24,405 epoch 59 - iter 7/79 - loss 0.30438998 - time (sec): 1.67 - samples/sec: 4734.43 - lr: 0.000195\n",
            "2023-04-21 02:25:26,074 epoch 59 - iter 14/79 - loss 0.30850758 - time (sec): 3.34 - samples/sec: 4952.96 - lr: 0.000195\n",
            "2023-04-21 02:25:27,864 epoch 59 - iter 21/79 - loss 0.31658635 - time (sec): 5.13 - samples/sec: 5098.34 - lr: 0.000195\n",
            "2023-04-21 02:25:29,477 epoch 59 - iter 28/79 - loss 0.32429240 - time (sec): 6.74 - samples/sec: 5151.46 - lr: 0.000195\n",
            "2023-04-21 02:25:31,040 epoch 59 - iter 35/79 - loss 0.32477130 - time (sec): 8.31 - samples/sec: 5127.70 - lr: 0.000195\n",
            "2023-04-21 02:25:32,789 epoch 59 - iter 42/79 - loss 0.32514634 - time (sec): 10.06 - samples/sec: 5122.33 - lr: 0.000195\n",
            "2023-04-21 02:25:34,667 epoch 59 - iter 49/79 - loss 0.32141662 - time (sec): 11.93 - samples/sec: 4967.59 - lr: 0.000195\n",
            "2023-04-21 02:25:37,077 epoch 59 - iter 56/79 - loss 0.31479085 - time (sec): 14.34 - samples/sec: 4749.08 - lr: 0.000195\n",
            "2023-04-21 02:25:38,679 epoch 59 - iter 63/79 - loss 0.31370467 - time (sec): 15.95 - samples/sec: 4762.33 - lr: 0.000195\n",
            "2023-04-21 02:25:40,367 epoch 59 - iter 70/79 - loss 0.31090784 - time (sec): 17.63 - samples/sec: 4779.39 - lr: 0.000195\n",
            "2023-04-21 02:25:42,252 epoch 59 - iter 77/79 - loss 0.31085186 - time (sec): 19.52 - samples/sec: 4730.38 - lr: 0.000195\n",
            "2023-04-21 02:25:42,712 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:25:42,714 EPOCH 59 done: loss 0.3122 - lr 0.000195\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.67it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:26:05,575 Evaluating as a multi-label problem: True\n",
            "2023-04-21 02:26:08,349 TRAIN : loss 0.2554610073566437 - f1-score (weighted avg)  0.367\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.62it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:26:11,191 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:26:11,261 DEV : loss 0.2738053500652313 - f1-score (weighted avg)  0.3429\n",
            "2023-04-21 02:26:11,271 BAD EPOCHS (no improvement): 0\n",
            "2023-04-21 02:26:11,274 saving best model\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:26:13,349 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:26:15,245 epoch 60 - iter 7/79 - loss 0.31186860 - time (sec): 1.89 - samples/sec: 4602.59 - lr: 0.000195\n",
            "2023-04-21 02:26:16,737 epoch 60 - iter 14/79 - loss 0.30737466 - time (sec): 3.39 - samples/sec: 4749.53 - lr: 0.000195\n",
            "2023-04-21 02:26:18,243 epoch 60 - iter 21/79 - loss 0.30390287 - time (sec): 4.89 - samples/sec: 5012.21 - lr: 0.000195\n",
            "2023-04-21 02:26:19,919 epoch 60 - iter 28/79 - loss 0.30454839 - time (sec): 6.57 - samples/sec: 4830.97 - lr: 0.000195\n",
            "2023-04-21 02:26:21,631 epoch 60 - iter 35/79 - loss 0.30680404 - time (sec): 8.28 - samples/sec: 4867.39 - lr: 0.000195\n",
            "2023-04-21 02:26:23,059 epoch 60 - iter 42/79 - loss 0.31829634 - time (sec): 9.71 - samples/sec: 4918.31 - lr: 0.000195\n",
            "2023-04-21 02:26:24,886 epoch 60 - iter 49/79 - loss 0.31634041 - time (sec): 11.53 - samples/sec: 4943.02 - lr: 0.000195\n",
            "2023-04-21 02:26:26,725 epoch 60 - iter 56/79 - loss 0.31833791 - time (sec): 13.37 - samples/sec: 4911.45 - lr: 0.000195\n",
            "2023-04-21 02:26:28,377 epoch 60 - iter 63/79 - loss 0.31405343 - time (sec): 15.03 - samples/sec: 4934.11 - lr: 0.000195\n",
            "2023-04-21 02:26:30,288 epoch 60 - iter 70/79 - loss 0.31464166 - time (sec): 16.94 - samples/sec: 4926.88 - lr: 0.000195\n",
            "2023-04-21 02:26:31,759 epoch 60 - iter 77/79 - loss 0.31724149 - time (sec): 18.41 - samples/sec: 4951.74 - lr: 0.000195\n",
            "2023-04-21 02:26:32,952 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:26:32,956 EPOCH 60 done: loss 0.3163 - lr 0.000195\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.66it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:26:54,706 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:26:57,472 TRAIN : loss 0.2575482130050659 - f1-score (weighted avg)  0.3654\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.58it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:27:01,491 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:27:01,558 DEV : loss 0.2756063640117645 - f1-score (weighted avg)  0.3486\n",
            "2023-04-21 02:27:01,567 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:27:01,571 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:27:03,096 epoch 61 - iter 7/79 - loss 0.33884506 - time (sec): 1.52 - samples/sec: 5100.53 - lr: 0.000195\n",
            "2023-04-21 02:27:04,611 epoch 61 - iter 14/79 - loss 0.32402882 - time (sec): 3.04 - samples/sec: 4968.72 - lr: 0.000195\n",
            "2023-04-21 02:27:06,043 epoch 61 - iter 21/79 - loss 0.32541626 - time (sec): 4.47 - samples/sec: 4897.63 - lr: 0.000195\n",
            "2023-04-21 02:27:07,626 epoch 61 - iter 28/79 - loss 0.31567293 - time (sec): 6.05 - samples/sec: 4966.70 - lr: 0.000195\n",
            "2023-04-21 02:27:09,007 epoch 61 - iter 35/79 - loss 0.32347214 - time (sec): 7.43 - samples/sec: 5104.00 - lr: 0.000195\n",
            "2023-04-21 02:27:10,785 epoch 61 - iter 42/79 - loss 0.31759931 - time (sec): 9.21 - samples/sec: 5111.09 - lr: 0.000195\n",
            "2023-04-21 02:27:12,628 epoch 61 - iter 49/79 - loss 0.32195840 - time (sec): 11.06 - samples/sec: 5108.79 - lr: 0.000195\n",
            "2023-04-21 02:27:14,589 epoch 61 - iter 56/79 - loss 0.32022252 - time (sec): 13.02 - samples/sec: 4998.03 - lr: 0.000195\n",
            "2023-04-21 02:27:16,922 epoch 61 - iter 63/79 - loss 0.32031147 - time (sec): 15.35 - samples/sec: 4826.16 - lr: 0.000195\n",
            "2023-04-21 02:27:18,435 epoch 61 - iter 70/79 - loss 0.31770949 - time (sec): 16.86 - samples/sec: 4841.41 - lr: 0.000195\n",
            "2023-04-21 02:27:20,588 epoch 61 - iter 77/79 - loss 0.31479942 - time (sec): 19.02 - samples/sec: 4841.05 - lr: 0.000195\n",
            "2023-04-21 02:27:21,098 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:27:21,100 EPOCH 61 done: loss 0.3127 - lr 0.000195\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.69it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:27:42,654 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:27:45,430 TRAIN : loss 0.25725242495536804 - f1-score (weighted avg)  0.3646\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.60it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:27:48,288 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:27:48,358 DEV : loss 0.2756286859512329 - f1-score (weighted avg)  0.3453\n",
            "2023-04-21 02:27:48,371 Epoch    61: reducing learning rate of group 0 to 9.7656e-05.\n",
            "2023-04-21 02:27:48,374 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:27:48,378 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:27:50,028 epoch 62 - iter 7/79 - loss 0.35226186 - time (sec): 1.65 - samples/sec: 4992.45 - lr: 0.000098\n",
            "2023-04-21 02:27:51,346 epoch 62 - iter 14/79 - loss 0.34728004 - time (sec): 2.97 - samples/sec: 5031.64 - lr: 0.000098\n",
            "2023-04-21 02:27:53,056 epoch 62 - iter 21/79 - loss 0.32440835 - time (sec): 4.68 - samples/sec: 4938.08 - lr: 0.000098\n",
            "2023-04-21 02:27:54,452 epoch 62 - iter 28/79 - loss 0.32594244 - time (sec): 6.07 - samples/sec: 4949.38 - lr: 0.000098\n",
            "2023-04-21 02:27:56,382 epoch 62 - iter 35/79 - loss 0.31822224 - time (sec): 8.00 - samples/sec: 4860.09 - lr: 0.000098\n",
            "2023-04-21 02:27:58,330 epoch 62 - iter 42/79 - loss 0.31603227 - time (sec): 9.95 - samples/sec: 4754.06 - lr: 0.000098\n",
            "2023-04-21 02:27:59,942 epoch 62 - iter 49/79 - loss 0.31758964 - time (sec): 11.56 - samples/sec: 4835.91 - lr: 0.000098\n",
            "2023-04-21 02:28:02,071 epoch 62 - iter 56/79 - loss 0.31266529 - time (sec): 13.69 - samples/sec: 4774.38 - lr: 0.000098\n",
            "2023-04-21 02:28:04,233 epoch 62 - iter 63/79 - loss 0.31001547 - time (sec): 15.85 - samples/sec: 4722.58 - lr: 0.000098\n",
            "2023-04-21 02:28:05,911 epoch 62 - iter 70/79 - loss 0.30932473 - time (sec): 17.53 - samples/sec: 4776.56 - lr: 0.000098\n",
            "2023-04-21 02:28:07,724 epoch 62 - iter 77/79 - loss 0.31085687 - time (sec): 19.35 - samples/sec: 4784.10 - lr: 0.000098\n",
            "2023-04-21 02:28:08,090 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:28:08,092 EPOCH 62 done: loss 0.3112 - lr 0.000098\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.66it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:28:31,054 Evaluating as a multi-label problem: True\n",
            "2023-04-21 02:28:33,836 TRAIN : loss 0.25689011812210083 - f1-score (weighted avg)  0.364\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.59it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:28:36,701 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:28:36,771 DEV : loss 0.275298148393631 - f1-score (weighted avg)  0.344\n",
            "2023-04-21 02:28:36,781 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:28:36,786 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:28:38,173 epoch 63 - iter 7/79 - loss 0.32255111 - time (sec): 1.38 - samples/sec: 5356.34 - lr: 0.000098\n",
            "2023-04-21 02:28:39,727 epoch 63 - iter 14/79 - loss 0.31481238 - time (sec): 2.94 - samples/sec: 5305.18 - lr: 0.000098\n",
            "2023-04-21 02:28:41,457 epoch 63 - iter 21/79 - loss 0.30932821 - time (sec): 4.67 - samples/sec: 5138.66 - lr: 0.000098\n",
            "2023-04-21 02:28:43,500 epoch 63 - iter 28/79 - loss 0.30375783 - time (sec): 6.71 - samples/sec: 4780.79 - lr: 0.000098\n",
            "2023-04-21 02:28:45,685 epoch 63 - iter 35/79 - loss 0.30140205 - time (sec): 8.90 - samples/sec: 4676.92 - lr: 0.000098\n",
            "2023-04-21 02:28:47,299 epoch 63 - iter 42/79 - loss 0.30448238 - time (sec): 10.51 - samples/sec: 4688.90 - lr: 0.000098\n",
            "2023-04-21 02:28:48,838 epoch 63 - iter 49/79 - loss 0.30457042 - time (sec): 12.05 - samples/sec: 4770.67 - lr: 0.000098\n",
            "2023-04-21 02:28:50,235 epoch 63 - iter 56/79 - loss 0.30624215 - time (sec): 13.45 - samples/sec: 4864.56 - lr: 0.000098\n",
            "2023-04-21 02:28:52,577 epoch 63 - iter 63/79 - loss 0.30461548 - time (sec): 15.79 - samples/sec: 4735.26 - lr: 0.000098\n",
            "2023-04-21 02:28:54,240 epoch 63 - iter 70/79 - loss 0.30366969 - time (sec): 17.45 - samples/sec: 4745.17 - lr: 0.000098\n",
            "2023-04-21 02:28:56,225 epoch 63 - iter 77/79 - loss 0.30616927 - time (sec): 19.44 - samples/sec: 4757.38 - lr: 0.000098\n",
            "2023-04-21 02:28:56,684 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:28:56,685 EPOCH 63 done: loss 0.3088 - lr 0.000098\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.67it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:29:18,334 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:29:21,130 TRAIN : loss 0.256653755903244 - f1-score (weighted avg)  0.3634\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.63it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:29:23,965 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:29:24,035 DEV : loss 0.275145560503006 - f1-score (weighted avg)  0.3439\n",
            "2023-04-21 02:29:24,044 Epoch    63: reducing learning rate of group 0 to 4.8828e-05.\n",
            "2023-04-21 02:29:24,046 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:29:24,050 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:29:25,806 epoch 64 - iter 7/79 - loss 0.30489494 - time (sec): 1.75 - samples/sec: 5527.14 - lr: 0.000049\n",
            "2023-04-21 02:29:27,279 epoch 64 - iter 14/79 - loss 0.29840004 - time (sec): 3.23 - samples/sec: 5467.60 - lr: 0.000049\n",
            "2023-04-21 02:29:29,039 epoch 64 - iter 21/79 - loss 0.30180673 - time (sec): 4.99 - samples/sec: 5111.55 - lr: 0.000049\n",
            "2023-04-21 02:29:30,961 epoch 64 - iter 28/79 - loss 0.30601432 - time (sec): 6.91 - samples/sec: 4997.26 - lr: 0.000049\n",
            "2023-04-21 02:29:32,777 epoch 64 - iter 35/79 - loss 0.30215893 - time (sec): 8.73 - samples/sec: 4927.43 - lr: 0.000049\n",
            "2023-04-21 02:29:34,467 epoch 64 - iter 42/79 - loss 0.30651611 - time (sec): 10.42 - samples/sec: 4936.88 - lr: 0.000049\n",
            "2023-04-21 02:29:36,292 epoch 64 - iter 49/79 - loss 0.30643739 - time (sec): 12.24 - samples/sec: 4894.72 - lr: 0.000049\n",
            "2023-04-21 02:29:37,684 epoch 64 - iter 56/79 - loss 0.30455879 - time (sec): 13.63 - samples/sec: 4920.42 - lr: 0.000049\n",
            "2023-04-21 02:29:39,655 epoch 64 - iter 63/79 - loss 0.30914634 - time (sec): 15.60 - samples/sec: 4823.16 - lr: 0.000049\n",
            "2023-04-21 02:29:41,513 epoch 64 - iter 70/79 - loss 0.31064333 - time (sec): 17.46 - samples/sec: 4787.27 - lr: 0.000049\n",
            "2023-04-21 02:29:43,190 epoch 64 - iter 77/79 - loss 0.30916354 - time (sec): 19.14 - samples/sec: 4793.92 - lr: 0.000049\n",
            "2023-04-21 02:29:44,127 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:29:44,129 EPOCH 64 done: loss 0.3091 - lr 0.000049\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.50it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:30:06,864 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:30:09,629 TRAIN : loss 0.25604048371315 - f1-score (weighted avg)  0.367\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.65it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:30:12,453 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:30:12,520 DEV : loss 0.2746654748916626 - f1-score (weighted avg)  0.347\n",
            "2023-04-21 02:30:12,530 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:30:12,537 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:30:13,987 epoch 65 - iter 7/79 - loss 0.29825899 - time (sec): 1.45 - samples/sec: 5194.75 - lr: 0.000049\n",
            "2023-04-21 02:30:15,866 epoch 65 - iter 14/79 - loss 0.31311362 - time (sec): 3.33 - samples/sec: 5067.81 - lr: 0.000049\n",
            "2023-04-21 02:30:17,917 epoch 65 - iter 21/79 - loss 0.31760081 - time (sec): 5.38 - samples/sec: 4607.53 - lr: 0.000049\n",
            "2023-04-21 02:30:19,452 epoch 65 - iter 28/79 - loss 0.32084275 - time (sec): 6.91 - samples/sec: 4721.28 - lr: 0.000049\n",
            "2023-04-21 02:30:21,008 epoch 65 - iter 35/79 - loss 0.32154933 - time (sec): 8.47 - samples/sec: 4896.54 - lr: 0.000049\n",
            "2023-04-21 02:30:22,483 epoch 65 - iter 42/79 - loss 0.31473351 - time (sec): 9.94 - samples/sec: 4936.69 - lr: 0.000049\n",
            "2023-04-21 02:30:24,059 epoch 65 - iter 49/79 - loss 0.31605230 - time (sec): 11.52 - samples/sec: 4979.68 - lr: 0.000049\n",
            "2023-04-21 02:30:25,600 epoch 65 - iter 56/79 - loss 0.31706628 - time (sec): 13.06 - samples/sec: 5065.25 - lr: 0.000049\n",
            "2023-04-21 02:30:27,441 epoch 65 - iter 63/79 - loss 0.31558318 - time (sec): 14.90 - samples/sec: 5012.33 - lr: 0.000049\n",
            "2023-04-21 02:30:29,599 epoch 65 - iter 70/79 - loss 0.31241246 - time (sec): 17.06 - samples/sec: 4935.89 - lr: 0.000049\n",
            "2023-04-21 02:30:31,602 epoch 65 - iter 77/79 - loss 0.31550823 - time (sec): 19.06 - samples/sec: 4867.12 - lr: 0.000049\n",
            "2023-04-21 02:30:32,001 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:30:32,002 EPOCH 65 done: loss 0.3172 - lr 0.000049\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.68it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:30:53,591 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:30:57,628 TRAIN : loss 0.2570628225803375 - f1-score (weighted avg)  0.3651\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.64it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:31:00,460 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:31:00,529 DEV : loss 0.27540668845176697 - f1-score (weighted avg)  0.347\n",
            "2023-04-21 02:31:00,539 Epoch    65: reducing learning rate of group 0 to 2.4414e-05.\n",
            "2023-04-21 02:31:00,541 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:31:00,544 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:31:02,729 epoch 66 - iter 7/79 - loss 0.32466213 - time (sec): 2.18 - samples/sec: 4040.55 - lr: 0.000024\n",
            "2023-04-21 02:31:04,623 epoch 66 - iter 14/79 - loss 0.32296931 - time (sec): 4.08 - samples/sec: 4459.36 - lr: 0.000024\n",
            "2023-04-21 02:31:06,150 epoch 66 - iter 21/79 - loss 0.32778423 - time (sec): 5.60 - samples/sec: 4731.41 - lr: 0.000024\n",
            "2023-04-21 02:31:07,876 epoch 66 - iter 28/79 - loss 0.32651338 - time (sec): 7.33 - samples/sec: 4643.52 - lr: 0.000024\n",
            "2023-04-21 02:31:09,396 epoch 66 - iter 35/79 - loss 0.32139772 - time (sec): 8.85 - samples/sec: 4747.61 - lr: 0.000024\n",
            "2023-04-21 02:31:11,134 epoch 66 - iter 42/79 - loss 0.31685400 - time (sec): 10.59 - samples/sec: 4619.09 - lr: 0.000024\n",
            "2023-04-21 02:31:12,559 epoch 66 - iter 49/79 - loss 0.31755559 - time (sec): 12.01 - samples/sec: 4687.68 - lr: 0.000024\n",
            "2023-04-21 02:31:14,177 epoch 66 - iter 56/79 - loss 0.31852393 - time (sec): 13.63 - samples/sec: 4731.67 - lr: 0.000024\n",
            "2023-04-21 02:31:15,910 epoch 66 - iter 63/79 - loss 0.32363991 - time (sec): 15.36 - samples/sec: 4763.39 - lr: 0.000024\n",
            "2023-04-21 02:31:18,081 epoch 66 - iter 70/79 - loss 0.31590007 - time (sec): 17.53 - samples/sec: 4771.97 - lr: 0.000024\n",
            "2023-04-21 02:31:19,826 epoch 66 - iter 77/79 - loss 0.31276783 - time (sec): 19.28 - samples/sec: 4794.34 - lr: 0.000024\n",
            "2023-04-21 02:31:20,244 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:31:20,245 EPOCH 66 done: loss 0.3143 - lr 0.000024\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.64it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:31:42,090 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:31:44,872 TRAIN : loss 0.2570132613182068 - f1-score (weighted avg)  0.3636\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.65it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:31:47,691 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:31:47,759 DEV : loss 0.27551859617233276 - f1-score (weighted avg)  0.3429\n",
            "2023-04-21 02:31:47,769 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:31:47,773 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:31:49,754 epoch 67 - iter 7/79 - loss 0.29952608 - time (sec): 1.98 - samples/sec: 4626.64 - lr: 0.000024\n",
            "2023-04-21 02:31:51,522 epoch 67 - iter 14/79 - loss 0.32018719 - time (sec): 3.75 - samples/sec: 4849.78 - lr: 0.000024\n",
            "2023-04-21 02:31:53,129 epoch 67 - iter 21/79 - loss 0.32277564 - time (sec): 5.35 - samples/sec: 4993.51 - lr: 0.000024\n",
            "2023-04-21 02:31:54,631 epoch 67 - iter 28/79 - loss 0.32813448 - time (sec): 6.86 - samples/sec: 4927.07 - lr: 0.000024\n",
            "2023-04-21 02:31:56,622 epoch 67 - iter 35/79 - loss 0.32395353 - time (sec): 8.85 - samples/sec: 4824.29 - lr: 0.000024\n",
            "2023-04-21 02:31:58,164 epoch 67 - iter 42/79 - loss 0.31322885 - time (sec): 10.39 - samples/sec: 4928.18 - lr: 0.000024\n",
            "2023-04-21 02:31:59,928 epoch 67 - iter 49/79 - loss 0.31637059 - time (sec): 12.15 - samples/sec: 4858.65 - lr: 0.000024\n",
            "2023-04-21 02:32:02,071 epoch 67 - iter 56/79 - loss 0.31160657 - time (sec): 14.30 - samples/sec: 4714.60 - lr: 0.000024\n",
            "2023-04-21 02:32:03,611 epoch 67 - iter 63/79 - loss 0.31500931 - time (sec): 15.84 - samples/sec: 4818.62 - lr: 0.000024\n",
            "2023-04-21 02:32:05,845 epoch 67 - iter 70/79 - loss 0.31747211 - time (sec): 18.07 - samples/sec: 4708.03 - lr: 0.000024\n",
            "2023-04-21 02:32:07,259 epoch 67 - iter 77/79 - loss 0.31298878 - time (sec): 19.48 - samples/sec: 4760.36 - lr: 0.000024\n",
            "2023-04-21 02:32:07,711 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:32:07,713 EPOCH 67 done: loss 0.3141 - lr 0.000024\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.50it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:32:30,412 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:32:33,195 TRAIN : loss 0.25704139471054077 - f1-score (weighted avg)  0.364\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.61it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:32:36,048 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:32:36,129 DEV : loss 0.27537089586257935 - f1-score (weighted avg)  0.3433\n",
            "2023-04-21 02:32:36,139 Epoch    67: reducing learning rate of group 0 to 1.2207e-05.\n",
            "2023-04-21 02:32:36,141 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:32:36,145 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:32:37,805 epoch 68 - iter 7/79 - loss 0.29504575 - time (sec): 1.66 - samples/sec: 5133.17 - lr: 0.000012\n",
            "2023-04-21 02:32:39,431 epoch 68 - iter 14/79 - loss 0.30378669 - time (sec): 3.28 - samples/sec: 5042.87 - lr: 0.000012\n",
            "2023-04-21 02:32:41,194 epoch 68 - iter 21/79 - loss 0.29799946 - time (sec): 5.05 - samples/sec: 4940.35 - lr: 0.000012\n",
            "2023-04-21 02:32:42,926 epoch 68 - iter 28/79 - loss 0.30582782 - time (sec): 6.78 - samples/sec: 4890.12 - lr: 0.000012\n",
            "2023-04-21 02:32:45,304 epoch 68 - iter 35/79 - loss 0.30226064 - time (sec): 9.16 - samples/sec: 4534.57 - lr: 0.000012\n",
            "2023-04-21 02:32:47,053 epoch 68 - iter 42/79 - loss 0.30275920 - time (sec): 10.91 - samples/sec: 4573.07 - lr: 0.000012\n",
            "2023-04-21 02:32:48,571 epoch 68 - iter 49/79 - loss 0.30412538 - time (sec): 12.42 - samples/sec: 4706.46 - lr: 0.000012\n",
            "2023-04-21 02:32:50,322 epoch 68 - iter 56/79 - loss 0.30571137 - time (sec): 14.18 - samples/sec: 4748.35 - lr: 0.000012\n",
            "2023-04-21 02:32:52,104 epoch 68 - iter 63/79 - loss 0.30368254 - time (sec): 15.96 - samples/sec: 4756.67 - lr: 0.000012\n",
            "2023-04-21 02:32:53,894 epoch 68 - iter 70/79 - loss 0.30897138 - time (sec): 17.75 - samples/sec: 4769.25 - lr: 0.000012\n",
            "2023-04-21 02:32:55,747 epoch 68 - iter 77/79 - loss 0.31001614 - time (sec): 19.60 - samples/sec: 4738.33 - lr: 0.000012\n",
            "2023-04-21 02:32:56,136 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:32:56,138 EPOCH 68 done: loss 0.3120 - lr 0.000012\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.68it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:33:17,732 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:33:21,761 TRAIN : loss 0.25673776865005493 - f1-score (weighted avg)  0.3643\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.58it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:33:24,633 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:33:24,701 DEV : loss 0.2751116156578064 - f1-score (weighted avg)  0.3446\n",
            "2023-04-21 02:33:24,711 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:33:24,715 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:33:26,414 epoch 69 - iter 7/79 - loss 0.32235992 - time (sec): 1.70 - samples/sec: 4491.66 - lr: 0.000012\n",
            "2023-04-21 02:33:28,225 epoch 69 - iter 14/79 - loss 0.30655792 - time (sec): 3.51 - samples/sec: 4489.86 - lr: 0.000012\n",
            "2023-04-21 02:33:30,610 epoch 69 - iter 21/79 - loss 0.30886373 - time (sec): 5.89 - samples/sec: 4345.86 - lr: 0.000012\n",
            "2023-04-21 02:33:32,400 epoch 69 - iter 28/79 - loss 0.30774513 - time (sec): 7.68 - samples/sec: 4399.79 - lr: 0.000012\n",
            "2023-04-21 02:33:34,175 epoch 69 - iter 35/79 - loss 0.31291556 - time (sec): 9.46 - samples/sec: 4445.75 - lr: 0.000012\n",
            "2023-04-21 02:33:35,855 epoch 69 - iter 42/79 - loss 0.30706023 - time (sec): 11.14 - samples/sec: 4497.72 - lr: 0.000012\n",
            "2023-04-21 02:33:37,592 epoch 69 - iter 49/79 - loss 0.30706294 - time (sec): 12.88 - samples/sec: 4553.70 - lr: 0.000012\n",
            "2023-04-21 02:33:39,224 epoch 69 - iter 56/79 - loss 0.30966225 - time (sec): 14.51 - samples/sec: 4649.42 - lr: 0.000012\n",
            "2023-04-21 02:33:41,370 epoch 69 - iter 63/79 - loss 0.30756757 - time (sec): 16.65 - samples/sec: 4629.44 - lr: 0.000012\n",
            "2023-04-21 02:33:42,845 epoch 69 - iter 70/79 - loss 0.31108427 - time (sec): 18.13 - samples/sec: 4684.70 - lr: 0.000012\n",
            "2023-04-21 02:33:44,275 epoch 69 - iter 77/79 - loss 0.31039281 - time (sec): 19.56 - samples/sec: 4714.35 - lr: 0.000012\n",
            "2023-04-21 02:33:44,894 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:33:44,896 EPOCH 69 done: loss 0.3109 - lr 0.000012\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.69it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:34:06,469 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:34:09,236 TRAIN : loss 0.25646814703941345 - f1-score (weighted avg)  0.3657\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.53it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:34:12,152 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:34:12,221 DEV : loss 0.27484560012817383 - f1-score (weighted avg)  0.346\n",
            "2023-04-21 02:34:12,230 Epoch    69: reducing learning rate of group 0 to 6.1035e-06.\n",
            "2023-04-21 02:34:12,232 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:34:12,235 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:34:14,032 epoch 70 - iter 7/79 - loss 0.35398288 - time (sec): 1.80 - samples/sec: 4617.08 - lr: 0.000006\n",
            "2023-04-21 02:34:15,461 epoch 70 - iter 14/79 - loss 0.33863775 - time (sec): 3.22 - samples/sec: 4909.13 - lr: 0.000006\n",
            "2023-04-21 02:34:17,738 epoch 70 - iter 21/79 - loss 0.33805257 - time (sec): 5.50 - samples/sec: 4532.01 - lr: 0.000006\n",
            "2023-04-21 02:34:19,372 epoch 70 - iter 28/79 - loss 0.32966111 - time (sec): 7.14 - samples/sec: 4718.24 - lr: 0.000006\n",
            "2023-04-21 02:34:21,106 epoch 70 - iter 35/79 - loss 0.32385505 - time (sec): 8.87 - samples/sec: 4665.70 - lr: 0.000006\n",
            "2023-04-21 02:34:23,074 epoch 70 - iter 42/79 - loss 0.31824165 - time (sec): 10.84 - samples/sec: 4655.62 - lr: 0.000006\n",
            "2023-04-21 02:34:24,675 epoch 70 - iter 49/79 - loss 0.31875829 - time (sec): 12.44 - samples/sec: 4738.21 - lr: 0.000006\n",
            "2023-04-21 02:34:26,544 epoch 70 - iter 56/79 - loss 0.31798946 - time (sec): 14.31 - samples/sec: 4735.74 - lr: 0.000006\n",
            "2023-04-21 02:34:28,361 epoch 70 - iter 63/79 - loss 0.31631642 - time (sec): 16.12 - samples/sec: 4757.18 - lr: 0.000006\n",
            "2023-04-21 02:34:29,902 epoch 70 - iter 70/79 - loss 0.31426937 - time (sec): 17.67 - samples/sec: 4793.13 - lr: 0.000006\n",
            "2023-04-21 02:34:31,491 epoch 70 - iter 77/79 - loss 0.31407827 - time (sec): 19.25 - samples/sec: 4795.04 - lr: 0.000006\n",
            "2023-04-21 02:34:31,892 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:34:31,894 EPOCH 70 done: loss 0.3145 - lr 0.000006\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.48it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:34:54,707 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:34:57,497 TRAIN : loss 0.2564685344696045 - f1-score (weighted avg)  0.3646\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.64it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:35:00,327 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:35:00,398 DEV : loss 0.27484130859375 - f1-score (weighted avg)  0.3454\n",
            "2023-04-21 02:35:00,408 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:35:00,412 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:35:02,143 epoch 71 - iter 7/79 - loss 0.29499605 - time (sec): 1.73 - samples/sec: 4825.82 - lr: 0.000006\n",
            "2023-04-21 02:35:03,898 epoch 71 - iter 14/79 - loss 0.29458844 - time (sec): 3.48 - samples/sec: 4897.38 - lr: 0.000006\n",
            "2023-04-21 02:35:05,770 epoch 71 - iter 21/79 - loss 0.30021075 - time (sec): 5.36 - samples/sec: 4900.14 - lr: 0.000006\n",
            "2023-04-21 02:35:07,664 epoch 71 - iter 28/79 - loss 0.29923878 - time (sec): 7.25 - samples/sec: 4955.69 - lr: 0.000006\n",
            "2023-04-21 02:35:09,171 epoch 71 - iter 35/79 - loss 0.30070454 - time (sec): 8.76 - samples/sec: 4965.90 - lr: 0.000006\n",
            "2023-04-21 02:35:11,241 epoch 71 - iter 42/79 - loss 0.30091917 - time (sec): 10.83 - samples/sec: 4833.96 - lr: 0.000006\n",
            "2023-04-21 02:35:12,816 epoch 71 - iter 49/79 - loss 0.29754670 - time (sec): 12.40 - samples/sec: 4860.56 - lr: 0.000006\n",
            "2023-04-21 02:35:14,430 epoch 71 - iter 56/79 - loss 0.30338780 - time (sec): 14.02 - samples/sec: 4918.27 - lr: 0.000006\n",
            "2023-04-21 02:35:16,121 epoch 71 - iter 63/79 - loss 0.30325234 - time (sec): 15.71 - samples/sec: 4914.44 - lr: 0.000006\n",
            "2023-04-21 02:35:17,773 epoch 71 - iter 70/79 - loss 0.30755363 - time (sec): 17.36 - samples/sec: 4891.82 - lr: 0.000006\n",
            "2023-04-21 02:35:19,282 epoch 71 - iter 77/79 - loss 0.31095914 - time (sec): 18.87 - samples/sec: 4906.99 - lr: 0.000006\n",
            "2023-04-21 02:35:19,664 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:35:19,666 EPOCH 71 done: loss 0.3116 - lr 0.000006\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.66it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:35:41,385 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:35:45,450 TRAIN : loss 0.25625917315483093 - f1-score (weighted avg)  0.3672\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.60it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:35:48,317 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:35:48,386 DEV : loss 0.2746853530406952 - f1-score (weighted avg)  0.3478\n",
            "2023-04-21 02:35:48,397 Epoch    71: reducing learning rate of group 0 to 3.0518e-06.\n",
            "2023-04-21 02:35:48,399 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:35:48,403 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:35:50,233 epoch 72 - iter 7/79 - loss 0.30149982 - time (sec): 1.83 - samples/sec: 4538.34 - lr: 0.000003\n",
            "2023-04-21 02:35:52,013 epoch 72 - iter 14/79 - loss 0.32608478 - time (sec): 3.61 - samples/sec: 4834.92 - lr: 0.000003\n",
            "2023-04-21 02:35:53,653 epoch 72 - iter 21/79 - loss 0.32188844 - time (sec): 5.25 - samples/sec: 4929.83 - lr: 0.000003\n",
            "2023-04-21 02:35:55,828 epoch 72 - iter 28/79 - loss 0.31466617 - time (sec): 7.42 - samples/sec: 4616.03 - lr: 0.000003\n",
            "2023-04-21 02:35:57,661 epoch 72 - iter 35/79 - loss 0.30281280 - time (sec): 9.26 - samples/sec: 4722.01 - lr: 0.000003\n",
            "2023-04-21 02:35:59,238 epoch 72 - iter 42/79 - loss 0.30273369 - time (sec): 10.83 - samples/sec: 4746.53 - lr: 0.000003\n",
            "2023-04-21 02:36:00,635 epoch 72 - iter 49/79 - loss 0.30061759 - time (sec): 12.23 - samples/sec: 4806.54 - lr: 0.000003\n",
            "2023-04-21 02:36:02,273 epoch 72 - iter 56/79 - loss 0.30292975 - time (sec): 13.87 - samples/sec: 4820.68 - lr: 0.000003\n",
            "2023-04-21 02:36:03,992 epoch 72 - iter 63/79 - loss 0.30632715 - time (sec): 15.59 - samples/sec: 4849.28 - lr: 0.000003\n",
            "2023-04-21 02:36:05,943 epoch 72 - iter 70/79 - loss 0.30575492 - time (sec): 17.54 - samples/sec: 4810.23 - lr: 0.000003\n",
            "2023-04-21 02:36:07,900 epoch 72 - iter 77/79 - loss 0.30737433 - time (sec): 19.50 - samples/sec: 4751.68 - lr: 0.000003\n",
            "2023-04-21 02:36:08,340 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:36:08,341 EPOCH 72 done: loss 0.3059 - lr 0.000003\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.67it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:36:30,017 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:36:32,811 TRAIN : loss 0.25622081756591797 - f1-score (weighted avg)  0.3673\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.60it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:36:35,667 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:36:35,736 DEV : loss 0.2746419310569763 - f1-score (weighted avg)  0.3491\n",
            "2023-04-21 02:36:35,747 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:36:35,751 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:36:37,553 epoch 73 - iter 7/79 - loss 0.29141806 - time (sec): 1.80 - samples/sec: 4847.80 - lr: 0.000003\n",
            "2023-04-21 02:36:39,116 epoch 73 - iter 14/79 - loss 0.28145140 - time (sec): 3.36 - samples/sec: 4849.33 - lr: 0.000003\n",
            "2023-04-21 02:36:40,974 epoch 73 - iter 21/79 - loss 0.32400264 - time (sec): 5.22 - samples/sec: 4760.31 - lr: 0.000003\n",
            "2023-04-21 02:36:42,648 epoch 73 - iter 28/79 - loss 0.33052245 - time (sec): 6.90 - samples/sec: 4887.75 - lr: 0.000003\n",
            "2023-04-21 02:36:44,546 epoch 73 - iter 35/79 - loss 0.32009502 - time (sec): 8.79 - samples/sec: 4810.35 - lr: 0.000003\n",
            "2023-04-21 02:36:46,116 epoch 73 - iter 42/79 - loss 0.31818378 - time (sec): 10.36 - samples/sec: 4830.63 - lr: 0.000003\n",
            "2023-04-21 02:36:47,724 epoch 73 - iter 49/79 - loss 0.31470949 - time (sec): 11.97 - samples/sec: 4921.06 - lr: 0.000003\n",
            "2023-04-21 02:36:49,145 epoch 73 - iter 56/79 - loss 0.31631929 - time (sec): 13.39 - samples/sec: 4917.12 - lr: 0.000003\n",
            "2023-04-21 02:36:50,857 epoch 73 - iter 63/79 - loss 0.31489889 - time (sec): 15.10 - samples/sec: 4948.25 - lr: 0.000003\n",
            "2023-04-21 02:36:52,976 epoch 73 - iter 70/79 - loss 0.31234470 - time (sec): 17.22 - samples/sec: 4905.37 - lr: 0.000003\n",
            "2023-04-21 02:36:54,924 epoch 73 - iter 77/79 - loss 0.31249962 - time (sec): 19.17 - samples/sec: 4818.93 - lr: 0.000003\n",
            "2023-04-21 02:36:55,471 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:36:55,473 EPOCH 73 done: loss 0.3126 - lr 0.000003\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:22<00:00,  3.48it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:37:18,346 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:37:21,128 TRAIN : loss 0.2562093734741211 - f1-score (weighted avg)  0.3664\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.63it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:37:23,963 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:37:24,030 DEV : loss 0.274594247341156 - f1-score (weighted avg)  0.3498\n",
            "2023-04-21 02:37:24,040 Epoch    73: reducing learning rate of group 0 to 1.5259e-06.\n",
            "2023-04-21 02:37:24,042 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:37:24,046 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:37:25,672 epoch 74 - iter 7/79 - loss 0.29530183 - time (sec): 1.63 - samples/sec: 4908.23 - lr: 0.000002\n",
            "2023-04-21 02:37:27,182 epoch 74 - iter 14/79 - loss 0.30761815 - time (sec): 3.13 - samples/sec: 5165.83 - lr: 0.000002\n",
            "2023-04-21 02:37:28,719 epoch 74 - iter 21/79 - loss 0.30554541 - time (sec): 4.67 - samples/sec: 5172.43 - lr: 0.000002\n",
            "2023-04-21 02:37:30,257 epoch 74 - iter 28/79 - loss 0.31561604 - time (sec): 6.21 - samples/sec: 5183.92 - lr: 0.000002\n",
            "2023-04-21 02:37:31,851 epoch 74 - iter 35/79 - loss 0.32108586 - time (sec): 7.80 - samples/sec: 5288.01 - lr: 0.000002\n",
            "2023-04-21 02:37:33,697 epoch 74 - iter 42/79 - loss 0.32604354 - time (sec): 9.65 - samples/sec: 5176.21 - lr: 0.000002\n",
            "2023-04-21 02:37:35,509 epoch 74 - iter 49/79 - loss 0.32210320 - time (sec): 11.46 - samples/sec: 5086.93 - lr: 0.000002\n",
            "2023-04-21 02:37:37,414 epoch 74 - iter 56/79 - loss 0.32773280 - time (sec): 13.37 - samples/sec: 4999.05 - lr: 0.000002\n",
            "2023-04-21 02:37:39,829 epoch 74 - iter 63/79 - loss 0.32344024 - time (sec): 15.78 - samples/sec: 4794.60 - lr: 0.000002\n",
            "2023-04-21 02:37:41,467 epoch 74 - iter 70/79 - loss 0.32273030 - time (sec): 17.42 - samples/sec: 4800.35 - lr: 0.000002\n",
            "2023-04-21 02:37:43,229 epoch 74 - iter 77/79 - loss 0.31802718 - time (sec): 19.18 - samples/sec: 4818.10 - lr: 0.000002\n",
            "2023-04-21 02:37:43,746 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:37:43,747 EPOCH 74 done: loss 0.3175 - lr 0.000002\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.69it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:38:05,313 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:38:09,403 TRAIN : loss 0.25622376799583435 - f1-score (weighted avg)  0.3664\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.58it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:38:12,284 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:38:12,354 DEV : loss 0.2746012806892395 - f1-score (weighted avg)  0.3479\n",
            "2023-04-21 02:38:12,365 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:38:12,369 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:38:14,288 epoch 75 - iter 7/79 - loss 0.28623282 - time (sec): 1.92 - samples/sec: 4783.38 - lr: 0.000002\n",
            "2023-04-21 02:38:16,229 epoch 75 - iter 14/79 - loss 0.30893455 - time (sec): 3.86 - samples/sec: 4590.57 - lr: 0.000002\n",
            "2023-04-21 02:38:18,010 epoch 75 - iter 21/79 - loss 0.30650463 - time (sec): 5.64 - samples/sec: 4714.39 - lr: 0.000002\n",
            "2023-04-21 02:38:19,811 epoch 75 - iter 28/79 - loss 0.30339899 - time (sec): 7.44 - samples/sec: 4667.23 - lr: 0.000002\n",
            "2023-04-21 02:38:21,337 epoch 75 - iter 35/79 - loss 0.30494570 - time (sec): 8.97 - samples/sec: 4719.71 - lr: 0.000002\n",
            "2023-04-21 02:38:22,894 epoch 75 - iter 42/79 - loss 0.30704377 - time (sec): 10.52 - samples/sec: 4804.36 - lr: 0.000002\n",
            "2023-04-21 02:38:24,518 epoch 75 - iter 49/79 - loss 0.31267726 - time (sec): 12.15 - samples/sec: 4856.79 - lr: 0.000002\n",
            "2023-04-21 02:38:25,958 epoch 75 - iter 56/79 - loss 0.31215562 - time (sec): 13.59 - samples/sec: 4896.33 - lr: 0.000002\n",
            "2023-04-21 02:38:27,552 epoch 75 - iter 63/79 - loss 0.31465958 - time (sec): 15.18 - samples/sec: 4908.76 - lr: 0.000002\n",
            "2023-04-21 02:38:29,306 epoch 75 - iter 70/79 - loss 0.31293313 - time (sec): 16.94 - samples/sec: 4875.40 - lr: 0.000002\n",
            "2023-04-21 02:38:31,606 epoch 75 - iter 77/79 - loss 0.31200033 - time (sec): 19.24 - samples/sec: 4819.65 - lr: 0.000002\n",
            "2023-04-21 02:38:31,952 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:38:31,954 EPOCH 75 done: loss 0.3138 - lr 0.000002\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.66it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:38:53,669 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:38:56,439 TRAIN : loss 0.25623443722724915 - f1-score (weighted avg)  0.3664\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.63it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:38:59,280 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:38:59,348 DEV : loss 0.2746208608150482 - f1-score (weighted avg)  0.3479\n",
            "2023-04-21 02:38:59,357 Epoch    75: reducing learning rate of group 0 to 7.6294e-07.\n",
            "2023-04-21 02:38:59,359 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:38:59,362 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:39:01,305 epoch 76 - iter 7/79 - loss 0.32272548 - time (sec): 1.94 - samples/sec: 4256.14 - lr: 0.000001\n",
            "2023-04-21 02:39:03,366 epoch 76 - iter 14/79 - loss 0.31457501 - time (sec): 4.00 - samples/sec: 4192.25 - lr: 0.000001\n",
            "2023-04-21 02:39:05,386 epoch 76 - iter 21/79 - loss 0.31819074 - time (sec): 6.02 - samples/sec: 4190.93 - lr: 0.000001\n",
            "2023-04-21 02:39:06,944 epoch 76 - iter 28/79 - loss 0.31699309 - time (sec): 7.58 - samples/sec: 4290.76 - lr: 0.000001\n",
            "2023-04-21 02:39:08,546 epoch 76 - iter 35/79 - loss 0.30980551 - time (sec): 9.18 - samples/sec: 4469.44 - lr: 0.000001\n",
            "2023-04-21 02:39:10,258 epoch 76 - iter 42/79 - loss 0.30560988 - time (sec): 10.89 - samples/sec: 4538.27 - lr: 0.000001\n",
            "2023-04-21 02:39:12,052 epoch 76 - iter 49/79 - loss 0.30978259 - time (sec): 12.69 - samples/sec: 4604.58 - lr: 0.000001\n",
            "2023-04-21 02:39:13,899 epoch 76 - iter 56/79 - loss 0.31031907 - time (sec): 14.54 - samples/sec: 4589.69 - lr: 0.000001\n",
            "2023-04-21 02:39:15,746 epoch 76 - iter 63/79 - loss 0.31109738 - time (sec): 16.38 - samples/sec: 4567.93 - lr: 0.000001\n",
            "2023-04-21 02:39:17,608 epoch 76 - iter 70/79 - loss 0.31298574 - time (sec): 18.24 - samples/sec: 4576.27 - lr: 0.000001\n",
            "2023-04-21 02:39:19,063 epoch 76 - iter 77/79 - loss 0.31442972 - time (sec): 19.70 - samples/sec: 4681.07 - lr: 0.000001\n",
            "2023-04-21 02:39:19,657 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:39:19,659 EPOCH 76 done: loss 0.3144 - lr 0.000001\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.68it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:39:42,457 Evaluating as a multi-label problem: True\n",
            "2023-04-21 02:39:45,233 TRAIN : loss 0.2562463581562042 - f1-score (weighted avg)  0.3663\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.62it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:39:48,074 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:39:48,142 DEV : loss 0.2746236026287079 - f1-score (weighted avg)  0.3479\n",
            "2023-04-21 02:39:48,152 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:39:48,156 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:39:49,947 epoch 77 - iter 7/79 - loss 0.28542610 - time (sec): 1.79 - samples/sec: 4488.99 - lr: 0.000001\n",
            "2023-04-21 02:39:51,368 epoch 77 - iter 14/79 - loss 0.29451433 - time (sec): 3.21 - samples/sec: 4831.71 - lr: 0.000001\n",
            "2023-04-21 02:39:53,140 epoch 77 - iter 21/79 - loss 0.30051844 - time (sec): 4.98 - samples/sec: 4838.12 - lr: 0.000001\n",
            "2023-04-21 02:39:54,727 epoch 77 - iter 28/79 - loss 0.31125601 - time (sec): 6.57 - samples/sec: 4827.75 - lr: 0.000001\n",
            "2023-04-21 02:39:56,632 epoch 77 - iter 35/79 - loss 0.31136587 - time (sec): 8.47 - samples/sec: 4773.67 - lr: 0.000001\n",
            "2023-04-21 02:39:58,493 epoch 77 - iter 42/79 - loss 0.30311878 - time (sec): 10.34 - samples/sec: 4852.38 - lr: 0.000001\n",
            "2023-04-21 02:40:00,193 epoch 77 - iter 49/79 - loss 0.30484447 - time (sec): 12.04 - samples/sec: 4838.14 - lr: 0.000001\n",
            "2023-04-21 02:40:01,855 epoch 77 - iter 56/79 - loss 0.30684012 - time (sec): 13.70 - samples/sec: 4838.73 - lr: 0.000001\n",
            "2023-04-21 02:40:03,521 epoch 77 - iter 63/79 - loss 0.31400494 - time (sec): 15.36 - samples/sec: 4895.48 - lr: 0.000001\n",
            "2023-04-21 02:40:06,004 epoch 77 - iter 70/79 - loss 0.31207601 - time (sec): 17.85 - samples/sec: 4755.79 - lr: 0.000001\n",
            "2023-04-21 02:40:07,425 epoch 77 - iter 77/79 - loss 0.31424058 - time (sec): 19.27 - samples/sec: 4806.94 - lr: 0.000001\n",
            "2023-04-21 02:40:07,786 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:40:07,787 EPOCH 77 done: loss 0.3137 - lr 0.000001\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.66it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:40:29,509 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:40:32,277 TRAIN : loss 0.25624728202819824 - f1-score (weighted avg)  0.3663\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:03<00:00,  2.54it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:40:36,293 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:40:36,365 DEV : loss 0.27462145686149597 - f1-score (weighted avg)  0.3479\n",
            "2023-04-21 02:40:36,378 Epoch    77: reducing learning rate of group 0 to 3.8147e-07.\n",
            "2023-04-21 02:40:36,380 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:40:36,384 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:40:38,018 epoch 78 - iter 7/79 - loss 0.30530310 - time (sec): 1.63 - samples/sec: 5118.89 - lr: 0.000000\n",
            "2023-04-21 02:40:40,128 epoch 78 - iter 14/79 - loss 0.30123786 - time (sec): 3.74 - samples/sec: 4428.36 - lr: 0.000000\n",
            "2023-04-21 02:40:41,575 epoch 78 - iter 21/79 - loss 0.31714221 - time (sec): 5.19 - samples/sec: 4560.86 - lr: 0.000000\n",
            "2023-04-21 02:40:43,406 epoch 78 - iter 28/79 - loss 0.31280390 - time (sec): 7.02 - samples/sec: 4707.69 - lr: 0.000000\n",
            "2023-04-21 02:40:44,929 epoch 78 - iter 35/79 - loss 0.31676541 - time (sec): 8.54 - samples/sec: 4693.56 - lr: 0.000000\n",
            "2023-04-21 02:40:46,574 epoch 78 - iter 42/79 - loss 0.31617064 - time (sec): 10.19 - samples/sec: 4738.52 - lr: 0.000000\n",
            "2023-04-21 02:40:48,249 epoch 78 - iter 49/79 - loss 0.31235856 - time (sec): 11.86 - samples/sec: 4844.45 - lr: 0.000000\n",
            "2023-04-21 02:40:49,966 epoch 78 - iter 56/79 - loss 0.31661441 - time (sec): 13.58 - samples/sec: 4844.48 - lr: 0.000000\n",
            "2023-04-21 02:40:51,478 epoch 78 - iter 63/79 - loss 0.31270830 - time (sec): 15.09 - samples/sec: 4908.28 - lr: 0.000000\n",
            "2023-04-21 02:40:53,297 epoch 78 - iter 70/79 - loss 0.31114360 - time (sec): 16.91 - samples/sec: 4907.42 - lr: 0.000000\n",
            "2023-04-21 02:40:55,220 epoch 78 - iter 77/79 - loss 0.30913398 - time (sec): 18.83 - samples/sec: 4879.52 - lr: 0.000000\n",
            "2023-04-21 02:40:55,822 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:40:55,824 EPOCH 78 done: loss 0.3096 - lr 0.000000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.68it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:41:17,409 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:41:20,174 TRAIN : loss 0.2562538981437683 - f1-score (weighted avg)  0.3663\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.54it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:41:23,078 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:41:23,147 DEV : loss 0.2746220827102661 - f1-score (weighted avg)  0.3482\n",
            "2023-04-21 02:41:23,156 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:41:23,160 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:41:25,096 epoch 79 - iter 7/79 - loss 0.30979849 - time (sec): 1.93 - samples/sec: 4087.35 - lr: 0.000000\n",
            "2023-04-21 02:41:26,710 epoch 79 - iter 14/79 - loss 0.32468243 - time (sec): 3.55 - samples/sec: 4385.92 - lr: 0.000000\n",
            "2023-04-21 02:41:28,401 epoch 79 - iter 21/79 - loss 0.31841748 - time (sec): 5.24 - samples/sec: 4637.99 - lr: 0.000000\n",
            "2023-04-21 02:41:30,742 epoch 79 - iter 28/79 - loss 0.31082022 - time (sec): 7.58 - samples/sec: 4436.22 - lr: 0.000000\n",
            "2023-04-21 02:41:32,784 epoch 79 - iter 35/79 - loss 0.31696433 - time (sec): 9.62 - samples/sec: 4343.58 - lr: 0.000000\n",
            "2023-04-21 02:41:34,164 epoch 79 - iter 42/79 - loss 0.31757769 - time (sec): 11.00 - samples/sec: 4547.68 - lr: 0.000000\n",
            "2023-04-21 02:41:35,839 epoch 79 - iter 49/79 - loss 0.31714125 - time (sec): 12.68 - samples/sec: 4555.90 - lr: 0.000000\n",
            "2023-04-21 02:41:37,644 epoch 79 - iter 56/79 - loss 0.31914081 - time (sec): 14.48 - samples/sec: 4644.24 - lr: 0.000000\n",
            "2023-04-21 02:41:39,255 epoch 79 - iter 63/79 - loss 0.31596874 - time (sec): 16.09 - samples/sec: 4694.67 - lr: 0.000000\n",
            "2023-04-21 02:41:40,851 epoch 79 - iter 70/79 - loss 0.31647597 - time (sec): 17.69 - samples/sec: 4719.26 - lr: 0.000000\n",
            "2023-04-21 02:41:42,797 epoch 79 - iter 77/79 - loss 0.31395945 - time (sec): 19.63 - samples/sec: 4709.10 - lr: 0.000000\n",
            "2023-04-21 02:41:43,298 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:41:43,300 EPOCH 79 done: loss 0.3154 - lr 0.000000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.69it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:42:06,092 Evaluating as a multi-label problem: True\n",
            "2023-04-21 02:42:08,874 TRAIN : loss 0.25624752044677734 - f1-score (weighted avg)  0.3663\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.68it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:42:11,675 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:42:11,743 DEV : loss 0.27461931109428406 - f1-score (weighted avg)  0.3482\n",
            "2023-04-21 02:42:11,753 Epoch    79: reducing learning rate of group 0 to 1.9073e-07.\n",
            "2023-04-21 02:42:11,755 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:42:11,758 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:42:13,312 epoch 80 - iter 7/79 - loss 0.29802553 - time (sec): 1.55 - samples/sec: 4975.93 - lr: 0.000000\n",
            "2023-04-21 02:42:14,913 epoch 80 - iter 14/79 - loss 0.29320941 - time (sec): 3.15 - samples/sec: 5178.30 - lr: 0.000000\n",
            "2023-04-21 02:42:16,568 epoch 80 - iter 21/79 - loss 0.29390605 - time (sec): 4.81 - samples/sec: 5152.77 - lr: 0.000000\n",
            "2023-04-21 02:42:18,645 epoch 80 - iter 28/79 - loss 0.31151413 - time (sec): 6.88 - samples/sec: 4889.65 - lr: 0.000000\n",
            "2023-04-21 02:42:20,909 epoch 80 - iter 35/79 - loss 0.30379914 - time (sec): 9.15 - samples/sec: 4725.31 - lr: 0.000000\n",
            "2023-04-21 02:42:22,593 epoch 80 - iter 42/79 - loss 0.31088619 - time (sec): 10.83 - samples/sec: 4745.77 - lr: 0.000000\n",
            "2023-04-21 02:42:24,080 epoch 80 - iter 49/79 - loss 0.31277958 - time (sec): 12.32 - samples/sec: 4777.24 - lr: 0.000000\n",
            "2023-04-21 02:42:25,976 epoch 80 - iter 56/79 - loss 0.30957043 - time (sec): 14.22 - samples/sec: 4764.07 - lr: 0.000000\n",
            "2023-04-21 02:42:27,463 epoch 80 - iter 63/79 - loss 0.31019791 - time (sec): 15.70 - samples/sec: 4819.01 - lr: 0.000000\n",
            "2023-04-21 02:42:29,101 epoch 80 - iter 70/79 - loss 0.30998745 - time (sec): 17.34 - samples/sec: 4850.09 - lr: 0.000000\n",
            "2023-04-21 02:42:30,837 epoch 80 - iter 77/79 - loss 0.31266391 - time (sec): 19.08 - samples/sec: 4822.38 - lr: 0.000000\n",
            "2023-04-21 02:42:31,380 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:42:31,382 EPOCH 80 done: loss 0.3114 - lr 0.000000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.67it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:42:53,067 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:42:55,855 TRAIN : loss 0.25624772906303406 - f1-score (weighted avg)  0.3663\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:03<00:00,  2.54it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:42:59,870 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:42:59,939 DEV : loss 0.2746177911758423 - f1-score (weighted avg)  0.3482\n",
            "2023-04-21 02:42:59,949 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:42:59,954 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:43:01,561 epoch 81 - iter 7/79 - loss 0.38487926 - time (sec): 1.61 - samples/sec: 4999.96 - lr: 0.000000\n",
            "2023-04-21 02:43:03,130 epoch 81 - iter 14/79 - loss 0.35350619 - time (sec): 3.18 - samples/sec: 5083.71 - lr: 0.000000\n",
            "2023-04-21 02:43:04,816 epoch 81 - iter 21/79 - loss 0.34742510 - time (sec): 4.86 - samples/sec: 5023.37 - lr: 0.000000\n",
            "2023-04-21 02:43:06,437 epoch 81 - iter 28/79 - loss 0.33358222 - time (sec): 6.48 - samples/sec: 4903.79 - lr: 0.000000\n",
            "2023-04-21 02:43:08,497 epoch 81 - iter 35/79 - loss 0.32298966 - time (sec): 8.54 - samples/sec: 4736.60 - lr: 0.000000\n",
            "2023-04-21 02:43:10,050 epoch 81 - iter 42/79 - loss 0.31859140 - time (sec): 10.09 - samples/sec: 4904.88 - lr: 0.000000\n",
            "2023-04-21 02:43:12,190 epoch 81 - iter 49/79 - loss 0.31203681 - time (sec): 12.23 - samples/sec: 4772.76 - lr: 0.000000\n",
            "2023-04-21 02:43:13,986 epoch 81 - iter 56/79 - loss 0.31143965 - time (sec): 14.03 - samples/sec: 4760.64 - lr: 0.000000\n",
            "2023-04-21 02:43:15,493 epoch 81 - iter 63/79 - loss 0.31093844 - time (sec): 15.54 - samples/sec: 4831.55 - lr: 0.000000\n",
            "2023-04-21 02:43:17,808 epoch 81 - iter 70/79 - loss 0.32115110 - time (sec): 17.85 - samples/sec: 4687.28 - lr: 0.000000\n",
            "2023-04-21 02:43:19,463 epoch 81 - iter 77/79 - loss 0.31932231 - time (sec): 19.51 - samples/sec: 4731.94 - lr: 0.000000\n",
            "2023-04-21 02:43:19,862 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:43:19,864 EPOCH 81 done: loss 0.3178 - lr 0.000000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.68it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:43:41,457 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:43:44,239 TRAIN : loss 0.2562499940395355 - f1-score (weighted avg)  0.3663\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.63it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:43:47,075 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:43:47,143 DEV : loss 0.2746165692806244 - f1-score (weighted avg)  0.3482\n",
            "2023-04-21 02:43:47,153 Epoch    81: reducing learning rate of group 0 to 9.5367e-08.\n",
            "2023-04-21 02:43:47,155 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:43:47,158 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:43:48,855 epoch 82 - iter 7/79 - loss 0.31936353 - time (sec): 1.70 - samples/sec: 4990.52 - lr: 0.000000\n",
            "2023-04-21 02:43:50,478 epoch 82 - iter 14/79 - loss 0.35005382 - time (sec): 3.32 - samples/sec: 5106.32 - lr: 0.000000\n",
            "2023-04-21 02:43:52,642 epoch 82 - iter 21/79 - loss 0.33954642 - time (sec): 5.48 - samples/sec: 4697.06 - lr: 0.000000\n",
            "2023-04-21 02:43:54,127 epoch 82 - iter 28/79 - loss 0.33749506 - time (sec): 6.97 - samples/sec: 4788.63 - lr: 0.000000\n",
            "2023-04-21 02:43:56,434 epoch 82 - iter 35/79 - loss 0.33622239 - time (sec): 9.27 - samples/sec: 4656.02 - lr: 0.000000\n",
            "2023-04-21 02:43:57,973 epoch 82 - iter 42/79 - loss 0.33240163 - time (sec): 10.81 - samples/sec: 4734.35 - lr: 0.000000\n",
            "2023-04-21 02:43:59,673 epoch 82 - iter 49/79 - loss 0.33072770 - time (sec): 12.51 - samples/sec: 4690.74 - lr: 0.000000\n",
            "2023-04-21 02:44:01,340 epoch 82 - iter 56/79 - loss 0.32405938 - time (sec): 14.18 - samples/sec: 4734.74 - lr: 0.000000\n",
            "2023-04-21 02:44:03,329 epoch 82 - iter 63/79 - loss 0.31992369 - time (sec): 16.17 - samples/sec: 4639.07 - lr: 0.000000\n",
            "2023-04-21 02:44:04,950 epoch 82 - iter 70/79 - loss 0.31956574 - time (sec): 17.79 - samples/sec: 4692.85 - lr: 0.000000\n",
            "2023-04-21 02:44:06,802 epoch 82 - iter 77/79 - loss 0.31738382 - time (sec): 19.64 - samples/sec: 4738.35 - lr: 0.000000\n",
            "2023-04-21 02:44:07,204 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:44:07,206 EPOCH 82 done: loss 0.3159 - lr 0.000000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.67it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:44:30,117 Evaluating as a multi-label problem: True\n",
            "2023-04-21 02:44:32,885 TRAIN : loss 0.25624722242355347 - f1-score (weighted avg)  0.3664\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.65it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:44:35,707 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:44:35,776 DEV : loss 0.2746174931526184 - f1-score (weighted avg)  0.3482\n",
            "2023-04-21 02:44:35,785 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:44:35,789 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:44:37,488 epoch 83 - iter 7/79 - loss 0.33052089 - time (sec): 1.70 - samples/sec: 4669.82 - lr: 0.000000\n",
            "2023-04-21 02:44:39,289 epoch 83 - iter 14/79 - loss 0.33757182 - time (sec): 3.50 - samples/sec: 4764.01 - lr: 0.000000\n",
            "2023-04-21 02:44:40,981 epoch 83 - iter 21/79 - loss 0.33439144 - time (sec): 5.19 - samples/sec: 4778.39 - lr: 0.000000\n",
            "2023-04-21 02:44:42,426 epoch 83 - iter 28/79 - loss 0.32940702 - time (sec): 6.64 - samples/sec: 5016.50 - lr: 0.000000\n",
            "2023-04-21 02:44:44,082 epoch 83 - iter 35/79 - loss 0.32783982 - time (sec): 8.29 - samples/sec: 5058.55 - lr: 0.000000\n",
            "2023-04-21 02:44:45,896 epoch 83 - iter 42/79 - loss 0.32725820 - time (sec): 10.11 - samples/sec: 4911.21 - lr: 0.000000\n",
            "2023-04-21 02:44:47,870 epoch 83 - iter 49/79 - loss 0.32588024 - time (sec): 12.08 - samples/sec: 4799.73 - lr: 0.000000\n",
            "2023-04-21 02:44:49,409 epoch 83 - iter 56/79 - loss 0.32280332 - time (sec): 13.62 - samples/sec: 4882.96 - lr: 0.000000\n",
            "2023-04-21 02:44:50,908 epoch 83 - iter 63/79 - loss 0.32144962 - time (sec): 15.12 - samples/sec: 4891.59 - lr: 0.000000\n",
            "2023-04-21 02:44:53,426 epoch 83 - iter 70/79 - loss 0.31943243 - time (sec): 17.64 - samples/sec: 4759.44 - lr: 0.000000\n",
            "2023-04-21 02:44:55,392 epoch 83 - iter 77/79 - loss 0.31476333 - time (sec): 19.60 - samples/sec: 4703.44 - lr: 0.000000\n",
            "2023-04-21 02:44:55,795 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:44:55,797 EPOCH 83 done: loss 0.3147 - lr 0.000000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.68it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:45:17,382 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:45:20,171 TRAIN : loss 0.25624778866767883 - f1-score (weighted avg)  0.3663\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:03<00:00,  2.55it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:45:24,169 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:45:24,238 DEV : loss 0.27461808919906616 - f1-score (weighted avg)  0.3482\n",
            "2023-04-21 02:45:24,248 Epoch    83: reducing learning rate of group 0 to 4.7684e-08.\n",
            "2023-04-21 02:45:24,250 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:45:24,254 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:45:26,079 epoch 84 - iter 7/79 - loss 0.30358435 - time (sec): 1.82 - samples/sec: 3884.88 - lr: 0.000000\n",
            "2023-04-21 02:45:27,888 epoch 84 - iter 14/79 - loss 0.30278594 - time (sec): 3.63 - samples/sec: 4150.26 - lr: 0.000000\n",
            "2023-04-21 02:45:29,678 epoch 84 - iter 21/79 - loss 0.29862481 - time (sec): 5.42 - samples/sec: 4521.52 - lr: 0.000000\n",
            "2023-04-21 02:45:31,598 epoch 84 - iter 28/79 - loss 0.29734640 - time (sec): 7.34 - samples/sec: 4596.80 - lr: 0.000000\n",
            "2023-04-21 02:45:33,097 epoch 84 - iter 35/79 - loss 0.28867990 - time (sec): 8.84 - samples/sec: 4742.54 - lr: 0.000000\n",
            "2023-04-21 02:45:35,326 epoch 84 - iter 42/79 - loss 0.28760586 - time (sec): 11.07 - samples/sec: 4503.59 - lr: 0.000000\n",
            "2023-04-21 02:45:37,235 epoch 84 - iter 49/79 - loss 0.30227545 - time (sec): 12.98 - samples/sec: 4535.02 - lr: 0.000000\n",
            "2023-04-21 02:45:38,596 epoch 84 - iter 56/79 - loss 0.30995309 - time (sec): 14.34 - samples/sec: 4608.78 - lr: 0.000000\n",
            "2023-04-21 02:45:40,238 epoch 84 - iter 63/79 - loss 0.31030017 - time (sec): 15.98 - samples/sec: 4644.78 - lr: 0.000000\n",
            "2023-04-21 02:45:42,088 epoch 84 - iter 70/79 - loss 0.30829903 - time (sec): 17.83 - samples/sec: 4677.31 - lr: 0.000000\n",
            "2023-04-21 02:45:43,845 epoch 84 - iter 77/79 - loss 0.31438524 - time (sec): 19.59 - samples/sec: 4724.84 - lr: 0.000000\n",
            "2023-04-21 02:45:44,282 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:45:44,284 EPOCH 84 done: loss 0.3140 - lr 0.000000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.68it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:46:05,881 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:46:08,651 TRAIN : loss 0.25624674558639526 - f1-score (weighted avg)  0.3663\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.66it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:46:11,462 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:46:11,530 DEV : loss 0.27461931109428406 - f1-score (weighted avg)  0.3482\n",
            "2023-04-21 02:46:11,541 BAD EPOCHS (no improvement): 1\n",
            "2023-04-21 02:46:11,545 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:46:13,093 epoch 85 - iter 7/79 - loss 0.33520194 - time (sec): 1.55 - samples/sec: 5536.54 - lr: 0.000000\n",
            "2023-04-21 02:46:14,898 epoch 85 - iter 14/79 - loss 0.33105697 - time (sec): 3.35 - samples/sec: 5097.75 - lr: 0.000000\n",
            "2023-04-21 02:46:16,680 epoch 85 - iter 21/79 - loss 0.31566206 - time (sec): 5.13 - samples/sec: 4923.43 - lr: 0.000000\n",
            "2023-04-21 02:46:18,346 epoch 85 - iter 28/79 - loss 0.30879253 - time (sec): 6.80 - samples/sec: 4909.62 - lr: 0.000000\n",
            "2023-04-21 02:46:20,141 epoch 85 - iter 35/79 - loss 0.31414794 - time (sec): 8.59 - samples/sec: 4831.16 - lr: 0.000000\n",
            "2023-04-21 02:46:22,238 epoch 85 - iter 42/79 - loss 0.31425041 - time (sec): 10.69 - samples/sec: 4663.34 - lr: 0.000000\n",
            "2023-04-21 02:46:23,883 epoch 85 - iter 49/79 - loss 0.31711686 - time (sec): 12.34 - samples/sec: 4725.99 - lr: 0.000000\n",
            "2023-04-21 02:46:25,382 epoch 85 - iter 56/79 - loss 0.31341859 - time (sec): 13.84 - samples/sec: 4760.27 - lr: 0.000000\n",
            "2023-04-21 02:46:27,516 epoch 85 - iter 63/79 - loss 0.30979050 - time (sec): 15.97 - samples/sec: 4741.06 - lr: 0.000000\n",
            "2023-04-21 02:46:29,142 epoch 85 - iter 70/79 - loss 0.31259852 - time (sec): 17.60 - samples/sec: 4767.90 - lr: 0.000000\n",
            "2023-04-21 02:46:31,234 epoch 85 - iter 77/79 - loss 0.31162778 - time (sec): 19.69 - samples/sec: 4728.16 - lr: 0.000000\n",
            "2023-04-21 02:46:31,574 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:46:31,576 EPOCH 85 done: loss 0.3123 - lr 0.000000\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:21<00:00,  3.66it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:46:54,515 Evaluating as a multi-label problem: True\n",
            "2023-04-21 02:46:57,310 TRAIN : loss 0.25624755024909973 - f1-score (weighted avg)  0.3663\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:02<00:00,  3.57it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:47:00,195 Evaluating as a multi-label problem: False\n",
            "2023-04-21 02:47:00,263 DEV : loss 0.27462145686149597 - f1-score (weighted avg)  0.3482\n",
            "2023-04-21 02:47:00,273 Epoch    85: reducing learning rate of group 0 to 2.3842e-08.\n",
            "2023-04-21 02:47:00,275 BAD EPOCHS (no improvement): 2\n",
            "2023-04-21 02:47:00,278 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:47:00,280 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:47:00,281 learning rate too small - quitting training!\n",
            "2023-04-21 02:47:00,283 ----------------------------------------------------------------------------------------------------\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:47:02,503 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:47:04,444 SequenceTagger predicts: Dictionary with 24 tags: <unk>, O, I-ADR, E-ADR, B-ADR, S-ADR, S-Drug, B-Finding, E-Finding, I-Finding, B-Disease, E-Disease, E-Symptom, S-Disease, S-Symptom, E-Drug, B-Symptom, B-Drug, S-Finding, I-Symptom, I-Drug, I-Disease, <START>, <STOP>\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 10/10 [00:03<00:00,  2.67it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:47:08,576 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:47:08,925 0.5188\t0.3469\t0.4158\t0.8665\n",
            "2023-04-21 02:47:08,927 \n",
            "Results:\n",
            "- F-score (micro) 0.4158\n",
            "- F-score (macro) 0.1152\n",
            "- Accuracy 0.8665\n",
            "\n",
            "By class:\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "       E-ADR     0.5361    0.5045    0.5198       442\n",
            "       I-ADR     0.2555    0.1543    0.1924       525\n",
            "       B-ADR     0.5361    0.5335    0.5348       418\n",
            "      S-Drug     0.9400    0.7344    0.8246       192\n",
            "       S-ADR     0.7097    0.1392    0.2328       158\n",
            "   S-Finding     0.0000    0.0000    0.0000        28\n",
            "   E-Finding     0.0000    0.0000    0.0000        27\n",
            "      B-Drug     0.0000    0.0000    0.0000        27\n",
            "      E-Drug     0.0000    0.0000    0.0000        27\n",
            "      I-Drug     0.0000    0.0000    0.0000        25\n",
            "   B-Finding     0.0000    0.0000    0.0000        24\n",
            "   I-Symptom     0.0000    0.0000    0.0000        20\n",
            "   B-Symptom     0.0000    0.0000    0.0000        13\n",
            "   E-Symptom     0.0000    0.0000    0.0000        13\n",
            "   S-Disease     0.0000    0.0000    0.0000        12\n",
            "   E-Disease     0.0000    0.0000    0.0000         9\n",
            "   I-Finding     0.0000    0.0000    0.0000         9\n",
            "   S-Symptom     0.0000    0.0000    0.0000         9\n",
            "   B-Disease     0.0000    0.0000    0.0000         8\n",
            "   I-Disease     0.0000    0.0000    0.0000         3\n",
            "\n",
            "   micro avg     0.5188    0.3469    0.4158      1989\n",
            "   macro avg     0.1489    0.1033    0.1152      1989\n",
            "weighted avg     0.4463    0.3469    0.3768      1989\n",
            " samples avg     0.0583    0.0583    0.0583      1989\n",
            "\n",
            "2023-04-21 02:47:08,928 ----------------------------------------------------------------------------------------------------\n"
          ]
        }
      ],
      "source": [
        "hist = trainer.train('flair_ner_model'+suf,\n",
        "              embeddings_storage_mode='gpu',\n",
        "              gold_label_dictionary_for_eval=tag_dictionary,\n",
        "              mini_batch_size = 12,\n",
        "              eval_batch_size = 12,\n",
        "              patience=1,\n",
        "              anneal_factor = 0.5,\n",
        "              anneal_against_dev_loss=True,\n",
        "              main_evaluation_metric = (\"weighted avg\", \"f1-score\"),\n",
        "              metrics_for_tensorboard=[(\"weighted avg\", 'precision'), (\"weighted avg\", 'recall'), (\"macro avg\", 'f1-score'), (\"macro avg\", 'precision'), (\"macro avg\", 'recall')],\n",
        "              use_tensorboard=True,\n",
        "              tensorboard_comment='Flair NER',\n",
        "              tensorboard_log_dir='runs/tb_logs'+suf,\n",
        "              min_learning_rate=3e-8,\n",
        "              optimizer=AdamW,\n",
        "              learning_rate=1e-1,\n",
        "              weight_decay=0.001,\n",
        "              max_epochs=100,\n",
        "              monitor_train=True,\n",
        "              save_optimizer_state=True,)\n",
        "# hist"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "id": "ExiLRreqXsDJ"
      },
      "outputs": [],
      "source": [
        "# print(hist.keys())\n",
        "# hist"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 25,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "P5Uu5qkN0pia",
        "outputId": "2034761a-2bdf-4478-db4b-1499af64bd51"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "_v2\n",
            "2023-04-21 02:48:20,350 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:48:20,351 WARNING: No LOSS found for test split in this data.\n",
            "2023-04-21 02:48:20,352 Are you sure you want to plot LOSS and not another value?\n",
            "2023-04-21 02:48:20,353 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:48:20,375 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:48:20,376 WARNING: No F1 found for test split in this data.\n",
            "2023-04-21 02:48:20,377 Are you sure you want to plot F1 and not another value?\n",
            "2023-04-21 02:48:20,378 ----------------------------------------------------------------------------------------------------\n",
            "2023-04-21 02:48:21,430 Loss and F1 plots are saved in flair_ner_model_v2/training.png\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1500x1000 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "100%|██████████| 79/79 [00:32<00:00,  2.45it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:48:54,085 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n",
            "100%|██████████| 10/10 [00:02<00:00,  3.83it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:48:59,516 Evaluating as a multi-label problem: False\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n",
            "100%|██████████| 10/10 [00:04<00:00,  2.24it/s]"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2023-04-21 02:49:04,081 Evaluating as a multi-label problem: True\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "\n",
              "  <div id=\"df-928659cd-88bd-4ad8-a705-804c70c3defe\">\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>split</th>\n",
              "      <th>index</th>\n",
              "      <th>E-ADR</th>\n",
              "      <th>I-ADR</th>\n",
              "      <th>B-ADR</th>\n",
              "      <th>S-Drug</th>\n",
              "      <th>S-ADR</th>\n",
              "      <th>B-Finding</th>\n",
              "      <th>E-Finding</th>\n",
              "      <th>I-Finding</th>\n",
              "      <th>...</th>\n",
              "      <th>B-Drug</th>\n",
              "      <th>B-Symptom</th>\n",
              "      <th>S-Finding</th>\n",
              "      <th>I-Symptom</th>\n",
              "      <th>I-Drug</th>\n",
              "      <th>I-Disease</th>\n",
              "      <th>micro avg</th>\n",
              "      <th>macro avg</th>\n",
              "      <th>weighted avg</th>\n",
              "      <th>samples avg</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>train</td>\n",
              "      <td>precision</td>\n",
              "      <td>0.523926</td>\n",
              "      <td>0.314389</td>\n",
              "      <td>0.511125</td>\n",
              "      <td>0.953757</td>\n",
              "      <td>0.542763</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.518621</td>\n",
              "      <td>0.242298</td>\n",
              "      <td>0.453802</td>\n",
              "      <td>0.055397</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>train</td>\n",
              "      <td>recall</td>\n",
              "      <td>0.490302</td>\n",
              "      <td>0.156664</td>\n",
              "      <td>0.500746</td>\n",
              "      <td>0.769829</td>\n",
              "      <td>0.121056</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.009091</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.333927</td>\n",
              "      <td>0.102835</td>\n",
              "      <td>0.333927</td>\n",
              "      <td>0.055360</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>train</td>\n",
              "      <td>f1-score</td>\n",
              "      <td>0.506557</td>\n",
              "      <td>0.209121</td>\n",
              "      <td>0.505882</td>\n",
              "      <td>0.851979</td>\n",
              "      <td>0.197960</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.018018</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.406268</td>\n",
              "      <td>0.115369</td>\n",
              "      <td>0.367001</td>\n",
              "      <td>0.055373</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>train</td>\n",
              "      <td>support</td>\n",
              "      <td>3506.000000</td>\n",
              "      <td>4449.000000</td>\n",
              "      <td>3349.000000</td>\n",
              "      <td>1286.000000</td>\n",
              "      <td>1363.000000</td>\n",
              "      <td>200.0</td>\n",
              "      <td>200.0</td>\n",
              "      <td>149.0</td>\n",
              "      <td>...</td>\n",
              "      <td>110.000000</td>\n",
              "      <td>110.0</td>\n",
              "      <td>109.0</td>\n",
              "      <td>104.0</td>\n",
              "      <td>50.0</td>\n",
              "      <td>50.0</td>\n",
              "      <td>15722.000000</td>\n",
              "      <td>15722.000000</td>\n",
              "      <td>15722.000000</td>\n",
              "      <td>15722.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>dev</td>\n",
              "      <td>precision</td>\n",
              "      <td>0.505643</td>\n",
              "      <td>0.252280</td>\n",
              "      <td>0.485327</td>\n",
              "      <td>0.919463</td>\n",
              "      <td>0.500000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.483453</td>\n",
              "      <td>0.133136</td>\n",
              "      <td>0.400421</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>dev</td>\n",
              "      <td>recall</td>\n",
              "      <td>0.498886</td>\n",
              "      <td>0.160542</td>\n",
              "      <td>0.486425</td>\n",
              "      <td>0.835366</td>\n",
              "      <td>0.062201</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.325266</td>\n",
              "      <td>0.102171</td>\n",
              "      <td>0.325266</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>dev</td>\n",
              "      <td>f1-score</td>\n",
              "      <td>0.502242</td>\n",
              "      <td>0.196217</td>\n",
              "      <td>0.485876</td>\n",
              "      <td>0.875399</td>\n",
              "      <td>0.110638</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.388889</td>\n",
              "      <td>0.108519</td>\n",
              "      <td>0.342883</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>dev</td>\n",
              "      <td>support</td>\n",
              "      <td>449.000000</td>\n",
              "      <td>517.000000</td>\n",
              "      <td>442.000000</td>\n",
              "      <td>164.000000</td>\n",
              "      <td>209.000000</td>\n",
              "      <td>39.0</td>\n",
              "      <td>34.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>...</td>\n",
              "      <td>20.000000</td>\n",
              "      <td>12.0</td>\n",
              "      <td>25.0</td>\n",
              "      <td>23.0</td>\n",
              "      <td>7.0</td>\n",
              "      <td>13.0</td>\n",
              "      <td>2066.000000</td>\n",
              "      <td>2066.000000</td>\n",
              "      <td>2066.000000</td>\n",
              "      <td>NaN</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>test</td>\n",
              "      <td>precision</td>\n",
              "      <td>0.536058</td>\n",
              "      <td>0.255521</td>\n",
              "      <td>0.536058</td>\n",
              "      <td>0.940000</td>\n",
              "      <td>0.709677</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.518797</td>\n",
              "      <td>0.148866</td>\n",
              "      <td>0.446338</td>\n",
              "      <td>0.058307</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>test</td>\n",
              "      <td>recall</td>\n",
              "      <td>0.504525</td>\n",
              "      <td>0.154286</td>\n",
              "      <td>0.533493</td>\n",
              "      <td>0.734375</td>\n",
              "      <td>0.139241</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.346908</td>\n",
              "      <td>0.103296</td>\n",
              "      <td>0.346908</td>\n",
              "      <td>0.058264</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>test</td>\n",
              "      <td>f1-score</td>\n",
              "      <td>0.519814</td>\n",
              "      <td>0.192399</td>\n",
              "      <td>0.534772</td>\n",
              "      <td>0.824561</td>\n",
              "      <td>0.232804</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>...</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0.415788</td>\n",
              "      <td>0.115218</td>\n",
              "      <td>0.376773</td>\n",
              "      <td>0.058278</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>test</td>\n",
              "      <td>support</td>\n",
              "      <td>442.000000</td>\n",
              "      <td>525.000000</td>\n",
              "      <td>418.000000</td>\n",
              "      <td>192.000000</td>\n",
              "      <td>158.000000</td>\n",
              "      <td>24.0</td>\n",
              "      <td>27.0</td>\n",
              "      <td>9.0</td>\n",
              "      <td>...</td>\n",
              "      <td>27.000000</td>\n",
              "      <td>13.0</td>\n",
              "      <td>28.0</td>\n",
              "      <td>20.0</td>\n",
              "      <td>25.0</td>\n",
              "      <td>3.0</td>\n",
              "      <td>1989.000000</td>\n",
              "      <td>1989.000000</td>\n",
              "      <td>1989.000000</td>\n",
              "      <td>1989.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>12 rows × 26 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-928659cd-88bd-4ad8-a705-804c70c3defe')\"\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-928659cd-88bd-4ad8-a705-804c70c3defe 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-928659cd-88bd-4ad8-a705-804c70c3defe');\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",
              "  "
            ],
            "text/plain": [
              "    split      index        E-ADR        I-ADR        B-ADR       S-Drug  \\\n",
              "0   train  precision     0.523926     0.314389     0.511125     0.953757   \n",
              "1   train     recall     0.490302     0.156664     0.500746     0.769829   \n",
              "2   train   f1-score     0.506557     0.209121     0.505882     0.851979   \n",
              "3   train    support  3506.000000  4449.000000  3349.000000  1286.000000   \n",
              "4     dev  precision     0.505643     0.252280     0.485327     0.919463   \n",
              "5     dev     recall     0.498886     0.160542     0.486425     0.835366   \n",
              "6     dev   f1-score     0.502242     0.196217     0.485876     0.875399   \n",
              "7     dev    support   449.000000   517.000000   442.000000   164.000000   \n",
              "8    test  precision     0.536058     0.255521     0.536058     0.940000   \n",
              "9    test     recall     0.504525     0.154286     0.533493     0.734375   \n",
              "10   test   f1-score     0.519814     0.192399     0.534772     0.824561   \n",
              "11   test    support   442.000000   525.000000   418.000000   192.000000   \n",
              "\n",
              "          S-ADR  B-Finding  E-Finding  I-Finding  ...      B-Drug  B-Symptom  \\\n",
              "0      0.542763        0.0        0.0        0.0  ...    1.000000        0.0   \n",
              "1      0.121056        0.0        0.0        0.0  ...    0.009091        0.0   \n",
              "2      0.197960        0.0        0.0        0.0  ...    0.018018        0.0   \n",
              "3   1363.000000      200.0      200.0      149.0  ...  110.000000      110.0   \n",
              "4      0.500000        0.0        0.0        0.0  ...    0.000000        0.0   \n",
              "5      0.062201        0.0        0.0        0.0  ...    0.000000        0.0   \n",
              "6      0.110638        0.0        0.0        0.0  ...    0.000000        0.0   \n",
              "7    209.000000       39.0       34.0       20.0  ...   20.000000       12.0   \n",
              "8      0.709677        0.0        0.0        0.0  ...    0.000000        0.0   \n",
              "9      0.139241        0.0        0.0        0.0  ...    0.000000        0.0   \n",
              "10     0.232804        0.0        0.0        0.0  ...    0.000000        0.0   \n",
              "11   158.000000       24.0       27.0        9.0  ...   27.000000       13.0   \n",
              "\n",
              "    S-Finding  I-Symptom  I-Drug  I-Disease     micro avg     macro avg  \\\n",
              "0         0.0        0.0     0.0        0.0      0.518621      0.242298   \n",
              "1         0.0        0.0     0.0        0.0      0.333927      0.102835   \n",
              "2         0.0        0.0     0.0        0.0      0.406268      0.115369   \n",
              "3       109.0      104.0    50.0       50.0  15722.000000  15722.000000   \n",
              "4         0.0        0.0     0.0        0.0      0.483453      0.133136   \n",
              "5         0.0        0.0     0.0        0.0      0.325266      0.102171   \n",
              "6         0.0        0.0     0.0        0.0      0.388889      0.108519   \n",
              "7        25.0       23.0     7.0       13.0   2066.000000   2066.000000   \n",
              "8         0.0        0.0     0.0        0.0      0.518797      0.148866   \n",
              "9         0.0        0.0     0.0        0.0      0.346908      0.103296   \n",
              "10        0.0        0.0     0.0        0.0      0.415788      0.115218   \n",
              "11       28.0       20.0    25.0        3.0   1989.000000   1989.000000   \n",
              "\n",
              "    weighted avg   samples avg  \n",
              "0       0.453802      0.055397  \n",
              "1       0.333927      0.055360  \n",
              "2       0.367001      0.055373  \n",
              "3   15722.000000  15722.000000  \n",
              "4       0.400421           NaN  \n",
              "5       0.325266           NaN  \n",
              "6       0.342883           NaN  \n",
              "7    2066.000000           NaN  \n",
              "8       0.446338      0.058307  \n",
              "9       0.346908      0.058264  \n",
              "10      0.376773      0.058278  \n",
              "11   1989.000000   1989.000000  \n",
              "\n",
              "[12 rows x 26 columns]"
            ]
          },
          "execution_count": 25,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "print(suf)\n",
        "\n",
        "os.mkdir('flair_ner_model'+suf+'/classification_report')\n",
        "plotter.plot_training_curves('flair_ner_model'+suf+'/loss.tsv')\n",
        "\n",
        "# evaluate the tagger on the test set\n",
        "result = tagger.evaluate(corpus.train, \n",
        "                         embeddings_storage_mode='gpu',\n",
        "                         mini_batch_size=12,\n",
        "                         main_evaluation_metric = (\"weighted avg\", \"f1-score\"),\n",
        "                         gold_label_type='ner',\n",
        "                         gold_label_dictionary=tag_dictionary)\n",
        "open('flair_ner_model'+suf+'/classification_report/train.txt', 'w').write(str(result))\n",
        "res1 = pd.DataFrame.from_dict(result.classification_report).reset_index()\n",
        "res1.insert(loc=0, column='split', value='train')\n",
        "\n",
        "result = tagger.evaluate(corpus.dev, \n",
        "                         embeddings_storage_mode='gpu',\n",
        "                         mini_batch_size=12,\n",
        "                         main_evaluation_metric = (\"weighted avg\", \"f1-score\"),\n",
        "                         gold_label_type='ner',\n",
        "                         gold_label_dictionary=tag_dictionary)\n",
        "open('flair_ner_model'+suf+'/classification_report/dev.txt', 'w').write(str(result))\n",
        "res2 = pd.DataFrame.from_dict(result.classification_report).reset_index()\n",
        "res2.insert(loc=0, column='split', value='dev')\n",
        "\n",
        "result = tagger.evaluate(corpus.test, \n",
        "                         embeddings_storage_mode='gpu',\n",
        "                         mini_batch_size=12,\n",
        "                         main_evaluation_metric = (\"weighted avg\", \"f1-score\"),\n",
        "                         gold_label_type='ner',\n",
        "                         gold_label_dictionary=tag_dictionary)\n",
        "open('flair_ner_model'+suf+'/classification_report/test.txt', 'w').write(str(result))\n",
        "res3 = pd.DataFrame.from_dict(result.classification_report).reset_index()\n",
        "res3.insert(loc=0, column='split', value='test')\n",
        "\n",
        "res = pd.concat([res1,res2,res3],ignore_index=True)\n",
        "res.to_csv('flair_ner_model'+suf+'/classification_report.csv',index=False)\n",
        "res"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "VvIDCtYJkY-G"
      },
      "outputs": [],
      "source": [
        "# import time\n",
        "# start_time = time.time()\n",
        "# while time.time() - start_time < 1:\n",
        "#     time.sleep(1)\n",
        "# print('Session has timed out (2 minutes). Disconnecting...')\n",
        "# !kill -9 -1"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tGILL4RFtGil"
      },
      "source": [
        "Layering can be a good idea if the NER tags in the first layer have their own set of dependent values in the second layer. This can help capture more information about the entities being recognized and provide more context for downstream tasks. However, it also increases the complexity of the annotation and the training process, so it should be used judiciously. In general, it's a good idea to start with a simple tag scheme and gradually increase complexity as needed."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JoZ5D6aEtGin"
      },
      "source": [
        "    does it really matter if u use the E header\n",
        "\n",
        "It depends on the specific task and the requirements of the downstream model. In some cases, using the E tag can help disambiguate between overlapping entities, while in other cases it may not be necessary or even desired. Ultimately, it's up to the developer to determine the appropriate tagging scheme for their specific use case."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BGt2RhM6tGin"
      },
      "source": [
        "    would adding POS as an extra NER layer help model performance or not\n",
        "\n",
        "Adding POS as an extra NER layer may or may not help model performance, depending on the specific use case and data being analyzed.\n",
        "\n",
        "On one hand, incorporating POS tags could provide additional context to the model and help it better distinguish between named entities and other parts of speech. For example, knowing that a word is a noun or verb could help the model determine if it is likely to be a named entity or not.\n",
        "\n",
        "On the other hand, adding an additional layer could also increase model complexity and training time without necessarily improving performance. It may also require additional manual annotation efforts to create a labeled dataset with both NER and POS tags.\n",
        "\n",
        "Ultimately, whether to include POS as an additional layer will depend on the specific requirements and goals of the project, as well as the available data and resources."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6zpEn1bgtGin"
      },
      "source": [
        "    besides bioes ordering (which is up-to-down), it doesn't matter what order the tags in the same domain appear in left-to-right?\n",
        "\n",
        "Yes, that is correct. As long as the BIOES order is maintained vertically, the order of tags in the same domain from left-to-right should not impact training in flai"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "yrmL98VatGin"
      },
      "source": [
        "    different typenames must be treated as separate tasks?\n",
        "\n",
        "Yes, that's correct. Different typenames should be treated as separate tasks because they represent different types of annotations that you want to predict. For example, in a biomedical NER task, you might want to predict the entity types of both drugs and diseases, which would be represented by two different typenames in your annotation data. Predicting both types of entities would require different models and possibly different evaluation metrics."
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "gpuType": "A100",
      "provenance": []
    },
    "gpuClass": "standard",
    "kernelspec": {
      "display_name": "flair",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.9.16"
    },
    "vscode": {
      "interpreter": {
        "hash": "381c8abc3392c5f20473858f182d9ac4a53ef053fb1f276fcb4936938616c3f3"
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}