{
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "!pip install transformers"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qi619MPB6_Pu",
        "outputId": "3dd04da1-b4a9-44a7-f570-913018266748"
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Collecting transformers\n",
            "  Downloading transformers-4.27.4-py3-none-any.whl (6.8 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m6.8/6.8 MB\u001b[0m \u001b[31m69.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.9/dist-packages (from transformers) (23.0)\n",
            "Collecting tokenizers!=0.11.3,<0.14,>=0.11.1\n",
            "  Downloading tokenizers-0.13.3-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (7.8 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m7.8/7.8 MB\u001b[0m \u001b[31m90.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: filelock in /usr/local/lib/python3.9/dist-packages (from transformers) (3.10.7)\n",
            "Requirement already satisfied: pyyaml>=5.1 in /usr/local/lib/python3.9/dist-packages (from transformers) (6.0)\n",
            "Requirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.9/dist-packages (from transformers) (2022.10.31)\n",
            "Requirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.9/dist-packages (from transformers) (4.65.0)\n",
            "Requirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.9/dist-packages (from transformers) (1.22.4)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.9/dist-packages (from transformers) (2.27.1)\n",
            "Collecting huggingface-hub<1.0,>=0.11.0\n",
            "  Downloading huggingface_hub-0.13.4-py3-none-any.whl (200 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m200.1/200.1 KB\u001b[0m \u001b[31m1.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.9/dist-packages (from huggingface-hub<1.0,>=0.11.0->transformers) (4.5.0)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.9/dist-packages (from requests->transformers) (3.4)\n",
            "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.9/dist-packages (from requests->transformers) (1.26.15)\n",
            "Requirement already satisfied: charset-normalizer~=2.0.0 in /usr/local/lib/python3.9/dist-packages (from requests->transformers) (2.0.12)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.9/dist-packages (from requests->transformers) (2022.12.7)\n",
            "Installing collected packages: tokenizers, huggingface-hub, transformers\n",
            "Successfully installed huggingface-hub-0.13.4 tokenizers-0.13.3 transformers-4.27.4\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
        "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
        "execution": {
          "iopub.execute_input": "2022-04-28T20:31:03.924205Z",
          "iopub.status.busy": "2022-04-28T20:31:03.923657Z",
          "iopub.status.idle": "2022-04-28T20:31:11.791038Z",
          "shell.execute_reply": "2022-04-28T20:31:11.789395Z",
          "shell.execute_reply.started": "2022-04-28T20:31:03.924086Z"
        },
        "id": "MK5WxLq16tov"
      },
      "outputs": [],
      "source": [
        "import numpy as np # linear algebra\n",
        "import pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n",
        "\n",
        "#For EDA\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "\n",
        "# Packages for general use throughout the notebook.\n",
        "import random\n",
        "import warnings\n",
        "import time\n",
        "%matplotlib inline\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "# to see columns properly\n",
        "pd.set_option('display.max_colwidth', None)\n",
        "\n",
        "# for build our model\n",
        "import tensorflow as tf\n",
        "from tensorflow.keras.layers import Add, GlobalAvgPool1D, MaxPool1D, Activation, BatchNormalization, Embedding, LSTM, Dense, Bidirectional, Input, SpatialDropout1D, Dropout, Conv1D\n",
        "from tensorflow.keras import Model\n",
        "from transformers import BertTokenizer, TFBertModel\n",
        "from tensorflow.keras.activations import relu\n",
        "\n",
        "from sklearn.metrics import confusion_matrix, classification_report, accuracy_score, precision_score, recall_score, f1_score\n",
        "\n",
        "\n",
        "# Input data files are available in the read-only \"../input/\" directory\n",
        "import os\n",
        "for dirname, _, filenames in os.walk('/kaggle/input'):\n",
        "    for filename in filenames:\n",
        "        print(os.path.join(dirname, filename))\n",
        "\n",
        "import torch\n",
        "import numpy as np\n",
        "from transformers import BertTokenizer, BertModel\n",
        "import time\n",
        "from datetime import datetime\n",
        "import matplotlib.pyplot as plt\n",
        "import torch\n",
        "import torch.nn as nn\n",
        "from torch.optim import Adam\n",
        "from tqdm import tqdm\n",
        "from torch.optim.lr_scheduler import ReduceLROnPlateau\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install session_info"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qKi_3QaBLt7n",
        "outputId": "1cb8b932-01aa-481b-9c1b-291a1b5363ec"
      },
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Requirement already satisfied: session_info in /usr/local/lib/python3.9/dist-packages (1.0.0)\n",
            "Requirement already satisfied: stdlib-list in /usr/local/lib/python3.9/dist-packages (from session_info) (0.8.0)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import session_info\n",
        "session_info.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 451
        },
        "id": "-7_PjaT-llAK",
        "outputId": "e9ac4d86-5d39-40bc-b4b6-d121404ebd43"
      },
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "<details>\n",
              "<summary>Click to view session information</summary>\n",
              "<pre>\n",
              "-----\n",
              "keras               2.12.0\n",
              "matplotlib          3.7.1\n",
              "numpy               1.22.4\n",
              "pandas              1.4.4\n",
              "seaborn             0.12.2\n",
              "session_info        1.0.0\n",
              "sklearn             1.2.2\n",
              "tensorflow          2.12.0\n",
              "torch               2.0.0+cu118\n",
              "tqdm                4.65.0\n",
              "transformers        4.27.4\n",
              "-----\n",
              "</pre>\n",
              "<details>\n",
              "<summary>Click to view modules imported as dependencies</summary>\n",
              "<pre>\n",
              "PIL                                         8.4.0\n",
              "aa8f2297d25b4dc6fd3d98411eb3ba53823c4f42    NA\n",
              "absl                                        NA\n",
              "astunparse                                  1.6.3\n",
              "attr                                        22.2.0\n",
              "backcall                                    0.2.0\n",
              "cachetools                                  5.3.0\n",
              "certifi                                     2022.12.07\n",
              "cffi                                        1.15.1\n",
              "chardet                                     4.0.0\n",
              "charset_normalizer                          2.0.12\n",
              "cloudpickle                                 2.2.1\n",
              "cryptography                                40.0.1\n",
              "cycler                                      0.10.0\n",
              "cython_runtime                              NA\n",
              "dateutil                                    2.8.2\n",
              "debugpy                                     1.6.6\n",
              "decorator                                   4.4.2\n",
              "defusedxml                                  0.7.1\n",
              "dot_parser                                  NA\n",
              "etils                                       1.1.1\n",
              "filelock                                    3.10.7\n",
              "flatbuffers                                 23.3.3\n",
              "fsspec                                      2023.3.0\n",
              "gast                                        NA\n",
              "google                                      NA\n",
              "google_auth_httplib2                        NA\n",
              "googleapiclient                             NA\n",
              "h5py                                        3.8.0\n",
              "httplib2                                    0.21.0\n",
              "huggingface_hub                             0.13.4\n",
              "idna                                        3.4\n",
              "importlib_resources                         NA\n",
              "ipykernel                                   5.5.6\n",
              "ipython_genutils                            0.2.0\n",
              "ipywidgets                                  7.7.1\n",
              "jax                                         0.4.7\n",
              "jaxlib                                      0.4.7\n",
              "joblib                                      1.1.1\n",
              "kiwisolver                                  1.4.4\n",
              "matplotlib_inline                           0.1.6\n",
              "ml_dtypes                                   0.0.4\n",
              "mpl_toolkits                                NA\n",
              "mpmath                                      1.3.0\n",
              "numexpr                                     2.8.4\n",
              "nvfuser                                     NA\n",
              "oauth2client                                4.1.3\n",
              "opt_einsum                                  v3.3.0\n",
              "packaging                                   23.0\n",
              "pexpect                                     4.8.0\n",
              "pickleshare                                 0.7.5\n",
              "pkg_resources                               NA\n",
              "platformdirs                                3.2.0\n",
              "portpicker                                  NA\n",
              "prompt_toolkit                              3.0.38\n",
              "psutil                                      5.9.4\n",
              "ptyprocess                                  0.7.0\n",
              "pyarrow                                     9.0.0\n",
              "pyasn1                                      0.4.8\n",
              "pyasn1_modules                              0.2.8\n",
              "pydev_ipython                               NA\n",
              "pydevconsole                                NA\n",
              "pydevd                                      2.9.5\n",
              "pydevd_file_utils                           NA\n",
              "pydevd_plugins                              NA\n",
              "pydevd_tracing                              NA\n",
              "pydot                                       1.4.2\n",
              "pydot_ng                                    2.0.0\n",
              "pygments                                    2.14.0\n",
              "pyparsing                                   3.0.9\n",
              "pytz                                        2022.7.1\n",
              "requests                                    2.27.1\n",
              "rich                                        NA\n",
              "rsa                                         4.9\n",
              "scipy                                       1.10.1\n",
              "sitecustomize                               NA\n",
              "six                                         1.16.0\n",
              "socks                                       1.7.1\n",
              "sphinxcontrib                               NA\n",
              "statsmodels                                 0.13.5\n",
              "storemagic                                  NA\n",
              "sympy                                       1.11.1\n",
              "tblib                                       1.7.0\n",
              "tensorboard                                 2.12.1\n",
              "termcolor                                   NA\n",
              "threadpoolctl                               3.1.0\n",
              "tokenizers                                  0.13.3\n",
              "tornado                                     6.2\n",
              "traitlets                                   5.7.1\n",
              "typing_extensions                           NA\n",
              "uritemplate                                 4.1.1\n",
              "urllib3                                     1.26.15\n",
              "wcwidth                                     0.2.6\n",
              "wrapt                                       1.14.1\n",
              "yaml                                        6.0\n",
              "zipp                                        NA\n",
              "zmq                                         23.2.1\n",
              "</pre>\n",
              "</details> <!-- seems like this ends pre, so might as well be explicit -->\n",
              "<pre>\n",
              "-----\n",
              "IPython             7.34.0\n",
              "jupyter_client      6.1.12\n",
              "jupyter_core        5.3.0\n",
              "notebook            6.4.8\n",
              "-----\n",
              "Python 3.9.16 (main, Dec  7 2022, 01:11:51) [GCC 9.4.0]\n",
              "Linux-5.10.147+-x86_64-with-glibc2.31\n",
              "-----\n",
              "Session information updated at 2023-04-07 10:09\n",
              "</pre>\n",
              "</details>"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 728
        },
        "id": "kEszsQy96to0",
        "outputId": "767240c5-1489-460f-db8a-0d59b6f9d1c6"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                         id  \\\n",
              "0      81c11060-a240-4d54-841b-9e2916039e85   \n",
              "1      be80ebbf-b322-4c3b-afa1-94932ea80731   \n",
              "2      f99e2513-83ed-4076-ac72-b9e2cff3f049   \n",
              "3      83ed2b2e-b815-4f36-9fc4-80a9050cf2d0   \n",
              "4      d93e05f7-bfdd-4cdb-99d8-3048761b30ff   \n",
              "...                                     ...   \n",
              "12462  71eedfa1-8fa6-425c-b982-258c3b29c003   \n",
              "12463  b38eed16-6501-4563-8b33-ff2e634bb8e5   \n",
              "12464  c8a051a8-94ef-4b64-a48e-54d0fa4f8323   \n",
              "12465  513a7e6d-4207-4a16-9b47-972f26e23cfe   \n",
              "12466  247834c9-ad37-4576-a094-69d70c69b124   \n",
              "\n",
              "                                                                                              text  \\\n",
              "0                                                                                      çürük dişli   \n",
              "1                              Bu adamın islama ve müslümanlara verdiği zararı Gavur bile yapmaz !   \n",
              "2                                                                             erkekler zora gelmez   \n",
              "3                                Utanmazın götüne kazık sokmuşlar bu tıkırtı nereden geliyor demiş   \n",
              "4                                                   otomasyon< sistemlerine= doğrudan bağlanabilir   \n",
              "...                                                                                            ...   \n",
              "12462                              uyuma taklidi yapan tehlikeli bir hayvanın gözüne benziyordu bu   \n",
              "12463                                                    yolda at kavga eden üç oğlan çocuğu görür   \n",
              "12464                           sizin köpeklerinizin burnu bile daha iyi koku alıyor bizimkilerden   \n",
              "12465   hayalleri gerçek etmek için birisinin delilik yapması lazım diyecektir onu uyaran kasabına   \n",
              "12466  deliklerden birini bulsan diğerini bulamıyorsun ve diğerini bulurken öncekini kaybediyorsun   \n",
              "\n",
              "       is_offensive     target  \\\n",
              "0                 1     INSULT   \n",
              "1                 1     RACIST   \n",
              "2                 1     SEXIST   \n",
              "3                 1  PROFANITY   \n",
              "4                 0      OTHER   \n",
              "...             ...        ...   \n",
              "12462             0      OTHER   \n",
              "12463             0      OTHER   \n",
              "12464             0      OTHER   \n",
              "12465             0      OTHER   \n",
              "12466             0      OTHER   \n",
              "\n",
              "                                                                                 preprocessed_text  \\\n",
              "0                                                                                      çürük dişli   \n",
              "1                                bu adamın islama ve müslümanlara verdiği zararı gavur bile yapmaz   \n",
              "2                                                                             erkekler zora gelmez   \n",
              "3                                utanmazın götüne kazık sokmuşlar bu tıkırtı nereden geliyor demiş   \n",
              "4                                                     otomasyon sistemlerine doğrudan bağlanabilir   \n",
              "...                                                                                            ...   \n",
              "12462                              uyuma taklidi yapan tehlikeli bir hayvanın gözüne benziyordu bu   \n",
              "12463                                                    yolda at kavga eden üç oğlan çocuğu görür   \n",
              "12464                           sizin köpeklerinizin burnu bile daha iyi koku alıyor bizimkilerden   \n",
              "12465   hayalleri gerçek etmek için birisinin delilik yapması lazım diyecektir onu uyaran kasabına   \n",
              "12466  deliklerden birini bulsan diğerini bulamıyorsun ve diğerini bulurken öncekini kaybediyorsun   \n",
              "\n",
              "                                                                                 first_p_sec_sw  \\\n",
              "0                                                                                   çürük dişli   \n",
              "1                                   adamın islama müslümanlara verdiği zararı gavur bile yapmaz   \n",
              "2                                                                          erkekler zora gelmez   \n",
              "3                                utanmazın götüne kazık sokmuşlar tıkırtı nereden geliyor demiş   \n",
              "4                                                  otomasyon sistemlerine doğrudan bağlanabilir   \n",
              "...                                                                                         ...   \n",
              "12462                                  uyuma taklidi yapan tehlikeli hayvanın gözüne benziyordu   \n",
              "12463                                                 yolda at kavga eden üç oğlan çocuğu görür   \n",
              "12464                             sizin köpeklerinizin burnu bile iyi koku alıyor bizimkilerden   \n",
              "12465     hayalleri gerçek etmek birisinin delilik yapması lazım diyecektir onu uyaran kasabına   \n",
              "12466  deliklerden birini bulsan diğerini bulamıyorsun diğerini bulurken öncekini kaybediyorsun   \n",
              "\n",
              "                                                                                                                      tkn_p_sw  \\\n",
              "0                                                                                                           ['çürük', 'dişli']   \n",
              "1                                         ['adamın', 'islama', 'müslümanlara', 'verdiği', 'zararı', 'gavur', 'bile', 'yapmaz']   \n",
              "2                                                                                               ['erkekler', 'zora', 'gelmez']   \n",
              "3                                      ['utanmazın', 'götüne', 'kazık', 'sokmuşlar', 'tıkırtı', 'nereden', 'geliyor', 'demiş']   \n",
              "4                                                                    ['otomasyon', 'sistemlerine', 'doğrudan', 'bağlanabilir']   \n",
              "...                                                                                                                        ...   \n",
              "12462                                           ['uyuma', 'taklidi', 'yapan', 'tehlikeli', 'hayvanın', 'gözüne', 'benziyordu']   \n",
              "12463                                                       ['yolda', 'at', 'kavga', 'eden', 'üç', 'oğlan', 'çocuğu', 'görür']   \n",
              "12464                                   ['sizin', 'köpeklerinizin', 'burnu', 'bile', 'iyi', 'koku', 'alıyor', 'bizimkilerden']   \n",
              "12465  ['hayalleri', 'gerçek', 'etmek', 'birisinin', 'delilik', 'yapması', 'lazım', 'diyecektir', 'onu', 'uyaran', 'kasabına']   \n",
              "12466     ['deliklerden', 'birini', 'bulsan', 'diğerini', 'bulamıyorsun', 'diğerini', 'bulurken', 'öncekini', 'kaybediyorsun']   \n",
              "\n",
              "                                                                                         lemma_tokenized_swp  \\\n",
              "0                                                                                         ['çürük', 'dişli']   \n",
              "1                                 ['adam', 'islam', 'müslüman', 'verdik', 'zarar', 'gavur', 'bil', 'yapmaz']   \n",
              "2                                                                                 ['erkek', 'zor', 'gelmez']   \n",
              "3                                       ['utanmaz', 'göt', 'kazık', 'sok', 'tıkır', 'nere', 'geliyor', 'de']   \n",
              "4                                                           ['otomasyon', 'sistem', 'doğru', 'bağlanabilir']   \n",
              "...                                                                                                      ...   \n",
              "12462                                      ['uyum', 'takli', 'yapan', 'tehlike', 'hayva', 'göz', 'benziyor']   \n",
              "12463                                            ['yol', 'at', 'kavg', 'e', 'üç', 'oğlan', 'çocuk', 'görür']   \n",
              "12464                                     ['siz', 'köpek', 'burn', 'bil', 'iyi', 'kok', 'alıyor', 'bizimki']   \n",
              "12465  ['hayal', 'gerçek', 'etmek', 'biri', 'delilik', 'yapma', 'lazım', 'diyecek', 'on', 'uyaran', 'kasap']   \n",
              "12466              ['delik', 'bir', 'bul', 'diğer', 'bulamıyor', 'diğer', 'bulurken', 'öncek', 'kaybediyor']   \n",
              "\n",
              "                                                   str_lemma_tokenized_swp  \n",
              "0                                                              çürük dişli  \n",
              "1                        adam islam müslüman verdik zarar gavur bil yapmaz  \n",
              "2                                                         erkek zor gelmez  \n",
              "3                              utanmaz göt kazık sok tıkır nere geliyor de  \n",
              "4                                      otomasyon sistem doğru bağlanabilir  \n",
              "...                                                                    ...  \n",
              "12462                          uyum takli yapan tehlike hayva göz benziyor  \n",
              "12463                                   yol at kavg e üç oğlan çocuk görür  \n",
              "12464                            siz köpek burn bil iyi kok alıyor bizimki  \n",
              "12465  hayal gerçek etmek biri delilik yapma lazım diyecek on uyaran kasap  \n",
              "12466        delik bir bul diğer bulamıyor diğer bulurken öncek kaybediyor  \n",
              "\n",
              "[12467 rows x 9 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-b5ce3fd1-6c2c-445f-994d-5b213cacbb44\">\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>id</th>\n",
              "      <th>text</th>\n",
              "      <th>is_offensive</th>\n",
              "      <th>target</th>\n",
              "      <th>preprocessed_text</th>\n",
              "      <th>first_p_sec_sw</th>\n",
              "      <th>tkn_p_sw</th>\n",
              "      <th>lemma_tokenized_swp</th>\n",
              "      <th>str_lemma_tokenized_swp</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>81c11060-a240-4d54-841b-9e2916039e85</td>\n",
              "      <td>çürük dişli</td>\n",
              "      <td>1</td>\n",
              "      <td>INSULT</td>\n",
              "      <td>çürük dişli</td>\n",
              "      <td>çürük dişli</td>\n",
              "      <td>['çürük', 'dişli']</td>\n",
              "      <td>['çürük', 'dişli']</td>\n",
              "      <td>çürük dişli</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>be80ebbf-b322-4c3b-afa1-94932ea80731</td>\n",
              "      <td>Bu adamın islama ve müslümanlara verdiği zararı Gavur bile yapmaz !</td>\n",
              "      <td>1</td>\n",
              "      <td>RACIST</td>\n",
              "      <td>bu adamın islama ve müslümanlara verdiği zararı gavur bile yapmaz</td>\n",
              "      <td>adamın islama müslümanlara verdiği zararı gavur bile yapmaz</td>\n",
              "      <td>['adamın', 'islama', 'müslümanlara', 'verdiği', 'zararı', 'gavur', 'bile', 'yapmaz']</td>\n",
              "      <td>['adam', 'islam', 'müslüman', 'verdik', 'zarar', 'gavur', 'bil', 'yapmaz']</td>\n",
              "      <td>adam islam müslüman verdik zarar gavur bil yapmaz</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>f99e2513-83ed-4076-ac72-b9e2cff3f049</td>\n",
              "      <td>erkekler zora gelmez</td>\n",
              "      <td>1</td>\n",
              "      <td>SEXIST</td>\n",
              "      <td>erkekler zora gelmez</td>\n",
              "      <td>erkekler zora gelmez</td>\n",
              "      <td>['erkekler', 'zora', 'gelmez']</td>\n",
              "      <td>['erkek', 'zor', 'gelmez']</td>\n",
              "      <td>erkek zor gelmez</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>83ed2b2e-b815-4f36-9fc4-80a9050cf2d0</td>\n",
              "      <td>Utanmazın götüne kazık sokmuşlar bu tıkırtı nereden geliyor demiş</td>\n",
              "      <td>1</td>\n",
              "      <td>PROFANITY</td>\n",
              "      <td>utanmazın götüne kazık sokmuşlar bu tıkırtı nereden geliyor demiş</td>\n",
              "      <td>utanmazın götüne kazık sokmuşlar tıkırtı nereden geliyor demiş</td>\n",
              "      <td>['utanmazın', 'götüne', 'kazık', 'sokmuşlar', 'tıkırtı', 'nereden', 'geliyor', 'demiş']</td>\n",
              "      <td>['utanmaz', 'göt', 'kazık', 'sok', 'tıkır', 'nere', 'geliyor', 'de']</td>\n",
              "      <td>utanmaz göt kazık sok tıkır nere geliyor de</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>d93e05f7-bfdd-4cdb-99d8-3048761b30ff</td>\n",
              "      <td>otomasyon&lt; sistemlerine= doğrudan bağlanabilir</td>\n",
              "      <td>0</td>\n",
              "      <td>OTHER</td>\n",
              "      <td>otomasyon sistemlerine doğrudan bağlanabilir</td>\n",
              "      <td>otomasyon sistemlerine doğrudan bağlanabilir</td>\n",
              "      <td>['otomasyon', 'sistemlerine', 'doğrudan', 'bağlanabilir']</td>\n",
              "      <td>['otomasyon', 'sistem', 'doğru', 'bağlanabilir']</td>\n",
              "      <td>otomasyon sistem doğru bağlanabilir</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12462</th>\n",
              "      <td>71eedfa1-8fa6-425c-b982-258c3b29c003</td>\n",
              "      <td>uyuma taklidi yapan tehlikeli bir hayvanın gözüne benziyordu bu</td>\n",
              "      <td>0</td>\n",
              "      <td>OTHER</td>\n",
              "      <td>uyuma taklidi yapan tehlikeli bir hayvanın gözüne benziyordu bu</td>\n",
              "      <td>uyuma taklidi yapan tehlikeli hayvanın gözüne benziyordu</td>\n",
              "      <td>['uyuma', 'taklidi', 'yapan', 'tehlikeli', 'hayvanın', 'gözüne', 'benziyordu']</td>\n",
              "      <td>['uyum', 'takli', 'yapan', 'tehlike', 'hayva', 'göz', 'benziyor']</td>\n",
              "      <td>uyum takli yapan tehlike hayva göz benziyor</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12463</th>\n",
              "      <td>b38eed16-6501-4563-8b33-ff2e634bb8e5</td>\n",
              "      <td>yolda at kavga eden üç oğlan çocuğu görür</td>\n",
              "      <td>0</td>\n",
              "      <td>OTHER</td>\n",
              "      <td>yolda at kavga eden üç oğlan çocuğu görür</td>\n",
              "      <td>yolda at kavga eden üç oğlan çocuğu görür</td>\n",
              "      <td>['yolda', 'at', 'kavga', 'eden', 'üç', 'oğlan', 'çocuğu', 'görür']</td>\n",
              "      <td>['yol', 'at', 'kavg', 'e', 'üç', 'oğlan', 'çocuk', 'görür']</td>\n",
              "      <td>yol at kavg e üç oğlan çocuk görür</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12464</th>\n",
              "      <td>c8a051a8-94ef-4b64-a48e-54d0fa4f8323</td>\n",
              "      <td>sizin köpeklerinizin burnu bile daha iyi koku alıyor bizimkilerden</td>\n",
              "      <td>0</td>\n",
              "      <td>OTHER</td>\n",
              "      <td>sizin köpeklerinizin burnu bile daha iyi koku alıyor bizimkilerden</td>\n",
              "      <td>sizin köpeklerinizin burnu bile iyi koku alıyor bizimkilerden</td>\n",
              "      <td>['sizin', 'köpeklerinizin', 'burnu', 'bile', 'iyi', 'koku', 'alıyor', 'bizimkilerden']</td>\n",
              "      <td>['siz', 'köpek', 'burn', 'bil', 'iyi', 'kok', 'alıyor', 'bizimki']</td>\n",
              "      <td>siz köpek burn bil iyi kok alıyor bizimki</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12465</th>\n",
              "      <td>513a7e6d-4207-4a16-9b47-972f26e23cfe</td>\n",
              "      <td>hayalleri gerçek etmek için birisinin delilik yapması lazım diyecektir onu uyaran kasabına</td>\n",
              "      <td>0</td>\n",
              "      <td>OTHER</td>\n",
              "      <td>hayalleri gerçek etmek için birisinin delilik yapması lazım diyecektir onu uyaran kasabına</td>\n",
              "      <td>hayalleri gerçek etmek birisinin delilik yapması lazım diyecektir onu uyaran kasabına</td>\n",
              "      <td>['hayalleri', 'gerçek', 'etmek', 'birisinin', 'delilik', 'yapması', 'lazım', 'diyecektir', 'onu', 'uyaran', 'kasabına']</td>\n",
              "      <td>['hayal', 'gerçek', 'etmek', 'biri', 'delilik', 'yapma', 'lazım', 'diyecek', 'on', 'uyaran', 'kasap']</td>\n",
              "      <td>hayal gerçek etmek biri delilik yapma lazım diyecek on uyaran kasap</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12466</th>\n",
              "      <td>247834c9-ad37-4576-a094-69d70c69b124</td>\n",
              "      <td>deliklerden birini bulsan diğerini bulamıyorsun ve diğerini bulurken öncekini kaybediyorsun</td>\n",
              "      <td>0</td>\n",
              "      <td>OTHER</td>\n",
              "      <td>deliklerden birini bulsan diğerini bulamıyorsun ve diğerini bulurken öncekini kaybediyorsun</td>\n",
              "      <td>deliklerden birini bulsan diğerini bulamıyorsun diğerini bulurken öncekini kaybediyorsun</td>\n",
              "      <td>['deliklerden', 'birini', 'bulsan', 'diğerini', 'bulamıyorsun', 'diğerini', 'bulurken', 'öncekini', 'kaybediyorsun']</td>\n",
              "      <td>['delik', 'bir', 'bul', 'diğer', 'bulamıyor', 'diğer', 'bulurken', 'öncek', 'kaybediyor']</td>\n",
              "      <td>delik bir bul diğer bulamıyor diğer bulurken öncek kaybediyor</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>12467 rows × 9 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-b5ce3fd1-6c2c-445f-994d-5b213cacbb44')\"\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-b5ce3fd1-6c2c-445f-994d-5b213cacbb44 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-b5ce3fd1-6c2c-445f-994d-5b213cacbb44');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ],
      "source": [
        "\n",
        "dataset = pd.read_csv(r\"train_with_preprocess.csv\")\n",
        "dataset"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df=dataset[[ \"first_p_sec_sw\",\"target\"]]\n",
        "df.columns=[\"text\",\"target\"]\n",
        "df"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "N2hkK77A9mTu",
        "outputId": "5bc75c7d-0702-4a50-8043-02a14f9406ee"
      },
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                                                                                           text  \\\n",
              "0                                                                                   çürük dişli   \n",
              "1                                   adamın islama müslümanlara verdiği zararı gavur bile yapmaz   \n",
              "2                                                                          erkekler zora gelmez   \n",
              "3                                utanmazın götüne kazık sokmuşlar tıkırtı nereden geliyor demiş   \n",
              "4                                                  otomasyon sistemlerine doğrudan bağlanabilir   \n",
              "...                                                                                         ...   \n",
              "12462                                  uyuma taklidi yapan tehlikeli hayvanın gözüne benziyordu   \n",
              "12463                                                 yolda at kavga eden üç oğlan çocuğu görür   \n",
              "12464                             sizin köpeklerinizin burnu bile iyi koku alıyor bizimkilerden   \n",
              "12465     hayalleri gerçek etmek birisinin delilik yapması lazım diyecektir onu uyaran kasabına   \n",
              "12466  deliklerden birini bulsan diğerini bulamıyorsun diğerini bulurken öncekini kaybediyorsun   \n",
              "\n",
              "          target  \n",
              "0         INSULT  \n",
              "1         RACIST  \n",
              "2         SEXIST  \n",
              "3      PROFANITY  \n",
              "4          OTHER  \n",
              "...          ...  \n",
              "12462      OTHER  \n",
              "12463      OTHER  \n",
              "12464      OTHER  \n",
              "12465      OTHER  \n",
              "12466      OTHER  \n",
              "\n",
              "[12467 rows x 2 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-c44d5cdf-6976-463d-a37a-452ac6521a7d\">\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>text</th>\n",
              "      <th>target</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>çürük dişli</td>\n",
              "      <td>INSULT</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>adamın islama müslümanlara verdiği zararı gavur bile yapmaz</td>\n",
              "      <td>RACIST</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>erkekler zora gelmez</td>\n",
              "      <td>SEXIST</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>utanmazın götüne kazık sokmuşlar tıkırtı nereden geliyor demiş</td>\n",
              "      <td>PROFANITY</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>otomasyon sistemlerine doğrudan bağlanabilir</td>\n",
              "      <td>OTHER</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12462</th>\n",
              "      <td>uyuma taklidi yapan tehlikeli hayvanın gözüne benziyordu</td>\n",
              "      <td>OTHER</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12463</th>\n",
              "      <td>yolda at kavga eden üç oğlan çocuğu görür</td>\n",
              "      <td>OTHER</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12464</th>\n",
              "      <td>sizin köpeklerinizin burnu bile iyi koku alıyor bizimkilerden</td>\n",
              "      <td>OTHER</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12465</th>\n",
              "      <td>hayalleri gerçek etmek birisinin delilik yapması lazım diyecektir onu uyaran kasabına</td>\n",
              "      <td>OTHER</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12466</th>\n",
              "      <td>deliklerden birini bulsan diğerini bulamıyorsun diğerini bulurken öncekini kaybediyorsun</td>\n",
              "      <td>OTHER</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>12467 rows × 2 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-c44d5cdf-6976-463d-a37a-452ac6521a7d')\"\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-c44d5cdf-6976-463d-a37a-452ac6521a7d 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-c44d5cdf-6976-463d-a37a-452ac6521a7d');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "tokenizer = BertTokenizer.from_pretrained(\"dbmdz/bert-base-turkish-128k-uncased\")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 197,
          "referenced_widgets": [
            "98d52d85159340beaa841dd93000693d",
            "b0753bb9e8fa48ac94e09d3554982e04",
            "e9d0f7aadc1f4bd884dffb408801c594",
            "1d43f272228144b78da88b84973ff52c",
            "6fef1206fb87437c9470c241c8a728ba",
            "ec26ce0de028473ebf458bc103ca7c67",
            "ec3df2b46c9e4b7a9f56e4b95acb78e9",
            "5f7329df42fc4171a7ae2e66b33ba062",
            "b4c8ad7e5b404d65b813fb89c628dad2",
            "7305ee44271d478eb7747ee233d9ac7c",
            "a5b805a2c32a4f88b575d32edd88de07",
            "239fbf31456543ed9f45c2feea4cb617",
            "146753eb82dd4251af66a62046545e24",
            "7ba0d78f51f9461299cd385167f5fe5b",
            "e795f26afb1f4b20b581f5e325f6b32f",
            "101ddce618334cf6874b0d473dd82fe8",
            "f93851550a954d9d80a2c308ed0199bd",
            "aa46630642d04dbca738575b650d86a1",
            "3ab878e4c538412cbe13d409a11e31b4",
            "85636f736a534a47811513098264d8a2",
            "48086df862604f618d778f86ef195475",
            "62dd7bdbcaf04cef8987e8689d2d74a9",
            "2bf25e6999804f4e9e0b845e3e79c5b0",
            "8e691692c11349469eb39327c23fead5",
            "045dc68924084dc195fdb5a277753a5e",
            "a422ea5a73aa469eb4121ccf9cb8ca1e",
            "c4fb9b4f26d04bd0bff010382b9eb460",
            "9750c6ff12d34b9cb99b3ba5b4ea27fc",
            "0f495859b0b54cfe85f580f5850bc535",
            "50e3784f0fb340e2a2c14fe2d2fa4f90",
            "8b35ee65c7d441478b137f71c565263f",
            "9f8a65a50cd240e9b206d45a07e2c70b",
            "8c1125f582db4ebbbfb48bb1f104a7ae"
          ]
        },
        "id": "kjAO35cB_UG2",
        "outputId": "a907b622-8fab-4c68-e404-7bb02c2ab2e0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)solve/main/vocab.txt:   0%|          | 0.00/1.23M [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "98d52d85159340beaa841dd93000693d"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)okenizer_config.json:   0%|          | 0.00/59.0 [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "239fbf31456543ed9f45c2feea4cb617"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading (…)lve/main/config.json:   0%|          | 0.00/386 [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "2bf25e6999804f4e9e0b845e3e79c5b0"
            }
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "labels = {'INSULT':0,\n",
        "          'OTHER':1,\n",
        "          'PROFANITY':2,\n",
        "          'RACIST':3,\n",
        "          'SEXIST':4\n",
        "          }"
      ],
      "metadata": {
        "id": "LwwlXQE69mWR"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "class Dataset(torch.utils.data.Dataset):\n",
        "\n",
        "    def __init__(self, df):\n",
        "\n",
        "        self.labels = [labels[label] for label in df['target']]\n",
        "        self.texts = [tokenizer(text, \n",
        "                               padding='max_length', max_length = 512, truncation=True,\n",
        "                                return_tensors=\"pt\") for text in df['text']]\n",
        "\n",
        "    def classes(self):\n",
        "        return self.labels\n",
        "\n",
        "    def __len__(self):\n",
        "        return len(self.labels)\n",
        "\n",
        "    def get_batch_labels(self, idx):\n",
        "        # Fetch a batch of labels\n",
        "        return np.array(self.labels[idx])\n",
        "\n",
        "    def get_batch_texts(self, idx):\n",
        "        # Fetch a batch of inputs\n",
        "        return self.texts[idx]\n",
        "\n",
        "    def __getitem__(self, idx):\n",
        "\n",
        "        batch_texts = self.get_batch_texts(idx)\n",
        "        batch_y = self.get_batch_labels(idx)\n",
        "\n",
        "        return batch_texts, batch_y"
      ],
      "metadata": {
        "id": "v02KwM949mYr"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "np.random.seed(112)\n",
        "df_train, df_val, df_test = np.split(df.sample(frac=1, random_state=42), \n",
        "                                     [int(.8*len(df)), int(.9*len(df))])\n",
        "\n",
        "print(len(df_train),len(df_val), len(df_test))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HsTBP0F89ma_",
        "outputId": "f094130e-41de-4c33-d23d-b0cc9c29cc9e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "9973 1247 1247\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "\n",
        "class BertClassifierConv1D(nn.Module):\n",
        "    def __init__(self, dropout=0.5, num_classes=5):\n",
        "        super(BertClassifierConv1D, self).__init__()\n",
        "        \n",
        "        self.bert = BertModel.from_pretrained('dbmdz/bert-base-turkish-128k-uncased', return_dict=True)\n",
        "        self.conv1d = nn.Conv1d(in_channels=self.bert.config.hidden_size, out_channels=128, kernel_size=5)\n",
        "        self.bilstm = nn.LSTM(input_size=128, hidden_size=64, num_layers=1, bidirectional=True, batch_first=True)\n",
        "        self.dropout = nn.Dropout(dropout)\n",
        "        self.linear = nn.Linear(128, num_classes)\n",
        "\n",
        "    def forward(self, input_id, mask):\n",
        "        output = self.bert(input_ids=input_id, attention_mask=mask).last_hidden_state\n",
        "        output = output.permute(0, 2, 1) # swap dimensions to prepare for Conv1d layer\n",
        "        output = self.conv1d(output)\n",
        "        output, _ = self.bilstm(output.transpose(1, 2))\n",
        "        output = self.dropout(output)\n",
        "        output = self.linear(output.mean(dim=1))\n",
        "        return output\n",
        "\n"
      ],
      "metadata": {
        "id": "0I8HzNP29mdU"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "\n",
        "def plot_graphs(history, string):\n",
        "    plt.plot(history[string])\n",
        "    plt.plot(history['val_'+string])\n",
        "    plt.xlabel(\"Epochs\")\n",
        "    plt.ylabel(string)\n",
        "    plt.legend([string, 'val_'+string])\n",
        "    plt.show()\n",
        "\n",
        "def train(model, train_data, val_data, learning_rate, epochs,patience=3):\n",
        "\n",
        "    train, val = Dataset(train_data), Dataset(val_data)\n",
        "\n",
        "    train_dataloader = torch.utils.data.DataLoader(train, batch_size=32, shuffle=True)\n",
        "    val_dataloader = torch.utils.data.DataLoader(val, batch_size=32)\n",
        "\n",
        "    use_cuda = torch.cuda.is_available()\n",
        "    device = torch.device(\"cuda\" if use_cuda else \"cpu\")\n",
        "\n",
        "    criterion = nn.CrossEntropyLoss()\n",
        "    optimizer = Adam(model.parameters(), lr= learning_rate)\n",
        "\n",
        "    if use_cuda:\n",
        "        model = model.cuda()\n",
        "        criterion = criterion.cuda()\n",
        "\n",
        "    history = {'loss': [], 'accuracy': [], 'val_loss': [], 'val_accuracy': []}\n",
        "    best_val_loss = float('inf')\n",
        "    counter = 0\n",
        "    scheduler = torch.optim.lr_scheduler.ReduceLROnPlateau(optimizer, mode='min', patience=5, factor=0.1, verbose=True, cooldown=0)\n",
        "\n",
        "    for epoch_num in range(epochs):\n",
        "\n",
        "        total_acc_train = 0\n",
        "        total_loss_train = 0\n",
        "\n",
        "        for train_input, train_label in tqdm(train_dataloader):\n",
        "\n",
        "            train_label = train_label.to(device)\n",
        "            mask = train_input['attention_mask'].to(device)\n",
        "            input_id = train_input['input_ids'].squeeze(1).to(device)\n",
        "\n",
        "            output = model(input_id, mask)\n",
        "            \n",
        "            batch_loss = criterion(output, train_label.long())\n",
        "            total_loss_train += batch_loss.item()\n",
        "            \n",
        "            acc = (output.argmax(dim=1) == train_label).sum().item()\n",
        "            total_acc_train += acc\n",
        "\n",
        "            model.zero_grad()\n",
        "            batch_loss.backward()\n",
        "            optimizer.step()\n",
        "        \n",
        "        total_acc_val = 0\n",
        "        total_loss_val = 0\n",
        "\n",
        "        with torch.no_grad():\n",
        "\n",
        "            for val_input, val_label in val_dataloader:\n",
        "\n",
        "                val_label = val_label.to(device)\n",
        "                mask = val_input['attention_mask'].to(device)\n",
        "                input_id = val_input['input_ids'].squeeze(1).to(device)\n",
        "\n",
        "                output = model(input_id, mask)\n",
        "\n",
        "                batch_loss = criterion(output, val_label.long())\n",
        "                total_loss_val += batch_loss.item()\n",
        "                \n",
        "                acc = (output.argmax(dim=1) == val_label).sum().item()\n",
        "                total_acc_val += acc\n",
        "        \n",
        "        train_loss = total_loss_train / len(train_data)\n",
        "        train_acc = total_acc_train / len(train_data)\n",
        "        val_loss = total_loss_val / len(val_data)\n",
        "        val_acc = total_acc_val / len(val_data)\n",
        "        history['loss'].append(train_loss)\n",
        "        history['accuracy'].append(train_acc)\n",
        "        history['val_loss'].append(val_loss)\n",
        "        history['val_accuracy'].append(val_acc)\n",
        "        print(f'Epochs: {epoch_num + 1} | Train Loss: {train_loss:.3f} | Train Accuracy: {train_acc:.3f} | Val Loss: {val_loss:.3f} | Val Accuracy: {val_acc:.3f}')\n",
        "        if val_loss < best_val_loss:\n",
        "            best_val_loss = val_loss\n",
        "            counter = 0\n",
        "        else:\n",
        "            counter += 1\n",
        "            if counter >= patience:\n",
        "                print(f'Early stopping at epoch {epoch_num+1}')\n",
        "                break\n",
        "        scheduler.step(val_loss)\n",
        "\n",
        "    plot_graphs(history, \"accuracy\")\n",
        "    plot_graphs(history, \"loss\")\n",
        "EPOCHS = 15\n",
        "model = BertClassifierConv1D()\n",
        "LR = 1e-6\n",
        "              \n",
        "train(model, df_train, df_val, LR, EPOCHS)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000,
          "referenced_widgets": [
            "af32f29cb21d4142a43acdbabe4921ff",
            "3585acb5179c4b89b6cb7e357674e083",
            "2951182944e94fc299dea78b56c4f195",
            "3a6e4df4c1644ecb9d36e720cea44e30",
            "b65f93de00404a05b4691709da027f69",
            "6f5f79f50a564d30b1cc13e179cd48ef",
            "9d2147283d9040efbdd5d1af509890bf",
            "ed6d68a54c0f49bead78ac7bbec4d589",
            "49312a16f029417c95b9d3491f09ab9b",
            "86bc31432baf47e6b3e4669e6331a20f",
            "eed8b4a68f914bd8a0c3340b15536001"
          ]
        },
        "id": "uZ2VQQBb9mfz",
        "outputId": "23bb52e5-826e-4305-d08f-81dcfbbaaf4e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Downloading pytorch_model.bin:   0%|          | 0.00/740M [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "af32f29cb21d4142a43acdbabe4921ff"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Some weights of the model checkpoint at dbmdz/bert-base-turkish-128k-uncased were not used when initializing BertModel: ['cls.predictions.decoder.weight', 'cls.predictions.transform.dense.bias', 'cls.seq_relationship.weight', 'cls.predictions.bias', 'cls.seq_relationship.bias', 'cls.predictions.transform.LayerNorm.weight', 'cls.predictions.transform.LayerNorm.bias', 'cls.predictions.transform.dense.weight', 'cls.predictions.decoder.bias']\n",
            "- This IS expected if you are initializing BertModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
            "- This IS NOT expected if you are initializing BertModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
            "100%|██████████| 312/312 [03:32<00:00,  1.47it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epochs: 1 | Train Loss: 0.044 | Train Accuracy: 0.527 | Val Loss: 0.036 | Val Accuracy: 0.744\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 312/312 [03:27<00:00,  1.51it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epochs: 2 | Train Loss: 0.028 | Train Accuracy: 0.841 | Val Loss: 0.023 | Val Accuracy: 0.881\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 312/312 [03:27<00:00,  1.51it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epochs: 3 | Train Loss: 0.019 | Train Accuracy: 0.903 | Val Loss: 0.017 | Val Accuracy: 0.901\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 312/312 [03:27<00:00,  1.51it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epochs: 4 | Train Loss: 0.015 | Train Accuracy: 0.926 | Val Loss: 0.015 | Val Accuracy: 0.911\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 312/312 [03:27<00:00,  1.51it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epochs: 5 | Train Loss: 0.012 | Train Accuracy: 0.941 | Val Loss: 0.013 | Val Accuracy: 0.913\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 312/312 [03:27<00:00,  1.51it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epochs: 6 | Train Loss: 0.010 | Train Accuracy: 0.954 | Val Loss: 0.012 | Val Accuracy: 0.922\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 312/312 [03:27<00:00,  1.51it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epochs: 7 | Train Loss: 0.009 | Train Accuracy: 0.964 | Val Loss: 0.012 | Val Accuracy: 0.915\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 312/312 [03:27<00:00,  1.51it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epochs: 8 | Train Loss: 0.008 | Train Accuracy: 0.971 | Val Loss: 0.011 | Val Accuracy: 0.921\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 312/312 [03:27<00:00,  1.51it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epochs: 9 | Train Loss: 0.007 | Train Accuracy: 0.977 | Val Loss: 0.011 | Val Accuracy: 0.921\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 312/312 [03:27<00:00,  1.51it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epochs: 10 | Train Loss: 0.006 | Train Accuracy: 0.982 | Val Loss: 0.010 | Val Accuracy: 0.923\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 312/312 [03:27<00:00,  1.51it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epochs: 11 | Train Loss: 0.005 | Train Accuracy: 0.985 | Val Loss: 0.010 | Val Accuracy: 0.926\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 312/312 [03:27<00:00,  1.51it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epochs: 12 | Train Loss: 0.005 | Train Accuracy: 0.985 | Val Loss: 0.010 | Val Accuracy: 0.917\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 312/312 [03:27<00:00,  1.51it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epochs: 13 | Train Loss: 0.004 | Train Accuracy: 0.988 | Val Loss: 0.010 | Val Accuracy: 0.914\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 312/312 [03:27<00:00,  1.51it/s]\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epochs: 14 | Train Loss: 0.004 | Train Accuracy: 0.990 | Val Loss: 0.010 | Val Accuracy: 0.917\n",
            "Early stopping at epoch 14\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install datetime"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mYUGPxzUDqQd",
        "outputId": "fc1c0087-7428-4630-e5ed-fc50b0dd87a8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Collecting datetime\n",
            "  Downloading DateTime-5.1-py3-none-any.whl (52 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m52.1/52.1 KB\u001b[0m \u001b[31m2.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: pytz in /usr/local/lib/python3.9/dist-packages (from datetime) (2022.7.1)\n",
            "Collecting zope.interface\n",
            "  Downloading zope.interface-6.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl (246 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m246.1/246.1 KB\u001b[0m \u001b[31m10.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: setuptools in /usr/local/lib/python3.9/dist-packages (from zope.interface->datetime) (67.6.1)\n",
            "Installing collected packages: zope.interface, datetime\n",
            "Successfully installed datetime-5.1 zope.interface-6.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "now = datetime.now()\n",
        "seed = int(now.strftime(\"%Y%m%d%H%M%S\"))  # daily\n",
        "print(seed)\n",
        "random.seed(seed)\n",
        "random_time=random.randint(0, 350)\n",
        "model_path= 'model_weights'+str(random_time)+\".pth\"\n",
        "torch.save(model.state_dict(), model_path)\n",
        "print(model_path)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vHxzjuP__w2J",
        "outputId": "4e8d2865-7f0f-4bed-ea9e-501e67b99436"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "20230404022018\n",
            "model_weights86.pth\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def evaluate(model, test_data):\n",
        "\n",
        "    test = Dataset(test_data)\n",
        "\n",
        "    test_dataloader = torch.utils.data.DataLoader(test, batch_size=32)\n",
        "\n",
        "    use_cuda = torch.cuda.is_available()\n",
        "    device = torch.device(\"cuda\" if use_cuda else \"cpu\")\n",
        "\n",
        "    if use_cuda:\n",
        "\n",
        "        model = model.cuda()\n",
        "\n",
        "    total_acc_test = 0\n",
        "    output_indices = []\n",
        "    test_labels=[]\n",
        "    with torch.no_grad():\n",
        "\n",
        "        for test_input, test_label in test_dataloader:\n",
        "\n",
        "              test_label = test_label.to(device)\n",
        "              mask = test_input['attention_mask'].to(device)\n",
        "              input_id = test_input['input_ids'].squeeze(1).to(device)\n",
        "\n",
        "              output = model(input_id, mask)\n",
        "             \n",
        "              acc = (output.argmax(dim=1) == test_label).sum().item()\n",
        "              total_acc_test += acc\n",
        "\n",
        "              batch_indices = output.argmax(dim=1).tolist()\n",
        "              output_indices.extend(batch_indices)\n",
        "              test_labels.extend(test_label)\n",
        "\n",
        "    \n",
        "    print(f'Test Accuracy: {total_acc_test / len(test_data): .3f}')\n",
        "    return output_indices, test_labels\n",
        "y_pred,y_test=evaluate(model, df_test)"
      ],
      "metadata": {
        "id": "_x71hMy1BTwO",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "fec072ba-fdb1-4540-d864-0efc625e4e94"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Test Accuracy:  0.925\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "y_pred_tensor = torch.tensor(y_pred)\n",
        "y_test_tensor = torch.tensor(y_test)"
      ],
      "metadata": {
        "id": "FrkkJ38SBTyr"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(classification_report(np.array(y_pred_tensor.cpu()), np.array(y_test_tensor.cpu()), output_dict=True))"
      ],
      "metadata": {
        "id": "1xMqUW7GBT05",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3fe5938b-895f-4c0d-ab4e-8a7f95a9f6a8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{'0': {'precision': 0.8508064516129032, 'recall': 0.871900826446281, 'f1-score': 0.8612244897959183, 'support': 242}, '1': {'precision': 0.9473684210526315, 'recall': 0.9243243243243243, 'f1-score': 0.935704514363885, 'support': 370}, '2': {'precision': 0.9071729957805907, 'recall': 0.9598214285714286, 'f1-score': 0.9327548806941431, 'support': 224}, '3': {'precision': 0.9731182795698925, 'recall': 0.9476439790575916, 'f1-score': 0.9602122015915119, 'support': 191}, '4': {'precision': 0.9534883720930233, 'recall': 0.9318181818181818, 'f1-score': 0.942528735632184, 'support': 220}, 'accuracy': 0.92542101042502, 'macro avg': {'precision': 0.9263909040218083, 'recall': 0.9271017480435615, 'f1-score': 0.9264849644155284, 'support': 1247}, 'weighted avg': {'precision': 0.9264324469871397, 'recall': 0.92542101042502, 'f1-score': 0.925678382088049, 'support': 1247}}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.metrics import f1_score\n",
        "f1_score(np.array(y_test_tensor.cpu()),np.array(y_pred_tensor.cpu()), average='macro')"
      ],
      "metadata": {
        "id": "9aM5sAqYJMch",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "93bc70ab-0e33-43bc-eb4c-b7cd849f0743"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.9264849644155284"
            ]
          },
          "metadata": {},
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def conf_matrix(y_test,y_pred):\n",
        "  cm = confusion_matrix(y_test,y_pred, normalize=\"true\")\n",
        "  sns.heatmap(cm, annot=True, cmap=\"Blues\",xticklabels=[\"INSULT\",\"OTHER\",\"PROFANITY\",\"RACIST\",\"SECIST\"],yticklabels=[\"INSULT\",\"OTHER\",\"PROFANITY\",\"RACIST\",\"SECIST\"] )\n",
        "  plt.xlabel('Tahmin Edilen Sınıf')\n",
        "  plt.ylabel('Gerçek Sınıf')\n",
        "  plt.show()\n",
        "conf_matrix(np.array(y_pred_tensor.cpu()), np.array(y_test_tensor.cpu()))"
      ],
      "metadata": {
        "id": "XSQhBGNv9miR",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 449
        },
        "outputId": "b907139d-1a1f-4fd5-c2ac-6915267ed6b1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhcAAAGwCAYAAAAaKEeDAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAACGQElEQVR4nO3dd1gTyRsH8G/oKkoLxY6K9GZF7AVFLGeXs4EdFSsWQAUUf4p6dhE7goqCvZxYsZ40Kyh2BTtCaIrSk98faDQSkEDCEng/9+zzXGZnd2fGhLyZmZ1l8Xg8HgghhBBCxESG6QIQQgghpGqh4IIQQgghYkXBBSGEEELEioILQgghhIgVBReEEEIIESsKLgghhBAiVhRcEEIIIUSsKLgghBBCiFjJMV0ASahht57pIki1D0dnMl0EqZWbz2W6CFJLpaY800WQalxaD7HMasqzJH6NGi2mi+U8Wfd8xXIeSaOeC0IIIYSIVZXsuSCEEEIqFVb1+i1PwQUhhBAiaSzJD71UJhRcEEIIIZJWzXouqldtCSGEECJx1HNBCCGESBoNixBCCCFErGhYhBBCCCGk7KjnghBCCJE0GhYhhBBCiFjRsAghhBBCSNlRzwUhhBAiaTQsQgghhBCxomERQgghhJCyo54LQgghRNJoWIQQQgghYkXDIhVn/Pjx+PLlC5NFIIQQQiSPxRLPJiUYDS4CAwORlZXFZBEIIYQQImaMDovweDwmL08IIYRUjGo2LML4nIsvX75ASUmpxDx16tSpoNIQQgghEkDBRcXS19cvdh+PxwOLxUJBQUEFlogQQggh5cF4cHHkyBGoq6szXQxCCCFEcmSkZzKmODAeXHTo0AFaWlpMF4MQQgiRnGo2LFKpa/vkyZMSh00IIYQQUvkw2nPRuHFjyMrKFrs/JycHL1++rMASEUIIIRIgRWtUiAOjwUV8fDyTlyeEEEIqBg2LEEIIIYSUHeMTOgkhhJAqj4ZFKo6amhpYJTR4fn5+BZaGEEIIkZBqNizCaHCxYcMGJi9PCCGEVAzquag4jo6OTF6eEEIIIRLAaD/N58+fhW7StNy3Uz8LPAkYj7STM3B9/d9ora9dYv7pA1sgZqcjUk/MwPO9E7F6chcoyv+8HfdJwHhknZ1TZFs/rZukq1IhjoQcwMA+NuhsZYnxY+wR9zC2xPxhF8/BflBfdLayxKhhAxB+41qxeVf9bwnatTBGcNBegfSBfWzQroWxwLbXf6dY6lORjh06iOF/9YJNh5ZwGjsCj+IelJj/yqXzGD20P2w6tITj34MQcfO6wP7ObUyFbgf3+Rc5V25uLsaPHILObUzx/OkTsdZLGgQfCIJdz+5o08IMo/4ehgexJb9vpVnIwSD06dUdVi3NMWbEcDx8UHJdL54/h0H97WDV0hzDBvXHjeuCn1Eejwc/303o2bUT2rWygNPEcXj9OkEgz+uEeMyeMQ3dOrZDR6tWGDdmJG5FR/L3p6enwdlpInp264S2LczQu0dXrFzujczMTLHVW+JYMuLZpASjJVVVVYWamlqRrUaNGjAwMMDOnZX7C2BoZ32smtwZy4MiYT0jCLHxHJz632BoqtQQmt++qwGWjeuIFUGRsJwciCkbLmBoZ314j+3Az9Nx1kHojtzO3/q4HwUAHLvxvELqJEkXz5/FxrWrMNFpGgIPHEFzfUPMnjYZqakpQvPH3r8HT/f56D9wMAIPHkXnrj2wwGUGXr4o2hZXL1/Cwwcx0NQUvtrr5KkzcObiNf42bMQosdZN0sIunMWWDasxduJU7Np3GHrNDTBvhhPSimm7BzH34L14AfoOGIRd+w+jU5fuWDRvJl790nbHz14V2Nw8loHFYqFLt55Fzrd101poFNO2Vd25s6FYs9oHTtOcEXz4OAwMDDHVaQJSUoS3vTQ7fzYUa1evhNNUZxw4fAz6BgaY5jQRqcXU9f69u3BfMBcDBw3FwcPH0bW7DVxmTseL58/4eQL8d+Fg0D4s9FyCvQcOoUaNGnB2moicnBx+npnOU1CQX4DtuwMRdOgo9A0MMdN5KjicZACADEsGXbr1wIbNfjhx5hyWLvdBVGQElnt7SbZBxInFEs8mJRgNLq5cuYLLly8X2Q4fPgx7e3vMnz8fe/bsYbKIJZo5qCX2nH2IfRcf4cmbVMzYfAlZOflw7GUqNH87o3qIePQBIVef4k3SZ4TdfYNDV5+itYEOPw8nIwuf0r7xtz5WTfDyQzpuPHhXUdWSmIP7AzBg8DD0GzAYTZrpwXWRF5SUlPDviWNC84cc3Id27TtitOMENGnaDE7OM2FgZIwjwUEC+ZKSPmHtquVYumI1ZOWEj/TVrFULGmxN/lajRk2x10+SDh3Yi34Dh6LPX4Og27QZ5rp7QklJCWdOHRea/0jwfrS17oARY8ZDt0kzTJw6A/qGxjh2+AA/jwabLbD9d/0KWrRqi3oNGgqcK/LmDdyKCofzrHkSrWNltS9wDwYPHY6Bg4agmZ4eFnsthZKSEk4cO8p00cRu/94ADB46DAMGDUGzZnpY5Pm9rseF1/Xg/n1o36EjHMdPQNNmzeA8YxaMjI0RfKDwM8rj8XBg315MmjwF3br3gL6BAZatWIXkpCRcCbsEAEhLS8Ob168xbuIk6BsYoHFjXcyc44LsrCy8eF4YDNdRUcHwv0fAxNQM9erVh1U7awyzH4F7d+5UTMMQkTEaXHTp0kXoNmDAAHh7e2PdunXYvHkzk0UslrycDFo018bl+2/4aTwecPn+G7Q1qiv0mMjHH9BCT4s/dKKrowLbNro4d0v4YmLycjL4u5sRAi88FH8FKlheXi6ePn6ENlbt+GkyMjJoY2WNB7H3hR7zMPY+2lhZC6S1s+6AB7Ex/NdcLhdLF7thtON4NG3WvNjr792zE726WsPh78HYH7hbqu5EysvLw7Mnj9C6rWDbtWrbDnEPYoQeE/cgBq3aCLZd23bti82fmsJBxH/X0XfA4CLp/6xYgsVLfaCopFTOmkifvNxcPH4Uh3bW7flpMjIyaNeuPWJj7jFYMvHLyyusq1U7wbpatbNGbMx9ocfExtyH1S9tAwDW7Tvw879/9w4cTrJAntq1a8PU3JyfR1VVFbpNmuDfUyeR9e0b8vPzcfRQCNTVNWBsbCL0uklJn3D50kW0at2m7BWuaNVsWKRSr3PRpUsXzJ49u8Q8OTk5At1rAMDj5oMlI9mqsevUgJysDJLSvgmkJ6V9g0EDNaHHhFx9Co06NRC2xh4sFiAvJ4sdZ2LwT8gtofn/staDqrIi9l98JPbyV7T0tHQUFBRAXZ0tkK6moYGEhFdCj0nhcKCurvFbfjZSUjj81/v27IKsrCyGjxhd7LWHjxgNAyNj1Kmjggcx97B18wZwkjmYPc+1HDWqOBnpaSgoKIDab22hrq6BNwnCA9PUFA7UNX5rO3U2Un9pu1+dO3MKNWvVROduNvw0Ho8Hn6WL8dfg4TA0NsXHD+/LWRPpk/a97TV+a0sNDQ3Exwt/30qrtLTCuv7+vtHQYCOhmNWUOZyi7zMNNhspHM73/YXDGsLO+SMPi8XCtp17MGemMzpYtYKMjAzU1NWxZftO1FFRETjObb4Lrl25jOzsbHTu2g2e3v8re4UrmhQNaYhDpQ6DMjIyoPLbm+t3Pj4+UFFREdjyX16qoBKKppNZA8y3b4tZWy7DekYQ7Jedgl2bJnAbYSU0v6OtCc7fTsDH1K8VXFLp8ORRHEIO7oPH0hUlrpcycsxYtGrdFs31DTB42N+Y6TIfh0OCkJubW4GlrdxCTx1Hz979oKioyE87GhKEb9++YvTYiQyWjFR1PB4PPsu9oa6hAf/AIOw7eAjduttg1vSpSE5OEsg7z9UdBw4dw/rNfnj39i3Wrl7JUKnJn1Tanou8vDz8888/sLIS/sX7g7u7O1xcXATStIZtl2TRAACcz1nIL+BCS01w7F5LrSYSf+vN+MHLoT0OXn6MgPOFwxxxCSmoqSiPLTNtsCo4Cjzez7yNtGqju2Uj/P2/0xKrQ0VSVVOFrKwsUlMFfzmnpaRAQ4Mt9BgNNrvIZM+0FA4///17d5CWmoqBfXrw9xcUFGDTutUIDtqLE6HCg0wTM3MU5Ofj44f3aKzbpDzVqhAqqmqQlZUtMnkzNTUF6sW0nboGu8gkvLRUjtD8Mffu4M3reCxZ8Y9A+t3b0Yh7EAObDi0F0ic72sOmd18sWrKiLNWRKmrf2/73yZspKSlgs4W3vbRSUyus6+/vm5QUDjSKqSubXfR9lsL5mZ/N1gQApKakCEy2TknhwMDACAAQHRWJG9eu4lp4NJSVlQEARsYmiIwIx+mTJzB+4uRfrqcJNlsTTZo2hYqKCsY7jMKkKVOLnchdqUjRkIY4MBpcDB48WGh6RkYG4uLiwGKxcOPGjRLPoaioKPBrC4DEh0QAIC+fi3vPP6GbZUOcjih8ciuLBXSzbIhtp4SPa9dQlAP31wgCAJfL+34sC7xf9o3paYKkjCycja4aD3eTl1eAgZExbkVFosv3rncul4tb0ZEYZj9S6DGm5pa4FR2Jv0c58NOiIyNgZm4BALDr+1eRORmzp01C775/od+AQcWW5dnTJ/yuV2kgLy8PfUNj3LkVhU5dCwMpLpeLu7eiMGjYCKHHmJhZ4O6tSAwfOYafdisqAiZmFkXynjl5DAZGxtDTNxRInzXPHROnzOC/5nCSMG+GE7xWrIGxiZk4qlbpySsowMjYBFGREeje4+f7NioqAn+XMBQnjeTlv9c1KgLdfqlrdFQk7Iu5u8rcwhLRkREYNebnmkWREeEwt7AEANRv0ABstiaiIiNgYFgYTGRmZuJhbCyGDS9872ZnZwEAZGQEex9lZFjgcbnFlpf7fV+etPRAUnBRcYob8mjYsCGGDBmCUaNG/XFYhEmbjt/Fzrm2uPM8CbefJmL6wBaoqSiPvRfjAAC75triQ0omPANuAgBCo15h5uCWiHmZhOgniWhWTxWeDu0RGvWKH2QAhUGKQ08TBF16hAIuT+i1pdGI0WOxzNMdRsamMDY1Q8iBvcjOykLf74HA0sVu0NTSwrSZhT1R9iPGYOokRwTt3YMOnbrg4vlQPH70EG4eSwEAKqqqUFFVFbiGrJwcNNhsfo/Eg5j7iHsYi1at26JmrVp4EHsfG9esQu8+/VGnTuV9b/1u+EgH+CxdBAMjExiZmOLwwf3IyspCn/4DAQDLvdzB1tSC0/Q5AIChf4/GTKdxCN4fAOuOnRF24SyePo7D/IVLBM77NTMTV8MuwHl20TtBtHUEJybXqFnYS1e/fkNoaesUyV9VjXEcB4+FrjAxMYWpmTn27wtEVlYWBg4S/uNImo12GAvPRW4wNjGFqak5DuwvrOuAgYV1XezuCi0tLcycMxcAMGL0GEwa54C9Af7o1Lkrzp89g0dxcfBY4g2g8EfTyDEO2LVjGxo11kX9+vXh57sJmlpa/ADG3KIF6tSpA4+Fbpg8xRlKSoo4duQw3r97j46duwIAbly/htQUDkxMzVCzZk28fPEC69f+A8sWLVGvfoOKbyjyR4wGF5X5NtPSOHL9GdgqNeA52hra6jUR+zIZAzyOIym9cFikoVZtgZ6KlQcLhz68HDqgnoYyOBnfcCbqFZYEhguct3uLRmikXadK3CXyq562dkhPS8XOrZuRksJBcwNDrN+ynT/MkZj4ESyZn9G9uWULeK9Yje1bNmGb7wY0bNQYq9dtRjO94u8K+Z28ggIung/Frm1bkJeXi7r16uPvUQ4YMWasuKsnUT162SE9PQ3+232RmsKBnr4h1mzaxh/m+JT4EaxffhmZWbSA5/9WYdfWzdjptxENGjbG8jWb0PS3tgu7cBY8Hg89bPtUaH2kSW+7PkhLTYWf7yZwOMkwMDSC3/ZdxQ4VSDNbuz5IS0vFVt/NSPle1y3bdvLrmvjxg0APg2WLllixag22bN4A343r0aixLtZt8oVec31+nrHjJyIrKwv/W+KJL18+w7JlK2zZtpPf46ympgbfbTuxZdMGOE1wRH5+Pprq6WH95i0wMCzsTfsRcKxZvRJ5ubnQ1tFBd5teGD9hUgW2TjlVswmdLB6PV+l+Gl+7dg1fv36FtbU11NSE33lRkhp26yVQqurjw9GZTBdBauXmF9+NS0qmUlOe6SJItd+HXEnp1ZSX/Bd/jQHimQuYddJJLOeRNEZ7LlatWoXMzEwsW7YMQOGsYTs7O1y4cAEAoKWlhbCwMJiYCL/XmRBCCJEK1azngtEZJiEhITA1/bma5ZEjR3D9+nXcuHEDHA4HrVu3xtKlSxksISGEEEJExWjPRXx8PMzNzfmvQ0NDMXToUHToUPisjcWLF2PYsGFMFY8QQggRj2p2twijtc3Pzxe4jTQiIgLt2/9cJrZevXrgcISvKEgIIYRIDXpwWcVp1qwZrl8vfAz0mzdv8OzZM3Tu3Jm//927d0WW3SWEEEJI5cbosIizszOmT5+OGzduIDIyEtbW1jA2Nubvv3z5Mlq0aMFgCQkhhJDyK+kRBVURo8HFpEmTICsri9OnT6Nz587w8vIS2P/hwweMHz+eodIRQggh4lHdgotKuc5FedE6F+VD61yUHa1zUXa0zkX50DoXZVcR61zUGiqeRSO/HhknlvNIWqV9cBkhhBBSZVSvjgtmgwsZGZk/dhWxWCzk5+dXUIkIIYQQ8atuwyKMBhfHjx8vdl9ERAQ2bdrEf/IdIYQQQqQDo8HFgAEDiqQ9ffoUbm5uOH36NEaNGgVvb28GSkYIIYSIT3Xruag0S4Z9+PABkyZNgpmZGfLz83H//n0EBgaicePGTBeNEEIIKRcWiyWWTVowHlxkZGTA1dUVenp6iIuLQ1hYGE6fPi3wzBFCCCFEmlW34ILRYZHVq1dj1apV0NHRwcGDB4UOkxBCCCFEujDac+Hm5obs7Gzo6ekhMDAQgwcPFroRQgghUo0lpq0MtmzZAl1dXSgpKcHKygrR0dEl5t+wYQMMDAxQo0YNNGzYEHPmzEF2drZI12S058LBwUGqunkIIYSQsmDquy4kJAQuLi7Ytm0brKyssGHDBtja2uLp06fQ0tIqkv/AgQNwc3ODv78/2rdvj2fPnmHs2LFgsVhYt25dqa/LaHAREBDA5OUJIYSQKm3dunWYNGkSxo0rXNlz27ZtOHPmDPz9/eHm5lYkf3h4ODp06ICRI0cCAHR1dTFixAhERUWJdF1Gg4vSDHmwWCwcPXq0AkpDCCGESIa4ei5ycnKQk5MjkKaoqAhFRcUieXNzc3Hnzh24u7vz02RkZGBjY4OIiAih52/fvj3279+P6OhotG3bFq9evUJoaCjGjBkjUjkZDS5UVFSYvDwhhBBSIcQVXPj4+GDp0qUCaV5eXliyZEmRvBwOBwUFBdDW1hZI19bWxpMnT4Sef+TIkeBwOOjYsSN4PB7y8/MxZcoULFy4UKRyMhpc7Nkjnge5EEIIIdWBu7s7XFxcBNKE9VqU1dWrV7FixQr4+fnBysoKL168wKxZs7Bs2TJ4eHiU+jz04DJCCCFEwsTVc1HcEIgwbDYbsrKy+PTpk0D6p0+foKOjI/QYDw8PjBkzBhMnTgQAmJmZ4evXr5g8eTIWLVoEGZnS3WTK+CJahBBCSJXHwK2oCgoKaNWqFcLCwvhpXC4XYWFhsLa2FnrMt2/figQQsrKyAAAej1fqa1PPBSGEEFJFubi4wNHREa1bt0bbtm2xYcMGfP36lX/3iIODA+rXrw8fHx8AQP/+/bFu3Tq0aNGCPyzi4eGB/v3784OM0qDgghBCCJEwpta5sLe3R3JyMjw9PZGYmAhLS0ucO3eOP8nzzZs3Aj0VixcvBovFwuLFi/H+/Xtoamqif//+WL58uUjXZfFE6eeQEjXs1jNdBKn24ehMposgtXLzuUwXQWqp1JRnughSjVv1/pRXmJrykv/i1xwXIpbzJO+xF8t5JI16LgghhBAJq26rUdOETkIIIYSIFfVcEEIIIZJWvTouKLgghBBCJI2GRQghhBBCyqFK9lx8PEZ3O5RH3R6lX+KVCEq7voLpIkgtLpfudigPar7Krbr1XFTJ4IIQQgipTKpbcEHDIoQQQggRK+q5IIQQQiSsuvVcUHBBCCGESFr1ii1oWIQQQggh4kU9F4QQQoiE0bAIIYQQQsSKggtCCCGEiFV1Cy5ozgUhhBBCxIp6LgghhBBJq14dFxRcEEIIIZJGwyKEEEIIIeVAPReEEEKIhFW3ngsKLgghhBAJq27BBQ2LEEIIIUSsKm1wkZ2djTVr1jBdDEIIIaTcWCyWWDZpwWhwkZycjH///RcXLlxAQUEBACAvLw8bN26Erq4uVq5cyWTxCCGEEPFgiWmTEozNufjvv//Qr18/fP78GSwWC61bt8aePXswcOBAyMnJYcmSJXB0dGSqeIQQQggpI8Z6LhYvXow+ffogNjYWLi4uuHXrFgYNGoQVK1bg0aNHmDJlCmrUqMFU8QghhBCxoWGRCvLgwQMsXrwYpqam8Pb2BovFwurVqzF06FCmikQIIYRIRHULLhgbFklLSwObzQYA1KhRAzVr1oSpqSlTxSGEEEIkRoriArFgdJ2LR48eITExEQDA4/Hw9OlTfP36VSCPubk5E0UjhBBCSBkxGlz06NEDPB6P/7pfv34ACruPeDweWCwW/y4SQgghRFpJ05CGODAWXMTHxzN1aUIIIaRCVbPYgrngonHjxkxdmhBCCCESxNjdIqtXr0ZWVhb/9c2bN5GTk8N//eXLF0ybNo2JohFCCCFiVd3uFmEsuHB3d8eXL1/4r+3s7PD+/Xv+62/fvmH79u1MFI0QQggRKxZLPJu0YCy4+HUip7DXhBBCCJFO9Mh1QgghRMJkZKSo20EMKLgghBBCJEyahjTEgdHgYteuXVBWVgYA5OfnIyAggL9q56/zMQghhBAiPRgLLho1aoSdO3fyX+vo6GDfvn1F8lQ2h4MPICjQHykpHDTXN8Bc10UwMSt+FdGwC+ew3W8zPn54j4aNGsN5lgs6dOoiNO/K/y3B8SOHMHueG0aMduCn79m5DTdvXMezZ08gLyePsP+ixF4vpjgNboc5ozpBW10ZD14kwmXdadx+/E5oXjlZGcx36IrRfVqgHrsOnr3hYLHfOVyMes7PM29MFwzsagL9RprIys1D1IM3WOR3Ds/fcCqqSpVS8IEgBO7ZDQ4nGfoGhnBb6AGzarT6bcjBIAQG7EYKhwN9A0O4ui+GaQmf24vnz8HPdyM+fHiPRo0aY+aceejU+efnNuzSBRw5FIzHj+KQkZGB4MPHYWBoJPRcPB4P06dORvjNG1i3wRfdetiIvX6Sdig4CHu/t19zfUMs+FP7XTiHrb4b+X/3Zs6Zh47f/+7l5eVhq+9G/HfjGt6/ewfl2sqwsmqPGbNdoKmlDQD48P4ddu3YiltRkUhJ4YCtqYU+fftjwuQpkJdXqJA6i5s03ekhDoxN6ExISEB8fPwft8rk4vmz2Lh2FSY4TUPgwSPQ0zfErGmTkZqaIjR/7P178HCfj/4DB2Nv8FF07tYDC+bMwMsXz4vkvXr5Eh7GxkBTU6vIvry8PPToaYshw+zFXicmDe1hhlUz+2C5fxisx21B7IuPOLV+HDTVagnNv8SpJyYObAOXdafRYtQG7DoRjZCVo2GhX5efp1OLJth2NBJdJm9Fv1n+kJOTwb8bxqGmknxFVavSOXc2FGtW+8BpmnPhl6CBIaY6TUBKivD3bVVz/lwo1v6zEk5TnHHg0DHo6xtgmtNEpBZT//v378LddS4GDh6Kg4ePo2t3G7jMmo4Xz5/x82RlZcGyRSvMnDPvj9cP2hco1V8sF86FYt0/KzF5ijOCQo5B38AA06cU334x9+9iketcDBw0FAcOFbbf3F/aLzs7G08eP8JEp2kICjmKNes2IyEhHnNm/lx6ICE+HlwuFws9l+LQ8X8xd747jh4Oge/G9RVSZ0mgu0VIsQ7uC8CAwcPQf+BgNG2mB7fFXlBSUsLpE8eE5g85sA/t2nfEmLET0KRpM0xxngkDI2McDg4SyJf06RPWrFwO7xWrISdXtDNp8rQZGDHGEc309CVSL6bM/Lsj9py6hX1n7uJJQhJmrD6JrJxcOPZrJTT/SNsWWB14DecjniHhQxp2Ho/C+fCnmDWiIz/PAJcA7A+9i8fxSXjwIhGT/3cUjXTU0MKwfkVVq9LZF7gHg4cOx8BBQ9BMTw+LvZZCSUkJJ44dZbpoFWL/3gAMHjIMAwYNQbNmeljkuRRKNZRw4rjw+h/cvw/tO3SE47gJaNq0GZxnzIKRsTGCD/783PbrPwBOU53Rrp11idd++uQx9gXuwZJly8Vap4q0f28ABg0Zhr8GDkHTZnpY6FHYfidPFNN+Qftg3aEjHMYV/t2bNn0WDI2Mcej7373atWvDb4c/etnaQbdJU5hZWMJ1oQceP4rDx48fAADtO3bCkmU+sG7fEQ0aNESXbt0xxnE8roRdrLB6i1t1W+eCsWGRTZs2lSrfzJkzJVyS0snLy8WTx4/gOH4SP01GRgZtrKzxIPa+0GMexN7HiNFjBdLaWXfAtauX+a+5XC6WLHbDaMfxaKrXXBJFr5Tk5WTRwqAe/tl3lZ/G4/Fw+dZLtDUVPhymoCCH7Nw8gbSs3Dy0N9ct9jp1aikCANI+ZxWbpyrLy83F40dxmDDJiZ8mIyODdu3aIzbmHoMlqxh5eYX1Hz9hMj9NRkYGVu2sERtzX+gxsTH3MdphrECadfsOuHI5TKRrZ2Vlwd11HtwWeYLN1hS16JVC4d+9OIybKNh+ba2s8aCk9hszViDNun0HXL1SfPtlZn4Bi8VC7dp1SsxTR0VFpPIT5jAWXKxfL9i99fbtW9StW1fglzuLxfpjcJGTkyOwsicA5HDloKioKL7CAkhPS0dBQQHUNdgC6eoaGnid8EroMSkcDtQ1NH7Lz0YK5+f4/949uyArKwv7kaPFWt7Kjq1aE3JyskhKzRRIT0rNhEFj4X+IL0U9x8y/O+K/+wl49T4V3Vo3w4AuJpCVEd4Bx2Kx8M/sfgiPScCjV5/EXgdpkJaehoKCAmj89j7U0NBAfLzw921VkpaW9v1z+3v92UgoZtiVI+Rzq/Hb57Y01q72gYVlC3Tr3kO0Qlci6WnFvX+Kb7/S/N37VU5ODjatXwNbu778Cf6/e/vmNYIP7sdslwVlqEXlIE29DuJQaR5cVrt2bVy7dg1NmzYV6Tw+Pj5YunSpQJrrQg+4LfYqdxkl7fGjOIQc2Ie9B49WuzdeWczb8C/83AYh5uAc8Hg8vHqfir1n7hY7jLJh7l8waaqNHlNopVdSsa5euYzo6CgEHxY+ZEoK5eXlwW3ebPB4gPviJULzJH36hOlTJ8GmZ28MHjq8YgsoRtXtT7zUr3Ph7u4OFxcXgbQsrvirpaqmCllZWaSmCEbfqSkpUGezhR6jwWYXmfSUmsKBxvf89+/eQVpqKgbY/fxlU1BQgE3rViMkaC9OnL0k5lpUHpz0b8jPL4CWuuAvFS11ZSSmCr8NmZP+FcPd9kNRQQ4adWriA+cz/jfNFvHvU4vkXe/SH306GMBm2k68T/4skTpIAzVVNcjKyhaZvJmSksK/7bsqU1NT+/65/b3+HGhoCK8/W8jnNuWXz21p3IqOxLu3b9C5fVuB9HkuM9GiZSvs2rOvmCMrF1W14t4/nGLfP3/6u/dDXl4e3ObPwcePH7BtV4DQXovkpE9wmugAC4sWWOzlXc7akIok9RM6FRUVUadOHYFN3EMiACAvrwBDI2Pcio7kp3G5XNyKjoSZuaXQY8zMLXH7l/wAEB0ZATNzCwBAn35/IejwCewLOcbfNDW1MNpxPDZu3SnslFVGXn4B7j39gG6t9PhpLBYL3Vo3Q/TDNyUem5Objw+cz5CTlcHArqb498Zjgf3rXfrjry7G6D1jN15/TJNI+aWFvIICjIxNEBUZwU/jcrmIioqAuUULBktWMeTlv9c/SrD+0ZGRMLewFHqMuYUlon/JDwCREeHF5hdm3IRJOHT0JIIPH+dvADB3gRuWLvMRuR5MKfy7Z4Jbv7XfrahImInQflGRgu33I7B4+/o1tu7YA1VVtSLnSfr0CZMnOMDIyARey1ZAppjhT2lBEzpJsUaMGQtvD3cYGZvC2NQMwUF7kZ2VhX4DBgEAlix2g6aWFpxnFvak2I8cgykTHRG0dw86dOqCi+dC8fjRQ7h7Fg7jqKiqQkVVVeAacnJyUNdgo7FuE35a4scP+JyRgcTEj+ByC/DsSeGXaYNGjVCzpvDbNqXBpuD/sHPxUNx58g63H73DdPsOqKmkgL3/3gUA7PIYig/Jn+G57QIAoI1xA9TTVEHM8w+or6mCRRN6QIbFwrqg6/xzbpj3F+x7WmCY635kfsuB9veekYzMbGTn5ld8JSuBMY7j4LHQFSYmpjA1M8f+fYHIysrCwEGDmS5ahRjtMBaei9xg/L3+B77Xf8DAwvovXugKLS0tzJw9FwAwYvQYTBrngL2B/ujUqSvOnzuDR3Fx8Pjll3NGRjoSP35EUlISACAhoXCYV4PNBputyd9+V1enHuo3aCDpKovVaIex8FrsBiPj7+23v7D9/vrefp4LXaGprYUZs76336gxmDTeAfsC/dGxc1dcOFvYfos8C9svLy8PrnNn4cnjR9jguw0F3AJwOMkAABUVFcjLK/ADi7p162H2XFekpf3snZTWybFSFBeIBWPBxefPgl3VLBYLmZmZRdLr1Cl+9nBF62lrh/S0VOzYupm/GM8Gv+387tVPHz9ChvUzuja3bIFlK1Zj25ZN2Lp5Axo2aozV6zejmYh3hezw88WZ0yf4r8f8PQQA4LczAK3atC3mqMrvSNgDsFVrwXOSDbTVayP2+UcMcNmDpLTCSZ4NtVXB5f58oJ2igjy8JvdEk3pqyMzKxfmIp5jgfQgZmdn8PE6D2wEALvpNErjWpP8dwf7QuxVQq8qnt10fpKWmws93EzicZBgYGsFv+y6RuvmlmW3vwvpv3bIZKd/rv2XbTn79Ez9+gMwvf/ktLVtixco12OK7Ab4b16NRY12s2+gLveY/bwW/duUyvDwW8l+7zS/8QeE01RlTps2ooJpVjF69+yAtLRXb/ArbT9/ACJu37uT/3UtM/ADWL8/NsLBsieUr12Dr5g3Ysmk9GjXSxdpf2i856RP/jrkRwwYKXGv77kC0bmOFyMibePvmNd6+eQ27noKLDt6JfSLB2hJxYfEYehypjIyMQBcPj8cT+rqgoEDkc6dniX4M+aluDw+miyC10q6vYLoIUuvXQJKIjpqv7JQVJd+t0GrZFbGc545HN7GcR9IY67m4ckU8DU0IIYRUdjQsUkFev34Ne3t7iUy+JIQQQghzGJt+O27cOGRkZDB1eUIIIaTC0N0iFYShqR6EEEJIhZOiuEAsGL0VVZqiMEIIIaSsqtv3HaPBRY8ePYQ+BfRXd+9Wz9sHCSGEEGnFaHBha2tb7INqCCGEkKqimnVcMBtczJ8/H1paWkwWgRBCCJG46jYswtjdItWtoQkhhJDqotLcLcLhFD5ttDo8qZEQQkj1Ut1+TzPWcxEfHw95eXk4OzuDzWZDW1sb2traYLPZmD59OtLT05kqGiGEECJWtM5FBalduzbatWuH9+/fY9SoUTAyMgIAPHr0CAEBAQgLC0N4eDjU1Io+ipcQQgghlRdjwYW3tzcUFBTw8uVLaGtrF9nXq1cveHt7Y/369QyVkBBCCBEPKep0EAvGhkVOnDiBNWvWFAksAEBHRwerV6/G8ePHGSgZIYQQIl7VbViEseDi48ePMDExKXa/qakpEhMTK7BEhBBCCBEHxoILNpuNhISEYvfHx8dDXV294gpECCGESAj1XFQQW1tbLFq0CLm5uUX25eTkwMPDA71792agZIQQQoh4sVji2aQFoxM6W7dujebNm8PZ2RmGhobg8Xh4/Pgx/Pz8kJOTg3379jFVPEIIIURspKnXQRwY67lo0KABIiIiYGxsDHd3dwwcOBCDBg3CokWLYGxsjJs3b6Jhw4ZMFY8QQgipErZs2QJdXV0oKSnBysoK0dHRJeZPT0+Hs7Mz6tatC0VFRejr6yM0NFSkazL6bJEmTZrg7NmzSEtLw/PnzwEAenp6NNeCEEJIlcJUx0VISAhcXFywbds2WFlZYcOGDbC1tcXTp0+FPtsrNzcXPXv2hJaWFo4cOYL69evj9evXUFVVFem6jAYXP6ipqaFt27ZMF4MQQgiRCKaGRdatW4dJkyZh3LhxAIBt27bhzJkz8Pf3h5ubW5H8/v7+SE1NRXh4OOTl5QEAurq6Il+XsWERQgghhIgmJycHnz9/FthycnKE5s3NzcWdO3dgY2PDT5ORkYGNjQ0iIiKEHnPq1ClYW1vD2dkZ2traMDU1xYoVK1BQUCBSOSm4IIQQQiRMXHeL+Pj4QEVFRWDz8fERek0Oh4OCgoIii1Vqa2sXu47Uq1evcOTIERQUFCA0NBQeHh5Yu3Yt/ve//4lU30oxLEIIIYRUZTJiGhZxd3eHi4uLQJqioqJYzg0AXC4XWlpa2LFjB2RlZdGqVSu8f/8e//zzD7y8vEp9HgouCCGEECmhqKhY6mCCzWZDVlYWnz59Ekj/9OkTdHR0hB5Tt25dyMvLQ1ZWlp9mZGSExMRE5ObmQkFBoVTXpmERQgghRMKYWERLQUEBrVq1QlhYGD+Ny+UiLCwM1tbWQo/p0KEDXrx4AS6Xy0979uwZ6tatW+rAAqDgghBCCJE4ppb/dnFxwc6dOxEYGIjHjx9j6tSp+Pr1K//uEQcHB7i7u/PzT506FampqZg1axaePXuGM2fOYMWKFXB2dhbpujQsQgghhEiYDEPrXNjb2yM5ORmenp5ITEyEpaUlzp07x5/k+ebNG8jI/OxnaNiwIc6fP485c+bA3Nwc9evXx6xZs+Dq6irSdVk8Ho/3p0wtW7ZEWFgY1NTU4O3tjXnz5qFmzZoiVrHipGeJdssMEVS3hwfTRZBaaddXMF0EqcXl/vFPESkBNV/ZKStK/pvfbmuUWM5zdqqVWM4jaaUaFnn8+DG+fv0KAFi6dCkyMzMlWihCCCGkKqluT0Ut1bCIpaUlxo0bh44dO4LH42HNmjVQVlYWmtfT01OsBSSEEEKknRTFBWJRqmGRp0+fwsvLCy9fvsTdu3dhbGwMObmicQmLxcLdu3clUlBR0LBI+SjJy/45ExFKrc10posgtdJu+TJdBFJNKVXA7MO+20t+WFhpnXGSjkdllKpJDQwMEBwcDKBw6dCwsDChDzwhhBBCSFEsVK+uC5HjtV/vfS2vDx8+oF69emI7HyGEEFIZMXW3CFNEDi5OnTpV4v6//vqr1OcyMTHBli1bMHLkSFGLQQghhJBKSuTgYuDAgcXuY7FYIj05bfny5XBycsLx48exfft2qKuri1ocQgghpNKTpjs9xEHkFTq5XG6xm6iPZJ02bRpiY2ORkpICY2NjnD59WtTiEEIIIZUeE8t/M4nxFTqbNGmCy5cvw9fXF4MHD4aRkVGRO1Eqwx0ohBBCCCmdMgUXYWFhCAsLQ1JSUpEJnv7+/iKf7/Xr1zh27BjU1NQwYMAAobe5EkIIIdJKXI9clxYif4svXboU3t7eaN26NerWrVvucaSdO3di7ty5sLGxQVxcHDQ1Nct1PkIIIaSyqWaxhejBxbZt2xAQEIAxY8aU++K9e/dGdHQ0fH194eDgUO7zEUIIIZVRdZvQKXJwkZubi/bt24vl4gUFBYiNjUWDBg3Ecj5CCCGEME/ku0UmTpyIAwcOiOXily9fhoKCgljORQghhFRWdLfIH2RnZ2PHjh24dOkSzM3NIS8vL7B/3bp1pT5XKR5rQgghhEg9mtD5B7GxsbC0tAQAPHz4UGBfdRtTIoQQQkhRIgcXV65cEWsBdu3aVezj23+YOXOmWK9JCCGEVKTq9tOb8QUltm3bBlnZ4h/xzWKxKLgghBAi1apbz77IwcXgwYNL3H/s2DGRznf79m16fDshhBBShYgcXKioqIjt4tUtkiOEEFI90SPX/2DPnj1iuzjdLUIIIaQ6qG4/pss95+LatWv4+vUrrK2toaamJtKxXl5ef5zMSQghhBDpUurgYtWqVcjMzMSyZcsAFPY62NnZ4cKFCwAALS0thIWFwcTEpNQXHzRoEF68ePHHfObm5qU+JyGEEFLZVLOOi9IHFyEhIXB1deW/PnLkCK5fv44bN27AyMgIDg4OWLp0KQ4dOlTqi1taWoLFYgkdHvmRzmKxUFBQUOpzEkIIIZUNDYsUIz4+XqAHITQ0FEOHDkWHDh0AAIsXL8awYcNEunh8fLxI+QkhhBBpRBM6i5Gfnw9FRUX+64iICMyePZv/ul69euBwOCJdvHHjxiLlJ4QQQkjlV+rgolmzZrh+/TqaNm2KN2/e4NmzZ+jcuTN//7t376ChoSHSxd+8eVOqfI0aNRLpvIQQQkhlQsMixXB2dsb06dNx48YNREZGwtraGsbGxvz9ly9fRosWLUS6uK6urtAG/zHXAij8B8nPzxfpvIQQQkhlUr1CCxGCi0mTJkFWVhanT59G586d4eXlJbD/w4cPGD9+vEgXv3fvntB0Ho+H4OBgbNq0iW5VJYQQQqQMi1fJVrK6dOkS3Nzc8OzZM7i4uGDu3LmoXbu2SOdIz6K7S8pDSb74Z72Qkqm1mc50EaRW2i1fpotAqimlCnjK1sSQh3/OVAq77E3Fch5JY/zBZT/cvXsXrq6uuHHjBiZOnIjQ0FB65gghhJAqoZpNuYAM0wV4+fIl7O3t0bZtW2hqauLRo0fw9fWlwIIQQgiRUowGF9OmTYOxsTEyMjJw+/ZtHDhwAE2bNmWySIQQQojYsVgssWzSgtFhkW3btkFJSQlJSUklTga9e/duBZaKEEIIES8pigvEQuTg4sqVK+jWrZvQfVu2bIGzs3Opz/X7HSfS6nDwAQQF+iMlhYPm+gaY67oIJmbFPw8l7MI5bPfbjI8f3qNho8ZwnuWCDp26CM278n9LcPzIIcye54YRox0kVQWpEHwgCIF7doPDSYa+gSHcFnrArBo/d8ZpeGfMcewBbY06ePDsPVxWHcbtuNdC88rJyWD++F4Y3c8K9bRU8ez1JyzeeBIXwx8L5KunqYL/zRqAXh1MUFNJHi/fcuC0ZD/uPirdmjRVFb33yo7arnoSObgYPHgwLl26hFatWgmkb9y4ER4eHtUuuLh4/iw2rl0F10VeMDEzR3DQPsyaNhmHTp6BunrRRcVi79+Dh/t8TJ0xGx07d8X5s2ewYM4M7A0+imZ6zQXyXr18CQ9jY6CpSfNPzp0NxZrVPljstRRmZhYI2heIqU4TcPLfcyIv3lYVDO3VEqvmDsKM5SG49TAB00d2wyk/Z1gM9EZyWmaR/Eum9ceIvm0wbdkBPI3/hJ7tjRCydhK6jV2HmKfvAACqtWvgcoALrt16joHT/ZCclgm9RppI+/ytoqtXqdB7r+yo7X6SqWZdFyLPufjnn39gZ2eHJ0+e8NPWrl0LT09PnDlzRmwF+/z5M7Zu3YrWrVuL7ZyScHBfAAYMHob+AwejaTM9uC32gpKSEk6fOCY0f8iBfWjXviPGjJ2AJk2bYYrzTBgYGeNwcJBAvqRPn7Bm5XJ4r1gNOblKc1MPY/YF7sHgocMxcNAQNNPTw2KvpVBSUsKJY0eZLhojZo7ujj3HwrHvVCSevErEjOXByMrOheNAa6H5R/Zri9W7L+D8f4+Q8D4FOw//h/M3H2HWmO78PHPH9cS7xDQ4LdmP23Gv8fpDCsIinyD+nWjL+lc19N4rO2q7n1gs8WzSQuTgYuLEiZg3bx5sbGyQkJCAVatWwdvbG6GhoejUqVO5C3TlyhWMGTMGdevWxbJly2BlZVXuc0pKXl4unjx+hLZW7fhpMjIyaGNljQex94Ue8yD2PtpYCX4BtLPugAexMfzXXC4XSxa7YbTjeDT9rTejOsrLzcXjR3FoZ92enyYjI4N27dojNkb4QmxVmbycLFoYNcTlqKf8NB6Ph8tRT9HWvInQYxTk5ZCdmyeQlpWdi/YtmvFf9+1ihruP3iBo9Xi8DvNBxEFXjBvU/vdTVSv03is7ajtBNKGzFBYsWICUlBS0bt0aBQUFOH/+PNq1a/fnA4vx/v17BAQEYM+ePUhPT0daWhoOHDiA4cOH/7Exc3JykJOTI5jGlRN4yJqkpKelo6CgAOoabIF0dQ0NvE54JfSYFA4H6r91B6prsJHyy0Pf9u7ZBVlZWdiPHC3+QkuhtPQ0FBQUFOlG1dDQQHy88HauythqypCTk0VS6heB9KSUzzDQ1RZ6zKWIx5g5ujv+u/sCr95y0K2tAQZ0t4Ss7M/PV5P6bEwa1gmb9l/G6t0X0MqkMdYuGIrc/AIEnY6SaJ0qK3rvlR21XfVWquBi06ZNRdLq16+PmjVronPnzoiOjkZ0dDQAYObMmaW++NGjR7F7925cv34ddnZ2WLt2Lezs7FCrVi2YmZmVKkrz8fHB0qVLBdJcF3rAbbF0zud4/CgOIQf2Ye/Bo1IVpZLKbd4/R+DnMQIxxzzA4/Hw6h0He09FwnHAr71uLNx99AZevqcBADFP38FEry4mDe1YbYMLQsSF8UWlKlipgov169cLTZeVlcXNmzdx8+ZNAIXdPqIEF/b29nB1dUVISIjIS3z/4O7uDhcXF4G0LG7FzFFQVVOFrKwsUlMEx6RTU1KgzmYLPUaDzUZqSspv+TnQ+J7//t07SEtNxQC7Hvz9BQUF2LRuNUKC9uLE2UtirkXlp6aqBllZWaT81m4pKSlgF9POVRknLRP5+QXQUhf8zGhp1EFiyudijxnushOKCnLQUKmFD8kZ+N/MAYh//7NNEzmf8fhVosBxT+ITMbCHpdjrIC3ovVd21HaCqtuPxVIFU/Hx8aXaXr0SratrwoQJ2LJlC3r37o1t27YhLS1N5AooKiqiTp06AltFDIkAgLy8AgyNjHErOpKfxuVycSs6EmbmlkKPMTO3xO1f8gNAdGQEzMwtAAB9+v2FoMMnsC/kGH/T1NTCaMfx2Lh1p8TqUpnJKyjAyNgEUZER/DQul4uoqAiYW4j2JN6qIC+/APcev0U3KwN+GovFQre2+oiOjS/x2JzcfHxIzoCcnAwG9rDEv1dj+fsi7r+CfmPBO5OaN9LCm4+p4q2AFKH3XtlR21VvZe6pyc3NxdOnT8v1OPTt27fj48ePmDx5Mg4ePIi6detiwIAB4PF44HK5ZT5vRRoxZixOHjuCM6dOIP7VS6xavhTZWVnoN2AQAGDJYjds2bSOn99+5BhEhP+HoL17kBD/Cju3+uLxo4cY9vcoAICKqiqa6TUX2OTk5KCuwUZjXeGT9aqDMY7jcOzIIZw6cRyvXr7E/7yXICsrCwMHDWa6aIzYtP8yxg1qj1H9rWDQRBubFtqjZg1F7D1ZGLjuWjYG3jP+4udvY9oYA7pbQLe+Bjq0aIZTvs6QkWFhXcDPnrDN+y+jrVkTzB/fC00bsmHfuzXGD+mA7SHXK7x+lQm998qO2u4nGZZ4Nmkh8vjBt2/fMGPGDAQGBgIAnj17hqZNm2LGjBmoX78+3NzcRDpfjRo14OjoCEdHRzx//hx79uzB7du30aFDB/Tt2xdDhw7F4MGV943Y09YO6Wmp2LF1M1I4HOgbGGKD33ZofJ/k+enjR8iwfsZw5pYtsGzFamzbsglbN29Aw0aNsXr95iJrXBBBve36IC01FX6+m8DhJMPA0Ah+23fxh5OqmyMX7oKtpgzPqX2hrVEbsU/fY4DzFv4kz4Y66uByfz7wWFFRHl7O/dCkPhuZ33Jw/mYcJnjsRUZmFj/PnUdvYD93J7xn/IWFk+2Q8D4F8/85iuCztyu8fpUJvffKjtruJ2kKDMRB5Eeuz5o1Czdv3sSGDRvQu3dvxMbGomnTpjh58iSWLFmCe/dKf4sRl8uFjEzRzhMul4szZ85g9+7dOHv2bJG7Qf6EHrlePvTI9bKjR66XHT1ynTClIh657nLqyZ8zlcK6vwzFch5JE3lY5MSJE/D19UXHjh0FJqiYmJjg5cuXIp1LXl4eSUlJ/Nfz589HamoqZGRk0L9/f5w4cQJv374VtYiEEEJIpVLd1rkQObhITk4W+jj0r1+/ilzx3ztNtm/fjvT0dIE0evQ6IYQQaVfd5lyIHFy0bt1aYJnvHwHFrl27YG0tfOnh0hJxhIYQQgghlZDII00rVqyAnZ0dHj16hPz8fGzcuBGPHj1CeHg4rl27JokyEkIIIVJNikY0xELk4KJjx464f/8+Vq5cCTMzM1y4cAEtW7ZEREQEzMzMRC6Ap6cnatasCaDw9tbly5dDRUVFIM+6deuEHUoIIYRIher2VNQyzZFt1qwZdu4s/4JOnTt3xtOnPx++1L59+yILcUnTBBZCCCFEGFr++w9CQ0MhKysLW1tbgfTz58+Dy+XCzs6u1Oe6evWqqJcnhBBCSCUncjDl5uaGgoKi60jweDy4ubnh2bNn2LhxIx48eCByYTgcDjgczp8zEkIIIVKExRLPJi1EDi6eP38OY2PjIumGhoZ48OAB5s2bh+vXr6N///6lOl96ejqcnZ3BZrOhra0NbW1tsNlsTJ8+vchtqYQQQog0kmGxxLJJC5GHRVRUVPDq1Svo6uoKpL948QLq6uo4deoU4uPjYWpq+sdzpaamwtraGu/fv8eoUaNgZGQEAHj06BECAgIQFhaG8PBwqKmpiVpMQgghhDBE5OBiwIABmD17No4fP45mzZoBKAws5s6di4EDBwIAlJWVcfLkyT+ey9vbGwoKCnj58iW0tbWL7OvVqxe8vb2LfeQ7IYQQIg2kqNNBLEQeFlm9ejVq1aoFQ0NDNGnSBE2aNIGRkRE0NDSwZs0aAICmpiZsbGz+eK4TJ05gzZo1RQILANDR0cHq1atx/PhxUYtICCGEVCrVbYXOMg2LhIeH4+LFi4iJiUGNGjVgbm6Ozp07i3zxjx8/wsTEpNj9pqamSExMFPm8hBBCCGGOSMFFXl4eatSogfv376NXr17o1atXuS7OZrORkJCABg0aCN0fHx8PdXX1cl2DEEIIYZo0TcYUB5GGReTl5dGoUSOht6KWha2tLRYtWoTc3Nwi+3JycuDh4YHevXuL5VqEEEIIU6rbragiD4ssWrQICxcuxL59+8rdq+Dt7Y3WrVujefPmcHZ2hqGhIXg8Hh4/fgw/Pz/k5ORg37595boGIYQQQiqWyMGFr68vXrx4gXr16qFx48aoVauWwP67d++W+lwNGjRAREQEpk2bBnd3d/5TUVksFnr27AlfX180bNhQ1CISQgghlYo0TcYUB5GDix+3m4pLkyZNcPbsWaSlpeH58+cAAD09PZprQQghpMpgoXpFFyIHF15eXpIoB9TU1NC2bdsi6UeOHMHQoUMlck1CCCGkIlS3nosyPagtPT0du3btgru7O1JTUwEUDoe8f/9e5HPl5+fj4cOHePbsmUD6yZMnYWFhgVGjRpWliIQQQghhiMjBRWxsLPT19bFq1SqsWbOG//yPY8eOwd3dXaRzPXz4EHp6erCwsICRkREGDx6MT58+oUuXLhg/fjzs7Ozw8uVLUYtICCGEVCrVbREtkYMLFxcXjB07Fs+fP4eSkhI/vU+fPrh+/bpI53J1dYWenh5OnjyJv//+GydOnEDXrl3Rv39/vHv3DitXrix2DQxCCCFEWrBYLLFsZbFlyxbo6upCSUkJVlZWiI6OLtVxwcHBYLFYZZprKXJwcevWLTg5ORVJr1+/vsirad66dQtr1qxBv3794OfnBwBYuHAh5s2bhxo1aohaNEIIIYT8IiQkBC4uLvDy8sLdu3dhYWEBW1tbJCUllXhcQkIC5s2bh06dOpXpuiIHF4qKivj8+XOR9GfPnkFTU1Okc3E4HNSrVw9A4bLitWrVQrt27UQtEiGEEFKpMTUssm7dOkyaNAnjxo2DsbExtm3bhpo1a8Lf37/YYwoKCjBq1CgsXboUTZs2LVt9RT3gr7/+gre3N/Ly8gAUdvW8efMGrq6uGDJkiEjnYrFY+PLlCz5//oyMjAywWCxkZWXh8+fPAhshhBAizcS1QmdOTk6R78icnByh18zNzcWdO3cEHiQqIyMDGxsbREREFFtWb29vaGlpYcKECWWur8jBxdq1a5GZmQktLS1kZWWhS5cu0NPTQ+3atbF8+XKRzsXj8aCvrw81NTWoq6sjMzMTLVq0gJqaGtTU1KCqqgo1NTVRi0gIIYRUST4+PlBRURHYfHx8hOblcDgoKCgo8uRxbW3tYqcx/Pfff9i9ezd27txZrnKW6amoFy9exH///YfY2FhkZmaiZcuWpXrE+u+uXLki8jGEEEKItBHXg8vc3d3h4uIikKaoqCiWc3/58gVjxozBzp07wWazy3UukYOLHzp27IiOHTuW6+IdO3bEmjVrcOrUKeTm5qJHjx7w8vKiyZyEEEKqFHHdRqqoqFjqYILNZkNWVhafPn0SSP/06RN0dHSK5H/58iUSEhLQv39/fhqXywUAyMnJ4enTp2jWrFmprl3q4CIrKwthYWHo168fgMLo6ddxHllZWSxbtkzg9tQ/WbFiBZYsWQIbGxvUqFEDGzduRFJSUokTTQghhBDyZwoKCmjVqhXCwsL4t5NyuVyEhYVh+vTpRfIbGhriwYMHAmmLFy/Gly9fsHHjRpGe9VXq4CIwMBBnzpzhBxe+vr4wMTHh9zI8efIE9erVw5w5c0p98b1798LPz49/a+ulS5fQt29f7Nq1CzIyZVo8lBBCCKl0mHpcuouLCxwdHdG6dWu0bdsWGzZswNevXzFu3DgAgIODA+rXrw8fHx8oKSnB1NRU4HhVVVUAKJL+J6UOLoKCgrBgwQKBtAMHDvBvU9m/fz+2bNkiUnDx5s0b9OnTh//axsYGLBYLHz58oMWzCCGEVBkyDD24zN7eHsnJyfD09ERiYiIsLS1x7tw5/iTPN2/eSOTHPIv34znnf1C3bl1ERERAV1cXAKCpqYlbt27xXz979gxt2rRBRkZGqS8uKyuLxMREgfUxateujdjYWDRp0qT0tfjN52xumY8lgKw0rTFbyYhr0lZ1pG5d+h8mpKi0yPVMF0FqKZV59mHp+YUniOU809rriuU8klbqJk1PTxeYY5GcnCywn8vlFnuvbXF4PB7Gjh0rMDklOzsbU6ZMQa1atfhpx44dE+m8hBBCCGFOqYOLBg0a4OHDhzAwMBC6PzY2VuShDEdHxyJpo0ePFukchBBCSGVX3TqESx1c9OnTB56enujbt2+RO0KysrKwdOlS9O3bV6SL79mzR6T8hBBCiDSqbkOmpQ4uFi5ciEOHDsHAwADTp0+Hvr4+AODp06fw9fVFfn4+Fi5cKLGCEkIIIUQ6lDq40NbWRnh4OKZOnQo3Nzf8mAfKYrHQs2dP+Pn5FVlilBBCCCHM3YrKFJHmyDZp0gTnzp1DamoqXrx4AQDQ09ODurq6RApHCCGEVAU0LFIK6urqaNu2rbjLQgghhJAqoALu7iWEEEKqt2rWcUHBBSGEECJp1e2BFtWtvoQQQgiRMOq5IIQQQiSMVc3GRSi4IIQQQiSseoUWFFwQQgghElfdbkWlOReEEEIIEStGg4u9e/eK/CRVQgghRNqwxLRJC0aDi3HjxiEjI4PJIhBCCCESx2KJZ5MWjAYXP55PQgghhJCqg/EJndXt9hxCCCHVT3X7rmM8uOjRowfk5Eouxt27dyuoNIQQQoj4Vbe7JxgPLmxtbaGsrMx0MQghhBAiJowHF/Pnz4eWlhbTxSCEEEIkhoZFKlB1a2xCCCHVU3X7tqO7RQghhBAiVoz2XMTHx0NTU7NIen5+PrKzs2kuBiGEkCqhuvXUM9pzERsbi8DAQIG05cuXQ1lZGaqqqujVqxfS0tIYKh0hhBAiHjJi2qQFo2Vdu3Ytvn79yn8dHh4OT09PeHh44NChQ3j79i2WLVvGYAkJIYSQ8mOxWGLZpAWjwyKPHj1C+/bt+a+PHDmCnj17YtGiRQAAJSUlzJo1C+vWrWOqiIQQQggREaM9F1++fIGGhgb/9X///YcePXrwX5uYmODDhw9MFI0QQggRG3pwWQWqX78+Hj9+DADIzMxETEyMQE9GSkoKatasyVTxCCGEELGgB5dVoGHDhmH27NnYt28fJk2aBB0dHbRr146///bt2zAwMGCwhIQQQggRFaNzLjw9PfH+/XvMnDkTOjo62L9/P2RlZfn7Dx48iP79+zNYQkIIIaT8ZKRqUKP8GO25qFGjBvbu3Yu0tDQ8fvwYnTp1Eth/5coVuLq6MlQ64Q4FB+Evux7o0MYCY0fZI+5BbIn5L104h6ED+qBDGwv8PeQv3LxxTWD/jq2+GDqgDzpZtUT3jlaYNnkcHsbGCOTx37kN4x1GoKNVC3Tr2FbsdaooIQeD0Ne2O9q1MofDyOF4+Ie2u3j+HAb3t0O7VuYYPqg//rsu2HZhly5g2uTx6NbRCi3NDPH0yWOh54m5fw+TJziifdsW6NSuFSY4jkZ2drbY6iUJwQeDYNerO9q2NMPoEcPw4A9tdeH8WQzs3xttW5ph6KD+uPFbW/F4PPj5boRN146wamUOp4lj8fp1gkAeu17dYWlqILD579rB338rOgqzZ0yFTdeOaNfGEsOHDMCZf0+Jrc4VzWlYBzw55YG0m6txPWA2Wps0KjavnKwM3Cf2QtyJRUi7uRpRB+ahp7WhQJ5Fk22RdXu9wHb/iJukq1HpBR8Igl3P7mjTwgyj/h6GB7Elv5erKhoWIcW6cC4UG9aswkQnZ+wLPormBgaYMXUSUlNShOaPuX8Pi93mYcCgIdgfcgxduvXAvNkz8OL5M36eRo11Md99MQ4ePYmdAftRr159TJ86EWmpqfw8eXl5sOlpiyHD/pZ4HSXl/LlQrPtnJSZPccaBQ8fQXN8Azk4TS2i7u1joOhcDBg/FgcPH0bW7DVxmTRdou6ysLFi2aIWZc+YVe92Y+/cwY+okWFt3wL4Dh7Dv4GHYjxgFGZnK+9Y/fzYUa1f7wGmqMw4ePg59A0NMc5pQbFvdv3cX7gvmYuCgoQg+fALduvfAnJnOAm0V4L8TB4L2YZHnEuw7cAg1atTANKcJyMnJETjXtOkzcenqf/xtxMjR/H0x9++hub4B1qzfhMNHT2HAwMHwWOiK61evSKYhJGhoT0usmjMQy3eeh/XotYh99gGnNjtBU034wn1LpvXBxMHWcPnnGFoMX4VdR8MR8s84WBjUF8gX9/IjdG09+VuPCZsrojqV1rmzoViz2gdO05wRfPg4DAwMMdVpAlKKeS+TqoPFY3AN7hYtWpTqvl1RH7n+OZtb1iKVaOwoexibmGLBQg8AAJfLRb9e3TB8xGiMnTCpSH73+XOQnZWF9b7b+GnjRttD38AI7h5LhF4jMzMT3Tq0wZYd/mhrZS2w7/TJ41j3jw+u/BctvkoJISsj/vDYYeRwGJuYwm2RJ4DCtrPr2RV/jxiNcRMnF8nvOm8OsrK+YdOW7T/PMcoeBgaGWOS5VCDvh/fv0K+3DQ4ePg4DQyPB646yR7t27TFtxiyx10kYGTH8tBg9YhhMTM3g/ktb2dp0wYiRYzBeSFstmDsbWVlZ2Oz3s63GjBwOAwNDLPbyBo/HQ89unTDGcRwcx00AUHinVo8u7eH9v5Xo3acvgMKei1FjHDB6zNhSl3X61MnQ0NDA0v/5lKPGhdSt55T7HKV1PWA27jx6gzmrjwEoXIPgxRlPbA35D2sCw4rkf3V2CVb5X8T2wzf5aQdXj0VWdh7GewYBKOy56N/FDO1GramYSvwmLXI9I9ctyai/C9/LCxf/fC/36lH4Xp4wqeh7mSlKFTBB4MzDJLGcp6+pdDzok9E5FwMHDmTy8iLJy8vFk8dxAkGEjIwM2razxoPY+0KPeRAbg5FjHAXS2rXviGtXiv7x+nGN40cPQbl2bejrGwrNI43y8nLx+FEcxk34+cdERkYGVu2sERtzX+gxD2LuY5TDWIE06/YdcPWy8LYTJjUlBQ9jY9CnTz+MHf033r19C90mTeA8cw5atGxVlqpI3I+2Gj/RiZ9W2FbtERtzT+gxsTH3MdpxrECadfuOuHr5EgDg/bt34HCSYWX9806s2rVrw8zcAjEx9/jBBQDs2bUTO7dthU7durDr0w+jHcZCTq74PxOZmV/QtGmzslSVMfJysmhh2AD/7LnET+PxeLgc/RxtzRsLPUZBXg7ZufkCaVnZeWhv2VQgTa8RG6/OLkF2Tj6iHiTA0/dfvP2ULvY6SIO83ML38oRJgu/ldiW8l6syaRrSEAdGgwsvL69ynyMnJ6dI124OTx6KiorlPvev0tPSUVBQAPVf1uUAAHUNDSTExws9JoXDgYYGu0j+FA5HIO3GtStY5DoP2dlZYLM14bttN1TV1MRafialp6UV03bsYtuOw+EIrIECABoa7CJtV5J3794CALZv9cXsuQtgYGiEf0+dxJSJY3H4+Gk0aqwrWkUqQNr3tipadw0kxL8SegxHyPtMg60Bzve24nCS+ef41e/vxZGjxsDQyBgqKiqIuX8PmzauA4eTjHkL3IVe9/y5UMQ9fIDFXt6iVZJhbNVakJOTRVLqF4H0pNQvMNAV/qvwUuQTzBzZFf/dfYlX71LQrW1zDOhuDtlfhtduPXyNyUsO4tnrJOiw62DRJFtc2jUDrexXI/NbjtDzVmVp6cW/l+OLeS+TqoPRgefs7GycOnUKX758KbLv8+fPOHXqVJHA4Xc+Pj5QUVER2Nb9s1JSRZaI1m2sEHToGHbvPQDrDh2xcP6cYsfXSenxeIXDY4OH2WPAoCEwNDLGPFd3NNZtgpPHjzJcuspnjOM4tGlrBX0DQwyzH4G581wRfGA/cnNzi+S9FR0JL4+F8FzyP+jpNWegtBVr3prjePk2GTFH3PE54h+sXzAEe09Fg8v9OQR7IfwJjoXF4OGLj7gU+RQDZ+2ASu0aGNLTkrmCk0pDBiyxbNKC0eBi+/bt2LhxI2rXrl1kX506dbBp0ybs3LmzxHO4u7sjIyNDYHOZL/4Z2qpqqpCVlS3ypZ+akgINNlvoMRpsNlJSOH/MX6NmTTRs1Bhm5pbwWLocsnKyOHmi6nz5qaqpFdN2RX9x/8Bms4tM+kpJ4RTb1sLPUfgrtGlTPYH0Jk2bIfHjx1KfpyKpfW+ronVPAbuYurOFvM9SOD/zs9ma/HP8qqT3LgCYmlsgPz8fH96/E0i/fSsaM52nYt4Cd/QfMLBU9apMOOlfkZ9fAC11wb87Wuq1kZjyudhjhs/zh0YnVxj0XwaLIT74mpWD+PepQvMDQEZmNl68TkazBqV/z1Ylaqqiv5erMrpbpAIFBQVh9uzZxe6fPXs29u7dW+I5FBUVUadOHYFN3EMiACAvrwBDIxPciorkp3G5XNyKioSZuaXQY8zMLQTyA0BUZHix+X+el4c8Ib8WpZW8vAKMjE0QHRXBT+NyuYiOjIS5haXQY8wsLAXyA0BURHix+YWpV78+NLW08DpBcOjlzesE6NSrV+rzVKRi2yoqAuYWLYQeY25hiehIwfdZ5C9tVb9BA7DZmoiO/HnOzMxMPIiNgUUx5wSAp08eQ0ZGBurqP7u1b0VHYcY0J8xymYehw+zLUkXG5eUX4N6Td+jWVp+fxmKx0K1Nc0THvi7x2JzcfHxIzoCcrAwGdjfHv9ceFJu3Vg0FNGmggUSO8IClqpNXKHwvR0UKvpejSngvV2XVLbhgdM7F8+fPYWFhUex+c3NzPH/+vAJLVLKRYxyx1MMdRiamMDE1w8H9e5GVlYX+AwcBALwWuUJTSxvTZ7kAAP4e5QCnCQ7YH7gHHTt3wYVzoXgcF4eFHoV3O2R9+wb/XdvRuWs3sNmaSE9Px+HgA0hO+oQePW351038+AEZGRlI/PgB3IIC/noODRs1Qs2atSq4FcpmlMNYeC1yg7GJKUzMzHFgXyCysrLw18DBAACPha7Q0tLCjNlzAQAjR4/BpHEO2Bfoj46duuL8uTN4FBcnML6fkZGOxI8fkZxUOAs74XsQocFmg83WBIvFgsPYCdjutxn6BgbQNzTCvydPICH+FVav21jBLVB6YxzGwWORK4xNTGFqao6g/YVtNeB7Wy12XwAtLW3MnPOjrRwwcdwY7A3wR6fOXXDubCgexT2E55LCtmKxWBg1xgE7d2xFo8aNUb9+A2zx3QhNLS1062EDoPA20wcPYtCmTTvUqlULMTH3sGa1D/r0+wt1VFQAFA6FzHCegpGjHGDTsxd/Loe8vDxUVFQruJXKZ1PQVexcMhJ3Hr3F7bjXmD6yC2rWUMDe01EAgF1LR+JDUgY8t5wBALQxaYR6WiqIefYB9TVVsGiyLWRYMli39zL/nD6z/sKZG3F48zEV9TRVsNipNwq4PBw6L9rdblXJGMdx8FjoChMTU5iamWP/98/9wEGDmS4akTBGg4v8/HwkJyejUSPhi9ckJycjPz9f6D4m9OrdB+lpadjutwkpHA70DYywyW8Hv2s/MfEjWL9M8LKwbIH/+fyDrb4b4bd5PRo2aow1GzZDr3nhLyYZWVkkxL/CmVMnkJ6eBhVVVRibmGHHnv1o9ss49ja/zThz6gT/9Wj7wg/mtl2BaNVGOhbVsu3dB2mpqdi6ZTNSOMkwMDSC77ad/G75xI8fBG7jtLBsieUr18DPdwN8N65Ho8a6WLfRl992AHDtymUs8VjIf+0+vzComzzVGVOmzQAAjBrjiNycHKxdvRIZnzOgr28Avx3+aNiw+AWTmGZr1wdpaanY6rsJnO9t5bdtF7+tPn4UfJ9ZtmiJFavWYMvmDdi8cR0aNdbF+k1bBNpq7PhJyMrKwrIlnvjy5TNatGwFv227+L18CgoKOH82FNv8fJGXm4v69Rtg9JixGOM4jn+OUydPIDsrC/67tsN/18/bXlu1bovdAfsk3SxideTifbDVlOE5pTe0Neog9tl7DJixHUmpmQCAhjpq4HJ/3qWvqCgPr6l90KS+BjKzcnD+5mNM8AxCRubPxdjqa6tg7/IxUFepBU5aJsJjXqHL2A3gpH+t8PpVFr3tCj/3fr++l7fvEml4s6pgSdF8CXFgdJ2Ldu3aYdCgQcWuwunj44OTJ08i8rcu3z+R1DoX1YUk1rmoLsSxzkV1VZHrXFRFlXGdC2lREetchD0p/Z1uJelhKB2BGaNzLsaPH49ly5bh33//LbLv9OnTWL58OcaPH89AyQghhBBSVowOi0yePBnXr1/HX3/9BUNDQ/4TUJ88eYJnz55h+PDhmDy58qziRgghhJRFdRsWYfwBC/v370dwcDD09fXx7NkzPH36FAYGBjh48CAOHjzIdPEIIYSQcqO7RRgwfPhwDB8+vEg6l8tFaGgo+vXrx0CpCCGEEFIWlSK4+N2LFy/g7++PgIAAJCcnIy8vj+kiEUIIIWVGwyIMycrKwt69e9G5c2cYGBggPDwcnp6eePfu3Z8PJoQQQioxGZZ4NmnBeM/FrVu3sGvXLgQHB6NZs2YYNWoUwsPD4efnB2NjY6aLRwghhBARMRpcmJub4/Pnzxg5ciTCw8NhYmICAHBzE/+zQQghhBCm0LBIBXr69Ck6d+6Mbt26US8FIYSQKqu63S3CaHDx6tUrGBgYYOrUqWjQoAHmzZuHe/fugSVNLUgIIYT8AUtMm7RgNLioX78+Fi1ahBcvXmDfvn1ITExEhw4dkJ+fj4CAADx79ozJ4hFCCCGkDCrN3SLdu3fH/v378fHjR/j6+uLy5cswNDSEnp4e00UjhBBCykWGxRLLJi0qTXDxg4qKCqZNm4b//vsP//zzD96+fct0kQghhJByoWGRCpSTkwN3d3e0bt0a7du3x4kTJwAAe/bsQbNmzbBx40Z4e3szWURCCCGEiIjRW1E9PT2xfft22NjYIDw8HMOGDcO4ceMQGRmJtWvXYtiwYZCVlWWyiIQQQkj5SVO3gxgwGlwcPnwYe/fuxV9//YWHDx/C3Nwc+fn5iImJoTtGCCGEVBm0zkUFevfuHVq1agUAMDU1haKiIubMmUOBBSGEECLFGO25KCgogIKCAv+1nJwclJWVGSwRIYQQIn7V7Tczo8EFj8fD2LFjoaioCADIzs7GlClTUKtWLYF8x44dY6J4hBBCiFhUs9iC2eDC0dFR4PXo0aMZKgkhhBBCxIXR4GLPnj1MXp4QQgipGNWs64LxR64TQgghVV11u1uEggtCCCFEwqrbhM5Kt/w3IYQQQqQbBReEEEKIhDH5bJEtW7ZAV1cXSkpKsLKyQnR0dLF5d+7ciU6dOkFNTQ1qamqwsbEpMX9xKLgghBBCJI2h6CIkJAQuLi7w8vLC3bt3YWFhAVtbWyQlJQnNf/XqVYwYMQJXrlxBREQEGjZsiF69euH9+/eiVZfH4/FEL27l9jmby3QRpJqsTDUbHBQjaXokcmWjbj2H6SJItbTI9UwXQWopVcDsw7uvP4vlPC0b1xEpv5WVFdq0aQNfX18AAJfLRcOGDTFjxgy4ubn98fiCggKoqanB19cXDg4Opb4uTegkhBBCJExcd4vk5OQgJydHIE1RUZG/GOWvcnNzcefOHbi7u/PTZGRkYGNjg4iIiFJd79u3b8jLy4O6urpI5aRhEUIIIUTCWCzxbD4+PlBRURHYfHx8hF6Tw+GgoKAA2traAuna2tpITEwsVbldXV1Rr1492NjYiFRf6rkghBBCpIS7uztcXFwE0oT1WojDypUrERwcjKtXr0JJSUmkYym4IIQQQiRMXLOxihsCEYbNZkNWVhafPn0SSP/06RN0dHRKPHbNmjVYuXIlLl26BHNzc5HLScEFKYJb9eb4Vhiazll2NCGxfNQ6ujJdBKmVFblK8hdh4I+DgoICWrVqhbCwMAwcOBBA4YTOsLAwTJ8+vdjjVq9ejeXLl+P8+fNo3bp1ma5NwQUhhBBSRbm4uMDR0RGtW7dG27ZtsWHDBnz9+hXjxo0DADg4OKB+/fr8eRurVq2Cp6cnDhw4AF1dXf7cDGVlZSgrK5f6uhRcEEIIIRLG1LNF7O3tkZycDE9PTyQmJsLS0hLnzp3jT/J88+YNZGR+3tuxdetW5ObmYujQoQLn8fLywpIlS0p9XVrnghRBSzWUnSw1XpnJ0Poq5ULDImVXEcMiD95liuU8Zg1K33vAJOq5IIQQQiSsuoXOtM4FIYQQQsSKei4IIYQQSatmXRcUXBBCCCESxtSETqbQsAghhBBCxIp6LgghhBAJq243klFwQQghhEhYNYstaFiEEEIIIeJFPReEEEKIpFWzrgsKLgghhBAJo7tFCCGEEELKgbHgwtvbG9++fWPq8oQQQkiFYbHEs0kLxoKLpUuXIjNTPA9yIYQQQiozlpg2acHYnIsq+DBWQgghRDhpigzEgNE5Fyxp6uMhhBBCSKkwereIvr7+HwOM1NTUCioNIYQQIhnV7W4RRoOLpUuXQkVFhckiEEIIIRJX3TrqGQ0u/v77b2hpaTFZBEIIIYSIGWPBBc23IIQQUl1Ut288uluEEEIIkbRqFl0wFlxwuVyh6a9fv8bXr19haGgIGRlaQJQQQgiRNox9e/v7+2PdunUCaZMnT0bTpk1hZmYGU1NTvH37lqHSEUIIIeLDEtN/0oKx4GLHjh1QU1Pjvz537hz27NmDvXv34tatW1BVVcXSpUuZKh4hhBAiNtVt+W/GhkWeP3+O1q1b81+fPHkSAwYMwKhRowAAK1aswLhx45gqHiGEEELKiLGei6ysLNSpU4f/Ojw8HJ07d+a/btq0KRITE5koGiGEECJW1e3ZIowFF40bN8adO3cAABwOB3FxcejQoQN/f2JiIi2wRQghpGqoZtEFY8Mijo6OcHZ2RlxcHC5fvgxDQ0O0atWKvz88PBympqZMFY8QQggRG2majCkOjAUXCxYswLdv33Ds2DHo6Ojg8OHDAvtv3ryJESNGMFQ6QgghhJQVi1cFV7P6nC18DQ1xOBQchP2B/kjhcNBc3xDz3RbBxMy82PyXLpzDti2b8PHDezRs1BgzZs9Fh05d+Pt3bPXFhXOh+JSYCHl5eRgaG2Pa9NkwNbcAAHx4/x67d/jhdnQUUlI4YGtqwa5vf4yf5AR5eQWJ1FGSM5IPBQdhX8Av7ee+CKZ/aL+tvr+035y56Pi9/fLz8uDnuxE3b1zH+3fvoFxbGW2trDFj9lxo/rKs/JwZ0/Ds6ROkpaagdp06aNvOGjNnzxPIIy6yEmy8kINBCAzYjRQOB/oGhnB1X1xi2108fw5+vhvx4cN7NGrUGDPnzEOnzj/fe2GXLuDIoWA8fhSHjIwMBB8+DgNDI6Hn4vF4mD51MsJv3sC6Db7o1sNG7PWTkam8v+yCDwQhcM9ucDjJ0DcwhNtCD5iZF9/2TFDr6Fph13IaYo05oztDW702Hrz4CJe1J3H70TuheeVkZTDfsRtG92mFepp18OxNMhZvOYuLkc/4eSYNbodJg9uhcd3COwgfv/qEFf5huBDxtELqkxW5SuLXeJOaI5bzNFJXFMt5JI1WqRLBhXOh2LBmFSY6OWNf8FE0NzDAjKmTkJqSIjR/zP17WOw2DwMGDcH+kGPo0q0H5s2egRfPf36oGjXWxXz3xTh49CR2BuxHvXr1MX3qRKR9fxpsQsIrcLk8uHssRfCx05gz3w3HDodgy6YNFVFlsbpwLhTr/1mFSVOcsT/kKPQNDDBjSsntt8i1sP2CDh1D1+49MG/Wz/bLzs7Gk8ePMNFpKvaHHMU/6zbhdUICXGZOEzhP67ZtsfKfdTh6KhSr123C+7dv4Tp3lsTrK07nz4Vi7T8r4TTFGQcOHYO+vgGmOU0stu3u378Ld9e5GDh4KA4ePo6u3W3gMmu6wHsvKysLli1aYeaceX+8ftC+wGq7ZP+5s6FYs9oHTtOcCwMwA0NMdZqAlGLavqobamOOVbP6YfmuMFg7bkLs8484tWECNNVqCc2/ZIotJg60gsvak2gxYh12HY9CyEoHWOjX4+d5n5QBjy1n0X7sJnQYuxlX77zE4dUOMGqiXVHVkrhqNuWCuZ4LNTW1Uv2xKssj1yXVczF2lD2MTUyxYKEHgMJVRvv16obhI0Zj7IRJRfK7z5+D7KwsrPfdxk8bN9oe+gZGcPdYIvQamZmZ6NahDbbs8EdbK2uhefYF7MaRQ8E4GXqx/JUSQlLfIY4j7WFsagrXX9qvb69usC+h/bKysrDhl/YbO8oe+oZGWFhM+8U9fADHkcPx7/kw6NStJzTPtSuXMW/2dETcjoGcvHz5K/YLSfVcjBk5HCYmpnBb5AmgsO169+yKv0eMxviJk4vkd503B1lZ37Bpy3Z+msMoe+gbGGKxp+D6MR/ev0Pf3jbF9lw8ffIYM52nICjkCHp261Ttei5G/T0MJqZmWLj4Z9v36tEFI0aOwYRJRdueKRXVc3F9tzPuPHqHOWtPAih8TtSLk+7Yejgca/ZdLZL/1elFWBVwGduPRvDTDvqMRlZOHsYvCSn2Ou/Pe2GhbygCT98SdxWKqIiei7di6rloKCU9F4zNudiwYQNTly6TvLxcPHkcJ/AlKCMjg7btrPEg9r7QYx7ExmDkGEeBtHbtO+LalbBir3H86CEo164NfX3DYsuSmflF6u6k+dF+4yb+1n5W1oiNuS/0mNiYGIz6rf2s23fE1WLaDyhsGxaLBeXadYTuz8hIx7nQ0zC3bCH2wEJS8vJy8fhRHMZP+PlFJiMjA6t2JbXdfYx2GCuQZt2+A65cLr7thMnKyoK76zy4LfIEm60patGlXl5uYdtPmOTET5ORkUG7du0RG3OPwZIxQ15OFi0M6uOfwCv8NB6Ph8u3XqCtWSOhxygoyCI7N18gLSsnD+0tdIXml5FhYUh3c9SqoYCoB6/FVnamVbeOP0bvFhGHnJwc5OQIRoQ5PHkoKoo3uktPS0dBQQHUNTQE0tU1NJAQHy/0mBQOBxoa7CL5UzgcgbQb165gkes8ZGdngc3WhO+23VD9ZfXSX7198xohB4Mwy2V+OWpT8crafuqlaL8fcnJysHn9Wtja9YWysrLAvk3r1+DQwQPIzs6CmbkF1vtuLUdtKlZaWprQttPQYBfbdhwOR2j+4tquOGtX+8DCsgW6de8hWqGriLT0wrbXKNKWGoiPf8VQqZjDVq0JOTlZJKVmCqQnpX2Bga7w4PNS5DPMHNEJ/91/hVfvUtGtjR4GdDWF7G/PjjJppoOrO6dBSUEOmVm5sHfdiycJSRKrS8WrXtEFY3Mu0tLSsHnzZnz+/LnIvoyMjGL3/c7HxwcqKioC27p/VkqiyBLTuo0Vgg4dw+69B2DdoSMWzp8jdCw96dMnzJw2GTY9bTFoyHAGSlp55eflwW3eHPB4PLgt9iqy32HsBAQdOgrf7bsgIysLr0Vu9GTeP7h65TKio6Mw39Wd6aIQKTZv/Wm8fMtBTPA8fL6xHOvnDsDef2+DyxX8/D17nQwrh43oPGELdh6LxE7P4TDUFf+ka1IxGOu58PX1RWxsLGbMmFFkn4qKCm7cuIHPnz9j0aJFJZ7H3d0dLi4uAmk5PPF3d6uqqUJWVrbIl35qSgo02Gyhx2iw2UhJ4fwxf42aNdGwUWM0bNQYZuaWGNzfFidPHMW4X7rBk5OSMHWiI8wtLLHQ01tMtao4ZW2/1FK0X35eHtzmz0Hixw/YumtPkV6LwuurQVVNDY11m6BJk2bo26sbHsTeh7lFi3LWTPLU1NSEtl1KStGesR/YbLbw/MW0tTC3oiPx7u0bdG7fViB9nstMtGjZCrv27Cv1uaSVmmph2/8+eTMlJQVsEdqyquCkf0N+fgG01AU/Y1pqtZGY8qWYY75iuOteKCrIQUOlJj4kf8b/nO0Q/0FwPl1efgFevSts53tP36OVcQM423fEjFXHJFOZClbdhkUY67k4evQopkyZUux+JycnHDly5I/nUVRURJ06dQQ2cQ+JAIC8vAIMjUxwKyqSn8blcnErKhJm5pZCjzEztxDIDwBRkeHF5v95Xh7ycnP5r5M+fcKUCQ4wNDaBp/cKqXwU/Y/2ixbSfuYWlkKPMbcopv1+yf8jsHjz+jX8dvhDVVX4cNKveLzCCb+5uXmiV4QB8vIKMDI2QVTUzwlxXC4X0ZEltZ0lon/JDwCREeHF5hdm3IRJOHT0JIIPH+dvADB3gRuWLvMRuR7SSF7he9tHCrZ9VFSEVASm4paXX4B7T9+jWxs9fhqLxUK3NnqIfvCmxGNzcvPxIfkz5GRlMLCrKf69HldifhkWC4oKsmIpd2VQ3e4WYazn4uXLl2jevHmx+5s3b46XL19WYIn+bOQYRyz1cIeRiSlMTM1wcP9eZGVlof/AQQAAr0Wu0NTSxvRZhT0pf49ygNMEB+wP3IOOnbvgwrlQPI6Lw0KPwtn6Wd++wX/XdnTu2g1stibS09NxOPgAkpM+oUdPWwDfA4uJDtCpWw+zXBYgLe1ntC9tE+xGOThiyWJ3GBubwsTMDAd+az/Pha7Q0hZsv8njf7bf+bOheBQXh4Xf73bIz8vDgrmz8fTxI6z33YoCbgE4nGQAhb1f8vIKeBgbg7i4h7Bs0RJ16tTBu7dvsXXLJjRo2EikL1qmjXYYC89FbjA2MYWpmTkO7AtEVlYWBgwcDABYvNAVWlpamDl7LgBgxOgxmDTOAXsD/dGpU1ecP3cGj+Li4OH1s9crIyMdiR8/IimpcFw7IaFw/oYGmw02W5O//a6uTj3Ub9BA0lWuNMY4joPHQleYfG/7/d/bfuCgwUwXjRGbDt7ATo/huPP4HW4/eofp9h1RU0kee8/cBgDs8hyOD8mf4bn1HACgjUlD1NOsg5hnH1Ffsw4WTewJGRkW1u2/xj+n99TeOB/xFG8/paN2TUXY97JE55ZN0X+2PyN1JOXHWHAhKyuLDx8+oFEj4TOMP3z4UOl+offq3QfpaWnY7rfp+0JGRtjkt4PfNZ2Y+BGsX8psYdkC//P5B1t9N8Jv83o0bNQYazZshl5zfQCAjKwsEuJf4cypE0hPT4OKqiqMTcywY89+NNMrDLyiIsPx9s0bvH3zBn17dRUoz62YxxVTcTHp1bsP0tLSsO2X9tu8VbD9ZH5rv+Ur/4Hf5o3Ysul7+2382X5JSUm4fvUyAGDksEEC19q2OxCt27SFklINXLl0ETv8NiMrq3DCrHWHjpjwz1QoKEhmETJJsO3dB2mpqdi6ZTNSOMkwMDTClm07+cMciR8/QOaXfldLy5ZYsXINtvhugO/G9WjUWBfrNvry2w4ovCXXy2Mh/7Xb/MKgzmmqM6ZMKzpcWV31titsez/fTeB8b3u/7btEGmKqSo5cigVbtRY8J/WCtkZtxD7/gAFz/PmTPBvqqIL7y3wmRQU5eDnZokk9dWRm5eJ8+BNMWBqMjMxsfh5NNWXs9hoOHY06yMjMxsOXH9F/tj8uRz+v8PpJSnUbFmFsnYtu3brBysoKK1cKn3zp6uqK6OhoXLlyRej+kkhyhc7qoLp9CMRJkit0VnWVdZ0LaVGRK3RWNRWxzkVihniGYXVUpOMWesZ6LqZPn46///4bDRo0wNSpUyErWzi2VlBQAD8/P6xfvx4HDhxgqniEEEKI+FSz2Jmx4GLIkCFYsGABZs6ciUWLFqFp06YAgFevXiEzMxPz58/H0KFDmSoeIYQQQsqI8QeXRUdHIygoCC9evACPx4O+vj5GjhyJtm3b/vngYtCwSPlQz37Z0bBI2dGwSPnQsEjZVcSwyKfP4hkW0a5DwyKl0rZt23IFEoQQQkhlV91+dzB2O8bq1auRlZXFf33z5k2BZby/fPmCadOmCTuUEEIIIZUYY8GFu7s7vnz5uaKbnZ0d3r9/z3/97ds3bN++XdihhBBCiFRhiek/acHYsMjvUz3oOQ+EEEKqLOmJC8Sicq1SRQghhBCpx/iETkIIIaSqq2YdF8wGF7t27eI/wTI/Px8BAQH8Jw3+Oh+DEEIIkWbV7W4Rxta50NXVBasUrR0fHy/yuWmdi/Kpbh8CcaJ1LsqO1rkoH1rnouwqYp2LlK/5YjmPRi3pGHBgrJQJCQlMXZoQQgipUNJ0p4c4MDahMyIiAv/++69A2t69e9GkSRNoaWlh8uTJAuteEEIIIdKKxRLPJi0YCy6WLl2KuLg4/usHDx5gwoQJsLGxgZubG06fPg0fHx+mikcIIYSQMmIsuIiJiUGPHj34r4ODg2FlZYWdO3fCxcUFmzZtwqFDh5gqHiGEEELKiLE5F2lpadDW1ua/vnbtGuzs7Piv27Rpg7dv3zJRNEIIIUSspGlIQxwY67nQ1tbm3wmSm5uLu3fvol27dvz9X758gby8dDz9jRBCCClJdVv+m7Hgok+fPnBzc8ONGzfg7u6OmjVrolOnTvz9sbGxaNasGVPFI4QQQkgZMTYssmzZMgwePBhdunSBsrIyAgMDoaCgwN/v7++PXr16MVU8QgghRGyq27AIY8EFm83G9evXkZGRAWVlZcjKygrsP3z4MH/1TkIIIUSaVbPYgvlni6ioqAhNV1dXr+CSEEIIIUQcGA8uCCGEkCqvmnVdUHBBCCGESJg03ekhDozdLUIIIYSQqol6LgghhBAJo7tFCCGEECJW1Sy2oOCCEEIIkbhqFl3QnAtCCCGkCtuyZQt0dXWhpKQEKysrREdHl5j/8OHDMDQ0hJKSEszMzBAaGiryNSm4IIQQQiSMqWeLhISEwMXFBV5eXrh79y4sLCxga2uLpKQkofnDw8MxYsQITJgwAffu3cPAgQMxcOBAPHz4ULT68ng8nsilreQ+Z3OZLoJUq24Tj8RJlhqvzGRkqO3KQ62jK9NFkFpZkaskfo3sfPGcR0nEyQxWVlZo06YNfH19AQBcLhcNGzbEjBkz4ObmViS/vb09vn79in///Zef1q5dO1haWmLbtm2lvi71XBBCCCFSIicnB58/fxbYcnJyhObNzc3FnTt3YGNjw0+TkZGBjY0NIiIihB4TEREhkB8AbG1ti81fnCo5obOOUuWNmXJycuDj4wN3d3coKioyXRypQm1XPtR+ZScNbVcRv77LQhrariKI2uNQnCX/88HSpUsF0ry8vLBkyZIieTkcDgoKCqCtrS2Qrq2tjSdPngg9f2JiotD8iYmJIpWz8n4LV1E5OTlYunRpsZEmKR61XflQ+5UdtV3ZUduJl7u7OzIyMgQ2d3d3potVRJXsuSCEEEKqIkVFxVL3ALHZbMjKyuLTp08C6Z8+fYKOjo7QY3R0dETKXxzquSCEEEKqIAUFBbRq1QphYWH8NC6Xi7CwMFhbWws9xtraWiA/AFy8eLHY/MWhngtCCCGkinJxcYGjoyNat26Ntm3bYsOGDfj69SvGjRsHAHBwcED9+vXh4+MDAJg1axa6dOmCtWvXom/fvggODsbt27exY8cOka5LwUUFU1RUhJeXV7We2FRW1HblQ+1XdtR2ZUdtxyx7e3skJyfD09MTiYmJsLS0xLlz5/iTNt+8eQMZmZ+DGO3bt8eBAwewePFiLFy4EM2bN8eJEydgamoq0nWr5DoXhBBCCGEOzbkghBBCiFhRcEEIIYQQsaLgghBCCCFiRcEFIYQQQsSKgosSjB07FgMHDuT/P4vFwsqVKwXynDhxAqzfHla1c+dOWFhYQFlZGaqqqmjRogX/Np/fz/urq1evgsViIT09HQAQEBAAVVVVgTy6urpgsVjFbmPHji1vtcXm7du3GD9+POrVqwcFBQU0btwYs2bNQkpKChISEkqsB4vFQkBAQJE2+ZWuri42bNgg8FrYeX78m/1+TXV1dXTp0gU3btyQSP1/vGdYLBYUFBSgp6cHb29v5Ofn8+v1Y9PU1ESfPn3w4MEDkdrxV127dhVa//z8n09McnJygqysLA4fPlzkOkuWLAGLxcKUKVME0u/fvw8Wi4WEhAQAP9vx/v37/GNK2mxsbGBra1vken5+flBVVcW7d+/K0rxC/drm8vLyaNKkCRYsWIDs7GyBfO/evYOCgkKxM+B5PB527NgBKysr/ue4devW2LBhA759+wagsL0sLS35x3z79g3u7u5o1qwZlJSUoKmpiS5duuDkyZOlfr8zLTk5GVOnTkWjRo2gqKgIHR0d2Nra4ubNmwD+/Bn74ejRo+jatStUVFSgrKwMc3NzeHt7IzU1FUDRv20FBQVYuXIlDA0NUaNGDairq8PKygq7du0CgD+2nbClrwmz6FZUESgpKWHVqlVwcnKCmpqa0Dz+/v6YPXs2Nm3ahC5duiAnJwexsbEiP662OLdu3UJBQQGAwkfjDhkyBE+fPkWdOnUAADVq1BDLdcrr1atXsLa2hr6+Pg4ePIgmTZogLi4O8+fPx9mzZxEREYGPHz/y869Zswbnzp3DpUuX+GkqKiqIiooS6bre3t6YNGmSQFrt2rUFXl+6dAkmJibgcDhYvnw5+vXrh2fPnhVZT18cevfujT179iAnJwehoaFwdnaGvLw8f0GaH/92Hz58wPz589G3b1+8ePECCgoKAP7cjpGRkVBXV+dfb9KkSfD29hYog5xc4cf827dvCA4OxoIFC+Dv749hw4YVKa+SkhJ2796NuXPnonnz5n+s37x58wSCkTZt2mDy5MkC/wZ5eXkwMzPD9u3b4eTkBACIj4/HggULsHXrVjRo0KC0zVkqP9o8Ly8Pd+7cgaOjI1gsFlat+vnsjYCAAAwfPhzXr19HVFQUrKysBM4xZswYHDt2DIsXL4avry80NTURExODDRs2QFdXV+iPgylTpiAqKgqbN2+GsbExUlJSEB4ejpSUFDRs2LBU73emDRkyBLm5uQgMDETTpk3x6dMnhIWFCQSyf/qMLVq0CKtWrcKcOXOwYsUK1KtXD8+fP8e2bduwb98+zJo1q8h1ly5diu3bt8PX1xetW7fG58+fcfv2baSlpQGAQNuFhITA09MTT58+5acpKyuLrQ2ImPBIsRwdHXkDBgzg/3+/fv14hoaGvPnz5/PzHD9+nPdrMw4YMIA3duzYUp/3V1euXOEB4KWlpfF4PB5vz549PBUVlWLP83v+yqR37968Bg0a8L59+yaQ/vHjR17NmjV5U6ZMEUj38vLiWVhYFDlPSXVs3Lgxb/369cW+/l18fDwPAO/evXv8tNjYWB4A3smTJ0tTLZEI+3fu2bMnr127dkLrderUKR4AXkxMDD9NlHbs0qULb9asWcWWJyAggNeuXTteeno6r2bNmrw3b94I7P/xb9CzZ0/esGHD+On37t3jAeDFx8fzeDzh7fhDcf8GAQEBPGVlZd6rV694XC6X161bN96gQYOKLWtZCWvzwYMH81q0aMF/zeVyeU2bNuWdO3eO5+rqyps0aZJA/pCQEB4A3okTJ4qcn8vl8tLT03k8XtH3rIqKCi8gIKBU5Szu/c6ktLQ0HgDe1atXi83zp89YVFQUDwBvw4YNxV6Dxyv6t83CwoK3ZMmSUpXzT38XSeVAwyIikJWVxYoVK7B58+Ziu3J1dHQQGRmJ169fV3DpKo/U1FScP38e06ZNK9KToqOjg1GjRiEkJAQ8hpdYycrKwt69ewGA31MgaTVq1EBubm6R9IyMDAQHBwuURdztuHv3bowePRoqKiqws7Mrtht+5cqVOHr0KG7fvi1CzUrm6OiIHj16YPz48fD19cXDhw+xfft2sZ2/OA8fPkR4eLjAv++VK1fw7ds32NjYYPTo0QgODsbXr1/5+4OCgmBgYIABAwYUOR+LxSq2h0FHRwehoaH48uWL+CtSAZSVlaGsrIwTJ06U+SFjQUFBUFZWxrRp04Tu/32Y9wcdHR1cvnwZycnJZbouqXwouBDRoEGDYGlpCS8vL6H7vby8oKqqCl1dXRgYGGDs2LE4dOgQuFxuBZeUOc+fPwePx4ORkZHQ/UZGRkhLSxPpD0mDBg34f/x+bG/evCmSz9XVtUi+3+dUtG/fHsrKyqhVqxbWrFmDVq1aoUePHqJVUkQ8Hg+XLl3C+fPn0b179yL1UlVVxYEDB/DXX3/B0NAQQNna0c/PT6Duc+fO5Z8rMjIS9vb2AIDRo0djz549QgOTli1bYvjw4XB1dRVb/QFgx44dePjwIWbPno0dO3ZAU1NTrOf/4d9//4WysjKUlJRgZmaGpKQkzJ8/n79/9+7d+PvvvyErKwtTU1M0bdpUYA7K8+fPYWBgIPJ1d+zYgfDwcGhoaKBNmzaYM2cOf66CNJCTk0NAQAACAwOhqqqKDh06YOHChYiNjRXIV9Jn7Pnz52jatCnk5eVFuva6deuQnJwMHR0dmJubY8qUKTh79qzY6kYqHgUXZbBq1SoEBgbi8ePHRfbVrVsXERERePDgAWbNmoX8/Hw4Ojqid+/e1SrAACDWnokbN27g/v37Alu9evWK5Js/f36RfK1btxbIExISgnv37uHo0aPQ09NDQECAyH8MS+vXLzo7OzvY29sLTD67ceMG7ty5g4CAAOjr62Pbtm1FziFKO44aNUqg7j8exezv7w9bW1uw2WwAQJ8+fZCRkYHLly8LPc///vc/3LhxAxcuXBChtiXT0tKCk5MTjIyMhM5ZEJdu3brh/v37iIqKgqOjI8aNG4chQ4YAANLT03Hs2DGMHj2an3/06NHYvXs3/3VZ37edO3fGq1evEBYWhqFDhyIuLg6dOnXCsmXLylehCjRkyBB8+PABp06dQu/evXH16lW0bNlSoJerpM9YWdvO2NgYDx8+RGRkJMaPH4+kpCT0798fEydOFEe1CANoQmcZdO7cGba2tnB3dy/27gxTU1OYmppi2rRpmDJlCjp16oRr166hW7duqFOnjtBhk/T0dMjKyqJWrVoSroFk6enpgcVi4fHjxxg0aFCR/Y8fP4aamppIv1ybNGlSpEv1x0TFX7HZbOjp6ZV4roYNG6J58+Zo3rw58vPzMWjQIDx8+FAizz7o1q0btm7dCgUFBdSrV69ImX/Uy8DAAElJSbC3t8f169cBlK0dVVRUitS/oKAAgYGBSExMFLh+QUEB/P39hfbaNGvWDJMmTYKbm5vAF295ycnJCf13E6datWrx28Df3x8WFhbYvXs3JkyYgAMHDiA7O1tgAiePxwOXy8WzZ8+gr68PfX19PHnypEzXlpeXR6dOndCpUye4urrif//7H7y9veHq6lphQ2/lpaSkhJ49e6Jnz57w8PDAxIkT4eXlxf9bV9JnTF9fH//99x/y8vJEDthlZGTQpk0btGnTBrNnz8b+/fsxZswYLFq0CE2aNClvtUgFo56LMlq5ciVOnz6NiIiIP+Y1NjYGAP64roGBAeLi4oqMa969exdNmjSR2K/oiqKhoYGePXvCz88PWVlZAvsSExMRFBQEe3v7IrfwMmHo0KGQk5ODn5+fRM7/44uuUaNGf/xSdXZ2xsOHD3H8+HEA4mvHH/MA7t27J/Br8+DBgzh27JjQ23wBwNPTE8+ePePPBZFGMjIyWLhwIRYvXoysrCz+nTC/tkNMTAw6deoEf39/AMDIkSPx7NkznDx5ssj5eDweMjIySn19Y2Nj5OfnF7kVVpoYGxsLzEkpyciRI5GZmVns56m491px1wVQ6muTyoWCizIyMzPDqFGjsGnTJoH0qVOnYtmyZbh58yZev36NyMhIODg4QFNTk3/74ahRo8BiseDg4IA7d+7gxYsX8Pf3x4YNG/hj5D8UFBQU6YIUNhxT2fj6+iInJwe2tra4fv063r59i3PnzqFnz56oX78+li9fLpHrfvnyBYmJiQLb58+fi83PYrEwc+ZMrFy5kr9+AVNq1qyJSZMmwcvLi9+9LI523L17N/r27QsLCwt+j5qpqSmGDx8OVVVVBAUFCT1OW1sbLi4uRd7j0mbYsGGQlZXFli1bcPfuXUycOFGgHUxNTTFixAgEBgYiPz8fw4cPh729PUaMGIEVK1bg9u3beP36Nf7991/Y2NjgypUrQq/TtWtXbN++HXfu3EFCQgJCQ0OxcOFCfm9lZZeSkoLu3btj//79iI2NRXx8PA4fPozVq1cLTG4t6TNmZWWFBQsWYO7cuViwYAEiIiLw+vVrhIWFYdiwYQgMDBR67aFDh2L9+vWIiorC69evcfXqVTg7O0NfX58/B4lIFwouysHb27vIPAobGxtERkZi2LBh0NfXx5AhQ6CkpISwsDBoaGgAKJwxfePGDeTl5eGvv/6CpaUlNm3ahHXr1vHXAfghMzMTLVq0ENj69+9fYXUsq+bNm+P27dto2rQphg8fjmbNmmHy5Mno1q0bIiIiBNZmECdPT0/UrVtXYFuwYEGJxzg6OiIvLw++vr4SKZMopk+fjsePH/MnGJa3HT99+oQzZ87w5xz8SkZGBoMGDSpx2GPevHlSv4aAnJwcpk+fDnd3d+jq6gr9sho0aBCSkpIQGhoKFouFAwcOYN26dThx4gS6dOkCc3NzLFmyBAMGDBC6IBgA2NraIjAwEL169YKRkRFmzJgBW1tbHDp0SNJVFAtlZWVYWVlh/fr16Ny5M0xNTeHh4YFJkyYJfDb+9BlbtWoVDhw4gKioKNja2sLExAQuLi4wNzeHo6Oj0Gvb2tri9OnT6N+/P/T19eHo6AhDQ0NcuHBB4sNoRDLokeuEEEIIESvquSCEEEKIWFFwQQghhBCxouCCEEIIIWJFwQUhhBBCxIqCC0IIIYSIFQUXhBBCCBErCi4IIYQQIlYUXBBCCCFErCi4IKQUdHV1sWHDBrGfd+zYsRJ9QmhJfq8Ti8XCiRMnAAAJCQlgsVi4f/8+I2UTtxMnTkBPTw+ysrKYPXs208UhpMqj4IJUCywWq8Tt18egV6SNGzcKPM66LJYsWSK0TqI+k+Hjx4+ws7MrV1lE9e3bN7i7u6NZs2ZQUlKCpqYmunTpIvShYeXh5OSEoUOH4u3bt1L1CHRCpBUt2k6qhY8fP/L/PyQkBJ6ennj69Ck/jannZ6ioqIjlPCYmJrh06ZJAmqjPZNDR0RFLWUQxZcoUREVFYfPmzTA2NkZKSgrCw8ORkpIitmtkZmYiKSkJtra2qFevntjOSwgpHvVckGpBR0eHv6moqIDFYvFff/36FaNGjYK2tjaUlZXRpk2bIl/UQOGv7PHjx6N27dpo1KgRduzYwd/3Yxjh0KFD6NSpE2rUqIE2bdrg2bNnuHXrFlq3bg1lZWXY2dkhOTmZf9zvwyJdu3bFzJkzsWDBAqirq0NHR6dUvSpycnICddTR0QGbzebvT0pKQv/+/VGjRg00adJE6JNQfx0WEebhw4ews7ODsrIytLW1MWbMGHA4nHKV/dSpU1i4cCH69OkDXV1dtGrVCjNmzMD48eP5eYQN3+zatQuDBg1CzZo10bx5c5w6dYq//+rVq2CxWEhPT8fVq1dRu3ZtAED37t3BYrFw9erVEstECCk/Ci5ItZeZmYk+ffogLCwM9+7dQ+/evdG/f3+8efNGIN/atWvRunVr3Lt3D9OmTcPUqVMFej8AwMvLC4sXL8bdu3chJyeHkSNHYsGCBdi4cSNu3LiBFy9ewNPTs8TyBAYGolatWoiKisLq1avh7e2NixcvlquOY8eOxdu3b3HlyhUcOXIEfn5+SEpKKvXx6enp6N69O1q0aIHbt2/j3Llz+PTpE4YPH16usuvo6CA0NBRfvnwRqT5Lly7F8OHDERsbiz59+mDUqFFITU0tkq99+/b8f6OjR4/i48ePaN++vUjXIoSUAY+QambPnj08FRWVEvOYmJjwNm/ezH/duHFj3ujRo/mvuVwuT0tLi7d161Yej8fjxcfH8wDwdu3axc9z8OBBHgBeWFgYP83Hx4dnYGDAf+3o6MgbMGAA/3WXLl14HTt2FChLmzZteK6ursWW1cvLiycjI8OrVauWwObk5MTj8Xi8p0+f8gDwoqOj+cc8fvyYB4C3fv16fhoA3vHjxwXqc+/ePR6Px+MtW7aM16tXL4Hrvn37lgeA9/Tp0zKX/dq1a7wGDRrw5OXlea1bt+bNnj2b999//wnkady4cZFyLl68mP86MzOTB4B39uxZHo/H4125coUHgJeWlsbj8Xi8tLQ0HgDelStXii0HIUS8aM4FqfYyMzOxZMkSnDlzBh8/fkR+fj6ysrKK9FyYm5vz///HsMrvv/5/zaOtrQ0AMDMzE0j7U4/Br+cAgLp16/7xGAMDA4GhAQCoU6cOAODx48eQk5NDq1at+PsMDQ2hqqpa4jl/FRMTgytXrgidm/Ly5Uvo6+uXqeydO3fGq1evEBkZifDwcISFhWHjxo1YunQpPDw8ij3u1+vUqlULderUEaknhhAiWRRckGpv3rx5uHjxItasWQM9PT3UqFEDQ4cORW5urkA+eXl5gdcsFgtcLrfYPCwWS2ja78f8rjTX+Z2CggL09PRKzFMemZmZ6N+/P1atWlVkX926dfn/X5ayy8vLo1OnTujUqRNcXV3xv//9D97e3nB1dYWCgkKxx4h6HUJIxaHgglR7N2/exNixYzFo0CAAhV+kCQkJzBZKjAwNDZGfn487d+6gTZs2AICnT58iPT291Odo2bIljh49Cl1dXZHvQhGVsbEx8vPzkZ2dXWxwQQip3GhCJ6n2mjdvjmPHjuH+/fuIiYnByJEjpe5XcH5+PhITEwW2T58+ASgcMunduzecnJwQFRWFO3fuYOLEiahRo0apz+/s7IzU1FSMGDECt27dwsuXL3H+/HmMGzcOBQUFZS53165dsX37dty5cwcJCQkIDQ3FwoUL0a1bN/6wDiFE+lBwQaq9devWQU1NDe3bt0f//v1ha2uLli1bMl0skcTFxaFu3boCW+PGjfn79+zZg3r16qFLly4YPHgwJk+eDC0trVKfv169erh58yYKCgrQq1cvmJmZYfbs2VBVVYWMTNn/jNja2iIwMBC9evWCkZERZsyYAVtbWxw6dKjM5ySEMI/F4/F4TBeCEEIIIVUH9VwQQgghRKwouCCEEEKIWFFwQQghhBCxouCCEEIIIWJFwQUhhBBCxIqCC0IIIYSIFQUXhBBCCBErCi4IIYQQIlYUXBBCCCFErCi4IIQQQohYUXBBCCGEELH6P97nRjcxysMgAAAAAElFTkSuQmCC\n"
          },
          "metadata": {}
        }
      ]
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3 (ipykernel)",
      "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"
    },
    "colab": {
      "provenance": []
    },
    "gpuClass": "premium",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "98d52d85159340beaa841dd93000693d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_b0753bb9e8fa48ac94e09d3554982e04",
              "IPY_MODEL_e9d0f7aadc1f4bd884dffb408801c594",
              "IPY_MODEL_1d43f272228144b78da88b84973ff52c"
            ],
            "layout": "IPY_MODEL_6fef1206fb87437c9470c241c8a728ba"
          }
        },
        "b0753bb9e8fa48ac94e09d3554982e04": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ec26ce0de028473ebf458bc103ca7c67",
            "placeholder": "​",
            "style": "IPY_MODEL_ec3df2b46c9e4b7a9f56e4b95acb78e9",
            "value": "Downloading (…)solve/main/vocab.txt: 100%"
          }
        },
        "e9d0f7aadc1f4bd884dffb408801c594": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_5f7329df42fc4171a7ae2e66b33ba062",
            "max": 1233088,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_b4c8ad7e5b404d65b813fb89c628dad2",
            "value": 1233088
          }
        },
        "1d43f272228144b78da88b84973ff52c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_7305ee44271d478eb7747ee233d9ac7c",
            "placeholder": "​",
            "style": "IPY_MODEL_a5b805a2c32a4f88b575d32edd88de07",
            "value": " 1.23M/1.23M [00:00&lt;00:00, 7.51MB/s]"
          }
        },
        "6fef1206fb87437c9470c241c8a728ba": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "ec26ce0de028473ebf458bc103ca7c67": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "ec3df2b46c9e4b7a9f56e4b95acb78e9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "5f7329df42fc4171a7ae2e66b33ba062": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b4c8ad7e5b404d65b813fb89c628dad2": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "7305ee44271d478eb7747ee233d9ac7c": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "a5b805a2c32a4f88b575d32edd88de07": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "239fbf31456543ed9f45c2feea4cb617": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_146753eb82dd4251af66a62046545e24",
              "IPY_MODEL_7ba0d78f51f9461299cd385167f5fe5b",
              "IPY_MODEL_e795f26afb1f4b20b581f5e325f6b32f"
            ],
            "layout": "IPY_MODEL_101ddce618334cf6874b0d473dd82fe8"
          }
        },
        "146753eb82dd4251af66a62046545e24": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_f93851550a954d9d80a2c308ed0199bd",
            "placeholder": "​",
            "style": "IPY_MODEL_aa46630642d04dbca738575b650d86a1",
            "value": "Downloading (…)okenizer_config.json: 100%"
          }
        },
        "7ba0d78f51f9461299cd385167f5fe5b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_3ab878e4c538412cbe13d409a11e31b4",
            "max": 59,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_85636f736a534a47811513098264d8a2",
            "value": 59
          }
        },
        "e795f26afb1f4b20b581f5e325f6b32f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_48086df862604f618d778f86ef195475",
            "placeholder": "​",
            "style": "IPY_MODEL_62dd7bdbcaf04cef8987e8689d2d74a9",
            "value": " 59.0/59.0 [00:00&lt;00:00, 2.85kB/s]"
          }
        },
        "101ddce618334cf6874b0d473dd82fe8": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f93851550a954d9d80a2c308ed0199bd": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "aa46630642d04dbca738575b650d86a1": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "3ab878e4c538412cbe13d409a11e31b4": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "85636f736a534a47811513098264d8a2": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "48086df862604f618d778f86ef195475": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "62dd7bdbcaf04cef8987e8689d2d74a9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "2bf25e6999804f4e9e0b845e3e79c5b0": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_8e691692c11349469eb39327c23fead5",
              "IPY_MODEL_045dc68924084dc195fdb5a277753a5e",
              "IPY_MODEL_a422ea5a73aa469eb4121ccf9cb8ca1e"
            ],
            "layout": "IPY_MODEL_c4fb9b4f26d04bd0bff010382b9eb460"
          }
        },
        "8e691692c11349469eb39327c23fead5": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_9750c6ff12d34b9cb99b3ba5b4ea27fc",
            "placeholder": "​",
            "style": "IPY_MODEL_0f495859b0b54cfe85f580f5850bc535",
            "value": "Downloading (…)lve/main/config.json: 100%"
          }
        },
        "045dc68924084dc195fdb5a277753a5e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_50e3784f0fb340e2a2c14fe2d2fa4f90",
            "max": 386,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_8b35ee65c7d441478b137f71c565263f",
            "value": 386
          }
        },
        "a422ea5a73aa469eb4121ccf9cb8ca1e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_9f8a65a50cd240e9b206d45a07e2c70b",
            "placeholder": "​",
            "style": "IPY_MODEL_8c1125f582db4ebbbfb48bb1f104a7ae",
            "value": " 386/386 [00:00&lt;00:00, 25.8kB/s]"
          }
        },
        "c4fb9b4f26d04bd0bff010382b9eb460": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "9750c6ff12d34b9cb99b3ba5b4ea27fc": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "0f495859b0b54cfe85f580f5850bc535": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "50e3784f0fb340e2a2c14fe2d2fa4f90": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "8b35ee65c7d441478b137f71c565263f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "9f8a65a50cd240e9b206d45a07e2c70b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "8c1125f582db4ebbbfb48bb1f104a7ae": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "af32f29cb21d4142a43acdbabe4921ff": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_3585acb5179c4b89b6cb7e357674e083",
              "IPY_MODEL_2951182944e94fc299dea78b56c4f195",
              "IPY_MODEL_3a6e4df4c1644ecb9d36e720cea44e30"
            ],
            "layout": "IPY_MODEL_b65f93de00404a05b4691709da027f69"
          }
        },
        "3585acb5179c4b89b6cb7e357674e083": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_6f5f79f50a564d30b1cc13e179cd48ef",
            "placeholder": "​",
            "style": "IPY_MODEL_9d2147283d9040efbdd5d1af509890bf",
            "value": "Downloading pytorch_model.bin: 100%"
          }
        },
        "2951182944e94fc299dea78b56c4f195": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ed6d68a54c0f49bead78ac7bbec4d589",
            "max": 740314769,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_49312a16f029417c95b9d3491f09ab9b",
            "value": 740314769
          }
        },
        "3a6e4df4c1644ecb9d36e720cea44e30": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_86bc31432baf47e6b3e4669e6331a20f",
            "placeholder": "​",
            "style": "IPY_MODEL_eed8b4a68f914bd8a0c3340b15536001",
            "value": " 740M/740M [00:20&lt;00:00, 46.5MB/s]"
          }
        },
        "b65f93de00404a05b4691709da027f69": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "6f5f79f50a564d30b1cc13e179cd48ef": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "9d2147283d9040efbdd5d1af509890bf": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "ed6d68a54c0f49bead78ac7bbec4d589": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "49312a16f029417c95b9d3491f09ab9b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "86bc31432baf47e6b3e4669e6331a20f": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "eed8b4a68f914bd8a0c3340b15536001": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    },
    "accelerator": "GPU"
  },
  "nbformat": 4,
  "nbformat_minor": 0
}