TRAINING.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "trW-06ieoj8y"
      },
      "source": [
        "Установка библиотек"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "k-4cIwVtivDk"
      },
      "outputs": [],
      "source": [
        "! git clone https://github.com/cbluebenchmark/cblue"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "iy_6-bWOi5gc"
      },
      "outputs": [],
      "source": [
        "! pip install jieba"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "5C6K9dsLi5jE"
      },
      "outputs": [],
      "source": [
        "! pip install gensim"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nCcFkGzSi5lf",
        "outputId": "b1904b7b-fb84-44ef-e9d0-5f4c39743b1a"
      },
      "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: scikit-learn in /usr/local/lib/python3.10/dist-packages (1.2.2)\n",
            "Requirement already satisfied: numpy>=1.17.3 in /usr/local/lib/python3.10/dist-packages (from scikit-learn) (1.22.4)\n",
            "Requirement already satisfied: scipy>=1.3.2 in /usr/local/lib/python3.10/dist-packages (from scikit-learn) (1.10.1)\n",
            "Requirement already satisfied: joblib>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from scikit-learn) (1.2.0)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn) (3.1.0)\n"
          ]
        }
      ],
      "source": [
        "! pip install scikit-learn"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fUKyRXBZslpm",
        "outputId": "abc1d992-de7e-4cc8-db3b-005592c91c2e"
      },
      "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 git+https://github.com/sinovation/ZEN\n",
            "  Cloning https://github.com/sinovation/ZEN to /tmp/pip-req-build-jaueqsd8\n",
            "  Running command git clone --filter=blob:none --quiet https://github.com/sinovation/ZEN /tmp/pip-req-build-jaueqsd8\n",
            "  Resolved https://github.com/sinovation/ZEN to commit d5d9008b19643ae06647a390266030cbf1c1c63c\n",
            "  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.10/dist-packages (from ZEN==0.1.0) (1.22.4)\n",
            "Collecting boto3 (from ZEN==0.1.0)\n",
            "  Downloading boto3-1.26.148-py3-none-any.whl (135 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m135.6/135.6 kB\u001b[0m \u001b[31m2.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from ZEN==0.1.0) (2.27.1)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.10/dist-packages (from ZEN==0.1.0) (4.65.0)\n",
            "Requirement already satisfied: regex in /usr/local/lib/python3.10/dist-packages (from ZEN==0.1.0) (2022.10.31)\n",
            "Collecting botocore<1.30.0,>=1.29.148 (from boto3->ZEN==0.1.0)\n",
            "  Downloading botocore-1.29.148-py3-none-any.whl (10.8 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m10.8/10.8 MB\u001b[0m \u001b[31m38.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hCollecting jmespath<2.0.0,>=0.7.1 (from boto3->ZEN==0.1.0)\n",
            "  Downloading jmespath-1.0.1-py3-none-any.whl (20 kB)\n",
            "Collecting s3transfer<0.7.0,>=0.6.0 (from boto3->ZEN==0.1.0)\n",
            "  Downloading s3transfer-0.6.1-py3-none-any.whl (79 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m79.8/79.8 kB\u001b[0m \u001b[31m10.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests->ZEN==0.1.0) (1.26.15)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests->ZEN==0.1.0) (2022.12.7)\n",
            "Requirement already satisfied: charset-normalizer~=2.0.0 in /usr/local/lib/python3.10/dist-packages (from requests->ZEN==0.1.0) (2.0.12)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->ZEN==0.1.0) (3.4)\n",
            "Requirement already satisfied: python-dateutil<3.0.0,>=2.1 in /usr/local/lib/python3.10/dist-packages (from botocore<1.30.0,>=1.29.148->boto3->ZEN==0.1.0) (2.8.2)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil<3.0.0,>=2.1->botocore<1.30.0,>=1.29.148->boto3->ZEN==0.1.0) (1.16.0)\n",
            "Building wheels for collected packages: ZEN\n",
            "  Building wheel for ZEN (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for ZEN: filename=ZEN-0.1.0-py3-none-any.whl size=32503 sha256=d81b0020aa19f886830938b7c1a25a9c2a6156559cb53c7a6b767ee7b6638228\n",
            "  Stored in directory: /tmp/pip-ephem-wheel-cache-_1sm_1di/wheels/12/fc/a2/9bc51ebe2f1d8e7ff3e6b09e98051ac69c710a4fc34aa72b41\n",
            "Successfully built ZEN\n",
            "Installing collected packages: jmespath, botocore, s3transfer, boto3, ZEN\n",
            "Successfully installed ZEN-0.1.0 boto3-1.26.148 botocore-1.29.148 jmespath-1.0.1 s3transfer-0.6.1\n"
          ]
        }
      ],
      "source": [
        "!pip install git+https://github.com/sinovation/ZEN"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_QzQ05hK2NGp",
        "outputId": "9200d92e-1309-4e23-c6a0-90e9e55adfa4"
      },
      "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-4.29.2-py3-none-any.whl (7.1 MB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m7.1/7.1 MB\u001b[0m \u001b[31m34.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from transformers) (3.12.0)\n",
            "Collecting huggingface-hub<1.0,>=0.14.1 (from transformers)\n",
            "  Downloading huggingface_hub-0.15.1-py3-none-any.whl (236 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m236.8/236.8 kB\u001b[0m \u001b[31m25.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: numpy>=1.17 in /usr/local/lib/python3.10/dist-packages (from transformers) (1.22.4)\n",
            "Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.10/dist-packages (from transformers) (23.1)\n",
            "Requirement already satisfied: pyyaml>=5.1 in /usr/local/lib/python3.10/dist-packages (from transformers) (6.0)\n",
            "Requirement already satisfied: regex!=2019.12.17 in /usr/local/lib/python3.10/dist-packages (from transformers) (2022.10.31)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from transformers) (2.27.1)\n",
            "Collecting tokenizers!=0.11.3,<0.14,>=0.11.1 (from transformers)\n",
            "  Downloading tokenizers-0.13.3-cp310-cp310-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[31m69.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: tqdm>=4.27 in /usr/local/lib/python3.10/dist-packages (from transformers) (4.65.0)\n",
            "Requirement already satisfied: fsspec in /usr/local/lib/python3.10/dist-packages (from huggingface-hub<1.0,>=0.14.1->transformers) (2023.4.0)\n",
            "Requirement already satisfied: typing-extensions>=3.7.4.3 in /usr/local/lib/python3.10/dist-packages (from huggingface-hub<1.0,>=0.14.1->transformers) (4.5.0)\n",
            "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests->transformers) (1.26.15)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests->transformers) (2022.12.7)\n",
            "Requirement already satisfied: charset-normalizer~=2.0.0 in /usr/local/lib/python3.10/dist-packages (from requests->transformers) (2.0.12)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->transformers) (3.4)\n",
            "Installing collected packages: tokenizers, huggingface-hub, transformers\n",
            "Successfully installed huggingface-hub-0.15.1 tokenizers-0.13.3 transformers-4.29.2\n",
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Requirement already satisfied: torch in /usr/local/lib/python3.10/dist-packages (2.0.1+cu118)\n",
            "Requirement already satisfied: torchvision in /usr/local/lib/python3.10/dist-packages (0.15.2+cu118)\n",
            "Requirement already satisfied: torchaudio in /usr/local/lib/python3.10/dist-packages (2.0.2+cu118)\n",
            "Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from torch) (3.12.0)\n",
            "Requirement already satisfied: typing-extensions in /usr/local/lib/python3.10/dist-packages (from torch) (4.5.0)\n",
            "Requirement already satisfied: sympy in /usr/local/lib/python3.10/dist-packages (from torch) (1.11.1)\n",
            "Requirement already satisfied: networkx in /usr/local/lib/python3.10/dist-packages (from torch) (3.1)\n",
            "Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from torch) (3.1.2)\n",
            "Requirement already satisfied: triton==2.0.0 in /usr/local/lib/python3.10/dist-packages (from torch) (2.0.0)\n",
            "Requirement already satisfied: cmake in /usr/local/lib/python3.10/dist-packages (from triton==2.0.0->torch) (3.25.2)\n",
            "Requirement already satisfied: lit in /usr/local/lib/python3.10/dist-packages (from triton==2.0.0->torch) (16.0.5)\n",
            "Requirement already satisfied: numpy in /usr/local/lib/python3.10/dist-packages (from torchvision) (1.22.4)\n",
            "Requirement already satisfied: requests in /usr/local/lib/python3.10/dist-packages (from torchvision) (2.27.1)\n",
            "Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /usr/local/lib/python3.10/dist-packages (from torchvision) (8.4.0)\n",
            "Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->torch) (2.1.2)\n",
            "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (1.26.15)\n",
            "Requirement already satisfied: certifi>=2017.4.17 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (2022.12.7)\n",
            "Requirement already satisfied: charset-normalizer~=2.0.0 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (2.0.12)\n",
            "Requirement already satisfied: idna<4,>=2.5 in /usr/local/lib/python3.10/dist-packages (from requests->torchvision) (3.4)\n",
            "Requirement already satisfied: mpmath>=0.19 in /usr/local/lib/python3.10/dist-packages (from sympy->torch) (1.3.0)\n"
          ]
        }
      ],
      "source": [
        "!pip install transformers\n",
        "!pip install torch torchvision torchaudio"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IDM-vrgRo-kc"
      },
      "source": [
        "Импортируем зависимости"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "wJ3CCRmfwuqa"
      },
      "outputs": [],
      "source": [
        "import torch\n",
        "import torch.nn as nn\n",
        "import numpy as np\n",
        "import argparse\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "from numpy import arange\n",
        "\n",
        "from ZEN.modeling import ZenModel\n",
        "\n",
        "from cblue.data import EEDataProcessor, EEDataset\n",
        "from cblue.trainer import EETrainer\n",
        "from cblue.metrics import ee_metric, ee_commit_prediction\n",
        "\n",
        "from cblue.models import CDNForCLSModel\n",
        "from cblue.trainer import CDNForCLSTrainer, CDNForNUMTrainer\n",
        "from cblue.utils import init_logger, seed_everything\n",
        "from cblue.data import CDNDataset, CDNDataProcessor\n",
        "from cblue.models import save_zen_model, ZenModel, ZenForSequenceClassification, ZenNgramDict\n",
        "\n",
        "from transformers import BertTokenizer, BertModel, AlbertModel, BertForSequenceClassification, \\\n",
        "    AlbertForSequenceClassification, BertForTokenClassification, AlbertForTokenClassification, RobertaTokenizer, AutoTokenizer"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "J2qK9FSdAwDa"
      },
      "source": [
        "# Новый раздел"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rTvG5xIdpRg-"
      },
      "source": [
        "Загружаем датасет"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "dbPDy-GUybXl"
      },
      "outputs": [],
      "source": [
        "data_processor = EEDataProcessor('/content/')\n",
        "train_samples = data_processor.get_train_sample()\n",
        "eval_samples = data_processor.get_dev_sample()\n",
        "test_samples = data_processor.get_test_sample()\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "fm5n-otnBzoy"
      },
      "outputs": [],
      "source": [
        "from transformers import BertTokenizer, BertModel, AlbertModel, BertForSequenceClassification, AlbertForSequenceClassification, BertForTokenClassification, AlbertForTokenClassification\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "I9cjyh-XpW3L"
      },
      "source": [
        "Загружаем модель"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "2OIZWrRWCk6N"
      },
      "outputs": [],
      "source": [
        "from transformers import (\n",
        "  BertTokenizerFast,\n",
        "  AutoModel,\n",
        ")\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "r2LQrosiCymx"
      },
      "outputs": [],
      "source": [
        "from transformers import (RobertaForTokenClassification )"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MMzkBUepF6mk",
        "outputId": "b23a7791-f825-48ae-ef1e-4f58f5cfbff7"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "You are using a model of type bert to instantiate a model of type roberta. This is not supported for all configurations of models and can yield errors.\n",
            "Some weights of the model checkpoint at hfl/chinese-roberta-wwm-ext-large were not used when initializing RobertaForTokenClassification: ['bert.encoder.layer.0.attention.output.dense.weight', 'bert.encoder.layer.1.output.LayerNorm.bias', 'bert.encoder.layer.7.output.LayerNorm.weight', 'bert.encoder.layer.2.output.dense.weight', 'bert.encoder.layer.1.attention.output.dense.bias', 'bert.encoder.layer.0.attention.self.key.weight', 'bert.encoder.layer.17.output.dense.weight', 'bert.encoder.layer.5.output.dense.bias', 'bert.encoder.layer.10.attention.self.value.bias', 'bert.encoder.layer.10.output.dense.weight', 'bert.encoder.layer.19.attention.self.query.bias', 'bert.encoder.layer.3.output.LayerNorm.weight', 'bert.encoder.layer.2.attention.output.dense.weight', 'bert.encoder.layer.15.attention.output.LayerNorm.weight', 'bert.encoder.layer.14.attention.self.key.bias', 'bert.encoder.layer.2.intermediate.dense.weight', 'bert.encoder.layer.0.attention.self.value.bias', 'bert.encoder.layer.5.attention.self.query.bias', 'bert.encoder.layer.4.attention.self.value.bias', 'bert.encoder.layer.14.attention.self.value.bias', 'bert.encoder.layer.14.attention.output.LayerNorm.bias', 'bert.encoder.layer.3.intermediate.dense.weight', 'bert.encoder.layer.16.attention.self.key.weight', 'bert.encoder.layer.22.attention.self.value.weight', 'bert.encoder.layer.23.intermediate.dense.bias', 'bert.encoder.layer.21.intermediate.dense.weight', 'bert.encoder.layer.23.attention.self.key.bias', 'bert.encoder.layer.7.attention.self.key.bias', 'bert.encoder.layer.23.output.LayerNorm.weight', 'bert.encoder.layer.5.output.LayerNorm.bias', 'bert.encoder.layer.8.attention.self.value.weight', 'bert.encoder.layer.3.attention.output.dense.bias', 'bert.encoder.layer.15.output.LayerNorm.bias', 'bert.encoder.layer.13.attention.self.query.weight', 'bert.encoder.layer.16.intermediate.dense.bias', 'bert.encoder.layer.1.output.dense.weight', 'bert.encoder.layer.12.attention.self.value.weight', 'bert.encoder.layer.21.attention.output.dense.bias', 'bert.encoder.layer.21.attention.output.LayerNorm.bias', 'bert.encoder.layer.3.output.dense.weight', 'bert.encoder.layer.16.intermediate.dense.weight', 'bert.encoder.layer.19.attention.self.key.weight', 'bert.encoder.layer.23.attention.output.LayerNorm.bias', 'bert.encoder.layer.2.attention.self.value.bias', 'bert.encoder.layer.12.attention.self.query.weight', 'bert.encoder.layer.20.attention.output.dense.bias', 'bert.encoder.layer.12.output.LayerNorm.bias', 'bert.encoder.layer.23.output.LayerNorm.bias', 'bert.encoder.layer.0.intermediate.dense.weight', 'cls.seq_relationship.weight', 'cls.predictions.transform.LayerNorm.weight', 'bert.encoder.layer.3.attention.self.value.bias', 'bert.encoder.layer.1.attention.self.value.bias', 'bert.encoder.layer.19.attention.output.dense.weight', 'bert.encoder.layer.14.attention.self.key.weight', 'bert.encoder.layer.10.attention.output.LayerNorm.weight', 'bert.encoder.layer.2.attention.output.dense.bias', 'bert.encoder.layer.20.output.LayerNorm.bias', 'bert.encoder.layer.5.intermediate.dense.bias', 'bert.encoder.layer.0.attention.output.dense.bias', 'bert.encoder.layer.4.output.dense.bias', 'bert.encoder.layer.6.attention.self.query.bias', 'bert.encoder.layer.0.attention.output.LayerNorm.bias', 'bert.encoder.layer.7.output.dense.weight', 'bert.encoder.layer.1.attention.self.query.bias', 'bert.encoder.layer.22.attention.self.key.bias', 'bert.encoder.layer.10.output.dense.bias', 'bert.encoder.layer.9.attention.output.dense.weight', 'bert.encoder.layer.7.intermediate.dense.bias', 'bert.encoder.layer.20.intermediate.dense.weight', 'bert.encoder.layer.9.output.LayerNorm.weight', 'bert.encoder.layer.1.attention.self.key.bias', 'cls.predictions.transform.dense.bias', 'bert.encoder.layer.3.attention.self.key.weight', 'bert.encoder.layer.20.attention.self.key.weight', 'bert.encoder.layer.14.attention.output.dense.bias', 'bert.encoder.layer.16.attention.self.value.weight', 'bert.encoder.layer.17.attention.self.query.bias', 'bert.encoder.layer.13.output.LayerNorm.bias', 'bert.encoder.layer.13.attention.self.value.weight', 'bert.encoder.layer.0.intermediate.dense.bias', 'bert.encoder.layer.0.attention.self.query.bias', 'bert.encoder.layer.13.attention.output.dense.weight', 'bert.encoder.layer.17.attention.self.value.weight', 'bert.encoder.layer.15.attention.self.key.weight', 'bert.encoder.layer.22.attention.output.dense.weight', 'bert.encoder.layer.4.attention.self.query.weight', 'cls.seq_relationship.bias', 'bert.encoder.layer.1.attention.output.LayerNorm.weight', 'bert.encoder.layer.5.attention.self.query.weight', 'bert.encoder.layer.13.attention.output.LayerNorm.bias', 'bert.encoder.layer.23.attention.output.LayerNorm.weight', 'bert.encoder.layer.10.attention.self.value.weight', 'bert.encoder.layer.19.output.dense.bias', 'bert.encoder.layer.4.attention.output.dense.bias', 'bert.encoder.layer.8.attention.self.key.bias', 'bert.encoder.layer.15.intermediate.dense.weight', 'cls.predictions.bias', 'bert.encoder.layer.2.output.LayerNorm.bias', 'bert.encoder.layer.23.attention.self.key.weight', 'bert.encoder.layer.21.attention.output.dense.weight', 'bert.encoder.layer.12.attention.self.key.weight', 'bert.encoder.layer.5.output.dense.weight', 'bert.encoder.layer.10.intermediate.dense.bias', 'bert.encoder.layer.16.attention.self.query.weight', 'bert.encoder.layer.20.attention.self.value.bias', 'bert.encoder.layer.10.output.LayerNorm.weight', 'bert.encoder.layer.7.output.dense.bias', 'bert.encoder.layer.19.attention.output.LayerNorm.weight', 'bert.encoder.layer.21.attention.self.value.bias', 'bert.encoder.layer.5.intermediate.dense.weight', 'bert.encoder.layer.8.attention.output.LayerNorm.weight', 'bert.encoder.layer.12.intermediate.dense.weight', 'bert.embeddings.word_embeddings.weight', 'bert.encoder.layer.9.attention.self.value.weight', 'bert.encoder.layer.3.attention.output.dense.weight', 'bert.encoder.layer.19.attention.output.LayerNorm.bias', 'bert.encoder.layer.11.attention.self.value.bias', 'bert.encoder.layer.3.output.LayerNorm.bias', 'bert.encoder.layer.5.attention.output.dense.weight', 'bert.encoder.layer.13.intermediate.dense.weight', 'bert.encoder.layer.17.attention.output.LayerNorm.bias', 'bert.encoder.layer.2.attention.self.query.weight', 'bert.encoder.layer.11.output.LayerNorm.bias', 'bert.encoder.layer.7.attention.self.key.weight', 'bert.encoder.layer.11.attention.output.dense.bias', 'bert.encoder.layer.5.attention.output.dense.bias', 'bert.encoder.layer.4.output.dense.weight', 'bert.encoder.layer.12.attention.output.dense.weight', 'bert.encoder.layer.0.attention.self.query.weight', 'bert.encoder.layer.10.output.LayerNorm.bias', 'bert.encoder.layer.5.attention.self.value.weight', 'bert.encoder.layer.6.intermediate.dense.bias', 'bert.encoder.layer.16.attention.output.LayerNorm.bias', 'bert.encoder.layer.14.attention.output.dense.weight', 'bert.encoder.layer.19.attention.output.dense.bias', 'bert.encoder.layer.11.attention.output.LayerNorm.bias', 'bert.encoder.layer.9.output.dense.bias', 'bert.encoder.layer.20.attention.output.LayerNorm.bias', 'bert.encoder.layer.16.attention.output.LayerNorm.weight', 'bert.encoder.layer.22.output.dense.weight', 'bert.encoder.layer.2.attention.output.LayerNorm.bias', 'bert.encoder.layer.10.attention.self.key.bias', 'bert.encoder.layer.9.intermediate.dense.weight', 'bert.encoder.layer.1.intermediate.dense.weight', 'bert.encoder.layer.9.attention.output.LayerNorm.weight', 'bert.encoder.layer.20.attention.self.value.weight', 'bert.encoder.layer.2.intermediate.dense.bias', 'bert.encoder.layer.19.output.LayerNorm.weight', 'bert.encoder.layer.9.attention.self.query.bias', 'bert.encoder.layer.6.output.dense.bias', 'bert.encoder.layer.13.intermediate.dense.bias', 'bert.encoder.layer.3.intermediate.dense.bias', 'bert.encoder.layer.20.attention.output.LayerNorm.weight', 'bert.encoder.layer.1.attention.self.query.weight', 'bert.encoder.layer.18.attention.self.value.weight', 'bert.encoder.layer.1.attention.self.value.weight', 'bert.encoder.layer.7.attention.output.dense.bias', 'bert.encoder.layer.3.attention.self.value.weight', 'bert.encoder.layer.7.attention.output.LayerNorm.bias', 'bert.encoder.layer.16.attention.self.key.bias', 'bert.encoder.layer.15.attention.self.value.weight', 'bert.encoder.layer.19.output.LayerNorm.bias', 'bert.encoder.layer.15.output.LayerNorm.weight', 'bert.encoder.layer.17.attention.self.value.bias', 'bert.encoder.layer.21.output.LayerNorm.bias', 'bert.encoder.layer.22.attention.self.query.weight', 'bert.encoder.layer.12.attention.output.LayerNorm.bias', 'bert.encoder.layer.17.intermediate.dense.bias', 'bert.encoder.layer.11.output.dense.weight', 'bert.encoder.layer.17.attention.self.key.weight', 'bert.encoder.layer.21.output.dense.weight', 'bert.encoder.layer.14.attention.output.LayerNorm.weight', 'bert.encoder.layer.6.attention.self.key.bias', 'bert.encoder.layer.18.output.dense.weight', 'bert.encoder.layer.22.attention.output.dense.bias', 'bert.encoder.layer.5.attention.self.key.bias', 'bert.encoder.layer.14.output.dense.bias', 'cls.predictions.decoder.weight', 'bert.encoder.layer.3.attention.output.LayerNorm.bias', 'bert.encoder.layer.6.attention.output.dense.weight', 'bert.encoder.layer.21.attention.self.value.weight', 'bert.encoder.layer.16.output.dense.weight', 'bert.encoder.layer.23.attention.self.query.bias', 'bert.encoder.layer.22.output.LayerNorm.bias', 'bert.encoder.layer.11.attention.output.LayerNorm.weight', 'bert.encoder.layer.8.attention.output.LayerNorm.bias', 'bert.encoder.layer.13.attention.output.dense.bias', 'bert.encoder.layer.4.attention.self.key.weight', 'bert.encoder.layer.13.attention.self.value.bias', 'bert.encoder.layer.14.output.LayerNorm.bias', 'bert.encoder.layer.14.intermediate.dense.weight', 'bert.encoder.layer.9.output.LayerNorm.bias', 'bert.encoder.layer.22.attention.self.key.weight', 'bert.encoder.layer.9.attention.self.key.bias', 'bert.encoder.layer.0.output.dense.weight', 'bert.encoder.layer.15.attention.output.dense.weight', 'bert.encoder.layer.17.intermediate.dense.weight', 'bert.encoder.layer.22.intermediate.dense.bias', 'bert.encoder.layer.12.attention.self.value.bias', 'bert.encoder.layer.12.attention.self.query.bias', 'bert.encoder.layer.9.attention.self.key.weight', 'bert.encoder.layer.17.attention.self.query.weight', 'bert.encoder.layer.15.intermediate.dense.bias', 'bert.encoder.layer.14.output.dense.weight', 'bert.encoder.layer.11.attention.output.dense.weight', 'bert.encoder.layer.4.attention.self.key.bias', 'bert.encoder.layer.15.output.dense.weight', 'bert.encoder.layer.21.attention.self.query.bias', 'bert.encoder.layer.15.attention.self.query.bias', 'bert.encoder.layer.4.attention.output.LayerNorm.weight', 'bert.encoder.layer.23.attention.self.query.weight', 'bert.encoder.layer.23.attention.output.dense.weight', 'bert.encoder.layer.8.output.dense.bias', 'bert.encoder.layer.16.attention.output.dense.bias', 'bert.encoder.layer.18.attention.output.dense.bias', 'bert.encoder.layer.6.output.LayerNorm.bias', 'bert.encoder.layer.2.output.LayerNorm.weight', 'bert.encoder.layer.18.attention.self.key.weight', 'bert.encoder.layer.17.attention.self.key.bias', 'bert.encoder.layer.23.output.dense.bias', 'bert.encoder.layer.0.attention.self.key.bias', 'bert.encoder.layer.19.intermediate.dense.weight', 'bert.encoder.layer.13.output.dense.weight', 'bert.encoder.layer.18.intermediate.dense.bias', 'bert.encoder.layer.21.attention.self.key.bias', 'bert.encoder.layer.23.attention.self.value.bias', 'bert.encoder.layer.2.attention.self.value.weight', 'bert.encoder.layer.7.attention.self.value.weight', 'bert.encoder.layer.2.attention.self.query.bias', 'bert.encoder.layer.10.attention.self.query.weight', 'bert.encoder.layer.22.output.LayerNorm.weight', 'bert.encoder.layer.7.attention.self.query.weight', 'bert.encoder.layer.22.attention.self.query.bias', 'bert.encoder.layer.18.intermediate.dense.weight', 'bert.encoder.layer.22.attention.self.value.bias', 'bert.encoder.layer.4.attention.self.query.bias', 'bert.encoder.layer.16.attention.output.dense.weight', 'bert.encoder.layer.20.output.dense.weight', 'bert.encoder.layer.15.attention.self.query.weight', 'bert.encoder.layer.21.output.LayerNorm.weight', 'bert.embeddings.LayerNorm.weight', 'bert.encoder.layer.11.attention.self.query.bias', 'bert.encoder.layer.14.output.LayerNorm.weight', 'bert.encoder.layer.13.attention.self.query.bias', 'bert.encoder.layer.23.attention.output.dense.bias', 'bert.encoder.layer.4.output.LayerNorm.bias', 'bert.encoder.layer.9.output.dense.weight', 'bert.encoder.layer.6.intermediate.dense.weight', 'bert.encoder.layer.11.intermediate.dense.bias', 'bert.encoder.layer.11.attention.self.key.bias', 'bert.encoder.layer.20.attention.self.query.bias', 'bert.encoder.layer.10.attention.output.LayerNorm.bias', 'bert.encoder.layer.5.attention.self.key.weight', 'bert.encoder.layer.5.output.LayerNorm.weight', 'bert.encoder.layer.4.intermediate.dense.weight', 'bert.encoder.layer.12.output.dense.bias', 'bert.encoder.layer.18.output.dense.bias', 'bert.encoder.layer.16.attention.self.query.bias', 'bert.encoder.layer.7.attention.self.query.bias', 'bert.encoder.layer.15.attention.self.key.bias', 'bert.encoder.layer.6.attention.self.value.weight', 'bert.encoder.layer.13.output.dense.bias', 'bert.encoder.layer.2.attention.self.key.bias', 'bert.encoder.layer.11.output.dense.bias', 'bert.encoder.layer.3.attention.self.key.bias', 'bert.encoder.layer.6.attention.self.query.weight', 'bert.encoder.layer.19.attention.self.value.bias', 'bert.encoder.layer.18.attention.output.dense.weight', 'bert.encoder.layer.19.intermediate.dense.bias', 'bert.encoder.layer.15.attention.output.dense.bias', 'bert.encoder.layer.16.output.dense.bias', 'bert.encoder.layer.6.attention.output.LayerNorm.weight', 'cls.predictions.transform.LayerNorm.bias', 'bert.encoder.layer.3.attention.output.LayerNorm.weight', 'bert.encoder.layer.8.intermediate.dense.weight', 'bert.encoder.layer.16.attention.self.value.bias', 'bert.encoder.layer.11.attention.self.query.weight', 'bert.encoder.layer.0.output.LayerNorm.weight', 'bert.encoder.layer.7.output.LayerNorm.bias', 'bert.encoder.layer.4.output.LayerNorm.weight', 'bert.encoder.layer.2.attention.self.key.weight', 'bert.encoder.layer.12.output.dense.weight', 'bert.encoder.layer.23.intermediate.dense.weight', 'bert.encoder.layer.15.attention.self.value.bias', 'bert.encoder.layer.8.output.dense.weight', 'bert.encoder.layer.12.attention.self.key.bias', 'bert.encoder.layer.22.intermediate.dense.weight', 'bert.encoder.layer.8.output.LayerNorm.bias', 'bert.embeddings.token_type_embeddings.weight', 'bert.encoder.layer.12.output.LayerNorm.weight', 'bert.encoder.layer.13.attention.self.key.bias', 'bert.encoder.layer.15.attention.output.LayerNorm.bias', 'bert.encoder.layer.20.attention.self.key.bias', 'bert.encoder.layer.1.intermediate.dense.bias', 'bert.encoder.layer.23.attention.self.value.weight', 'bert.encoder.layer.18.attention.output.LayerNorm.weight', 'bert.encoder.layer.20.output.dense.bias', 'bert.encoder.layer.3.attention.self.query.bias', 'bert.encoder.layer.8.attention.self.value.bias', 'bert.encoder.layer.20.output.LayerNorm.weight', 'bert.encoder.layer.18.attention.self.key.bias', 'bert.encoder.layer.18.attention.self.value.bias', 'bert.encoder.layer.21.output.dense.bias', 'bert.encoder.layer.5.attention.output.LayerNorm.bias', 'bert.encoder.layer.18.output.LayerNorm.weight', 'bert.encoder.layer.3.output.dense.bias', 'bert.encoder.layer.19.attention.self.key.bias', 'bert.encoder.layer.5.attention.output.LayerNorm.weight', 'bert.encoder.layer.7.attention.self.value.bias', 'bert.encoder.layer.15.output.dense.bias', 'bert.encoder.layer.18.output.LayerNorm.bias', 'bert.encoder.layer.23.output.dense.weight', 'bert.encoder.layer.13.attention.output.LayerNorm.weight', 'bert.encoder.layer.8.intermediate.dense.bias', 'bert.encoder.layer.9.intermediate.dense.bias', 'bert.encoder.layer.12.intermediate.dense.bias', 'bert.encoder.layer.7.attention.output.dense.weight', 'bert.encoder.layer.4.attention.output.LayerNorm.bias', 'bert.encoder.layer.21.attention.self.query.weight', 'bert.encoder.layer.2.attention.output.LayerNorm.weight', 'bert.encoder.layer.17.output.dense.bias', 'bert.encoder.layer.8.attention.self.query.weight', 'bert.encoder.layer.11.intermediate.dense.weight', 'bert.encoder.layer.14.attention.self.query.bias', 'bert.encoder.layer.6.attention.self.key.weight', 'bert.encoder.layer.17.output.LayerNorm.weight', 'bert.encoder.layer.18.attention.self.query.bias', 'bert.encoder.layer.6.attention.output.LayerNorm.bias', 'bert.encoder.layer.4.attention.output.dense.weight', 'bert.encoder.layer.12.attention.output.dense.bias', 'bert.embeddings.LayerNorm.bias', 'bert.encoder.layer.16.output.LayerNorm.bias', 'bert.encoder.layer.1.attention.output.dense.weight', 'bert.encoder.layer.8.output.LayerNorm.weight', 'bert.encoder.layer.10.attention.output.dense.bias', 'bert.encoder.layer.8.attention.output.dense.weight', 'bert.encoder.layer.6.attention.self.value.bias', 'cls.predictions.transform.dense.weight', 'bert.encoder.layer.8.attention.self.query.bias', 'bert.encoder.layer.17.attention.output.dense.bias', 'bert.encoder.layer.14.intermediate.dense.bias', 'bert.encoder.layer.22.output.dense.bias', 'bert.encoder.layer.1.output.dense.bias', 'bert.encoder.layer.1.output.LayerNorm.weight', 'bert.encoder.layer.11.attention.self.value.weight', 'bert.encoder.layer.6.output.dense.weight', 'bert.encoder.layer.22.attention.output.LayerNorm.bias', 'bert.encoder.layer.6.attention.output.dense.bias', 'bert.encoder.layer.0.attention.self.value.weight', 'bert.encoder.layer.2.output.dense.bias', 'bert.encoder.layer.5.attention.self.value.bias', 'bert.encoder.layer.8.attention.output.dense.bias', 'bert.encoder.layer.20.intermediate.dense.bias', 'bert.encoder.layer.0.output.LayerNorm.bias', 'bert.encoder.layer.9.attention.self.query.weight', 'bert.encoder.layer.1.attention.output.LayerNorm.bias', 'bert.encoder.layer.1.attention.self.key.weight', 'bert.encoder.layer.9.attention.output.LayerNorm.bias', 'bert.encoder.layer.21.attention.output.LayerNorm.weight', 'bert.encoder.layer.17.attention.output.LayerNorm.weight', 'bert.encoder.layer.21.intermediate.dense.bias', 'bert.encoder.layer.4.intermediate.dense.bias', 'bert.encoder.layer.19.output.dense.weight', 'bert.encoder.layer.11.output.LayerNorm.weight', 'bert.encoder.layer.8.attention.self.key.weight', 'bert.encoder.layer.12.attention.output.LayerNorm.weight', 'bert.encoder.layer.7.intermediate.dense.weight', 'bert.encoder.layer.21.attention.self.key.weight', 'bert.encoder.layer.0.output.dense.bias', 'bert.encoder.layer.18.attention.output.LayerNorm.bias', 'bert.encoder.layer.9.attention.self.value.bias', 'bert.encoder.layer.10.attention.output.dense.weight', 'bert.encoder.layer.16.output.LayerNorm.weight', 'bert.encoder.layer.0.attention.output.LayerNorm.weight', 'bert.encoder.layer.19.attention.self.query.weight', 'bert.encoder.layer.14.attention.self.query.weight', 'bert.encoder.layer.7.attention.output.LayerNorm.weight', 'bert.encoder.layer.10.intermediate.dense.weight', 'bert.encoder.layer.20.attention.output.dense.weight', 'bert.encoder.layer.10.attention.self.key.weight', 'bert.encoder.layer.22.attention.output.LayerNorm.weight', 'bert.encoder.layer.17.attention.output.dense.weight', 'bert.encoder.layer.19.attention.self.value.weight', 'bert.encoder.layer.3.attention.self.query.weight', 'bert.encoder.layer.9.attention.output.dense.bias', 'bert.encoder.layer.13.output.LayerNorm.weight', 'bert.encoder.layer.4.attention.self.value.weight', 'bert.encoder.layer.10.attention.self.query.bias', 'bert.encoder.layer.6.output.LayerNorm.weight', 'bert.embeddings.position_embeddings.weight', 'bert.encoder.layer.20.attention.self.query.weight', 'bert.encoder.layer.17.output.LayerNorm.bias', 'bert.encoder.layer.14.attention.self.value.weight', 'bert.encoder.layer.11.attention.self.key.weight', 'bert.encoder.layer.18.attention.self.query.weight', 'bert.encoder.layer.13.attention.self.key.weight']\n",
            "- This IS expected if you are initializing RobertaForTokenClassification 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 RobertaForTokenClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
            "Some weights of RobertaForTokenClassification were not initialized from the model checkpoint at hfl/chinese-roberta-wwm-ext-large and are newly initialized: ['encoder.layer.18.intermediate.dense.bias', 'encoder.layer.1.attention.output.LayerNorm.weight', 'encoder.layer.21.attention.self.query.weight', 'encoder.layer.23.attention.output.LayerNorm.weight', 'encoder.layer.9.attention.output.LayerNorm.weight', 'encoder.layer.14.attention.self.query.weight', 'encoder.layer.6.attention.self.value.bias', 'encoder.layer.1.output.LayerNorm.bias', 'encoder.layer.5.output.dense.weight', 'encoder.layer.3.attention.output.dense.bias', 'encoder.layer.4.attention.output.LayerNorm.bias', 'encoder.layer.3.attention.output.LayerNorm.weight', 'encoder.layer.20.intermediate.dense.weight', 'encoder.layer.16.output.dense.weight', 'encoder.layer.4.attention.self.query.weight', 'encoder.layer.22.attention.self.query.bias', 'encoder.layer.3.output.dense.weight', 'encoder.layer.1.intermediate.dense.weight', 'encoder.layer.3.attention.self.key.bias', 'encoder.layer.23.attention.self.key.bias', 'encoder.layer.8.attention.self.key.weight', 'encoder.layer.7.attention.self.query.weight', 'encoder.layer.23.output.LayerNorm.bias', 'encoder.layer.6.output.LayerNorm.weight', 'encoder.layer.17.intermediate.dense.weight', 'encoder.layer.9.attention.output.dense.weight', 'encoder.layer.14.attention.output.LayerNorm.bias', 'encoder.layer.13.attention.self.key.weight', 'encoder.layer.23.attention.self.query.weight', 'encoder.layer.11.attention.output.LayerNorm.weight', 'encoder.layer.22.attention.self.value.weight', 'encoder.layer.16.attention.self.query.bias', 'encoder.layer.8.output.dense.bias', 'encoder.layer.0.attention.output.LayerNorm.weight', 'encoder.layer.7.attention.output.LayerNorm.weight', 'encoder.layer.18.attention.output.LayerNorm.weight', 'encoder.layer.9.attention.self.value.bias', 'encoder.layer.0.attention.output.dense.weight', 'encoder.layer.20.attention.output.LayerNorm.weight', 'encoder.layer.23.intermediate.dense.bias', 'encoder.layer.10.output.LayerNorm.bias', 'encoder.layer.1.attention.output.dense.bias', 'encoder.layer.1.output.dense.weight', 'encoder.layer.20.attention.output.dense.weight', 'encoder.layer.17.attention.self.key.bias', 'encoder.layer.23.attention.self.key.weight', 'encoder.layer.14.output.LayerNorm.weight', 'encoder.layer.22.attention.output.LayerNorm.bias', 'embeddings.LayerNorm.weight', 'encoder.layer.19.intermediate.dense.weight', 'encoder.layer.2.attention.self.query.bias', 'encoder.layer.9.intermediate.dense.weight', 'encoder.layer.18.output.dense.bias', 'encoder.layer.20.attention.self.key.bias', 'encoder.layer.15.output.dense.weight', 'encoder.layer.23.intermediate.dense.weight', 'encoder.layer.13.attention.self.value.bias', 'encoder.layer.17.attention.self.value.bias', 'encoder.layer.3.attention.self.query.bias', 'encoder.layer.18.output.dense.weight', 'encoder.layer.21.attention.self.query.bias', 'encoder.layer.10.output.dense.bias', 'encoder.layer.16.attention.self.key.bias', 'encoder.layer.12.output.dense.weight', 'encoder.layer.7.output.LayerNorm.bias', 'encoder.layer.6.attention.output.dense.bias', 'encoder.layer.0.intermediate.dense.bias', 'encoder.layer.13.output.dense.bias', 'encoder.layer.19.output.LayerNorm.weight', 'encoder.layer.22.output.LayerNorm.bias', 'encoder.layer.9.output.LayerNorm.bias', 'encoder.layer.20.attention.output.dense.bias', 'encoder.layer.20.attention.self.query.weight', 'encoder.layer.11.attention.output.dense.bias', 'encoder.layer.17.attention.output.LayerNorm.weight', 'encoder.layer.2.output.dense.bias', 'encoder.layer.8.attention.self.value.weight', 'encoder.layer.18.attention.self.query.weight', 'encoder.layer.11.attention.self.query.weight', 'encoder.layer.7.attention.output.LayerNorm.bias', 'encoder.layer.2.attention.output.dense.weight', 'encoder.layer.0.attention.self.query.weight', 'encoder.layer.17.intermediate.dense.bias', 'encoder.layer.1.attention.self.value.weight', 'encoder.layer.11.attention.self.key.weight', 'encoder.layer.1.attention.self.query.weight', 'encoder.layer.5.attention.output.dense.bias', 'encoder.layer.11.output.LayerNorm.weight', 'encoder.layer.9.attention.self.query.bias', 'encoder.layer.16.output.dense.bias', 'encoder.layer.15.attention.output.dense.bias', 'encoder.layer.22.intermediate.dense.bias', 'encoder.layer.23.attention.output.dense.weight', 'encoder.layer.17.attention.output.LayerNorm.bias', 'encoder.layer.0.attention.self.key.weight', 'encoder.layer.13.output.LayerNorm.bias', 'encoder.layer.17.output.dense.bias', 'encoder.layer.0.attention.self.key.bias', 'encoder.layer.7.intermediate.dense.weight', 'encoder.layer.14.attention.output.dense.weight', 'encoder.layer.4.output.LayerNorm.weight', 'encoder.layer.5.attention.self.query.bias', 'encoder.layer.16.attention.self.key.weight', 'encoder.layer.1.attention.self.value.bias', 'encoder.layer.11.attention.output.LayerNorm.bias', 'encoder.layer.7.output.dense.bias', 'encoder.layer.4.intermediate.dense.weight', 'encoder.layer.12.intermediate.dense.weight', 'encoder.layer.22.attention.output.LayerNorm.weight', 'encoder.layer.7.attention.output.dense.bias', 'encoder.layer.15.attention.self.key.weight', 'encoder.layer.16.attention.self.value.weight', 'encoder.layer.8.attention.output.dense.weight', 'encoder.layer.14.attention.self.key.bias', 'encoder.layer.5.attention.output.dense.weight', 'encoder.layer.19.attention.self.value.bias', 'encoder.layer.12.attention.output.LayerNorm.weight', 'encoder.layer.14.output.LayerNorm.bias', 'encoder.layer.8.attention.self.value.bias', 'encoder.layer.18.attention.self.key.weight', 'encoder.layer.20.output.LayerNorm.bias', 'encoder.layer.4.attention.output.LayerNorm.weight', 'encoder.layer.16.intermediate.dense.weight', 'encoder.layer.13.attention.output.LayerNorm.weight', 'encoder.layer.0.attention.self.value.bias', 'encoder.layer.1.intermediate.dense.bias', 'encoder.layer.21.output.dense.weight', 'encoder.layer.3.attention.self.value.bias', 'encoder.layer.20.output.dense.bias', 'encoder.layer.2.output.dense.weight', 'encoder.layer.18.attention.self.value.weight', 'encoder.layer.7.attention.self.value.weight', 'encoder.layer.17.attention.self.query.weight', 'encoder.layer.21.output.dense.bias', 'encoder.layer.20.output.LayerNorm.weight', 'encoder.layer.6.attention.self.key.weight', 'encoder.layer.10.attention.output.LayerNorm.weight', 'encoder.layer.0.output.LayerNorm.bias', 'encoder.layer.8.output.LayerNorm.bias', 'encoder.layer.9.attention.self.key.bias', 'encoder.layer.1.output.LayerNorm.weight', 'encoder.layer.12.output.LayerNorm.bias', 'encoder.layer.23.output.dense.bias', 'encoder.layer.23.output.LayerNorm.weight', 'encoder.layer.12.attention.self.query.weight', 'encoder.layer.9.output.LayerNorm.weight', 'encoder.layer.14.attention.self.query.bias', 'encoder.layer.22.attention.self.key.weight', 'encoder.layer.3.attention.output.LayerNorm.bias', 'encoder.layer.15.output.dense.bias', 'encoder.layer.10.attention.self.key.bias', 'encoder.layer.2.attention.self.key.bias', 'encoder.layer.22.output.LayerNorm.weight', 'encoder.layer.15.attention.output.LayerNorm.weight', 'encoder.layer.23.attention.output.LayerNorm.bias', 'encoder.layer.11.attention.self.value.weight', 'encoder.layer.12.output.LayerNorm.weight', 'encoder.layer.15.output.LayerNorm.bias', 'encoder.layer.11.intermediate.dense.bias', 'encoder.layer.5.output.LayerNorm.weight', 'encoder.layer.5.intermediate.dense.bias', 'encoder.layer.3.attention.self.query.weight', 'encoder.layer.11.output.dense.bias', 'encoder.layer.12.attention.self.value.weight', 'encoder.layer.18.attention.self.key.bias', 'encoder.layer.17.attention.self.query.bias', 'embeddings.token_type_embeddings.weight', 'encoder.layer.16.attention.self.value.bias', 'encoder.layer.19.attention.self.key.bias', 'encoder.layer.14.intermediate.dense.weight', 'encoder.layer.16.intermediate.dense.bias', 'encoder.layer.6.attention.output.LayerNorm.weight', 'encoder.layer.7.attention.output.dense.weight', 'encoder.layer.21.attention.output.dense.bias', 'encoder.layer.2.attention.output.LayerNorm.weight', 'encoder.layer.2.attention.output.LayerNorm.bias', 'encoder.layer.1.attention.output.LayerNorm.bias', 'encoder.layer.1.attention.self.key.bias', 'encoder.layer.6.intermediate.dense.bias', 'encoder.layer.9.attention.output.LayerNorm.bias', 'encoder.layer.19.output.LayerNorm.bias', 'encoder.layer.12.output.dense.bias', 'encoder.layer.5.intermediate.dense.weight', 'encoder.layer.17.output.LayerNorm.weight', 'encoder.layer.5.attention.self.value.bias', 'encoder.layer.2.intermediate.dense.weight', 'encoder.layer.3.attention.self.key.weight', 'embeddings.position_embeddings.weight', 'encoder.layer.18.attention.output.dense.weight', 'encoder.layer.19.attention.output.dense.bias', 'encoder.layer.14.attention.output.LayerNorm.weight', 'encoder.layer.4.output.dense.weight', 'encoder.layer.2.attention.self.query.weight', 'encoder.layer.1.attention.self.key.weight', 'encoder.layer.5.output.LayerNorm.bias', 'encoder.layer.16.attention.output.dense.weight', 'encoder.layer.0.output.dense.weight', 'encoder.layer.10.attention.self.value.bias', 'encoder.layer.23.attention.self.query.bias', 'encoder.layer.0.output.dense.bias', 'encoder.layer.11.intermediate.dense.weight', 'encoder.layer.6.attention.self.query.bias', 'encoder.layer.8.intermediate.dense.bias', 'encoder.layer.15.attention.output.dense.weight', 'encoder.layer.13.attention.self.value.weight', 'encoder.layer.9.attention.output.dense.bias', 'encoder.layer.3.intermediate.dense.bias', 'encoder.layer.8.output.dense.weight', 'encoder.layer.19.attention.output.LayerNorm.weight', 'encoder.layer.18.attention.self.value.bias', 'encoder.layer.13.attention.output.dense.weight', 'encoder.layer.21.attention.output.LayerNorm.bias', 'encoder.layer.6.output.dense.bias', 'encoder.layer.15.output.LayerNorm.weight', 'encoder.layer.11.output.LayerNorm.bias', 'encoder.layer.16.attention.output.LayerNorm.bias', 'encoder.layer.2.intermediate.dense.bias', 'encoder.layer.12.attention.output.dense.weight', 'encoder.layer.11.attention.self.value.bias', 'encoder.layer.14.attention.self.key.weight', 'encoder.layer.12.attention.output.LayerNorm.bias', 'encoder.layer.9.attention.self.value.weight', 'encoder.layer.8.intermediate.dense.weight', 'encoder.layer.8.attention.output.LayerNorm.weight', 'encoder.layer.13.attention.self.key.bias', 'encoder.layer.21.attention.output.dense.weight', 'encoder.layer.4.attention.self.key.weight', 'encoder.layer.7.attention.self.key.weight', 'encoder.layer.1.attention.self.query.bias', 'encoder.layer.14.output.dense.bias', 'encoder.layer.11.attention.self.key.bias', 'encoder.layer.16.attention.output.dense.bias', 'encoder.layer.12.attention.output.dense.bias', 'encoder.layer.2.attention.self.value.weight', 'encoder.layer.16.output.LayerNorm.weight', 'encoder.layer.7.intermediate.dense.bias', 'encoder.layer.17.attention.self.key.weight', 'encoder.layer.10.attention.output.dense.weight', 'encoder.layer.6.intermediate.dense.weight', 'encoder.layer.9.attention.self.query.weight', 'encoder.layer.14.attention.output.dense.bias', 'encoder.layer.10.attention.self.query.weight', 'encoder.layer.15.attention.self.value.weight', 'encoder.layer.10.attention.self.key.weight', 'encoder.layer.15.attention.self.value.bias', 'encoder.layer.14.intermediate.dense.bias', 'encoder.layer.5.attention.self.value.weight', 'encoder.layer.0.attention.self.query.bias', 'encoder.layer.16.attention.output.LayerNorm.weight', 'encoder.layer.0.attention.self.value.weight', 'encoder.layer.17.attention.output.dense.weight', 'encoder.layer.23.attention.self.value.weight', 'encoder.layer.8.attention.self.query.weight', 'encoder.layer.2.attention.self.key.weight', 'encoder.layer.3.intermediate.dense.weight', 'encoder.layer.4.output.dense.bias', 'encoder.layer.9.output.dense.weight', 'encoder.layer.23.attention.self.value.bias', 'encoder.layer.22.output.dense.weight', 'encoder.layer.19.attention.output.LayerNorm.bias', 'encoder.layer.18.intermediate.dense.weight', 'encoder.layer.20.attention.self.key.weight', 'embeddings.word_embeddings.weight', 'encoder.layer.0.output.LayerNorm.weight', 'encoder.layer.12.intermediate.dense.bias', 'encoder.layer.5.attention.output.LayerNorm.weight', 'encoder.layer.8.attention.output.LayerNorm.bias', 'encoder.layer.19.output.dense.bias', 'encoder.layer.13.intermediate.dense.bias', 'encoder.layer.13.attention.output.dense.bias', 'encoder.layer.2.output.LayerNorm.weight', 'encoder.layer.15.attention.self.query.weight', 'encoder.layer.21.attention.self.value.weight', 'encoder.layer.23.output.dense.weight', 'encoder.layer.11.attention.self.query.bias', 'encoder.layer.18.attention.output.dense.bias', 'encoder.layer.0.attention.output.dense.bias', 'encoder.layer.19.output.dense.weight', 'encoder.layer.21.attention.self.key.weight', 'encoder.layer.7.attention.self.value.bias', 'encoder.layer.5.attention.self.key.bias', 'encoder.layer.19.attention.self.query.weight', 'embeddings.LayerNorm.bias', 'encoder.layer.5.attention.output.LayerNorm.bias', 'encoder.layer.2.output.LayerNorm.bias', 'encoder.layer.15.attention.self.key.bias', 'encoder.layer.19.attention.self.query.bias', 'encoder.layer.13.attention.output.LayerNorm.bias', 'encoder.layer.10.attention.self.query.bias', 'encoder.layer.9.intermediate.dense.bias', 'encoder.layer.10.output.dense.weight', 'encoder.layer.13.output.LayerNorm.weight', 'encoder.layer.17.attention.output.dense.bias', 'encoder.layer.21.output.LayerNorm.weight', 'encoder.layer.22.attention.self.query.weight', 'encoder.layer.2.attention.output.dense.bias', 'encoder.layer.14.output.dense.weight', 'encoder.layer.1.attention.output.dense.weight', 'encoder.layer.19.attention.self.key.weight', 'encoder.layer.19.intermediate.dense.bias', 'encoder.layer.6.output.LayerNorm.bias', 'encoder.layer.5.output.dense.bias', 'encoder.layer.6.attention.output.LayerNorm.bias', 'encoder.layer.21.intermediate.dense.weight', 'encoder.layer.6.attention.self.value.weight', 'encoder.layer.10.attention.self.value.weight', 'encoder.layer.13.intermediate.dense.weight', 'encoder.layer.15.attention.self.query.bias', 'encoder.layer.10.intermediate.dense.weight', 'encoder.layer.17.output.LayerNorm.bias', 'encoder.layer.4.intermediate.dense.bias', 'encoder.layer.20.output.dense.weight', 'encoder.layer.3.attention.self.value.weight', 'encoder.layer.3.output.LayerNorm.weight', 'encoder.layer.6.attention.output.dense.weight', 'encoder.layer.11.attention.output.dense.weight', 'encoder.layer.16.output.LayerNorm.bias', 'encoder.layer.15.attention.output.LayerNorm.bias', 'encoder.layer.8.attention.self.query.bias', 'encoder.layer.15.intermediate.dense.weight', 'encoder.layer.11.output.dense.weight', 'encoder.layer.12.attention.self.key.bias', 'encoder.layer.20.intermediate.dense.bias', 'encoder.layer.1.output.dense.bias', 'encoder.layer.13.attention.self.query.bias', 'encoder.layer.13.attention.self.query.weight', 'encoder.layer.21.output.LayerNorm.bias', 'encoder.layer.5.attention.self.query.weight', 'classifier.weight', 'encoder.layer.21.attention.output.LayerNorm.weight', 'encoder.layer.23.attention.output.dense.bias', 'encoder.layer.10.attention.output.LayerNorm.bias', 'encoder.layer.18.output.LayerNorm.weight', 'encoder.layer.12.attention.self.value.bias', 'encoder.layer.4.attention.output.dense.weight', 'encoder.layer.22.attention.self.key.bias', 'encoder.layer.0.attention.output.LayerNorm.bias', 'encoder.layer.14.attention.self.value.bias', 'encoder.layer.4.attention.self.query.bias', 'encoder.layer.8.output.LayerNorm.weight', 'encoder.layer.18.attention.self.query.bias', 'encoder.layer.5.attention.self.key.weight', 'encoder.layer.8.attention.output.dense.bias', 'encoder.layer.4.attention.self.value.weight', 'encoder.layer.9.attention.self.key.weight', 'encoder.layer.21.intermediate.dense.bias', 'encoder.layer.19.attention.self.value.weight', 'encoder.layer.22.attention.output.dense.bias', 'encoder.layer.16.attention.self.query.weight', 'encoder.layer.20.attention.self.value.weight', 'encoder.layer.4.output.LayerNorm.bias', 'encoder.layer.3.output.dense.bias', 'encoder.layer.7.output.dense.weight', 'encoder.layer.20.attention.output.LayerNorm.bias', 'encoder.layer.2.attention.self.value.bias', 'encoder.layer.10.intermediate.dense.bias', 'encoder.layer.4.attention.self.key.bias', 'encoder.layer.10.attention.output.dense.bias', 'encoder.layer.12.attention.self.key.weight', 'classifier.bias', 'encoder.layer.6.attention.self.query.weight', 'encoder.layer.9.output.dense.bias', 'encoder.layer.4.attention.self.value.bias', 'encoder.layer.22.attention.output.dense.weight', 'encoder.layer.0.intermediate.dense.weight', 'encoder.layer.18.attention.output.LayerNorm.bias', 'encoder.layer.20.attention.self.value.bias', 'encoder.layer.21.attention.self.value.bias', 'encoder.layer.3.attention.output.dense.weight', 'encoder.layer.4.attention.output.dense.bias', 'encoder.layer.10.output.LayerNorm.weight', 'encoder.layer.14.attention.self.value.weight', 'encoder.layer.12.attention.self.query.bias', 'encoder.layer.17.attention.self.value.weight', 'encoder.layer.21.attention.self.key.bias', 'encoder.layer.20.attention.self.query.bias', 'encoder.layer.22.intermediate.dense.weight', 'encoder.layer.19.attention.output.dense.weight', 'encoder.layer.7.attention.self.key.bias', 'encoder.layer.17.output.dense.weight', 'encoder.layer.7.attention.self.query.bias', 'encoder.layer.3.output.LayerNorm.bias', 'encoder.layer.22.output.dense.bias', 'encoder.layer.6.attention.self.key.bias', 'encoder.layer.13.output.dense.weight', 'encoder.layer.8.attention.self.key.bias', 'encoder.layer.18.output.LayerNorm.bias', 'encoder.layer.6.output.dense.weight', 'encoder.layer.22.attention.self.value.bias', 'encoder.layer.7.output.LayerNorm.weight', 'encoder.layer.15.intermediate.dense.bias']\n",
            "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
          ]
        }
      ],
      "source": [
        "tokenizer = BertTokenizer.from_pretrained(\"hfl/chinese-roberta-wwm-ext-large\")\n",
        "model = RobertaForTokenClassification.from_pretrained(\"hfl/chinese-roberta-wwm-ext-large\", num_labels=data_processor.num_labels)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "X1mjhObSCw5D"
      },
      "outputs": [],
      "source": [
        "train_dataset = EEDataset(train_samples, data_processor, tokenizer=tokenizer, mode='train', model_type = 'roberta', max_length=128)\n",
        "eval_dataset = EEDataset(eval_samples, data_processor, tokenizer=tokenizer, mode='eval', model_type = 'roberta', max_length=128)\n",
        "test_dataset = EEDataset(test_samples, data_processor, tokenizer=tokenizer, mode='test', model_type = 'roberta', max_length=128)\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9RhR3Emdpyh-"
      },
      "source": [
        "Гиперпараметры"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ZDIHhIu0IX5V"
      },
      "outputs": [],
      "source": [
        "logger = init_logger('test.txt')\n",
        "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
        "\n",
        "class Args:\n",
        "  def __init__(self):\n",
        "    self.model_type = \"roberta\"\n",
        "    self.model_dir = \"hfl/chinese-roberta-wwm-ext-large\"\n",
        "    self.model_name = \"chinese-roberta-wwm-ext-large\"\n",
        "    self.output_dir = \"/content/output\"\n",
        "    self.result_output_dir = \"/content/output_result\"\n",
        "    self.task_name =  \"ee\"\n",
        "    self.device = device\n",
        "    self.train_batch_size = 24\n",
        "    self.eval_batch_size = 24\n",
        "    self.epochs = 3\n",
        "    self.warmup_proportion = 0.1\n",
        "    self.weight_decay = 0.01\n",
        "    self.learning_rate = 2e-5\n",
        "    self.adam_epsilon = 1e-8\n",
        "    self.max_grad_norm = 1.0\n",
        "    self.max_length = 128\n",
        "    self.task_name = \"ee\"\n",
        "    self.logging_steps = 50\n",
        "    self.earlystop_patience = 10\n",
        "    self.return_dict = False"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "BddT2gpxfs2H"
      },
      "outputs": [],
      "source": [
        "args = Args()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "kQYenDAd3Q5l"
      },
      "outputs": [],
      "source": [
        "!pip install -Uqq ipdb\n",
        "import ipdb"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "fkPJbEbr3T3b"
      },
      "outputs": [],
      "source": [
        "%pdb on"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Тренировка модели"
      ],
      "metadata": {
        "id": "DsjpJeQec6_A"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cTpWuI7jEPZJ",
        "outputId": "37d11ece-ff30-4acd-d2d2-2169602d25b7"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/transformers/optimization.py:407: FutureWarning: This implementation of AdamW is deprecated and will be removed in a future version. Use the PyTorch implementation torch.optim.AdamW instead, or set `no_deprecation_warning=True` to disable this warning\n",
            "  warnings.warn(\n",
            "06/07/2023 18:25:48 - INFO - root -   ***** Running training *****\n",
            "06/07/2023 18:25:48 - INFO - root -   Num samples 23025\n",
            "06/07/2023 18:25:48 - INFO - root -   Num epochs 3\n",
            "06/07/2023 18:25:48 - INFO - root -   Num training steps 2880\n",
            "06/07/2023 18:25:48 - INFO - root -   Num warmup steps 288\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 49/960 [>.............................] - ETA: 24:14  loss: 1.4310 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:27:08 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 18:27:08 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 50/960 [>.............................] - ETA: 24:12  loss: 1.5365 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:28:50 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6085872639531604 - recall: 0.6085872639531604 - f1 score: 0.6085872639531604\n",
            "06/07/2023 18:28:55 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-50\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 99/960 [==>...........................] - ETA: 38:20  loss: 1.0016 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:30:14 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 18:30:14 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 100/960 [==>...........................] - ETA: 38:08  loss: 1.2494 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:31:57 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6085872639531604 - recall: 0.6085872639531604 - f1 score: 0.6085872639531604\n",
            "06/07/2023 18:31:57 - INFO - root -   Earlystopper counter: 1 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 149/960 [===>..........................] - ETA: 40:30  loss: 1.1922 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:33:16 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 18:33:16 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 150/960 [===>..........................] - ETA: 40:19  loss: 1.1525 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:34:59 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6095554657660611 - recall: 0.6095554657660611 - f1 score: 0.6095554657660611\n",
            "06/07/2023 18:35:09 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-150\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 199/960 [=====>........................] - ETA: 40:42  loss: 1.1691 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:36:28 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 18:36:28 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 200/960 [=====>........................] - ETA: 40:32  loss: 1.0206 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:38:11 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6126011084767212 - recall: 0.6126011084767212 - f1 score: 0.6126011084767212\n",
            "06/07/2023 18:38:17 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-200\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 249/960 [======>.......................] - ETA: 39:21  loss: 1.0919 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:39:37 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 18:39:37 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 250/960 [======>.......................] - ETA: 39:12  loss: 0.9523 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:41:19 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.620342911161766 - recall: 0.620342911161766 - f1 score: 0.620342911161766\n",
            "06/07/2023 18:41:33 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-250\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 299/960 [========>.....................] - ETA: 37:41  loss: 1.2981 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:42:53 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 18:42:53 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 300/960 [========>.....................] - ETA: 37:33  loss: 1.0530 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:44:35 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6086558766800588 - recall: 0.6086558766800588 - f1 score: 0.6086558766800588\n",
            "06/07/2023 18:44:35 - INFO - root -   Earlystopper counter: 1 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 349/960 [=========>....................] - ETA: 35:09  loss: 1.0920 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:45:55 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 18:45:55 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 350/960 [=========>....................] - ETA: 35:03  loss: 0.9287 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:47:37 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6115871648458882 - recall: 0.6115871648458882 - f1 score: 0.6115871648458882\n",
            "06/07/2023 18:47:37 - INFO - root -   Earlystopper counter: 2 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 399/960 [===========>..................] - ETA: 32:30  loss: 0.6528 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:48:57 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 18:48:57 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 400/960 [===========>..................] - ETA: 32:23  loss: 1.2090 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:50:39 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6140953411958436 - recall: 0.6140953411958436 - f1 score: 0.6140953411958436\n",
            "06/07/2023 18:50:39 - INFO - root -   Earlystopper counter: 3 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 449/960 [=============>................] - ETA: 29:45  loss: 0.9786 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:51:59 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 18:51:59 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 450/960 [=============>................] - ETA: 29:39  loss: 0.8776 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:53:41 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.622523271149873 - recall: 0.622523271149873 - f1 score: 0.622523271149873\n",
            "06/07/2023 18:53:59 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-450\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 499/960 [==============>...............] - ETA: 27:14  loss: 0.7592 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:55:19 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 18:55:19 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 500/960 [==============>...............] - ETA: 27:09  loss: 0.5927 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:57:02 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.631309512011039 - recall: 0.631309512011039 - f1 score: 0.631309512011039\n",
            "06/07/2023 18:57:14 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-500\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 549/960 [================>.............] - ETA: 24:30  loss: 0.8555 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 18:58:34 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 18:58:34 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 550/960 [================>.............] - ETA: 24:25  loss: 0.8316 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:00:17 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6221420893337705 - recall: 0.6221420893337705 - f1 score: 0.6221420893337705\n",
            "06/07/2023 19:00:17 - INFO - root -   Earlystopper counter: 1 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 599/960 [=================>............] - ETA: 21:33  loss: 0.9156 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:01:36 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:01:36 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 600/960 [=================>............] - ETA: 21:28  loss: 1.1562 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:03:18 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.5564492151466407 - recall: 0.5564492151466407 - f1 score: 0.5564492151466407\n",
            "06/07/2023 19:03:18 - INFO - root -   Earlystopper counter: 2 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 649/960 [===================>..........] - ETA: 18:35  loss: 1.2220 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:04:38 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:04:38 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 650/960 [===================>..........] - ETA: 18:31  loss: 0.7362 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:06:20 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6047563867013288 - recall: 0.6047563867013288 - f1 score: 0.6047563867013288\n",
            "06/07/2023 19:06:20 - INFO - root -   Earlystopper counter: 3 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 699/960 [====================>.........] - ETA: 15:37  loss: 0.7726 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:07:40 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:07:40 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 700/960 [====================>.........] - ETA: 15:32  loss: 0.8280 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:09:22 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6353042974437948 - recall: 0.6353042974437948 - f1 score: 0.6353042974437948\n",
            "06/07/2023 19:09:32 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-700\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 749/960 [======================>.......] - ETA: 12:41  loss: 0.7630 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:10:52 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:10:52 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 750/960 [======================>.......] - ETA: 12:36  loss: 0.9374 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:12:34 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6360895319849662 - recall: 0.6360895319849662 - f1 score: 0.6360895319849662\n",
            "06/07/2023 19:12:47 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-750\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 799/960 [=======================>......] - ETA: 9:43  loss: 0.5026 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:14:06 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:14:06 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 800/960 [========================>.....] - ETA: 9:39  loss: 0.7894 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:15:49 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6400500110542726 - recall: 0.6400500110542726 - f1 score: 0.6400500110542726\n",
            "06/07/2023 19:15:59 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-800\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 849/960 [=========================>....] - ETA: 6:43  loss: 0.7444 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:17:18 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:17:18 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 850/960 [=========================>....] - ETA: 6:39  loss: 0.8288 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:19:01 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6361543328937036 - recall: 0.6361543328937036 - f1 score: 0.6361543328937036\n",
            "06/07/2023 19:19:01 - INFO - root -   Earlystopper counter: 1 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 899/960 [===========================>..] - ETA: 3:41  loss: 0.6139 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:20:21 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:20:21 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 900/960 [===========================>..] - ETA: 3:38  loss: 0.6066 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:22:03 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6500064800908737 - recall: 0.6500064800908737 - f1 score: 0.6500064800908737\n",
            "06/07/2023 19:22:13 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-900\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 949/960 [============================>.] - ETA: 40s  loss: 0.8863 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:23:32 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:23:32 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 950/960 [============================>.] - ETA: 36s  loss: 0.5313 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:25:15 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.643126148310221 - recall: 0.643126148310221 - f1 score: 0.643126148310221\n",
            "06/07/2023 19:25:15 - INFO - root -   Earlystopper counter: 1 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 39/960 [>.............................] - ETA: 24:17  loss: 0.3808 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:26:34 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:26:34 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 40/960 [>.............................] - ETA: 24:17  loss: 0.3838 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:28:17 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6357045383507025 - recall: 0.6357045383507025 - f1 score: 0.6357045383507025\n",
            "06/07/2023 19:28:17 - INFO - root -   Earlystopper counter: 2 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 89/960 [=>............................] - ETA: 39:50  loss: 1.0954 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:29:36 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:29:36 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 90/960 [=>............................] - ETA: 39:36  loss: 0.8475 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:31:19 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6466940101089418 - recall: 0.6466940101089418 - f1 score: 0.6466940101089418\n",
            "06/07/2023 19:31:19 - INFO - root -   Earlystopper counter: 3 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 139/960 [===>..........................] - ETA: 41:57  loss: 0.5538 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:32:38 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:32:38 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 140/960 [===>..........................] - ETA: 41:45  loss: 1.0003 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:34:21 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6460498128397283 - recall: 0.6460498128397283 - f1 score: 0.6460498128397283\n",
            "06/07/2023 19:34:21 - INFO - root -   Earlystopper counter: 4 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 189/960 [====>.........................] - ETA: 41:19  loss: 0.7023 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:35:40 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:35:40 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 190/960 [====>.........................] - ETA: 41:10  loss: 0.6816 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:37:22 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6486952146434807 - recall: 0.6486952146434807 - f1 score: 0.6486952146434807\n",
            "06/07/2023 19:37:22 - INFO - root -   Earlystopper counter: 5 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 239/960 [======>.......................] - ETA: 39:42  loss: 0.4473 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:38:42 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:38:42 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 240/960 [======>.......................] - ETA: 39:34  loss: 0.6142 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:40:24 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6558881155133376 - recall: 0.6558881155133376 - f1 score: 0.6558881155133376\n",
            "06/07/2023 19:40:29 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-1200\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 289/960 [========>.....................] - ETA: 37:47  loss: 0.9875 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:41:48 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:41:48 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 290/960 [========>.....................] - ETA: 37:39  loss: 0.5125 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:43:31 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6620289545707512 - recall: 0.6620289545707512 - f1 score: 0.6620289545707512\n",
            "06/07/2023 19:43:43 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-1250\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 339/960 [=========>....................] - ETA: 35:44  loss: 0.6781 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:45:02 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:45:02 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 340/960 [=========>....................] - ETA: 35:37  loss: 0.7753 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:46:45 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6615867836640721 - recall: 0.6615867836640721 - f1 score: 0.6615867836640721\n",
            "06/07/2023 19:46:45 - INFO - root -   Earlystopper counter: 1 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 389/960 [===========>..................] - ETA: 33:05  loss: 0.6559 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:48:04 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:48:04 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 390/960 [===========>..................] - ETA: 32:59  loss: 0.5353 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:49:47 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6650631618269283 - recall: 0.6650631618269283 - f1 score: 0.6650631618269283\n",
            "06/07/2023 19:50:01 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-1350\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 439/960 [============>.................] - ETA: 30:37  loss: 0.7880 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:51:20 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:51:20 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 440/960 [============>.................] - ETA: 30:31  loss: 0.5657 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:53:03 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6635041281990683 - recall: 0.6635041281990683 - f1 score: 0.6635041281990683\n",
            "06/07/2023 19:53:03 - INFO - root -   Earlystopper counter: 1 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 489/960 [==============>...............] - ETA: 27:46  loss: 0.3980 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:54:22 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:54:22 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 490/960 [==============>...............] - ETA: 27:41  loss: 0.5402 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:56:05 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6617773745721234 - recall: 0.6617773745721234 - f1 score: 0.6617773745721234\n",
            "06/07/2023 19:56:05 - INFO - root -   Earlystopper counter: 2 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 539/960 [===============>..............] - ETA: 24:53  loss: 0.7476 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:57:25 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 19:57:25 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 540/960 [===============>..............] - ETA: 24:48  loss: 0.7931 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 19:59:07 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6641940672862142 - recall: 0.6641940672862142 - f1 score: 0.6641940672862142\n",
            "06/07/2023 19:59:07 - INFO - root -   Earlystopper counter: 3 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 589/960 [=================>............] - ETA: 21:59  loss: 0.5609 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:00:27 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:00:27 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 590/960 [=================>............] - ETA: 21:54  loss: 0.4751 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:02:10 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.663446950926653 - recall: 0.663446950926653 - f1 score: 0.663446950926653\n",
            "06/07/2023 20:02:10 - INFO - root -   Earlystopper counter: 4 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 639/960 [==================>...........] - ETA: 19:03  loss: 0.7065 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:03:29 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:03:29 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 640/960 [===================>..........] - ETA: 18:59  loss: 0.7323 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:05:11 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6593644936761937 - recall: 0.6593644936761937 - f1 score: 0.6593644936761937\n",
            "06/07/2023 20:05:11 - INFO - root -   Earlystopper counter: 5 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 689/960 [====================>.........] - ETA: 16:07  loss: 0.6694 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:06:31 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:06:31 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 690/960 [====================>.........] - ETA: 16:02  loss: 0.3435 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:08:13 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6582971845911063 - recall: 0.6582971845911063 - f1 score: 0.6582971845911063\n",
            "06/07/2023 20:08:13 - INFO - root -   Earlystopper counter: 6 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 739/960 [======================>.......] - ETA: 13:09  loss: 0.6735 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:09:33 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:09:33 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 740/960 [======================>.......] - ETA: 13:05  loss: 0.7216 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:11:15 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6658483963680997 - recall: 0.6658483963680997 - f1 score: 0.6658483963680997\n",
            "06/07/2023 20:11:25 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-1700\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 789/960 [=======================>......] - ETA: 10:13  loss: 0.5296 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:12:45 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:12:45 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 790/960 [=======================>......] - ETA: 10:09  loss: 0.5483 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:14:27 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6717338436087245 - recall: 0.6717338436087245 - f1 score: 0.6717338436087245\n",
            "06/07/2023 20:14:55 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-1750\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 839/960 [=========================>....] - ETA: 7:18  loss: 0.3557 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:16:15 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:16:15 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 840/960 [=========================>....] - ETA: 7:14  loss: 0.4738 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:17:58 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6603555663980606 - recall: 0.6603555663980606 - f1 score: 0.6603555663980606\n",
            "06/07/2023 20:17:58 - INFO - root -   Earlystopper counter: 1 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 889/960 [==========================>...] - ETA: 4:17  loss: 0.5316 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:19:17 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:19:17 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 890/960 [==========================>...] - ETA: 4:13  loss: 0.7357 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:21:00 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6681088045375884 - recall: 0.6681088045375884 - f1 score: 0.6681088045375884\n",
            "06/07/2023 20:21:00 - INFO - root -   Earlystopper counter: 2 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 939/960 [============================>.] - ETA: 1:16  loss: 0.5409 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:22:19 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:22:19 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 940/960 [============================>.] - ETA: 1:12  loss: 0.5805 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:24:02 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6682193472642581 - recall: 0.6682193472642581 - f1 score: 0.6682193472642581\n",
            "06/07/2023 20:24:02 - INFO - root -   Earlystopper counter: 3 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 29/960 [..............................] - ETA: 24:32  loss: 0.7958 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:25:21 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:25:21 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 30/960 [..............................] - ETA: 24:30  loss: 0.6217 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:27:04 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6634507627448141 - recall: 0.6634507627448141 - f1 score: 0.6634507627448141\n",
            "06/07/2023 20:27:04 - INFO - root -   Earlystopper counter: 4 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 79/960 [=>............................] - ETA: 42:23  loss: 0.6936 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:28:23 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:28:23 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 80/960 [=>............................] - ETA: 42:06  loss: 0.5379 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:30:06 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6668242218173224 - recall: 0.6668242218173224 - f1 score: 0.6668242218173224\n",
            "06/07/2023 20:30:06 - INFO - root -   Earlystopper counter: 5 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 129/960 [===>..........................] - ETA: 44:04  loss: 0.7101 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:31:25 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:31:25 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 130/960 [===>..........................] - ETA: 43:50  loss: 0.4259 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:33:08 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.67138315633791 - recall: 0.67138315633791 - f1 score: 0.67138315633791\n",
            "06/07/2023 20:33:08 - INFO - root -   Earlystopper counter: 6 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 179/960 [====>.........................] - ETA: 43:06  loss: 0.4469 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:34:28 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:34:28 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 180/960 [====>.........................] - ETA: 42:55  loss: 0.5020 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:36:11 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6607710545776124 - recall: 0.6607710545776124 - f1 score: 0.6607710545776124\n",
            "06/07/2023 20:36:11 - INFO - root -   Earlystopper counter: 7 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 229/960 [======>.......................] - ETA: 41:14  loss: 0.6268 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:37:30 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:37:30 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 230/960 [======>.......................] - ETA: 41:05  loss: 0.6312 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:39:13 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6780576499378673 - recall: 0.6780576499378673 - f1 score: 0.6780576499378673\n",
            "06/07/2023 20:39:22 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-2150\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 279/960 [=======>......................] - ETA: 39:18  loss: 0.4149 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:40:41 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:40:41 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 280/960 [=======>......................] - ETA: 39:10  loss: 0.6097 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:42:24 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6766282181274824 - recall: 0.6766282181274824 - f1 score: 0.6766282181274824\n",
            "06/07/2023 20:42:24 - INFO - root -   Earlystopper counter: 1 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 329/960 [=========>....................] - ETA: 36:42  loss: 0.3675 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:43:43 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:43:43 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 330/960 [=========>....................] - ETA: 36:35  loss: 0.6151 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:45:26 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.676262283584024 - recall: 0.676262283584024 - f1 score: 0.676262283584024\n",
            "06/07/2023 20:45:26 - INFO - root -   Earlystopper counter: 2 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 379/960 [==========>...................] - ETA: 34:00  loss: 0.6190 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:46:46 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:46:46 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 380/960 [==========>...................] - ETA: 33:53  loss: 0.7486 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:48:28 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6737236126887803 - recall: 0.6737236126887803 - f1 score: 0.6737236126887803\n",
            "06/07/2023 20:48:28 - INFO - root -   Earlystopper counter: 3 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 429/960 [============>.................] - ETA: 31:12  loss: 0.4323 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:49:48 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:49:48 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 430/960 [============>.................] - ETA: 31:06  loss: 0.8167 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:51:31 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6645333191025455 - recall: 0.6645333191025455 - f1 score: 0.6645333191025455\n",
            "06/07/2023 20:51:31 - INFO - root -   Earlystopper counter: 4 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 479/960 [=============>................] - ETA: 28:22  loss: 0.7187 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:52:50 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:52:50 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 480/960 [==============>...............] - ETA: 28:16  loss: 0.4830 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:54:33 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6786141753893772 - recall: 0.6786141753893772 - f1 score: 0.6786141753893772\n",
            "06/07/2023 20:54:43 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-2400\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 529/960 [===============>..............] - ETA: 25:37  loss: 0.7310 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:56:02 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:56:02 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 530/960 [===============>..............] - ETA: 25:32  loss: 0.8997 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:57:45 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6737922254156787 - recall: 0.6737922254156787 - f1 score: 0.6737922254156787\n",
            "06/07/2023 20:57:45 - INFO - root -   Earlystopper counter: 1 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 579/960 [=================>............] - ETA: 22:41  loss: 0.6143 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 20:59:05 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 20:59:05 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 580/960 [=================>............] - ETA: 22:37  loss: 0.5218 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:00:47 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6782406172095966 - recall: 0.6782406172095966 - f1 score: 0.6782406172095966\n",
            "06/07/2023 21:00:47 - INFO - root -   Earlystopper counter: 2 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 629/960 [==================>...........] - ETA: 19:45  loss: 0.5134 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:02:07 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 21:02:07 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 630/960 [==================>...........] - ETA: 19:40  loss: 0.3634 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:03:50 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6797234144742359 - recall: 0.6797234144742359 - f1 score: 0.6797234144742359\n",
            "06/07/2023 21:04:00 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-2550\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 679/960 [====================>.........] - ETA: 16:51  loss: 0.5949 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:05:19 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 21:05:19 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 680/960 [====================>.........] - ETA: 16:47  loss: 0.5378 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:07:02 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6771618726700261 - recall: 0.6771618726700261 - f1 score: 0.6771618726700261\n",
            "06/07/2023 21:07:02 - INFO - root -   Earlystopper counter: 1 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 729/960 [=====================>........] - ETA: 13:52  loss: 0.7440 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:08:21 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 21:08:21 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 730/960 [=====================>........] - ETA: 13:48  loss: 0.4942 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:10:04 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6763728263106936 - recall: 0.6763728263106936 - f1 score: 0.6763728263106936\n",
            "06/07/2023 21:10:04 - INFO - root -   Earlystopper counter: 2 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 779/960 [=======================>......] - ETA: 10:52  loss: 0.4745 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:11:23 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 21:11:23 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 780/960 [=======================>......] - ETA: 10:48  loss: 0.5065 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:13:06 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6817017480998087 - recall: 0.6817017480998087 - f1 score: 0.6817017480998087\n",
            "06/07/2023 21:13:23 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-2700\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 829/960 [========================>.....] - ETA: 7:55  loss: 0.5512 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:14:43 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 21:14:43 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 830/960 [========================>.....] - ETA: 7:51  loss: 0.2836 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:16:26 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6798187099282615 - recall: 0.6798187099282615 - f1 score: 0.6798187099282615\n",
            "06/07/2023 21:16:26 - INFO - root -   Earlystopper counter: 1 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 879/960 [==========================>...] - ETA: 4:53  loss: 0.4913 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:17:45 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 21:17:45 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 880/960 [==========================>...] - ETA: 4:50  loss: 0.4271 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:19:28 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6826928208216755 - recall: 0.6826928208216755 - f1 score: 0.6826928208216755\n",
            "06/07/2023 21:19:38 - INFO - root -   Saving models checkpoint to /content/output/checkpoint-2800\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 929/960 [============================>.] - ETA: 1:52  loss: 0.3938 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:20:57 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 21:20:57 - INFO - root -   Num samples 5000\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 930/960 [============================>.] - ETA: 1:49  loss: 0.5600 \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:22:40 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6807830999230012 - recall: 0.6807830999230012 - f1 score: 0.6807830999230012\n",
            "06/07/2023 21:22:40 - INFO - root -   Earlystopper counter: 1 out of 10\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Training] 959/960 [============================>.] - ETA: 3s  loss: 0.4266 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:23:27 - INFO - root -   Training Stop! The best step 2800: 0.6826928208216755\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r[Training] 960/960 [==============================] 3.7s/step  loss: 0.8073 "
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:23:46 - INFO - root -   Saving models checkpoint to /content/output\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(2880, 2800)"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ],
      "source": [
        "trainer = EETrainer(args, model, data_processor, tokenizer, logger, RobertaForTokenClassification, train_dataset, eval_dataset)\n",
        "trainer.train()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "J4LR6cd49SyD",
        "outputId": "5bcb0591-d18c-4824-dd9c-645aa277a783"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:24:08 - INFO - root -   ***** Running prediction *****\n",
            "06/07/2023 21:24:08 - INFO - root -   Num samples 9868\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[Prediction] 412/412 [==============================] 489.4ms/step"
          ]
        }
      ],
      "source": [
        "trainer.predict(model, test_dataset)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "awu4Rty29nBR",
        "outputId": "32a67188-82b7-4c00-a598-885a744d732c"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "06/07/2023 21:28:10 - INFO - root -   ***** Running evaluation *****\n",
            "06/07/2023 21:28:10 - INFO - root -   Num samples 5000\n",
            "06/07/2023 21:29:54 - INFO - root -   ee-chinese-roberta-wwm-ext-large precision: 0.6824298053685647 - recall: 0.6824298053685647 - f1 score: 0.6824298053685647\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.6824298053685647"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ],
      "source": [
        "trainer.evaluate(model)"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Графики"
      ],
      "metadata": {
        "id": "W2d6KYK1dCFb"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "ayo6RRIuEeQ3"
      },
      "outputs": [],
      "source": [
        "loss_val_arr = []\n",
        "loss_by_epochs = []\n",
        "stopCount = int(len(trainer.predicts_outputs)/ args.epochs)\n",
        "\n",
        "lossEpoch = 0\n",
        "item_count = 0\n",
        "\n",
        "for item in trainer.predicts_outputs:\n",
        "  tensor = item[0]\n",
        "  item_count += 1\n",
        "\n",
        "  lossEpoch += tensor.item()\n",
        "  loss_val_arr.append(tensor.item())\n",
        "\n",
        "  if item_count == stopCount:\n",
        "    loss_by_epochs.append(lossEpoch/stopCount)\n",
        "    item_count = 0\n",
        "    lossEpoch = 0"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "eFVo0Mh11Xr4"
      },
      "outputs": [],
      "source": [
        "loss_arr = []\n",
        "loss_arr_by_epoch = []\n",
        "\n",
        "for item in trainer.loss_array:\n",
        "  tmp_loss = []\n",
        "  loss = 0\n",
        "  for tensor in item:\n",
        "      loss += tensor.item()\n",
        "      loss_arr.append(tensor.item())\n",
        "\n",
        "  loss_arr_by_epoch.append(loss/len(trainer.loss_array[0]))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "aVkHIcGCd3rV"
      },
      "outputs": [],
      "source": [
        "trainer.loss_array"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "F1mzXYMcF3iJ"
      },
      "outputs": [],
      "source": [
        "if len(loss_val_arr) < len(loss_arr):\n",
        "  for i in range(len(loss_val_arr), len(loss_arr)):\n",
        "    loss_val_arr.append(0)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 563
        },
        "id": "C8cqEm6f1GoN",
        "outputId": "4c5065cb-4cce-4678-bddb-fbb1ebd66649"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-47-38c6df0fae35>:6: UserWarning: color is redundantly defined by the 'color' keyword argument and the fmt string \"b\" (-> color=(0.0, 0.0, 1.0, 1)). The keyword argument will take precedence.\n",
            "  plt.plot(epochs, loss_arr, 'b', label='Training Loss', color='pink')\n",
            "<ipython-input-47-38c6df0fae35>:7: UserWarning: color is redundantly defined by the 'color' keyword argument and the fmt string \"b\" (-> color=(0.0, 0.0, 1.0, 1)). The keyword argument will take precedence.\n",
            "  plt.plot(epochs, loss_val_arr[0:len(loss_arr)], 'b', label='Val Loss', color='green')\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkUAAAHHCAYAAACx7iyPAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAACZOUlEQVR4nO3dd3zU9P8H8Feuu3RTOqClFFpKGWUUgQICCgioyHT/ZAj4VRniQK0LxVEVFQcKqEAdDEVlyB6yZO+9BcpoC5RuoKW9/P4Id83dJZdcLne5a9/P76NfuUvyyedyueSdz2RYlmVBCCGEEFLD6bTOACGEEEKIK6CgiBBCCCEEFBQRQgghhACgoIgQQgghBAAFRYQQQgghACgoIoQQQggBQEERIYQQQggACooIIYQQQgBQUEQIIYQQAoCCIkIIIYQQABQUEUJcTGZmJhiGwe7du7XOCiGkhqGgiBBCCCEEFBQRQgghhACgoIgQ4ob27duHPn36ICgoCAEBAejevTu2b99uss7t27fx3nvvITExEb6+vqhduzY6d+6MNWvWGNfJycnB8OHDERMTAx8fH0RHR6Nfv344d+6ckz8RIcQVeGqdAUIIscWRI0dw9913IygoCK+++iq8vLwwY8YMdOvWDRs3bkT79u0BAO+++y4yMjIwcuRItGvXDkVFRdi9ezf27t2Lnj17AgAGDRqEI0eOYOzYsWjQoAGuXLmCNWvWICsrCw0aNNDwUxJCtMCwLMtqnQlCCDHIzMzE8OHDsWvXLrRt29Zi+YABA7B8+XIcO3YMDRs2BABkZ2cjKSkJrVu3xsaNGwEArVq1QkxMDJYuXSq4n4KCAoSGhmLy5Ml45ZVXHPeBCCFug6rPCCFuo7KyEqtXr0b//v2NAREAREdH44knnsC///6LoqIiAEBISAiOHDmCU6dOCabl5+cHb29vbNiwAfn5+U7JPyHEtVFQRAhxG1evXsWNGzeQlJRksSw5ORl6vR4XLlwAAEyaNAkFBQVo3LgxWrRogQkTJuDgwYPG9X18fPDJJ59gxYoViIyMRJcuXfDpp58iJyfHaZ+HEOJaKCgihFRLXbp0wZkzZzBr1iw0b94cP/74I9q0aYMff/zRuM748eNx8uRJZGRkwNfXF2+//TaSk5Oxb98+DXNOCNEKBUWEELdRp04d+Pv748SJExbLjh8/Dp1Oh9jYWON7YWFhGD58OObNm4cLFy4gJSUF7777rsl2jRo1wssvv4zVq1fj8OHDKC8vx+eff+7oj0IIcUEUFBFC3IaHhwfuu+8+LF682KTbfG5uLubOnYvOnTsjKCgIAJCXl2eybUBAABISElBWVgYAuHHjBm7dumWyTqNGjRAYGGhchxBSs1CXfEKIS5o1axZWrlxp8f67776LNWvWoHPnznj++efh6emJGTNmoKysDJ9++qlxvaZNm6Jbt25ITU1FWFgYdu/ejT/++ANjxowBAJw8eRLdu3fHI488gqZNm8LT0xMLFy5Ebm4uHnvsMad9TkKI66Au+YQQl2Loki/mwoULuHr1KtLT07Flyxbo9Xq0b98eH374IdLS0ozrffjhh1iyZAlOnjyJsrIyxMXF4amnnsKECRPg5eWFvLw8TJw4EevWrcOFCxfg6emJJk2a4OWXX8bDDz/sjI9KCHExFBQRQgghhIDaFBFCCCGEAKCgiBBCCCEEAAVFhBBCCCEAKCgihBBCCAFAQREhhBBCCAAKigghhBBCANTAwRv1ej0uX76MwMBAMAyjdXYIIYQQIgPLsiguLkbdunWh0zmmTKfGBUWXL182mRuJEEIIIe7jwoULiImJcUjaNS4oCgwMBMAdVMMcSYQQQghxbUVFRYiNjTXexx2hxgVFhiqzoKAgCooIIYQQN+PIpi/U0JoQQgghBBQUEUIIIYQAoKCIEEIIIQRADWxTRAghRBt6vR7l5eVaZ4O4MG9vb4d1t5eDgiJCCCEOV15ejrNnz0Kv12udFeLCdDod4uPj4e3trcn+KSgihBDiUCzLIjs7Gx4eHoiNjdW0JIC4LsPgytnZ2ahfv74mAyxTUEQIIcShKioqcOPGDdStWxf+/v5aZ4e4sDp16uDy5cuoqKiAl5eX0/dP4TohhBCHqqysBADNqkSI+zCcI4ZzxtkoKCKEEOIUNN8kkaL1OUJBESGEEEIIKCgihBBCnKZBgwb48ssvZa+/YcMGMAyDgoICh+WJVKGgiBBCCDHDMIzVv3fffVdRurt27cIzzzwje/2OHTsiOzsbwcHBivYnFwVfHOp9pha9Hii/DYABfKkxISGEuLPs7Gzjv3/77Te88847OHHihPG9gIAA479ZlkVlZSU8PaVvqXXq1LEpH97e3oiKirJpG6IclRSppbgU2HEIOHhCel1CCCEuLSoqyvgXHBwMhmGMr48fP47AwECsWLECqamp8PHxwb///oszZ86gX79+iIyMREBAAO666y6sXbvWJF3z6jOGYfDjjz9iwIAB8Pf3R2JiIpYsWWJcbl6Ck5mZiZCQEKxatQrJyckICAhA7969TYK4iooKjBs3DiEhIahduzZee+01DB06FP3791d8PPLz8zFkyBCEhobC398fffr0walTp4zLz58/j759+yI0NBS1atVCs2bNsHz5cuO2Tz75JOrUqQM/Pz8kJiZi9uzZivPiSBQUqcXQYl7PapsPQghxdSwLVFZq88eqd41+/fXX8fHHH+PYsWNISUlBSUkJ7r//fqxbtw779u1D79690bdvX2RlZVlN57333sMjjzyCgwcP4v7778eTTz6J69evi65/48YNfPbZZ/jll1+wadMmZGVl4ZVXXjEu/+STTzBnzhzMnj0bW7ZsQVFRERYtWmTXZx02bBh2796NJUuWYNu2bWBZFvfffz9u374NABg9ejTKysqwadMmHDp0CJ988omxNO3tt9/G0aNHsWLFChw7dgzTpk1DeHi4XflxFKo+U4thhFYVf3CEEFIt6fXAv/u02Xfn1oCHhypJTZo0CT179jS+DgsLQ8uWLY2v33//fSxcuBBLlizBmDFjRNMZNmwYHn/8cQDARx99hK+//ho7d+5E7969Bde/ffs2pk+fjkaNGgEAxowZg0mTJhmXf/PNN0hPT8eAAQMAAFOnTjWW2ihx6tQpLFmyBFu2bEHHjh0BAHPmzEFsbCwWLVqEhx9+GFlZWRg0aBBatGgBAGjYsKFx+6ysLLRu3Rpt27YFwJWWuSoqKVKLoaSIgiJCCKkRDDd5g5KSErzyyitITk5GSEgIAgICcOzYMcmSopSUFOO/a9WqhaCgIFy5ckV0fX9/f2NABADR0dHG9QsLC5Gbm4t27doZl3t4eCA1NdWmz8Z37NgxeHp6on379sb3ateujaSkJBw7dgwAMG7cOHzwwQfo1KkTJk6ciIMHDxrXfe655zB//ny0atUKr776KrZu3ao4L45GJUVqoeozQgiRR6fjSmy02rdKatWqZfL6lVdewZo1a/DZZ58hISEBfn5+GDx4MMrLy62mYz6dBcMwVifOFVqf1fiBfOTIkejVqxeWLVuG1atXIyMjA59//jnGjh2LPn364Pz581i+fDnWrFmD7t27Y/To0fjss880zbMQKilSi85QUkQzQBNCiFUMw1VhafHnwBGTt2zZgmHDhmHAgAFo0aIFoqKicO7cOYftT0hwcDAiIyOxa9cu43uVlZXYu3ev4jSTk5NRUVGBHTt2GN/Ly8vDiRMn0LRpU+N7sbGxePbZZ/HXX3/h5Zdfxg8//GBcVqdOHQwdOhS//vorvvzyS3z//feK8+NIVFKkFiopIoSQGi0xMRF//fUX+vbtC4Zh8Pbbb1st8XGUsWPHIiMjAwkJCWjSpAm++eYb5Ofny5pC49ChQwgMDDS+ZhgGLVu2RL9+/TBq1CjMmDEDgYGBeP3111GvXj3069cPADB+/Hj06dMHjRs3Rn5+PtavX4/k5GQAwDvvvIPU1FQ0a9YMZWVlWLp0qXGZq6GgSC063snGsg59GiGEEOJ6vvjiCzz99NPo2LEjwsPD8dprr6GoqMjp+XjttdeQk5ODIUOGwMPDA8888wx69eoFDxkNzLt06WLy2sPDAxUVFZg9ezZeeOEFPPjggygvL0eXLl2wfPlyY1VeZWUlRo8ejYsXLyIoKAi9e/fGlClTAHBjLaWnp+PcuXPw8/PD3Xffjfnz56v/wVXAsFpXRDpZUVERgoODUVhYiKCgIPUSrqgEttzpTXF3G1XrrQkhxJ3dunULZ8+eRXx8PHx9fbXOTo2j1+uRnJyMRx55BO+//77W2bHK2rnisPs3D5UUqYVfUqRnqbUWIYQQTZw/fx6rV69G165dUVZWhqlTp+Ls2bN44okntM6ay6Nbt1oYs+ozQgghRAM6nQ6ZmZm466670KlTJxw6dAhr16512XY8roRKitTCD4o0aFhHCCGEAFwvsC1btmidDbdEJUVq0tEAjoQQQoi7oqBITTSqNSGEEOK2KChSE3PncNJYRYQQQojboaBITVR9RgghhLgtCorURKNaE0IIIW6LgiI10fxnhBBCiNvSNCiaNm0aUlJSEBQUhKCgIKSlpWHFihWi62dmZoJhGJM/lxodlUqKCCGE8HTr1g3jx4/XOhtEJk2DopiYGHz88cfYs2cPdu/ejXvvvRf9+vXDkSNHRLcJCgpCdna28e/8+fNOzLEE6n1GCCHVQt++fdG7d2/BZZs3bwbDMDh48KDd+8nMzERISIjd6RB1aDp4Y9++fU1ef/jhh5g2bRq2b9+OZs2aCW7DMAyioqKckT3bUUNrQgipFkaMGIFBgwbh4sWLiImJMVk2e/ZstG3bFikpKRrljjiKy7QpqqysxPz581FaWoq0tDTR9UpKShAXF4fY2FjJUiUAKCsrQ1FRkcmfw1CXfEIIqRYefPBB1KlTB5mZmSbvl5SUYMGCBRgxYgTy8vLw+OOPo169evD390eLFi0wb948VfORlZWFfv36ISAgAEFBQXjkkUeQm5trXH7gwAHcc889CAwMRFBQEFJTU7F7924A3Bxoffv2RWhoKGrVqoVmzZph+fLlquavutF8mo9Dhw4hLS0Nt27dQkBAABYuXIimTZsKrpuUlIRZs2YhJSUFhYWF+Oyzz9CxY0ccOXLEIpI3yMjIwHvvvefIj1CFqs8IIUQSy7K4cfuGJvv29/IHw5+WSYSnpyeGDBmCzMxMvPnmm8ZtFixYgMrKSjz++OMoKSlBamoqXnvtNQQFBWHZsmV46qmn0KhRI7Rr187uvOr1emNAtHHjRlRUVGD06NF49NFHsWHDBgDAk08+idatW2PatGnw8PDA/v374eXlBQAYPXo0ysvLsWnTJtSqVQtHjx5FQECA3fmqzjQPipKSkrB//34UFhbijz/+wNChQ7Fx40bBwCgtLc2kFKljx45ITk7GjBkz8P777wumn56ejpdeesn4uqioCLGxsep/EICqzwghRIYbt28gIEObm3NJeglqedeSte7TTz+NyZMnY+PGjejWrRsArups0KBBCA4ORnBwMF555RXj+mPHjsWqVavw+++/qxIUrVu3DocOHcLZs2eN962ff/4ZzZo1w65du3DXXXchKysLEyZMQJMmTQAAiYmJxu2zsrIwaNAgtGjRAgDQsGFDu/NU3Wlefebt7Y2EhASkpqYiIyMDLVu2xFdffSVrWy8vL7Ru3RqnT58WXcfHx8fYu83w5zDG3mfUJZ8QQtxdkyZN0LFjR8yaNQsAcPr0aWzevBkjRowAwDX7eP/999GiRQuEhYUhICAAq1atQlZWlir7P3bsGGJjY00e5Js2bYqQkBAcO3YMAPDSSy9h5MiR6NGjBz7++GOcOXPGuO64cePwwQcfoFOnTpg4caIqDcOrO81Liszp9XqUlZXJWreyshKHDh3C/fff7+BcyUQlRYQQIsnfyx8l6SWa7dsWI0aMwNixY/Htt99i9uzZaNSoEbp27QoAmDx5Mr766it8+eWXaNGiBWrVqoXx48ejvLzcEVkX9O677+KJJ57AsmXLsGLFCkycOBHz58/HgAEDMHLkSPTq1QvLli3D6tWrkZGRgc8//xxjx451Wv7cjaZBUXp6Ovr06YP69eujuLgYc+fOxYYNG7Bq1SoAwJAhQ1CvXj1kZGQAACZNmoQOHTogISEBBQUFmDx5Ms6fP4+RI0dq+TGq0DhFhBAiiWEY2VVYWnvkkUfwwgsvYO7cufj555/x3HPPGdsXbdmyBf369cP//d//AeAe6k+ePCnaLtZWycnJuHDhAi5cuGAsLTp69CgKCgpM9tG4cWM0btwYL774Ih5//HHMnj0bAwYMAADExsbi2WefxbPPPov09HT88MMPFBRZoWlQdOXKFQwZMgTZ2dkIDg5GSkoKVq1ahZ49ewLg6kN1uqoavvz8fIwaNQo5OTkIDQ1Famoqtm7dqtoJaDdDXqmkiBBCqoWAgAA8+uijSE9PR1FREYYNG2ZclpiYiD/++ANbt25FaGgovvjiC+Tm5tp8T6qsrMT+/ftN3vPx8UGPHj3QokULPPnkk/jyyy9RUVGB559/Hl27dkXbtm1x8+ZNTJgwAYMHD0Z8fDwuXryIXbt2YdCgQQCA8ePHo0+fPmjcuDHy8/Oxfv16JCcn23tIqjVNg6KZM2daXW5oXW8wZcoUTJkyxYE5shOVFBFCSLUzYsQIzJw5E/fffz/q1q1rfP+tt97Cf//9h169esHf3x/PPPMM+vfvj8LCQpvSLykpQevWrU3ea9SoEU6fPo3Fixdj7Nix6NKlC3Q6HXr37o1vvvkGAODh4YG8vDwMGTIEubm5CA8Px8CBA409risrKzF69GhcvHgRQUFB6N27t2vfQ10Aw7I1q1ijqKgIwcHBKCwsVL/R9cnzQPZVIK4u0KCu9PqEEFID3Lp1C2fPnkV8fLxrTc1EXI61c8Wh9+87NO99Vq0Yh76oUXEmIYQQUi1QUOQIFBMRQgghboeCIjXJGCWVEEIIIa6JgiJHqFnNtAghhJBqgYIiQgghTlHD+vUQBbQ+RygoUhNVnxFCiAUPDw8AcOpIz8Q9Gc4RwznjbC43zUe1QE9DhBBi5OnpCX9/f1y9ehVeXl4mg/ISYqDX63H16lX4+/vD01Ob8ISCIjVRSREhhFhgGAbR0dE4e/Yszp8/r3V2iAvT6XSoX7++cSoVZ6OgyBGopIgQQkx4e3sjMTGRqtCIVd7e3pqWJFJQpCYqKCKEEFE6nY5GtCYujSp2VXUnKqKCIkIIIcTtUFCkJprmgxBCCHFbFBQRQgghhICCIpVR9RkhhBDirigoUhNVnxFCCCFui4IiVVFJESGEEOKuKChSE5UUEUIIIW6LgiJHoJiIEEIIcTsUFKmJpvkghBBC3BYFRY5A03wQQgghboeCIjVRSREhhBDitigocgQqKCKEEELcDgVFaqLeZ4QQQojboqBIVTROESGEEOKuKChSE5UUEUIIIW6LgiJVUUkRIYQQ4q4oKFITdT4jhBBC3BYFRY5A4xQRQgghboeCIjXROEWEEEKI26KgyCGopIgQQghxNxQUOQLFRIQQQojboaBITVR9RgghhLgtCoocgRpaE0IIIW6HgiI1UUkRIYQQ4rYoKHIEKikihBBC3A4FRWqigiJCCCHEbWkaFE2bNg0pKSkICgpCUFAQ0tLSsGLFCqvbLFiwAE2aNIGvry9atGiB5cuXOym3ctA0H4QQQoi70jQoiomJwccff4w9e/Zg9+7duPfee9GvXz8cOXJEcP2tW7fi8ccfx4gRI7Bv3z70798f/fv3x+HDh52ccxE0ISwhhBDithiWda0GMGFhYZg8eTJGjBhhsezRRx9FaWkpli5danyvQ4cOaNWqFaZPny4r/aKiIgQHB6OwsBBBQUGq5RsAcC0fOHIGCKwFtElWN21CCCGkBnPo/fsOl2lTVFlZifnz56O0tBRpaWmC62zbtg09evQwea9Xr17Ytm2baLplZWUoKioy+XMcalRECCGEuCvNg6JDhw4hICAAPj4+ePbZZ7Fw4UI0bdpUcN2cnBxERkaavBcZGYmcnBzR9DMyMhAcHGz8i42NVTX/Jqj6jBBCCHFbmgdFSUlJ2L9/P3bs2IHnnnsOQ4cOxdGjR1VLPz09HYWFhca/CxcuqJa2JWpoTQghhLgrT60z4O3tjYSEBABAamoqdu3aha+++gozZsywWDcqKgq5ubkm7+Xm5iIqKko0fR8fH/j4+KibaTFUUkQIIYS4Lc1Liszp9XqUlZUJLktLS8O6detM3luzZo1oGyTNUExECCGEuB1NS4rS09PRp08f1K9fH8XFxZg7dy42bNiAVatWAQCGDBmCevXqISMjAwDwwgsvoGvXrvj888/xwAMPYP78+di9eze+//57LT9GFZrmgxBCCHFbmgZFV65cwZAhQ5CdnY3g4GCkpKRg1apV6NmzJwAgKysLOl1VYVbHjh0xd+5cvPXWW3jjjTeQmJiIRYsWoXnz5lp9BGGuNcoBIYQQQmRwuXGKHM2h4xwUFAEHTgL+vsBdLhaoEUIIIW6sRo1TVD1Q7zNCCCHEXVFQpCbqfUYIIYS4LQqKVEUlRYQQQoi7oqBITVRSRAghhLgtCooIIYQQQkBBkboYqj4jhBBC3BUFRQ5BUREhhBDibigoUhOVFBFCCCFui4IiQgghhBBQUOQYNWuQcEIIIaRaoKBITTQhLCGEEOK2KChyBCopIoQQQtwOBUVqopIiQgghxG1RUOQIVFBECCGEuB0KitRE03wQQgghbouCIlW57zhFE1ZPwD0/3YPblbe1zgohhBCiCQqK1OTGTYo+2/YZNpzbgBWnV2idFUIIIUQTFBQ5ghv3PqOSIkIIITUVBUVqqga9z5hq8BkIIYQQJSgocgQ3Lili3LkOkBBCCLEDBUVqolIWQgghxG1RUEQIIYQQAgqK1MUvKHLTKjRqU0QIIaSmoqBIVRRQEEIIIe6KgiI1VYeSIgrsCCGE1FAUFBFCCCGEgIIilfFKWdyzoIgQQgipsSgoUpNJzZN7RkXU0JoQQkhNRUGRqtw/oKA2RYQQQmoqCorUZNLQWrNcEEIIIUQBCoochqIiQgghxJ1QUKQmxv0bWlObIkIIITUVBUUO46ZRESGEEFJDUVCkpupQUkQNrQkhhNRQFBQRQgghhICCIsdx12k+qE0RIYSQGkrToCgjIwN33XUXAgMDERERgf79++PEiRNWt8nMzATDMCZ/vr6+TsqxDBRUEEIIIW5J06Bo48aNGD16NLZv3441a9bg9u3buO+++1BaWmp1u6CgIGRnZxv/zp8/76Qc28BdS4qoTREhhJAaylPLna9cudLkdWZmJiIiIrBnzx506dJFdDuGYRAVFeXo7CnDwG0bWRNCCCE1mUu1KSosLAQAhIWFWV2vpKQEcXFxiI2NRb9+/XDkyBFnZE+m6l/SUlZRhuWnlqOkvETrrBBCCCGqcZmgSK/XY/z48ejUqROaN28uul5SUhJmzZqFxYsX49dff4Ver0fHjh1x8eJFwfXLyspQVFRk8ucUblpaJKeh9atrXsUDcx/AwwsedkKOCCGEEOfQtPqMb/To0Th8+DD+/fdfq+ulpaUhLS3N+Lpjx45ITk7GjBkz8P7771usn5GRgffee0/1/FZXctoUTd8zHQCw8vRKiTUJIYQQ9+ESJUVjxozB0qVLsX79esTExNi0rZeXF1q3bo3Tp08LLk9PT0dhYaHx78KFC2pkWZwxpnCfoiLWTRuFE0IIIWrStKSIZVmMHTsWCxcuxIYNGxAfH29zGpWVlTh06BDuv/9+weU+Pj7w8fGxN6s2uBMVuWmcQeMUEUIIqak0DYpGjx6NuXPnYvHixQgMDEROTg4AIDg4GH5+fgCAIUOGoF69esjIyAAATJo0CR06dEBCQgIKCgowefJknD9/HiNHjtTsc7g71l0jOEIIIURFmgZF06ZNAwB069bN5P3Zs2dj2LBhAICsrCzodFW1fPn5+Rg1ahRycnIQGhqK1NRUbN26FU2bNnVWtq1zw4IWW6vPaCwjQggh1ZHm1WdSNmzYYPJ6ypQpmDJlioNypCb5gcYvB37Bm/+8iSWPL0GrqFaOy5IIfkkRBTyEEEJqKpdoaF0t2VD4MmTREFwouoAn/nzCcflREbU7IoQQUh1RUKQ2OwKG2/rbKmZEPluqzw5fOYxbFbccmBtCCCFEGxQUEdPqM4mgrt0P7RydHUIIIUQTFBQ5ihuN/cMvKZJqU3Sz4qajs0MIcaLyynKts0CIy6CgyIVo1cjZlpIiQkj1sfn8Zvh84IP3N1rOBkBITURBkdrcPKg4lHsIH23+CDdvcyVCz/z9DHr+0hN6Vq9xzgghant++fMAgHc2vKNxTghxDS4z9xmRp7yyHI8seATd47tjbPuxqqTJrz4bv2o8AOBWxS1MumcSftj7AwBgx8UdSItNE9qcEEIIqRaopMhRFDQpklN19cuBX7D4xGKMWzlOQaaECY1ovSd7j8lrKikihBBS3VFQpDYHTwhbVFakeppCXfJpEEdCCCE1DQVFqquewQQ1wCaEEFLdUVDkKA7qke+I4MRZE8LeuH0DS08uNTbiJoQQQlwJBUUO45hAw9bJW/lyS3Kx6fwm47/PF5yXnaYa1WnDFw9H33l9MervUXanRQghhKiNgiK1icQOB3MPYs/lPcILjZs6tooqZkoMumZ2xeozqxH1eRQafNUAhbcKBUuKHFEi9fuR3wEAcw7NUT1tQojt7HnIIqQ6oqDICSr0FWg5vSXa/tAWhbcK7UpLabCyN3svKvQVAIBVp1cZ3z9XcI4ujIQQQggoKHKAO0ELL87gD6N/7cY1J+cHKC4rRur3qcbX/O71YkEWA8Z0+g9qaE2I6r7b9R2WnVym2f7pd02IKRq8UW0SXfKtXYTUuEDduH0DPh4+8NB5GN8zD8TMxxwSqj5j7/yPEOIY+7L3YfTy0QAAdqI2vzUqJSbEFJUUOYGaFx5r7Y6u37yOWh/VQrsfrc9kbz4wo7MaWhNCqlwuvqx1FgghZigochSROMORwYWhrdDe7L1W19tyYYvJa8GG1mbVZ1RqRAghpLqjoEh1yoOe49eOo9HXjfDn0T8Vba80cBErKeKnd/3mdUVpE9eTXZyNtJlp+PnAz1pnhRBCXAoFRWoTiIlsCVb+y/8PgxcMBsC1D9p5aacm9f4MY1pS9NC8h5yeB+IYE9ZMwPaL2zF00VCts1KjUekrIa6HgiKHsb2htblumd3Q/sf2yNyfaVdOpPYpdnHmv1/JVtqVB+I6Cm4VaJ0Fl7Do+CK89c9b1NiYEGJEQZGjqHCd3XV5FwBg1v5ZxvccMs2HWPUZ3SxINTbgtwH4cPOH+Pvk31pnpUY7lHsI6WvTKVgnLoG65KtNIGixN7hwdM8vOSVFhFRX2cXZWmehRkuZngIAuFJ6BTP7zdQ4N6Smo5IiJ7A3uJBbOqQ0+BLazrz3GSFEXTTMham9OdZ7zRLiDBQUqUjP6nGrsox7we/Ozh8ZWsGF0JEXTwYMrt64KriMSooIIYTUJBQUqahbZjcErmqP/NtFJu87K7hQsp9tF7ehxbQWFu+b9z4DgNLyUsV5I4QQQlwdBUUq2py1GRVsBVZc32ryvtbVUNb2/8W2L8S3MwuyAjICVMtTTfbtzm/x0/6fNNt/TSoBZFkWI5eMxGdbP9M6KxZq0vdAiLugoMgBPBjTw8q/+CnpPcbfxlpVmtgy87nO+NTsan/m+hn87+//4fT106qlWd1cKrqEMSvGYNjiYZoHyzXBpvObMHPfTExYM0HrrBBC3AAFRQ7gAQ/RNkUAN0Dj93u+R3lluaz05LYp4gdfS08uFXzf3M3bN8XTs/Gm3fOXnvh+7/e456d7bNquJiksK9Q6Cy7hVN4pvL72dVwtFW7PppaS8hKHpu/uqLSKEFPUJd8BrJYUgUGjrxsB4KbOeL3z65LpMQyDm7dvovR2KcatHCcrD33n9YX+HT1m7JmBUN9Q0fUuFV8S3icYmy+YZwvOAgAuFl20abuaigWrSQ8kVyihSv0+FcXlxTiQewArnlzhsP3QTZ8QYgsKihzAIigSuQltPL9RVlC05/Ie+H/kL7me+X5+P/I7nlv2nOR2ctMjRC3F5cUAgK0Xtkqs6XiOGBCVuKbFxxcjISwBzSKaaZ0V4qIoKFIJv92OB+NhMqK1vW2KlFa5HMw9qGg74E7vMxlP2bcqbsHX01fxfmoylmXtmT+YEGKDf7P+Rf/f+gMA2In0wEeEUZsilVToK4z/9oB4SRH/3ytPr3RonuwdNl9OSdHl4st27YMQQpxhX/Y+k9dfbPsCid8k4lKRcBMCUjMpCoouXLiAixer2o3s3LkT48ePx/fff69axtzN7crbxn97MB7gFxVZK3HJ2JzhsDytP7feru2pPYb6qEqSENfw8uqXcfr6aby1/i2ts0JciKKg6IknnsD69dwNNycnBz179sTOnTvx5ptvYtKkSapm0F3wS4qeOfkRckpzja/5N0LzLvBv/POGw/LkofOwa3u6gTsWBZ2OR+cwkWLoBXwo9xAemvcQDuQcUC3tcwXn6Bx0M4qCosOHD6Ndu3YAgN9//x3NmzfH1q1bMWfOHGRmZqqZP7fBD3bO3bqMUevHG1/zb37xX8U7LA/mN1kdo7x2VG7vM/rBux9XCsZo/i/iLGLnvaE9aNfMrvj75N/oPLuzKvv7avtXiP8qHuNXjlclPeIciu6at2/fho+PDwBg7dq1eOihhwAATZo0QXZ2zZxx2jw4OJZ/UnSZs3DVeMpRwONYdHxdAwVmNRvLspi8ZTLyb+UDUG9sK8OAoV/v/FqV9IhzKAqKmjVrhunTp2Pz5s1Ys2YNevfuDQC4fPkyateuLTudjIwM3HXXXQgMDERERAT69++PEydOSG63YMECNGnSBL6+vmjRogWWL1+u5GOoyvwphP969+XdDt9/3o08DF001OQ9u0qKZPY+I7ahY0qIa2HB4tW1r6qeri3X3/LKcgz8bSC+3qFdALX1wlbU+6Ie/jj6h2Z5cAWK7pqffPIJZsyYgW7duuHxxx9Hy5YtAQBLliwxVqvJsXHjRowePRrbt2/HmjVrcPv2bdx3330oLRWfeHTr1q14/PHHMWLECOzbtw/9+/dH//79cfjwYSUfRTXWnvoN3UAdSWhuJ3uCIkBeSYa73eRXn1mNV9e8atIwXivuduwIscf6s+sx5+AcrbNhwVEltrZcf+cemouFxxfihZUvOCQvcjww9wFcLr6Mhxc8rFkeXIGicYq6deuGa9euoaioCKGhVaMlP/PMM/D3lx5k0GDlStMu6ZmZmYiIiMCePXvQpUsXwW2++uor9O7dGxMmcEWT77//PtasWYOpU6di+vTpCj6NOixKipxcNcJv6G1gb1BkC3epCur1ay8AQHxIPJ67S/nAloS4k/LKcpzKO4WmdZpqNljlvT/fCwBoFdXKpQZPdNTDiS3X3+KyYofkwRZyp52q7hTdNW/evImysjJjQHT+/Hl8+eWXOHHiBCIiIhRnprCQG6QwLCxMdJ1t27ahR48eJu/16tUL27ZtU7xfNbhiUGB37zMbLhYLjy+0a1/Odq7gnNZZcMlzprrhn8PTdk3TMCfa6je/H5pPa47Z+2drnRWXmwbIFUqKiOtQ9K3169cPP//8MwCgoKAA7du3x+eff47+/ftj2jRlFx69Xo/x48ejU6dOaN68ueh6OTk5iIyMNHkvMjISOTk5guuXlZWhqKjI5M8RrLUp0oozGlob1nG3oKgmc6VgzJmlFs8vf95p+3I1hoFiv9n5jcY5cT2uUFJEXIeib23v3r24++67AQB//PEHIiMjcf78efz888/4+mtlDcVGjx6Nw4cPY/78+Yq2F5ORkYHg4GDjX2xsrKrpG2h9oxH6Ydvzo/zj6B84k39G/v5d6Ebryug4EeJaHPWbtLeknmhD0V3zxo0bCAwMBACsXr0aAwcOhE6nQ4cOHXD+/Hmb0xszZgyWLl2K9evXIyYmxuq6UVFRyM3NNXkvNzcXUVFRguunp6ejsLDQ+HfhwgWb8yeHUJuikUtGou+8vg7Znxz2/ijvnn235DqGJ31XKBlzN3TMHE/ODa8mTQhrPvwABelUUkRMKfrWEhISsGjRIly4cAGrVq3CfffdBwC4cuUKgoKCZKfDsizGjBmDhQsX4p9//kF8vPTAhmlpaVi3bp3Je2vWrEFaWprg+j4+PggKCjL5cwTziwsLFjP3zcTSk0sdsj856EcpribdCM1RMEZckVbjRVGbIsKn6Ft755138Morr6BBgwZo166dMSBZvXo1WrduLTud0aNH49dff8XcuXMRGBiInJwc5OTk4ObNm8Z1hgwZgvT0dOPrF154AStXrsTnn3+O48eP491338Xu3bsxZswYJR9FNVq3KRK6oDjjR2m4oJhfWPJv5rtEt3dXRk/pRGuu9HCgVbBuGNFabRQUuSdF39rgwYORlZWF3bt3Y9WqVcb3u3fvjilTpshOZ9q0aSgsLES3bt0QHR1t/Pvtt9+M62RlZZmMkt2xY0fMnTsX33//PVq2bIk//vgDixYtsto42xnoBmcq7NMwNP2uqdbZIAp8tPkjvLBCu/FSiPNocd1ypUAMsC8o+u3wb9h2QbjnMwVF7knROEUA17YnKioKFy9y3StjYmJsGrgRkPeD3LBhg8V7Dz/8MB5+2LUGmNK6SkJo/8684Ant//T1007bvzty9Dmz7OQyvLb2Nfw84Ge0iW4je7s3/3kTAPBs22eRXCfZUdkDQFNsOJurBSTOJHY9VPo7PJBzAI/9+RiXxkR1O7oQ7Sj61vR6PSZNmoTg4GDExcUhLi4OISEheP/996HXO6Yo0tVZtClygZIjRxUL82kZDNrz+TRrv+DE4/XgvAdx5OoRi8b+cs/Nssoyk9eV+kpU6itF1ibEPSm9Vkv1zqWgyD0p+tbefPNNTJ06FR9//DH27duHffv24aOPPsI333yDt99+W+08ugWt2xQJcUZQpJX3NryHOpPr4Mx1+cMG8LnC9+OswPly8WXBEc+l8ANHPatHi2ktkPxtcrU+rxwh70Yexq0Yh73Ze7XOiktztwcVqfxSUOSeFH1rP/30E3788Uc899xzSElJQUpKCp5//nn88MMPyMzMVDmL7kGLkiKpfTjz5uXIz3vkyhEU3Cowee/dje/i+s3rSF+XLryRC5i6cyrqT6mPU3mntM4KDuYelLWe2Pd4/eZ1HLt2DKeun0JuSa7gOkTYuJXj8M3Ob5D6farJ+65QmuxKtHpQod5nhE/Rt3b9+nU0adLE4v0mTZrg+vXrdmfKHZn/oF2h6qqSdXxVh5oXlB/2/IBN5zeZvLc3ey+aT2uOmC+sj19lK2c8lY5dMRYXii5g3Mpxxvdc/UbIP6fE2p+4U7sUVygRPJBzQOssuBRXa0emuKRI4ndAQZF7UvSttWzZElOnTrV4f+rUqUhJSbE7U+5IaJwiZ+/TnFNLiuz8vJvPb8YzS59B18yuJu8bpicovV1qV/paEqu6coUbtjn+OcO/eSkN5liWxX/5/7l8MFhTuFpAwufMvPHPR6XnJlWfVU+Kep99+umneOCBB7B27VrjGEXbtm3DhQsXsHz5clUz6C5csU2RU3uf2bmvU9eFq5joZqoOa8dx6cml8NR5ondCb1k9dGy5eaWvS8cnWz7BB/d8gDe7vCk/w07kyoECcYzYKfZP9+ROJaZEPkWhbNeuXXHy5EkMGDAABQUFKCgowMCBA3HkyBH88ssvaufRLWjRpqi8stzqcleowpNLrFeTuzbq5edbNNDQKODjf2cFtwrQd15f9JnTB2UVZTZXn525fgZjlo/B2fyzgut+suUTAMBb699SI+s2o6Da1J7sPVpnQTP8c/tS8SXB921hKMUG6DyrThSPU1S3bl18+OGHJu8dOHAAM2fOxPfff293xtyNZUmR432942u81vk10eXOCCh6/tITk7pNsiuNP4/+iWeWPiO4TOqC5aj2APaavW+2Q9O3ldhxKrxVaPx3eWU5vDy8jK/llKDc+/O9yCrMwtr/1uL4mOM254uetjksyzrtWNy4fQP+Xv5O2Zc7UBrQTNs9rSoNsBa/F1tKIOl34Dqo0lMlWpQUbbsoPJKqgTOCootFF/H0kqftKjEavGCw6DJXLinSs3rREq7FJxZLbu/MKla55yN/vfLKcpwrOGfxPv9in1WYBQA4kXdChVy6r292fIOf9v8kuMwVqtL5yirKpFeqQdT4fuwdv4tKmlwHBUUqseh9BsffzBefWIx+8/uJ/qCc0fvMHsVlxVh1epXVdZReLIrLijF151RcLLqoaHspLMui/Y/t0WpGK5suiK5wg+Qf0/c2vme6jJe/9j+2R/xX8dh2YZusarWa6nzBeYxbOQ7DFg/TOitEATUCEle/1spBbes4FBSpRKsRrZecWILLxZfx5fYvLZa5+jhFD8x9AL3n9LaersIg4tE/HsXYFWORNjNN0fZSisuLsfvybhy+ctikfYIBP99in8EVng5n7zet5uOfM7f13IS+vx781WH7F7oQ86v03AF/DC3FvfScGCxTUGtK7rG39t0KPRjRcXZPNrUpGjhwoNXlBQUF9uTFrWkxTpHB1RtXBZ9UXOGmK6ZSX4nNWZsFl03eMhn5t/LxUfePFB3H0vJSrDi9AgAcVlJ0u/K28d9eOi+L5ddvutZ4XfZc+FmwDjuXSspLoGf1mHNwDj7Z8gkSaydi0fFF+Pvxv/Fg4wcdsk+1mTSqF2hb4mq0vi64WrAg53icKziHjjM7Ylz7cXi98+sWy6tDSRHh2BQUBQcHSy4fMmSIXRlyV1qMU2Rw8/ZNwfdduT3Ozwd+Fl326tpXAQBPt35a0QX8fOF5xfmSy1CKAgCeOtOfUf7NfGy9sFVwO5PxUVyxTZHIxMKOymtZZRk6z+psbB935OoRAMCENRPcMyhiWZjHRPzlV0uvok6tOhZpCG1HnEPOuf3a2teQXZKN9HXpgkGRnGutMxvTE+VsCopmz3atHjWuRMtxisT25cygSOpJybzk5PT105Jp3rh9Q7r3mUpPvfMOzcOm85sw9f6p8NB5SK7PLykyZ8scV2eun8HOSzvxWPPHrF4wWZZF6e1SBHgHyErXfFoUOViwsi/uapLqMGCPwjLpqjh7b1TWqkrLKspw/FpVr7wPN3+IL3t/adf+fjv8G1adWYXpD06Ht4e3XWnx8/vWP2/hrS5vwdfT16403Y2c81nqdyHVrrC4rBgtp7dE9/ju+OGhH2zKH3EualOkEq3aFFnblzODokXHF1ld/t6G96wuFyN1HMVKNqSYV3E88dcTmL5nOuYdnicrX/ySIvM8mN9kreUn4ZsEPPHXE5L7fWrhUwjMCJQ9h9mE1RNM8yAjSK/UV2L92fUW7zuy+kyMLVVQJeUlOJl3UnDZ8MXD1cqSKGtjUi04usDk9a2KW4JpCH0/q8+sxpjlYyy2eezPxzB7/2zM3DtTaZYFfbj5Q3y29TNV03QHqvQ+E3go5J/D8w/Px9mCs/hx349274s4FgVFKtGypEgs+HGl6rOrN66avJZ7fBz9GbZf3I5fDlQNOHq19KqVtavwS4rMb4Ryb+jtf2xv/PeWrC1W151zaA4A4IttX8hK+9CVQ6LLxI791J1TMfB34XaDchqOayX522QkTU3C7su7Ndm/eZsiPqkBVq3p9WsvfLvrW0zZNkVwuflvSi5+0G5+roqNLG+vr3d87ZB01SAV8J/NP4s/jv5hdR2pkiL+MX93w7ui1etEexQUqcSipEiv/bxjjrrA2cJwXORUSQltqyQosuWmnTYzDUMW2d4Ojn+zM8+j3OoYsdINazwYecdRyfHOPJApukyNuaIcxdCYfunJpZrs31pJkXnQYe3Ybb2wFVN3TrVYxzBWlCM4K8B9YeULTtmPElLHoEtmF8k0pJoP8OdBe2/je+g0qxOOXj1qfM8V2hq52sOOVigoUon5CXVT77wB0lztJsVnuGHIvZmbMymhkPE592Xvc8hTaU5JjklbIWslJ46cCFJu2ubHW86xE3vadWRDazFKbhJ+nn4OyImpf7P+RdrMNJNSKWuN5y2qUq0Mz9BpVieMXTFWdnAnt0Qy70aexb7soWf16PVrL8VVk4apX9zBybyTFj1Y+cGMgdDDG/+7F7r+/X3ibxVySNRGQZFKtAxMXDnCZ8Gi8Fah8qCId1y/2vGVZHVEm+/b4Ie96jdkjP48Gqnfpxrb9FgrObEoHbCh6ulq6VUsP7UclfpK5JTkWCyXGxSZrycnuKzQV4imZ62KyBYf//sxnl36rOR6R68etfmm4YypK+6efTe2X9yO3r9Wja8lZ547ucsB+aOD82+6V0uvou+8voJt+0YsGWGaByvfn5xA60DOAaw+sxqZ+zNl5dPc2v/WKtrOXkrmIEyammTxXrPvmlm8J1V9JvS7deXrdk1GQZFKtDzBXbmkaPru6Qj5JASz9s+yeVsWpiUUL656EUEZQSZPvs7+7JvPbzbmzZgHidIBOQzbtJrRCg/MfQBNvm2C6M+j8cGmD0zWM7+4XrtxTfCCrKT6zFpVpVrVZ+nr0jFjzwxZ6z40/yGL93458Asaf9MYx64es1jm4+mjOF+2yrtZdQ5aCxitBady3pfr9bWvY+nJpRjw2wCLZWLjgSnF72RQHagxuKpUQ2tHlh4TddE3pRIqKRI2dsVYwfcz/s2Qtb35cS2rLMNfx/6yO18f/fsR5h2y7PElFdAILVfa0FrI5eLLAKqGLHh7/dsmy/kX1wM5B1Bnch30+KWHRTqKqs9E2kWYB6dKKRkmwNyQRUNw6vopjPx7pN1p2YPfFd5aKZy1NkW/HhIeKdw8je/3Ck+wzU87tzRXNK+2tGuSw5U6cKhBjd67W7K24NMtn4qWGLlDUOTqg446i03jFBFxWgYm9k5G6MoceQF+4q8nFG/Lv5AqbWitBL8E6Me9XPfeDec2WKxnPqCkHGLVZ+YN3pWe633m9FG0nZBrN64huzgb0YHRxvdYlsWPe39Eg5AG6NHQMlBUEz8osuXY8Jf/fuR3u/Ig9zyT265JLlcIiliWxanrp5AYlmj3702N3ruGee8iakVgWKthFsurW+ladeb64aubULukyJbSkOo6xLwWDXwN9mbvFZyqQ+hpyqL6TOUncz7ZDa0VVJ9ZC675nyG7OBtzD821ubv59ovbbc6TmJN5J1H3i7q4duOa8b3dl3dj1N+j0POXnqrtR4xoUGReUmRDQ2v+OraeM7YEBtbSlvN7EwsWdlzcgYcXPOzQ3nIGkzZOQtLUJExYM0F6ZQlin1nJdfXY1WOYvW824r+Kx7FrVVW8QxcNVZw/4lxUUqQStW/eg34fJHvd6lxSpIUtWVvQeXZnBHgHoDi92GSZ4eZjtcrEyg1q/Mrxgu/LLbrmV4tZ249F9ZmM81O0pMis+qztD20BAKfytB/y4UDOAeO/lU7vsuzUMrSJbmMcWTzIJ0hyG7klRUoC5JLyEjSe2lhyPX7a1s4fue2a5BK73nSY2QEAkFWYhR0jd9i1DynvbnwXAPD5ts/x2X32DTip9uC3Ty952q79Em1RUKQSLU/w6lpSJIcjSpKWn1oOgLs5AcCSE0ss92utG7aVG5S1Rq/r/lsnmTfFvc94+V1/znLUasD6eSR0g1h6SrzreHFZMTae3yiVTUlXSq8golaE3elY89exv4wls/+c/QeJYYk4kHsACx9dKFriJtabUnFJEe/9uYfm4kz+Gcl8y64+U7mtiHnJmHk+XCFYFjJ7/2yb1lfysOnK7TuJPFR9phItfwyuUMfvKELBJv8irHYwKnQh7De/X9W+harPbCgpEnOu8Jxgg2lzcoMisTz8l/+f6DaiU1CwtlfnDPp9EPrO62vTNkIcNUp1VmGW4Pt/HvsTH2z+AH+f/Btr/lsjur3Y8VWjpEgoOG01vRX2XN4jmrZa1WdyAih+/twpCBAb5V3J3JGz99kWYLkSlmXxyupX8OX2L7XOikuioEglWpYU2foE5C7Eej058li/suYVrD0rPo6KYPWZ1I1Qxo3jzHXpkgHAtK2QtRuY2LLs4mxZ+zFn683PWkChJpMA2YY8CnVdN1dWYfsArKq0KRI4vw/kHlDcUN2WhtY/H/gZq06vwsm8k+j+c3fBufBsGZfJVVhtR6Wg+kxuFZkrOpB7AJ9v+xwvrnpR66y4JAqKVFI3sC7qBdbTZN9C1TvVgVYX3J2Xdooum7pzqsWIw+YXz0+3fuqQfAHKq0IMN0I/L2WjPmv1XTww9wE8NI8br+hK6RWL5UrnFuOPTi6GBYsrpVeQNjPNYvJVsfY8ckuK1pwxDRr5jfrFgpb8W/miebXlvJBqaN17Tm8M/n0w/jn7D+79+V5jdbKBWgN5mnNkr01r+VSzobUtrOWppLxEcPBWNRSXFUuvVINRUKSSuJA4/Db4N62zQazgT+Kq1KErh9B3Xl/RJ/uckhyLySNzS8THkLEVv/rM2k3EvAuwIY9KpsJgoWwOOrX8ffJvlJSXYNLGSRbLrJWeKMmzeWPzd9a/g+0Xt1uMiyRafWYWcDz+5+Omy8HiYO5B3PfrfSbvj1s5TjQNMQzDYNP5TTiVd8rquaAkkL5UfMn47wfmPmBSteqokiJHBt5yByaVu42tadkqYnIEoj+PdlhgRMRRUERcltrtFcxHh7aHWPWZULscZzeEP3LliGjpodKncWdXY5pjwODG7RtW1+Hn51TeKdT+tLbd+y0ul36qtlaFZ/GaZXEo17JtC/89uef9sWvH0DWzq2RPNTXGKeKXyNnaAFlJNaTarAZFCtoUOdrNipsAuEmC1eYKk8+6MgqKVEQnm7r0rF66obUNF/gf9/2oSr7M8fMoOMeRg4IHsRKA9za+Z5mHO8dJSWmZknFz1CantMowCjgAvL7udVVG0JbTsN2k+kxq7jOwkmnKvRnzAym1qs/E5N3IM1Y5SlWfmb/30qqXZO1jX84+2ZPh2sraMTXMZ2jOlYY62XN5D2btm6XK73DhsYUq5Kj6oi75xGWp/aTmqKJ+/k1A6Ian5udgGAaV+kqrbWnEupHn3chDyvQUm/dpPqK1FliWtTphLaB8nCJrxAIYQxBiHmRKjmjNspIDa8o9T2X3RLQ2QbHMfTX8uiEAYPeo3TZv/93u72TtAwBXNT1R/QBcyfmr9TkPVB1fw9hgkbUi8UDjB+xK84vtX9idr+qMSoqIyxK7KI36e5Tg+0INcfkcVY3Fz6fQ+DVqVwM2+64Z/D/yF63aEcwDy+LDzR8q3qfWXa/1rF6T8bhEgyKGwatrXoXvh744evWo8X2pXmQsWJMSLf77Qv+WmzerbYrMq8/seDgwH+NK6/PC3Ox9s/G/v/9nUcqj5DM7vKG1jDyZH98jV484PQ81DZUUEZdl6w9WatwNNZ/8xJ6WxarPfj34q6L5yIScyDsBANhxSXjUYKGSiK0XtmLK9imK9yn0XUiV3KhJz+pVq86Qe14N/G0gHmv+mOjyyVsnAwDe2fBOVdq880LofMstycWbR960mie55yk/2FEyPIN5GnK2MR+zyvDvBUcWWLynlJ7VK55A1dBV/r5G92FQ06pZAZT89j/bavtI2Z9ts290bXPmx9LaOGOK0gdLE8GaoZIi4rL0rF7ySXRL1haEfByCBUcWWL2Ql1WUOaX6TCi/129ex1MLn7LoiWQvsRuHUEnR1zu/VrwfsfGiDuQeEFjbMfSs3qYgzNp3/eRfT8pLAyzmHZ4nuExsfi9rI50DMClVMt+XUBrW8L//P4/9KbqetYbWtv4mhBqSF9wqwCN/PGJTOtaoEWybz1uoJCj6dte3dudDbTP2zLC6/MiVI3jrn7dQVFYkKz1XqCJ0NVRSRFyWWENrvrybeQCAR/54BLX9hHsbjVgywqEDXEpVmcjpwSTXybyTxn+LBYFCQZFU1aIUrYvZK9lK24IiK8G0WKCjBqmSIjmDN8q9UckpTVlzZo3F6N0sy+L4tePYcG6Doioi/mf4L/8/tJjWwuY0rLldedtkbjklzANBW2/+5mMzOYLhONrz2/p6x9fw8fDB/9r+DwB3rdtxaQd2X96Nlf+3UjoPVH1mgYIiFdEJpi5b2ysYAiRzjgiIrt64avy3kvYg5vgzalvz25GqsbDEbopqVdPxad12xNY2RVo9AUsFN2LXCKlgSojY/GsG+3P2W4yHZNhX8rfJsvZhsa1Z/oWqrO09VwzB7/Wb17E3ey/ujb9XcXWa0jw9MNe+xsxy6Fk98m/mI/X7VOSWCo9lZi3fpeWleGHlCwCA+xPvR2xwrLFKfdP5TbLzINeyk8sQ5heGtNg0k/cP5R5Cw9CGqOVdS3ZarkzT6rNNmzahb9++qFu3LhiGwaJFi6yuv2HDBjAMY/GXk0MDXFVHD817yNh+xtXw50PTatoDsfYgUr2bbCV37rOpO6c6bDJQW9sUOXsMJeN+JYIbsZsQf/gAuTdwqSFADuQIV28qaeAr9r4jgk/DwKOtZ7RGz196InN/ps1pmJeiumI10Y5LOxD2aRjOFpwVHYPL2nfFH6DVfJBHuSVtcs+1cwXn8OC8B9FxVkeT95efWo6U6SlI/T5VVjruQNOgqLS0FC1btsS339pWd3vixAlkZ2cb/yIiHDuLNtFGcXmx6IzuroR/4XJmEGdLmyJ7ybl4jl0xFo2nNkZpeanq+7e1TZEzb4L86TckS4pEjiM/KJLd0FqigaxY0KR0ahTA8ibtiON8vuA89KzeWO1nrb2UGPPP/sOeH1TJm5rsraKzFjB5eXgJvv/Xsb9MXsv9/s4XCA93MffQXADOve45mqbVZ3369EGfPrZPchgREYGQkBD1M0SIAldvXEW/+f0wovUIPPrHo07br9hNUe3qM1un+bhcfFnV/QNuVH0m0YhZVimNzFIupSWCYo3EleTBlipCudr+0BaPNqv6HZ3KO4XPtn6G5+96Hv5e/orSfOOfN+zKk1aUzNkGAF464aBo0O+DTF7L/a7E9lUdByx2y95nrVq1QnR0NHr27IktW7ZYXbesrAxFRUUmf45SHU8QIu31ta9jyYkl6De/n+A0H44iVlJU29/+6S345FafGZg/jarB5t5nGrWBkiopKiwrlE5DwThFQsSCZjntc6xVn/E/Y1mlY6bw4LedO3X9FCasmYC3/3lb9vbu3M2c33PM2u+Of35l/Jthssy8pKhCX4EXVrxgkQb/e7Z2/xLLh71tvVyRW32i6OhoTJ8+HX/++Sf+/PNPxMbGolu3bti7V3zG64yMDAQHBxv/YmNjHZY/amhdM9nbs0spsZKC3Zd3q74vW4KMz7d9rvr+XblNkcl+FTSYNif2Oc1v9EpvSLKCIrEG4azp8Ay/H/ldUR6UfD+bszYr2pe7eWNdVamW1ZIi3jFceNx06g7zNkWZ+zMFh+Wwdo6eKzhnDNDE1quOQZFb9T5LSkpCUlKS8XXHjh1x5swZTJkyBb/88ovgNunp6Xjppaq5d4qKihwaGLmyVgGNsb/kpPSKxCaO6O0lh9hkkeYXSHu5wtxnlfpKm6rPSspLbHpfLfzjxG8IawvDZKBSVp623uXanob4YjdBOcGxnHUuFV9CTFCM5HpiGDCa94h0lENXLCcMFmLt85sHKxeLLgqnIfK7Ppt/Fg2/bggfDx/ceuuWeFDkXuUqsrj9J2rXrh1On7YcNt/Ax8cHQUFBJn81VZJfHAbWteyiS8S9t8FyclVzWgVFzmTLDYg/XIFaKtlKmyaz3XJBuFq948yOgu+rhX+cIj+LdOi+lJIzsJ+c8ZSs+f3I75iyTXwEdUNJhi2NvivZSuP+pZoquHNTBqnvx9Do2ZYHFbF1xYKdDec2AKiqHhVbz52Psxi3D4r279+P6OhorbPhFliwaB2sbHySmurdje9KruNKQdHhK4cdkq7WJUUdZ3bEvpx9dqcj9ylcKUceJ7VK7F5e/bL0vuzcz6N/PIqXVr9kdViA9LXp8PnAR3aae7P34qH5D9mVL3fArz6t1Ffik38/MVne4KsG+OXAL6qUlLFgcbvyNjae24iyCvH2YWL7ouozlZWUlJiU8pw9exb79+9HWFgY6tevj/T0dFy6dAk///wzAODLL79EfHw8mjVrhlu3buHHH3/EP//8g9WrV2v1EdxOdYzstXbqumPG5lEi/2a+9Eo2Mm9HogV+t3dX5ujjZEv6Yo2NhSaklbsfWz9fdkm2aDofb/nYprQAYOnJpQBkDEfgxg2t+dfoXw7+gjX/rbFY54PNH6BHwx6y0xT73vSsHq+sfkVyGiCxINmdj7MYTYOi3bt345577jG+NrT9GTp0KDIzM5GdnY2srKoh6svLy/Hyyy/j0qVL8Pf3R0pKCtauXWuSBrFOVw1P4upOzlN7hb7CoU9tWpcUuQtHHydb0nfEA5CtPRF3Xdolmo49GIaBtfjMnR/++IHGkatHRNdTIwDXs3pZ8yLa0tC6Ul8JHaNz2+9A06CoW7duVn8cmZmZJq9fffVVvPrqqw7OVfVWHSP76k7OxS/xm0RE1IrAZz3VnaXbljxUd1INmwHHHydberTZ81u3Nh2JLZ/x5HXHdOyoKdcxsWEoblWIN34GuDkSN5zbgG4NugEQ/z5XnFohKx/m428ZAh7zwKdCX4EW01og3D8cm4e7Z29B12kMQRyONf4fcSdynqrPFZzDuYJzDrkpu0LvM1fQZ470QLOOPE7O7HFlT0Nr/jpijYbV/hz5N/MR6heqapquQCwoyirMwsDfBlrd9p6f7gE78c6ksyLHe+3ZtbLyYTKVEVhjUGpeUnTkyhEcv3acW48XPLmT6tdKSkPu8DTthudojWfLeWXLAIey9+8CbYrchSuVFNnD2iCktgR+5nNyKUnDnJ7Vm9xsv9z+JcI+DcM3O74xvufOJUn8c8ja73nXZeGqSVvI7STC/7747RbNgyL+oJFKh6TQGgVFNQ3d29yOLTdCOd2tlaCSInlYljXOfu6o9OVySJsiG6vPxHpD2hM86lm9SdDz4qoXAQDjVo5TnKYWxAI3/u/dEQ85fGLTgZifO/zvi98D0Dwo4gdZtgyh4UooKKph3PkJqqay5Ub4/LLn1d+/jTfCmowFi4fmPYTwyeEOS19LchpayynpsCfIljOyuTtU28ipolQrsBA73h6Mh6wOGvxAjT9orEVJkc79S4qoTVENwtUFE3djy41QrAu03XmgkiJZWJbFslPLHJa+sxpai7E1QBYbnNHukiKpwRt5n/2Po38o3pcW+MfGllHc5abJ56nzlHWeyOmSr2f1VFJE3I87PEERU4W3pCcRdaTfj/yOY9eOaZoHd7Hu7DqHpV3JVuK1Na9Jrrfs5DLsz9nvsEFFtQ6QK9lKmwK+hxc87MDcOJaj25AF+gTaXFLEx992+u7pJstsGa3clVBQVM2kBDcRXbazSHzMC+K65IxA7GiGdhvEuhdWWs5ErhY9q8d3u7+TXO/BeQ+i9YzWjhunSKKUx9YearZ665+3ZM8P546UHpu3/nnL5m3a1WsnKygquFUg+D7/HPv14K8m54a7Vp9RUFSDZJUJ9wQhrm3OoTlaZ4G4IUeUMrBgVUl34oaJirf9asdXkuu4c4m40oDvw80fii4THXdKRrf5Gbtn4Jmlzwgu4wdU5sEyVZ8RlyB1gmtd9E0IcQ6HBEU2NrQW89OBn9TKUrVzruCc6mlam7ZFqKTotbVV1bTPLntWNF3zakyTRuJUUkQo4CCEuIpDuepPfiunobUrXAcZMGBZFsVlxVpnxaWxLCvY9udK6RWr241cMhJ6Vm9aUmRWtUolRcQlSDVA1LpLLyHEOT7Y/IFD0pUKepw1wKQ1DMNg1N+jEPRxkNZZcQnWpm1RMhbSzH0z8c/Zf0xqJsoryzFt1zST1+6IgqJqxpHjEP0++HfJdUa0HuGw/RNCtOVOo5vP3DdT6yy4DLHvzJ4Atris2KSkaF/OPny69VPja6o+Iw43PKqv9Eoq1PeLqRtYV3IdGhySkOpLzjx47nozrInsqerUMTqrPdeo+ow43Kwm70ivVKnOYF9CZPXoKC512P4JIdpyl5Iiejjj/Hn0T9T7oh7+zfpXcLk93yXDMNaDIjcNjmlE6xrGnicDP08/yXWY0uo7fgghNZ1aXfKJcwxeMBgAcLn4suBye75L5s7/xFBJEXGJJyh7npDGxzxudftw/3BM6my9tIqe0Aip3t7b+J7WWZD0z9l/tM6CW5i8dbLibaVKiq7euKo4bS1RUESMRkX3x+v1h4ouZxgGb981wWoaUkHRN32+UZQ3Qoj2/j75t2ipgyv5fu/3WmfBLezP2a94WwaM1SYVQxeJ30tcGQVF1Yw9JTU66NDIL8b6SifPK04fAEJ9Q+3anhCinZN5J7XOAnERRWVF2HZxm9bZUB0FRdWMVFtoaxV8OobB0MgHrCdgZ5shZwy/3ymopcP3Qew3uOlgrbNACFHoib+ewNKTS7XOhuooKCJGOujsnlnbFeYcmtfU+qB1fz3yl5NyQqzpUr8LYoNitc4GIQCAbSOqX6kHsR0FRU70cJ0eDt+H5IjWVnqfhXgGqp0dC67QEDstNk3rLBBCXEzTOk21zgJxARQUOdEvydr32hDrIfdm/acR7h1iddtaXrUckCP1SQVezgjMZveb7fB9EELU48F4aJ0F4gIoKHIiVyglEWtV9HBEd6tb/dz/Z4T62d9I2hnVa65QhWetq6qzuEsQS4grcIXfLNEenQVO5ApBkdKRlJ5q+RSgd49B2yRLiqwETXIGqHQXQT7aTob5ThcZI7AT4iI8dFRSRCgoqnYkA69K4cBGcrvLV4Hbts+mbO5C4QW705CitPosKiAKRelFjshSjVTL23pJlSuU6BFiQCVFBKCgyKmkbgEH285TYR8Se6kQnhtNcrtT54Fy+4dtP5N/xq7te4S2w4TYp6yuI3WcxW7GlfpKu3vfEflcoeSUEAMKighAQZFLaRGQoNm+Zd2eSm7Yv6Or+aKLsp+VDpjaBzZH//CuVtdRWgKh11cCt8ok12sU2khR+s7mCtPOEOIuKCgiAAVFTqX0Zj008gEsav6ZrHU7BDW3ulzsRikrb3aOZg3AahVcVK0ohHuFWN28T+2O0Nl52oqVUOgrKoAdhyS3PzX2FNpEt7G6jj0T7xL1uMKYVN4e3lpngRAiEwVFKrLnRri19UzRZZnJ76KfROmIwQfxz+H9Bs9ifcvpNu3fVaoypI5hp+CWkgGcPQ2tpTzX9jkwDAMvnZfiNNxBdXlqvj/xfq2zgNc6vaZ1FgghMlWPK5+bELtZj6n3CNKCU1TZR4CnP95qMAJNa8ULLhctKXJSUCRVpSMnrLR3HCKx5VJ5y+iegS97fwkA1b7tkTM+H8MwuHFbhSpZK6pLcEfUN6DJAFXTo3OteqBv0QU4IiCp5WFb13JXKSmK8g6TXEen0eCMXeO6GqtCvDzsKylqHeHa87M5qyQs72aeQ9OnHm5EzPzB8y3e2z5iu+L0BiUPsic7xEVQUORETBvhYeTVahCb5Bdn/HctDz8sbv4ZFjabbLYvkby5yL3jj2afIMwzWHDZpw3HAZC+0Ul9FrHtban+9LRz9Ns/H7K/p6EkO4ZQuK23v6ehK1AaILer1w4TOk5QJQ/Uvsw1CZ0b7WPaIyVSnVJ74p4oKHKmQOFxW6RKPuRa03KqyeuHwruif51usvZly81jbctv8Uy0ukXPXCYYJNeKx6bWMywWdQpqiQn1ua74Ug2t7a0+69Wol2RWPcuEhzYwT0s8D/ZpENJAeiU7bsblleWKt3WGn/r/JGs9pSVFXjovfNrzU0XbEvcgdm7s+98+vNjhRSfnhrgKCopUFBUQZXW52I9QrbroWF/r+weA0fUeRgPfungs4j7TvNlwm+4e2g5TEl6yOX9q0dlZrCV1o1z6xFLhBTnXgB0HgbJyeLES39mV64rzIKfq6uwLZyXXcfUSCqWlOMNaDcPA5IEO3Qepnt7t+q7x32Lnho7R4YteX9icdoB3gNJsERdCQZGKksKTMOuhWVjW4kt8l/ga7gqUN+uyLSVFsT6RSrMHAAjzCsZ/7Rfh+8ZvmLxv683Dw4GNCtVuSG1tsMf+Tfob/20IIcQaGde7zgC3yoHz2fBkJBoiXy+0nkcr8Yq/l7/1tN2Erb0AbZml3NHBDsMwogOdEvfVqX4n47/Vbm/2UfePVE2PaIOCIpUNbz0c99fuhOfqDcbOVHlF/LbMzry25bf4v8g+SrMHgLsYBHrWwreJr5q8ZwuxNjUhnvY/LUnd8GxpaP1Xs0/xaaNxqOMVKri8tl9tywTMSli2tZ6F5X3mob6hJI5lMTnFvm7W1o63Gj2/nFFC0rl+Z4emH+orPgGx3PNVar2C1wrEFx61b/R1UuWhpIe0zgIA09+Fmr+R9vXaS9YUEPdAQZELsKWkqLF/HH5JnmTxfm2RxsnW3BPS1vhvWy8PYu16Xq8/zOZ8mBO6WPFvbtINreUvF7wwHjiBcfUeNb7sENwCfW4lmqzSKCAOZ9svtrofq3mwUlL07f3fKk7XmL6DW85HeIVh0/+ttysNqZvS9deu489H/lS0rVxW52fLF58HL+flHFX2X1MMbCKvutPR+E0VqGciEaJpULRp0yb07dsXdevWBcMwWLRokeQ2GzZsQJs2beDj44OEhARkZmY6PJ+OptX4Fjqp4MAKsQtKsBolRRJZkTOidfNa3FQc3UJSuTT5T4iXrlStF1E1ArixcXRhiaz+gPb0GhQ73m90fgN9Eu0rCVRDl7gukuswZy+quk+hYxLsIxzsq3VDE/se5PweqGRAPlcJQBx1rXWVz0fsp2lQVFpaipYtW+Lbb+U9GZ89exYPPPAA7rnnHuzfvx/jx4/HyJEjsWrVKgfnVKGocFmr2TttBSBv0ENz/Au/l1QbGQUa+cZgf9s5WNfyO5u2s7dNEQDsbzsHJXdvQqhX0J1teM5ewq42P+Gjru/j+bueF9xeTsBjX1AkLKJWhNOC5BGtRwi+v/XpregU20lwmYli6YEXjz5/FJ/1/AwdYjrYlLeEMOvzAPp4+KBJeBOb0hSi9GYW5BOELU9vweudXheugiUmXKXBu6N+W67eqYHIp2lQ1KdPH3zwwQcYMEBe9+7p06cjPj4en3/+OZKTkzFmzBgMHjwYU6ZMcXBOFQqQN4Civb2pAGCOQJWauQ/inzN5zb9Qeas4WN+kBv8DAExr/DpaBjRGoId4w2FDSY5YvoTIOVwejIfJAJbmVWZtg5oivcMEk0EY+Rc2ORc5PW8dWy+21m7GzgqKxI5zXEic4Pt8cgPC5DrJeLnjy3ii+RPS+eEdk92jdlu8Z77u4ecOY+GjC2XlQ3SfYiVFEieZn5cfGoY2REaPDNQNrGtXHmoCqeP5ePPHRYN0a2Y+JD49kjlnlOxV9+l/agK3alO0bds29OjRw+S9Xr16Ydu2baLblJWVoaioyOTP1djS0Npc5+BWKOuyFb1rd5Tej9nXXcFW9a7xZtT7Mb/dYCRu3v0veoa1t21DxvAfgTZFvPeUlKyZNrC8wyzw4d/obS0p2vPMHtsyZCV5WUFRmfQ4QpKfIF+4h5wjgrLn73oeSx9fio+7f2x8z9oxDvaVbiPnofOw+yakRrUHTe8gTepBZ3ir4fjxoR9tTtfX01f2ulnjsxxWzWVI195z4dJLl9TIjkvw8fDROguKuNWvOScnB5GRpl3SIyMjUVRUhJs3bwpuk5GRgeDgYONfbGysM7LKCZLXvsaenynLsopLecrZqlGL7SkpGhh+DwDgkTpVAasv7wdh7eYsVCIjdeHy95B/ITSmKdTrxGzX/ODUalBUylUb6aE3vpUYlii2tnB+KvXC7zOMvKoGke1tIhJYqVbVcS0fqOQCbw+dBx5o/ADq1KojurqrVLHYqm/jvlpnwW7xIcJzJdbyqoV29drZnb49wYhYSVygd6BN6dg7NY81hipU/ue8q+5dWPDwApP1nmhhvcTUnlJH/gOHGKlqaTWl1rWsBXAHbhUUKZGeno7CwkLj34ULFxy/0w4pQKsmoiNYm7OnTRH/xmwrfkmRj0BQNDD8HoyK7o89qb9YTefzRuNxpv1CzG36geK88EldPuv5ROCtuKcR5xNt347MAjJ+L0CLkIi/bvENAKzJ9+a154Tx303946VHtD5xTnSZtafNFzu8iK1Pb7WaNpdd5W0c5Dztcp9PYh9FpcD5bMX5kMNhT/5WzsJXO75q8vqtLm9ZDIbqbnw8hZ/qs1/ORri/vLaR9rD2exEL2BwZ5Niia1xXTHtgGgDT387Mh2aajDk25q4xDs1Hm+g2kutI1UqIHWsl3LWdlVsFRVFRUcjNzTV5Lzc3F0FBQfDzE26/4+Pjg6CgIJM/h/PxBoLl98ISm9FeFK8Btz2n3W191dxYQj+WSO8wfJ/0JtoECjdovZy2AofazkMDv7po6BcjuxqwVyjX6LZlLbPSFT33aeSUGLwf/xzeiBsma3/maRpvpEfOAFevW74PeT/oBL9YPFynB56JHgAv3rQf3yRKz5ll7TNaC0q+6PUF0mLTYN83bx3DMJKfPzWgibws3LJSzcey6BjDVfuKTaapRqP7Bxs/KLmOXOdeOIePe5g+kft4+mBo5AOq7cNWG4dttDuN2CDhEnRPnacqJXj2pHFPg3tUS1fN0siDzx5ExdsV2DBsA2KDYy3SbxHZwuSaeE+8+OeQIlXCBHClsVKkHnhsqY6UMrTlUNXScia3CorS0tKwbt06k/fWrFmDtLQ0jXJku+yXs3Gm23Lj5KZvJTyDAeHKfyz2RONN/Bso3hYAon3C0TxAujjW/ClwbtMP8Hmj8ViZ8rXpinkFAMTaFFkS704t8B7DX37nxc1bwNH/jO/zS37ktCliGAa/N8vAjKQ3TAMqOdta6Qouq12CxC7klKCInTo6RofE2uLVgePqPYrMJhMl05fcf+41LGkwCT/2mY5Z/WZZZi6vELhtfWJaa5/T0LB2yWNL8Fon+wbbNIgLiXO57tdyhk94ru1zVpeLnbM6RqfK52Vuline9s0ub4ou6xgr3ZbSSOWSixaRLSwCEfNjZTIuksKA7KUOL2HOwDmS67WKaiW5jr1jvNnif23/p1pazqRpUFRSUoL9+/dj//79ALgu9/v370dWVhYArupryJAhxvWfffZZ/Pfff3j11Vdx/PhxfPfdd/j999/x4osvapF9RaICotDQPxYT6j8FttsuvP/kDNETcUjLIfjyPus96+ypPgvxCkR22goUdLZvED5bhXkF46XYJxHlE256MS4oBiD8w6wvMK+b2EUmSGCsJDkXJJ2NgY01/P2lxdgWtKtxYWoYLF36KPYZGTAY2nKoyTxRfF8lvoIoH5lVKoz5S94bRaWo7RmMEREPIsgnyPRzX80HDp8CjknP8Sbm/oT7uX0WlcC3tEJ0vRGtRyDEN8Q0nyreHCID7JuaRw1f9v4S64fa/jv30HmoU1J0/rLV5dYe7sRKL1pHtTbpKflAokRp3eWrDu8paH6s+EGT0uq+/0v5P6vL817NQ/bL2QjzC7M5f7Yurwk0DYp2796N1q1bo3Xr1gCAl156Ca1bt8Y777wDAMjOzjYGSAAQHx+PZcuWYc2aNWjZsiU+//xz/Pjjj+jVS3pWc7cQZ/qD/an/T3ih/Tirm9h7847yCVdlwEVrZP/QDEERb/0fGr+JxyN64YtG8gLfdoHNJNcRfSq2saRINH2WxSMRPdE4LBHPtHkGm4dvRv5r+TalsXn4ZsX7B4ClgxZiftMP4cl4YGhz8bnfhDAMAw+dByZ2s680yFtiuhLzY2xynhjmjpN4urd2bhnT338CLwY8iDYBTfB83cEW6/340I+4OuGq1f1IsXa+PJP6DIa3Gm61GsjRvD280a1BN5u3k9ubSqp6hwGDwXW627x/A/Pu+sNbDccvA7i2jn8+8ide7PAiFj+2GOxEK+fLtXzEh8bjz0f+xIahGxTnxZoIX9OHBf7xUzp9j9S1KMwvzFgquuXpLVbXdZV2WK5M06CoW7duYFnW4s8wSnVmZiY2bNhgsc2+fftQVlaGM2fOYNiwYU7Pt+qaxAMNY4AGAk8xEk+s7tCWzdYAg/+Je4S2w9ymH6COd6jlegLHRmxfIZ5VPVXEGrablBTZeVz9PXxxfORBzOg7Ax46D4Qw8iZ51d3k2uBIzivGskjyEx9PKKl2Y9wT2hY37/4Xz7V+Rna+AfuGiOAbGd3f7jTseXLlnwvBngHY0/YXfNtYuBrN3vnmrJ0u3h7emNVvFh5p9ohd+xClwsS1YiU1DBjJUrOP7v0Ivw741eo6DBj83jQDxcP/E1wudY2Y8eAMkyrQ7x74DtGB0UD5bQys2xNf9PpCVpsaABiYPBBdG3SVta6tFqV9h7sCm2JFi68AmP6W1JjTUErH2I5Y/Nhi9EvqJ7h8ap+pVrdnGAYf3vuhI7LmNtyqTVG1FVkbiFU2sJi9JUVaE8q9yZhEKo0BE+4VYvy3l8jFSY2RxfmMN5Obt4DtB2VtoysulZ3+P62mYYpoCRp3ZLkLsW3niO58DrDzEHBbvMqJ24X1dPmDZwpubi1fuXnW932HtRu2nrWtavnHe6vauDG37Q80LMj8TDbbss/uJESrUiWGiFjw8AK80vEV6bYqd4KrAC/hUmmptpEeOg+TKXmM14Udh4D9J6pKFh3Alslsm4ckYWfqT8Zx41QpKbLxCe2hpIew6LFFgsuSwpMkt0+Nds+u9GqhoMgd8a4/9rQpstBYeiRjZzAdfdrKegJLxS7utjZ+llzfSq8q1vwfAhdsOVV4Uur61MH4WOleKa0jW6FBSAPZs9p7XLwC3CwDLl+RXtmF2RoUjWjO6y1Tyo179l/7RepkhmWBwhJ10nIyawHP4KaDNamSMQYb+jvfsQODIvP2ZoL0eqDI8vvll14pDYqsdXqwlVMHGtWreG9yIgqKtGBLI07J6jMWSG0KJDWwL08A4G16cVOrDMpk+gwbgxNrP+IQgbZQ9lR78avPJIPNEst5v1rUSoAn44G0oBaG3ADFpcAF+TOq627cMg54KEqvNw5fwNcjtB1aRrbkxtHhLfbWeeP02NPYNGyTvDxoNEKzksbNLSJaiC6zuWemrmr/hnZ28X71cF+o9Lxtjp4rzx6NQhtZvJdiNhyGtWOlRuNb/uj29QLrWSy3vYrdtjxNiDVrV8eyxsBXFYdPA/uOW5QG2ltSlPNyDoJ8uGFk2tZta18eYX9D68FNB+PltJfl7eySez5UUVCkhSbxgJenKiUzLAAE+MuefNY6x/c8sLg4CeZC3pQe/cK7YmjkA/guUaXu1r5Vg0EqGepgX9tfUdx5IwI877QfunId2HsMKLPerXxJ88+N/2b04HpciXVFr9Rz1SX7jlks8tP5YP+z+/FJz0/MlrBcLyKZQYeH8UIuvf6ph7dhZtLbstKVIjgUg7UsVFQi2jMUx0cfx1P1LEeVllVSlFfAPeHr9QAYzE3+AO0Cm+HrhKoLvz3BzMz7v1clHQMdo7N5LBmh8WIerC2v1BCQH6z+1P8n0WX8zhy2BL/8wQ/5FAXu/P1euQ7sPmJ7GmLyhaePMmlTxHhYDcQMXeo/v4+7Hvh5+pn0XNw+YjtK36iqXg/1DcWpsadsyibDMGhfz/r0S9a+n1kPzcLYdmPl7ezGLVuy5jIoKNJCgD+Q1hKIFp/yQC6TC2087wmsXoTtiTmhN2asj2n3ZKEbBSPyb3MejAcyk9/Fc/WqehSJVp9ZCXL+aTkN94d1ws9N3uWlYzsPxsNkehOcs94N2SAtKMX4bx2j427UR4UbpKL0pmApEWDlpiv6tkQVnozzISE4HveHVY0Vs7DZZPSIuxebW/0guL4qbYAMU5zsOAjsPIwkv/qI9rH8LekrJdpEAVVP+Me5rv+PR/bCjtRMxAoMAWENvyE/X7eYO8EHy5pMIKxUfEg8itOLse9/+xATFCNrvKj0u9ON/17d6luMj3kcY+rJb/Qtt1RmSMshgu+/3pjX0P/cZcEpasx/n1EBUXiixRPYPmK74DqKhkzg78OGUgx7Ssr4wZtXQSmaseKj8BuClZfSXsLh5w5bzIPmofMwCRI7xHSwedqOCn0Flj2xDLMemoV74++1aVtDHoQC0q1Pb3XKhLvOQEGRVlQaB8XkRuLLuyHrLL9ak5ug0IjbThiUTs7FjL+O3CfC1IBkAMDQKNtHFr4ntC2WpXyJeL+qoFKrqg7jVCN3hiewIHX8sq9KN5C+w1rjWln7YgGYDTaZXKsB1jy2HJ1DWknv37h7bj+ygyJD+xFDr6szF4R7Il7LFzwWDXwFenlezRcNAuWcCx15ga3YtkrOqXnJlj2BPHWeaBXVChdeOI+hUdKjdfOrbXqGdcCUhJckh0tQ06h4XgBWfhtJXpbVZ+YSwhIwZ+ActIgUqR6VMSGyWuwZs8qkTVF5JV6OfRJvxT0tWVrTLKIZQv0se9zyKTmf/L38Udu/Noa3Ho42UZbTgog1rO/RsAc+6fEJ/L38BY9HWmwaLr10CS91eMnmPLka5/0yiIpEGgSH8p5WGQZok8xV39ierMOY/+C+TXwVnfaNxFtxT8taX8z6VtOwp/g4uoS0tjuPgOODojpeoegU2wlMUSlqewUb35e8AFtZzLIATp7nqgYS65stUEDOoWcEvqOrto3JBAC4cQvsDXltPBp4mQ2GmF9kMm+dgRfjqUq7EZNSjBs3uWCwlmnPOlkjiNt4Tu1sk4mWAY0BsZ+wHaeo+XdmLW/mny01OhV7svdg5kMzre4jyS8OgV610MCvHlBW1Qg5s8lETCj5FcVlxVh2apmC3AM4fQFoZtlWytWYtClivOCj88b78c+h5OYc7Li0w660lVTx80uahB44xYZgWPPUGtHtDK8tq3bds2c0lRS5g5TGQH1esauOHxTxePEaSjPgJqStI/a0IXQRd0JJkdk+UgISUdD5H7xWv6rdg0kxucw8BXrWQrfQVM0aCduKYRhsTvsZm1p9b1oyJvKTnNxwHDZ1/tVq6Y3xxiZWysQzvs1ok9eC48z8d8nyPbM9ciVFZuM7/XdRfBNejxSTG/Gp89DrLRuY87//NSlT8XzdwYLt0vjrfZXwMhL96iOj4WjBgNDWm4lJPncd4dqiCFQBCW9b9Q9be8MBgLfOCyV3izWQr8rXM9ED5CVoKASU+1s/eR7MLdP2bS+lvYTC1wvxdOs7DzIix3N9q+nY2ToTugLTXlkxvpGYN2ieyUjvNj+EVIiXhn7a41MAwPQ7k7Qa0+f/dmwY+sKe6jPTcYqq/u0KQ6kIBkUygnv+dgOTB+Lo80eNryNqKWi24WLc4w5S04UGce2FGsVwT6hxVQGS+IVWoNEq/z3BmMjOoCiytuQqQnswHyzQEZcLWy9Czpjhmbly3eIiJPYNvFL/KdztmSTYwNrA5DPysy/wWe6KbmvSA0lwlnSpY3CzDMi6bH0k8MpKrkeOoUfdnfntLOj1kt9Qj7D2+Lbxa/D3sGxozL9Qj4t5DCfb/8m1CxJpACuoXPhGK5gvw0259CZw5gIA4PEIbmT9+xv1wX2hHZAW1AINQxsaU/ETyLc1jfxiAEiP9wQA0xunI7fjKpvS5xM937OvoouuscXbhh5RuHod2HZAcDR6BvKrntT8vU3oNAHsRBb/i7EcvRyA7K7iXjruIbNXo14Y1WaUoryYlhRVVcy82MFyfDHBY1BRYfr7UZHYA6R5OyXzXm/873hi14lVYx+V3MD/PLvh4To9MDvpHXUz60RUfeZOYqK4Px7W11t4XYFrkfGG5ePN/cnYRpC/r3DPgtrBggPU8W+UtlYzqNXMydZLbteQNphzZaU6O7eBZEmXlRIKWwO/MfUewS19Ge4L64D/lN6Uym5bf5LOK+T+IsKA5IYmgYdpfhm7np5F8yBUUiS2nyOnhZMQW7+wmBs48I4fkt7EgPBu6NXnGQTuPsPli/d9Do18AM+ezBBOC1wnhAtluQCAjxuOQRivWlUKwzCI8Jae+8q4vg3VZ8/WHQj/pk0wYonpVBtg2aoOAQdPAl1SsXn4ZnSe3dmYJ7nkzA9nkseiUrOAV2BfAsNmWJVfxD103hma5Pz489ifsx+9E3rjkWaPYGLXibhZcRPPLXsOb3R+Q1aSpuMUVf07LiQO9QLrIbc0FxV6K20Aj5zhAs46YUDThuLrKSB2rWkQ0gDrhqxDbb/aqBdUz2KcJtFJbg+ehO/tCvze7M45rn1hmCJUUuTmdB4iQ9sbL0gCFwuG4UqdRKvWOFXj7Zi5q7nw+zKiKjlF0SZBkUYTFD4d/RB+TZ6k3uB9Mqk9qraosnJ46TyRHjccqYHJdgUksibSvXKd+6+Vr1NJ9ZLB/+Iehg46Y2mNmgSf4FnWJCACuBKdhyN6IOj0Va7BqllQYNIzUcDJdn8a/93IN0ZGxqRXMaqoBHKuGYNqft4a+lpv+Oyp86yqKgPvN8kfwPTOMeJPSir122WvFxj/3a5eOwDApG6T4KnzxJe9vrTcoJxXjafXc4GYnV6o9xgA4MnEwVx62w4Y26FFB0ajT2If43dZL6geEsISsOapNbgn/h5Z6Zv0PtOZjgN39oWzKHpdoiTTUBV+9brFIovfGssC1wpkN0IXCloN39m98feiZVRLhPuHW4yvxP9MJnkw79TgqBHcHYyCIjc1u99sRAVEYc7AOcoS8PICmgo3VDzZ7k/83OQ9/F9kHztyKExWUKSgTZHaPBgPPBnZx6RHmlpifCzr3e8NaQt/nS/61O4osIU8JjdvfpsJoZu6WZdkNk9B4+g7ZE2ke6c3pLVRw4Uq0BL8YmXloZ5vJG502Yw5ye+bLnDU06rIsAgAuBIkA0Ojc5HVezbsiR0jd2BXj4WSQZNdTpzj/gR8kzgB98R1k52UtRIg09HoJXoF8Ef4vnM+vt31bdx88yZS66ZywyQc4w1NcTFXdh5F5VzjSpnumNzoBWwevhkzUyZVraPm+EU8nvy2WecuwevoWfjxGybb0M5J0LV8rrRT7pRCCttfmnz/pTeBgiJg52FFabkiqj5zMYHegSgul24oO6zVMAxtOVT8AmXonq8gpkj0r49E//rSK1qQHj/H5uozlYIiV2jYaM3alt+hnL0NH51IdagMJp/QvHozr0C8PQ8Afc41WfsI9rAcysHkyVGsGk6vFx8qgOH+hEqK6vrUwc42mZbjAJ2zbAQufOxsqD6zxS6ZN4EzF4CYSMF8AAD0eq6UpOi/qtI02Wz4HNdMg17+r6q2VzDS7xqOqOxyxPpG4sFDlu1d5DJtU2Tlt2seeJzOMo6t5qnz5M4VQ0lDQiz3EFdufRBUC+aHJ7/Ioo2Zl84Tnet3uBN82RmUCOCf0563KgHcKdk/n839l/+btLG6z+K3lm9639gxcgf2nt6CTUdXYd4Vy/ZmajS0xrH/gEBP+6YScDEUFLmY9UPXY9zKcZjcc7LkuoIncPME7klIompMEH+co+YJ3MB2tnDA78KeMULcCcMw8GGUBURpQS2wregQRkb34yXIW4FlJb9LqUDB0LvpxRjLudZkf0O7j1gfUkAkD3cFNbN803BTcQK7gyhrDXsLSwQbg+usnfe5edxYTY3klaIJMSmxY1n4ePjg2XqDAAB9a9+Nv/M2W09A6DNdzQdzs2oIBKs/XaE2iYdPc93sGcb0Jivn8Nt7mXDQdYblHSdPoZIZsd6YSph9hHb12qGdb2M8eKsxjpb+h9H3TDBZrrSkyGK7ahQQARQUuZzUuqnY8vQW5QnUDuH+BLQTurl4enI9HBiGa2DYsRX3vq2NFFUU6V3Viy1ARs8bOVrUaoR/C/erkparWd9qOk7fvIBmtfjVodYv8uZP8b4SJVRxvtF4I2644DLTNgZWmD3p9wkzrSp01Uur3fnavNf4u3or7mlsLzqMtfk7q5YfPAmEhwAAnq87GDuLj+ABkWk4GIYxjr4t2Fnijijv2sgpv1PSItAexaIUh/eyXWAzq0ERcyUfuL7Psvr96BkwnlXtSCTbFJkf2bwCrkSrjvwG49ZpfEaxLPTHzhhfcr3P1O9FVkX4eMf4RmL/XXOBVNNeZHX8BWZUkHHItGrS4CwUFNUAJ9v9iWM3zqJ7aDvLhanJXLsHw5QjXiKnhPnTmwN567xQ1HkDGDAW3fWVymg4Bn46H4c0xNWaj87bLCCCaSNHoXbCZm8+VLsL+oR1RIcg4Ub01i6DstoUCaTU2D+O+8ed6VDsaWhtyyCNNg/PoMbN9U57kffjnwMAMBvuMl1+rQAA8G1jG+bxszLR8OMRvZDs34DrLCE2ZcwdDMOYlLxJft68fCCCBc5ajkfFFFU9TCm6eQr1rtx1GAiT3xNPSJyP+PQaDqNn4VNa9SDAzSmoXlBk/J4u5nKN3oUOt1AV4vGzQOM4DG89HBtPrUVyeW28c24GAIARm6/s9m3uHIkKh87blt+7+6GgqAawaCPE//H4+gCxAnPWKCpOVu8HEuhZS7W0AG5Cys8TlLeVUJNTnrT4PVD2H5dc3VPnieUpXynalVqfRg87giIxKgTyqoyhI3PqFUk2ZGVUXfEBHS2qpXltykZF98dHWZkYFN1TeFsr37jshtZi9Cz3nfGPeUWldHurolJuLB+z3rgrWnyFVfnb8b+6A23Pi70YoIFfXbwU8wRCPAPvPODZ2C5KjjvjZCFAqFTd7IQx9Ng7fBreXdtibq8fcXHXv8agSLQDwcksrhTv0hXo0pqqkm1XRUERUQ8LoF0Lbtbxi7maVsER62y9YVlbX7SLrlA6ReKNWR1SECmQpq2fXZWnYZHB92xOW2ZjY8nvwcoxiPIJR1HnDfAMCrL+Gxbs0s1frCAoOnWeq+5rEm+xyOpnKioBDpzkpjbi6V27I3or6dHJsgofDM3SAKoexlQ6v/8v5f/w68FfLcdKKlE2pQ3/e5rcaJzwSrzzgNlxSNF+3AV1ySfCzK8HAhPMCvLz4Ua2rt7VzoTHvNGuUhMbjAQADIm0fVJfcfb3PlMlKFJ4w7IgczRme0u3vHSe0kGNQF7MR7S2RvS4FhQri5DNu7SrHWQ7ImrPl+5pbO7n/j8j79U8eWMl2dhGqFkt6QEidRW8xuHVrJE1QCVFNUtsFNcOQUmvFR2juDpclZ9NckPTMUvcmJqlNM4id2wae77rUXUH4N7QuxAvNIu9Uq5y0c6+6tTdSZcUKWc8H2+WiS+Dneet1t+bXm9aFZdzjZvTr3kiEGRWtS82D5v5R7gp0F6HV20pN/hmrhUg7PJVoEmA1cb2wpkwT4yx/M6ESsl4L4UmX65OqKTISUxnD9ZIwxigU2vR3mmmzE58XxkDy9najdYmLnJzU4Hci1+rAG7OqQHh3RyYG3XVtmFqCiGN/GLUndRXoPrH1vutK5x59by5MXzuD5NXFSTd3F35jc2hbYoMBD5Asr9llZrNiQgRKn0znCS3b3NtaU6c49qF8XqTGV1QYVBJc2XlwlWlLAscvTP1xxkrEy8b1uUfgkOnLNcxa+tm8p2J/FCq+zApVFLkYMufWI5xK8fhp/4/aZMB82lAPGX25uKf900bcRcGhSOuBnr4K9rOhCvcmZxsZcrX+PPqP3jSASOL20rqBrek+efIryhGfV+BRvs2pKM6FaqtMuJH454Dz2JMvUdUyJApudUPO1MzsSzvX9nngi1Vfmp+J9ZGLLeJwHFpF9QMvzX9yDhRrqDz2UBUba4ERe7uT563fM/Q4HjfcdMSMaE0RSdrVfj5KyrER6XmjzcmVkIFcHneewzw5t3irxdarmdW+m6Md3KuAWcvcSVjgabXb6dNRaQRCoocrE9iH5xKFIjQnaVBXS6YiQ5XnkZYEACW6wHiYfsPIiUgEa/FDkE9gektZAtUtzealuTehCK9a+P5eg9rmgd/nS9u6G+hV1gHq+v1De8iKz137MLbLTQVRZ03qN4j0hZ1fepY7U1mTirYMq3yFFuX977QDVUoXZOt7fiu84X390iEcI84o3OXgCt5XBMBmaO0IzePawfJZzh+5lWEej0XBB05w/27lh834bFguraNUN7BMNdksUjj9vLbpt/DzTLxhvfnLnFBk7XASaAU1Vh9ZpgS5vh/d+a65JUAqlRd7qooKKruvL2AVAVdKM2LSMNDgRaJQICVUh/+L8Rs848bjbU9DwYNY+SXcBFZ5N6wLqUtR3b5NSTXsrXqwnUpuVlrGRApITW8gaygmH9zPnQK0d7hyC6/hi4hrWWla9cNU6pqyJobt4Sriqwxn7z0Ym7VvHV8tyuAf/dVvebP32budJasXR9vtwBbCg9iaNQDwOkLQG2RKmjzCXBvlXET2AqRM2XMnqMAJNqX6fXceSDUHqqaqt7lYEQ9DMMNoOZtOtMzGvKLsnmXwcYNuIEgE+Ps37f5U1yw5fxb7sQVGk7LFeIVqGpA5Aqf/YUYbmb0B8KER412Jke1z7AlIJH7nZztsBh5ndYiwlt8xGn+MrVGo9fExVzZs83bK8m/AZ6Ofogbx+hSLlcKJcSGAUrtwq8OvFUO7D1qsUoj3xjU0vkh2b+Bc/LkRFRSRITZe7EO8AfSWnLpnBKos7c1L/zqgEax3IBu/CenlMaWT1JElFbBSRN/FYJkO70S+3/oFpKKlgGJWmdF9S7N3owXytnb6BVqvbpTyffvo/OWnLDYW+eFgs7rwQCqjUZf44i2UXKSrSKlTzzH2y1AJfR2TWDtqigoIvbhX1vN2/2IBVZBAUBifWPxrax9sOZvmAkNkpcW0dRdQc2woOnHiPdTsdu9jXSMTngewGrgfIclOFx6RnhKHx7V2v4ICPaUV5JbHduj1BSeOs9qGzxU189FVCUQhNSL4CZwjKoDhIdxRc3W2huZJyd3XcMGDGvykrivwRHdtc5CtRXlE44oH+lOFdW9WzVRxh07QqiN2hQRYfyLptD1M6E+N6WHpwfg6229nU+UWZsgWyd3ZMwywTDcyNnOpNMBCQoGvSTExWlVldotpI0m+yWmKEA2RSVFRDm5P6bGDYBG9bm68sISIDxEeht/X64XiXE/Zk8wcgaTVFP7FoIj+CpBFyFioNWTuSs0eO8akop/Wk5Doh89bBDXQSVFRJia10yG4UqUfLyBiDD586jxt7f22qBdC2X5k8PbC2DVmcW99Z2RqrVEcVnN5gpBEQDcE9oWMb6RWmejRgvgDa7rp3OBmRc0RiVFRIQTLpppLbkBxOSMKcKI/JvP0VVqevue6g+2nYdZOUvwZv2nVcoQIYTYp5aHH9akTAXDMPD3oKCISoqIdry9xHuNmQwEKVJSJLRtHfExVIwiwoDOCtoz2Nl9ukVAAqYkvIRw7xC70iEqaGhlqggn8tZ5Sa/kAFSFS/h6hLWX7LFYU1BQRIQ565rJMEC75sq2rRNq+V5ifXnbKpiuxKW6/UvOji0s5s5UK/eEtFUzN8RG0xuno6FvPXyTMEHrrBBCeKj6jAgzn0jWkfx8uQCHP6y++bQe1kqO+OS0V1Ja4GNrWyhH6pACbNxt82Zn2i9CaeVNhHq5UIBXA/2v7kD8r+5ArbMBgEa4IISPgiIizNuLq2LQ6ZwTDCQ3BK7uqXpdy48rmfExVC/IjGToCm+Vt85Lsyob4ppoZBpCqrjQoy9xObFR3CCNzsAw3HhHxn1HAvH1gLoRVcsNjFVfSiMgJ90GImS0bxKj5Lir1dA8JFCddAghxM24RFD07bffokGDBvD19UX79u2xc+dO0XUzMzPBMIzJn68vtZivFhje6ehvNpmkpycXJDWM4f4NCMdEchqQ+jvpfEluaPme3AaufgryaE8QxlfLjSfyJDajwlVCqmgeFP3222946aWXMHHiROzduxctW7ZEr169cOXKFdFtgoKCkJ2dbfw7f97OCUeJa9BJXJ7rR3OlVwZCwQ3DCAcHDAO0asJtHxstnRcviZplX4mGzlEiUy2kyByjiAHQrBFXjSmbWrc3gXSipaeOIIQQd6d5UPTFF19g1KhRGD58OJo2bYrp06fD398fs2bNEt2GYRhERUUZ/yIjafCvaqFJPBeMyO1BFhTAlca0STZ7nzcxrYcHFwx1SOGmImkYI93zzMuT28aaiNpcyZWYxiKzwdtSNRUeyo3lZD6FiqdIwObAmAgxUQJvasDaMXc0P1+geYJ2+yeEOJymQVF5eTn27NmDHj16GN/T6XTo0aMHtm3bJrpdSUkJ4uLiEBsbi379+uHIkSOi65aVlaGoqMjkj7ioAH8uCKhrQ3uaiDAgsJbpe/wmQz5eXFBhS4lLYpx0FRsD62mqOQ4M//NEhQOtkqxkisfLU732QfZ8HDWrK+tHA3e3sX3+PDUkxNpYcufaBobfg7aByWgTKPEAQEgNomlQdO3aNVRWVlqU9ERGRiInJ0dwm6SkJMyaNQuLFy/Gr7/+Cr1ej44dO+LixYuC62dkZCA4ONj4FxtL8+y4NDWCCf4gi00b2Z+emMjaQHQdrrRKbNygZglVQYFhnXDe+EohgdJVdfyoKKmBeJsf/qELC+YCzBaJEmnL5IzB/sJDZRwLcL0hmzSwb19KhpxgGMu2bm7sz+afYmebn+DBOHH4DUJcnObVZ7ZKS0vDkCFD0KpVK3Tt2hV//fUX6tSpgxkzZgiun56ejsLCQuPfhQsXnJxj4nT8oEhuo+Em8TbuhOFuko3juNIqsSq58BCgbTMuOEltyr2XHA/E1eUCoqQGwtuF8kpClHSWY8DlT6cTr8qzOUGlm8rctm4doGMrriRIipcXV2qkRFgwEFXb9u0YhvuenTmGl4PRyNaEmNJ0nKLw8HB4eHggNzfX5P3c3FxERclrw+Dl5YXWrVvj9OnTgst9fHzg4+PkGdWJtszb4IgJDwWu5XMlPRFhwPGz6uxfaFoSfnWPTgc0qCu8bfsW3PoKR6zm7bTqn9F1uFKtzXtVSU42H2+grByIiQROnJO/nfm4WIn1gaJSyyoz8wE+5agdorxdkOF71TFApbIkCFEsug6QfVXrXFR7mpYUeXt7IzU1FevWrTO+p9frsW7dOqSlpclKo7KyEocOHUJ0tMKnRlL9BAcCLRsD7VOsr9e0IReEKOnKbl7NE39nLi0fby5Npbw8LQMiQ6lGoL/l+nz2jvTNz4MaOqQAXdsqC1746kZwJXnm31OAxPFQm1jvSKmeiHKpVdVJqidXGlG/GtN8ROuXXnoJQ4cORdu2bdGuXTt8+eWXKC0txfDhwwEAQ4YMQb169ZCRkQEAmDRpEjp06ICEhAQUFBRg8uTJOH/+PEaOHKnlxyCuJkTGNBYMA/jaWIqY1AC4XmjZRT08hKv68fSwvf2N1PrRdbhqQFuCgAA72r4wDNC5NfDvPtP3tGBtv6FBXClf9lWgoLjqfW8voPy2+Gu186L02Ph6A7fKuX/rdPY3Hg8JND0OQjq1Am5XADl5QFa2ffsjTkZjjzuD5kHRo48+iqtXr+Kdd95BTk4OWrVqhZUrVxobX2dlZUHHi5Dz8/MxatQo5OTkIDQ0FKmpqdi6dSuaNm2q1Ucg1Y21e1xUuPgYREpLWBLrA0fO3Nm3wM4Zhiv9ksIAaJ0M5BUIj8UUVZu7GUbWBnLzrKflKk+lUvFGRBhQVGIaDJjnvXED4PAp7t/2jPotFvwoPVZJ8VzAdv6yZfuo0CAgX6KnrHng2iJRuorU07NqIFQKityLq/wmqznNgyIAGDNmDMaMGSO4bMOGDSavp0yZgilTpjghV6TGCqglvY6aVBtBmuHGaAoSyX9iHDe+UnCAZVBkfhPmBwBibXBSm3LtsEpv2pbNQH+g+IbMlWWUwlTqpddp1QS4el28LZesrNzJi5oP7P6+wiOfy+n67+oNviPCgCvXxZd7egAVGjbOMrR54wsOAApLLNeNCueW2dI2Tm2xUUBBkQ2/HaIEhZ6EGHRsBbRrrl4bEVej03HBj/kTZ+c21tuz2PuEaj6OVOtkbp9yyKmZ0ksFRSx3Q0uob18gYSxlsiEqamRlCBBrn40V2EfHlvL3C3A38gb1bP/McSq0z4yvxwV7YgE6IF7iymd+7qglMQ6IERgPrZnIA4Cfj3R+5Q46q5SXJ9BGhRqRZg4cpoRPKNh3AxQUEWLg5Wk655hhSpFYB4/mzC8VsKftjpJtG90Z4Zu/reCcciL7k7NLH29uehOG4UYUN3RtF0pPznuSHND2om4dZXnhtz2LrwfUCRVfl8+8sX0tP24YAmvM8xdfjwtwYu+MA1dbZpuluLrcw4Fc5u3r7mqufLgEc62SuAb7SlgLUjx0wqeJWBW4nFPKywuoJ3N2hUgFQ0K4k4T66s3F6GQUFBEiJr4eN8aQo6eW8PDgeqx1SHF+g2ZGxiXA24trhxIcIPDkLjO/oUHc+ENiAWZ4iMhQCjLSN/9+zCfTFates1YCZtHoWUnAGWtaSuPlKT2YaPME7oYpFVTYUnpXP5qrPpQ7kKmcISH4Y1/V51VJ6nTyRzDnn+tigY8hL6G8jhNdUuWlH12Ha1iuBrGeh9F1eC9Y+aVstrZt41dhqxVwEkEUFBEihmG4J3RnBCq+PvaPTSS3JIDP2jxwLRK5RsqGY9AyCWhtx5QQYsexlh9XbcFfbujGHyJjzClfH650wiCpgemTuFi7ldZNgDoCT7N+PtyTrr2UnDa1Q7jhB8yHMZCTlvnxNXy3DMMFnLYEUuZp8Y9nnVAuGEiM40r++NXN8eZttmQeBG8v678zQ0Dn5Wnb71FsnsA6odJt0eKiuYA7qBZXUsgXHMBVZyodGFVOOzh+1Su/52mDuvYN+0GscomG1oQQBWr5cY2cUxpz/7ZlXq5GMUBBiUgR952bjnlpieFm5OfDdeu2pycXn1C7kzbJXGNwudUR/r5cKZRhnKcm8VyD2VtlQJjI8AwB/txYVRvNGgO3bWYZQPDbn1irSqkdwvX+A6DeDL0CaZknHSQQPNoSBEkNWyDUxsk8UBDKmNB2Bvz8GapizVc3nHOeHlypjz2Bnfm+KyUaeet0XKmMUMlM3QjL6ky5tbaeHtyxuyA8lRUA7gGkli9w5s4MDCbV2woGd01tCuw5an0dhrH+fdlCzfkOnYyCIkLcVfME4Hal9KCOQmKigBiF+72rOXfx1OmAyDCg5AZ3EWwYW9X1XVY6zYAr+dyI1+b8fLlGwrZoaPaB7mrGlRIpncQ1LAi4XsQdZ/MqOYNGsUDutarXzROAjbuV7c8mMgIuuSUqjWKAvEL5YznZUlJjLfDw8wHqRcgfykKs1EdKVDiQc+c78vLkAg5A/HeTEAtczbdtYmoDa4fGMK2Pj7e8Y8iv2jYPBvnb+/kAN8vE0+nc2rk9FRvHmVZ3uhkKighxV74+gCMeyEIkxkRimKqLcr1IbpLUoFq2d8339wMaOHCCVZ0O8FbQQsDw2ZoncoGC+VN5XDTw30WuSikmUnrMJwN7HsLN76FqlNI1iedKtepGcEGRyf5sKPHhMx9gNDEOOHSqqodg6ybAvuNVy/nVlLYcHzljbRnzUJ8boyuwlmlwUScM0LNcns9d5kr5AO6cliqhtPX4JzWQ19uOn76nR9XciEKjwrdM4oLOsGDgYi53TprTOXCuvvh6wNlLpu9FhJm1s3I/FBQRQjgdW3Hjttgycrb5vG7uKCyYG6XcnFg1RUwkt41UFYFY41wlDCO0t24CXMi1LBUzMAzKmNJYOs3I2vJ7QbFs1dxbQtVJbZsBN29ZBtQhgVxJBcNwN3C5N+i6dcRL5wAu0KkdAmRd5oJr8/GQ+KVPOp3w4KcMUxWoyJ0Pr2USVyUrNFSAhw4m0athHKb2LaRHzg8OBArvDEDauknVeWctkOIf69go4aBI8BQUG5kdtgWm9aMtg6JqgIIiQgjHy9POec94F1vDE647aJ7ADVx5SGbVn6EBvpi6dYCSm1XttQJrAcWlXA87k3RklmKFBnHd5AGu7VAzK43PxdrASBEqCWrWqGqkdRZctUhCrHC7nlp+4sfEUOokFRA1T+D2l1hfulTFw4NrLG0Y4iA2imsz0yiWGyLAUZ0jQgIBmAVYjWK5KuSwYNMxs1Iacw8YsqrKeP8Wah+mmMS+/X2BG7eq1m3SQL2Jsd0UBUWEEPXZUlWgNbFxk+RqUBc4fLpq4t5Esx5JrZtwN0tDUFA/+k5pg8wSuXoR9uVPjuBAy5Gcw0OrSocMXc0dOdVEWHBVqZKtAvy5yYe1INQmzkDuZ4kI46aqUavzgnH/Nq4bWZsL4C9dUTcfboSCIkII4ReU2HpTrh0CpLUUL2VjGNNSErnjXgX4c+20rM17ZyiFMh9A0VZx0Vz+zatCG8fdGeBTxXYpkbW5AMy85AzQbuJhrUWFc1VstlRdy2LD8dTxhnCwxtenWo+VREERIUQdNfR+BkB5Dzdr2iRX9fIT07IxUHpLWQ9EPp1OvMRD7Ya6TeK5z1UtAyBG5N9SmzGO6bEllQWhhwFr7YoYxvoYSQ4YTN7ZKCgihBBXxO/lJ8bDw/r8Yq6qWgZEcI8HA5M88qIYY/BtJbKR6oVYDaYvoRGtCSHqsHdEbkKqE6kAqdGdHoRqz63IL9mzJfi0Z/JdL0+uB6KSUfVdDJUUEULU4ePNTQ0iNKYKIcRUTBTXmN0RDxNBtYCiUuGSG/NAqU1TbgBSQw9HoUAqKAAoKhEvCZLqkelGKCgihKjH3ccsIsQetlYLSo1fpFSLRG4aH/4UN7FRXMN9ftslFlx7tEDeIJr1o7hxu4xd9cENl3C90LJxvIeOm8dNcDJn90RBESGEEFKdeHpaBjBiA36a8/LipvI5exHIujM/m5encClRalMg9zo3bEQ1QUERIYQQUiNZ7WomvbmfLzdOVzVCDa0JIYSQmsQwppYbT9zqKFRSRAgh1aSRKNGYKww14CNjzKw2ycC1AvsH/ayGKCgihBAvT25UakdOY0Fqhlp+wO0K6QmD1dYyiZsMOKG+9Lq+PtanJwHUndDYjVBQRAghgGNGpSY1T2pT7r/OLjUKCbwzYa1K6kUAV/OFp2OpxigoIoQQQtTiClVoavC8MyBjDUNlxYQQQgghoKCIEEIIIQQABUWEEEIIIQAoKCKEEEIIAUBBESGEEEIIAAqKCCGEEEIAUFBECCGEEAKAgiJCCCGEEAAUFBFCCCGEAKCgiBBCCCEEAAVFhBBCCCEAKCgihBBCCAFAQREhhBBCCAAKigghhBBCAACeWmfA2ViWBQAUFRVpnBNCCCGEyGW4bxvu445Q44Ki4uJiAEBsbKzGOSGEEEKIrYqLixEcHOyQtBnWkSGXC9Lr9bh8+TICAwPBMIyqaRcVFSE2NhYXLlxAUFCQqmlXZ3TclKHjphwdO2XouClDx005/rELDAxEcXEx6tatC53OMa1/alxJkU6nQ0xMjEP3ERQURCe+AnTclKHjphwdO2XouClDx005w7FzVAmRATW0JoQQQggBBUWEEEIIIQAoKFKVj48PJk6cCB8fH62z4lbouClDx005OnbK0HFTho6bcs4+djWuoTUhhBBCiBAqKSKEEEIIAQVFhBBCCCEAKCgihBBCCAFAQREhhBBCCAAKilTz7bffokGDBvD19UX79u2xc+dOrbOkqXfffRcMw5j8NWnSxLj81q1bGD16NGrXro2AgAAMGjQIubm5JmlkZWXhgQcegL+/PyIiIjBhwgRUVFQ4+6M41KZNm9C3b1/UrVsXDMNg0aJFJstZlsU777yD6Oho+Pn5oUePHjh16pTJOtevX8eTTz6JoKAghISEYMSIESgpKTFZ5+DBg7j77rvh6+uL2NhYfPrpp47+aA4ndeyGDRtmcQ727t3bZJ2aduwyMjJw1113ITAwEBEREejfvz9OnDhhso5av80NGzagTZs28PHxQUJCAjIzMx398RxKzrHr1q2bxTn37LPPmqxT047dtGnTkJKSYhx8MS0tDStWrDAud7nzjSV2mz9/Puvt7c3OmjWLPXLkCDtq1Cg2JCSEzc3N1Tprmpk4cSLbrFkzNjs72/h39epV4/Jnn32WjY2NZdetW8fu3r2b7dChA9uxY0fj8oqKCrZ58+Zsjx492H379rHLly9nw8PD2fT0dC0+jsMsX76cffPNN9m//vqLBcAuXLjQZPnHH3/MBgcHs4sWLWIPHDjAPvTQQ2x8fDx78+ZN4zq9e/dmW7ZsyW7fvp3dvHkzm5CQwD7++OPG5YWFhWxkZCT75JNPsocPH2bnzZvH+vn5sTNmzHDWx3QIqWM3dOhQtnfv3ibn4PXr103WqWnHrlevXuzs2bPZw4cPs/v372fvv/9+tn79+mxJSYlxHTV+m//99x/r7+/PvvTSS+zRo0fZb775hvXw8GBXrlzp1M+rJjnHrmvXruyoUaNMzrnCwkLj8pp47JYsWcIuW7aMPXnyJHvixAn2jTfeYL28vNjDhw+zLOt65xsFRSpo164dO3r0aOPryspKtm7dumxGRoaGudLWxIkT2ZYtWwouKygoYL28vNgFCxYY3zt27BgLgN22bRvLstwNT6fTsTk5OcZ1pk2bxgYFBbFlZWUOzbtWzG/ser2ejYqKYidPnmx8r6CggPXx8WHnzZvHsizLHj16lAXA7tq1y7jOihUrWIZh2EuXLrEsy7LfffcdGxoaanLcXnvtNTYpKcnBn8h5xIKifv36iW5Dx45lr1y5wgJgN27cyLKser/NV199lW3WrJnJvh599FG2V69ejv5ITmN+7FiWC4peeOEF0W3o2HFCQ0PZH3/80SXPN6o+s1N5eTn27NmDHj16GN/T6XTo0aMHtm3bpmHOtHfq1CnUrVsXDRs2xJNPPomsrCwAwJ49e3D79m2TY9akSRPUr1/feMy2bduGFi1aIDIy0rhOr169UFRUhCNHjjj3g2jk7NmzyMnJMTlOwcHBaN++vclxCgkJQdu2bY3r9OjRAzqdDjt27DCu06VLF3h7exvX6dWrF06cOIH8/HwnfRptbNiwAREREUhKSsJzzz2HvLw84zI6dkBhYSEAICwsDIB6v81t27aZpGFYpzpdE82PncGcOXMQHh6O5s2bIz09HTdu3DAuq+nHrrKyEvPnz0dpaSnS0tJc8nyrcRPCqu3atWuorKw0+cIAIDIyEsePH9coV9pr3749MjMzkZSUhOzsbLz33nu4++67cfjwYeTk5MDb2xshISEm20RGRiInJwcAkJOTI3hMDctqAsPnFDoO/OMUERFhstzT0xNhYWEm68THx1ukYVgWGhrqkPxrrXfv3hg4cCDi4+Nx5swZvPHGG+jTpw+2bdsGDw+PGn/s9Ho9xo8fj06dOqF58+YAoNpvU2ydoqIi3Lx5E35+fo74SE4jdOwA4IknnkBcXBzq1q2LgwcP4rXXXsOJEyfw119/Aai5x+7QoUNIS0vDrVu3EBAQgIULF6Jp06bYv3+/y51vFBQRh+jTp4/x3ykpKWjfvj3i4uLw+++/u+WPmrifxx57zPjvFi1aICUlBY0aNcKGDRvQvXt3DXPmGkaPHo3Dhw/j33//1Torbkfs2D3zzDPGf7do0QLR0dHo3r07zpw5g0aNGjk7my4jKSkJ+/fvR2FhIf744w8MHToUGzdu1Dpbgqj6zE7h4eHw8PCwaC2fm5uLqKgojXLlekJCQtC4cWOcPn0aUVFRKC8vR0FBgck6/GMWFRUleEwNy2oCw+e0dm5FRUXhypUrJssrKipw/fp1OpZmGjZsiPDwcJw+fRpAzT52Y8aMwdKlS7F+/XrExMQY31frtym2TlBQkNs/FIkdOyHt27cHAJNzriYeO29vbyQkJCA1NRUZGRlo2bIlvvrqK5c83ygospO3tzdSU1Oxbt0643t6vR7r1q1DWlqahjlzLSUlJThz5gyio6ORmpoKLy8vk2N24sQJZGVlGY9ZWloaDh06ZHLTWrNmDYKCgtC0aVOn518L8fHxiIqKMjlORUVF2LFjh8lxKigowJ49e4zr/PPPP9Dr9cYLclpaGjZt2oTbt28b11mzZg2SkpLcuvrHVhcvXkReXh6io6MB1Mxjx7IsxowZg4ULF+Kff/6xqBpU67eZlpZmkoZhHXe+JkodOyH79+8HAJNzriYeO3N6vR5lZWWueb7Z3m6cmJs/fz7r4+PDZmZmskePHmWfeeYZNiQkxKS1fE3z8ssvsxs2bGDPnj3Lbtmyhe3RowcbHh7OXrlyhWVZrhtm/fr12X/++YfdvXs3m5aWxqalpRm3N3TDvO+++9j9+/ezK1euZOvUqVPtuuQXFxez+/btY/ft28cCYL/44gt237597Pnz51mW5brkh4SEsIsXL2YPHjzI9uvXT7BLfuvWrdkdO3aw//77L5uYmGjSrbygoICNjIxkn3rqKfbw4cPs/PnzWX9/f7ftVm5g7dgVFxezr7zyCrtt2zb27Nmz7Nq1a9k2bdqwiYmJ7K1bt4xp1LRj99xzz7HBwcHshg0bTLqN37hxw7iOGr9NQxfpCRMmsMeOHWO//fZbt+5WzrLSx+706dPspEmT2N27d7Nnz55lFy9ezDZs2JDt0qWLMY2aeOxef/11duPGjezZs2fZgwcPsq+//jrLMAy7evVqlmVd73yjoEgl33zzDVu/fn3W29ubbdeuHbt9+3ats6SpRx99lI2Ojma9vb3ZevXqsY8++ih7+vRp4/KbN2+yzz//PBsaGsr6+/uzAwYMYLOzs03SOHfuHNunTx/Wz8+PDQ8PZ19++WX29u3bzv4oDrV+/XoWgMXf0KFDWZbluuW//fbbbGRkJOvj48N2796dPXHihEkaeXl57OOPP84GBASwQUFB7PDhw9ni4mKTdQ4cOMB27tyZ9fHxYevVq8d+/PHHzvqIDmPt2N24cYO977772Dp16rBeXl5sXFwcO2rUKIsHlZp27ISOFwB29uzZxnXU+m2uX7+ebdWqFevt7c02bNjQZB/uSOrYZWVlsV26dGHDwsJYHx8fNiEhgZ0wYYLJOEUsW/OO3dNPP83GxcWx3t7ebJ06ddju3bsbAyKWdb3zjWFZlrW9fIkQQgghpHqhNkWEEEIIIaCgiBBCCCEEAAVFhBBCCCEAKCgihBBCCAFAQREhhBBCCAAKigghhBBCAFBQRAghhBACgIIiQgghhBAAFBQRQlzI1atX8dxzz6F+/frw8fFBVFQUevXqhS1btgAAGIbBokWLtM0kIaTa8tQ6A4QQYjBo0CCUl5fjp59+QsOGDZGbm4t169YhLy9P66wRQmoAKikihLiEgoICbN68GZ988gnuuecexMXFoV27dkhPT8dDDz2EBg0aAAAGDBgAhmGMrwFg8eLFaNOmDXx9fdGwYUO89957qKioMC5nGAbTpk1Dnz594Ofnh4YNG+KPP/4wLi8vL8eYMWMQHR0NX19fxMXFISMjw1kfnRDiIigoIoS4hICAAAQEBGDRokUoKyuzWL5r1y4AwOzZs5GdnW18vXnzZgwZMgQvvPACjh49ihkzZiAzMxMffvihyfZvv/02Bg0ahAMHDuDJJ5/EY489hmPHjgEAvv76ayxZsgS///47Tpw4gTlz5pgEXYSQmoEmhCWEuIw///wTo0aNws2bN9GmTRt07doVjz32GFJSUgBwJT4LFy5E//79jdv06NED3bt3R3p6uvG9X3/9Fa+++iouX75s3O7ZZ5/FtGnTjOt06NABbdq0wXfffYdx48bhyJEjWLt2LRiGcc6HJYS4HCopIoS4jEGDBuHy5ctYsmQJevfujQ0bNqBNmzbIzMwU3ebAgQOYNGmSsaQpICAAo0aNQnZ2Nm7cuGFcLy0tzWS7tLQ0Y0nRsGHDsH//fiQlJWHcuHFYvXq1Qz4fIcS1UVBECHEpvr6+6NmzJ95++21s3boVw4YNw8SJE0XXLykpwXvvvYf9+/cb/w4dOoRTp07B19dX1j7btGmDs2fP4v3338fNmzfxyCOPYPDgwWp9JEKIm6CgiBDi0po2bYrS0lIAgJeXFyorK02Wt2nTBidOnEBCQoLFn05XdYnbvn27yXbbt29HcnKy8XVQUBAeffRR/PDDD/jtt9/w559/4vr16w78ZIQQV0Nd8gkhLiEvLw8PP/wwnn76aaSkpCAwMBC7d+/Gp59+in79+gEAGjRogHXr1qFTp07w8fFBaGgo3nnnHTz44IOoX78+Bg8eDJ1OhwMHDuDw4cP44IMPjOkvWLAAbdu2RefOnTFnzhzs3LkTM2fOBAB88cUXiI6ORuvWraHT6bBgwQJERUUhJCREi0NBCNEKSwghLuDWrVvs66+/zrZp04YNDg5m/f392aSkJPatt95ib9y4wbIsyy5ZsoRNSEhgPT092bi4OOO2K1euZDt27Mj6+fmxQUFBbLt27djvv//euBwA++2337I9e/ZkfXx82AYNGrC//fabcfn333/PtmrViq1VqxYbFBTEdu/end27d6/TPjshxDVQ7zNCSLUn1GuNEELMUZsiQgghhBBQUEQIIYQQAoAaWhNCagBqJUAIkYNKigghhBBCQEERIYQQQggACooIIYQQQgBQUEQIIYQQAoCCIkIIIYQQABQUEUIIIYQAoKCIEEIIIQQABUWEEEIIIQAoKCKEEEIIAQD8P1/pK/zqmzn2AAAAAElFTkSuQmCC\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "epochs = range(1, len(loss_arr) + 1)\n",
        "\n",
        "\n",
        "plt.plot(epochs, loss_arr, 'b', label='Training Loss', color='pink')\n",
        "plt.plot(epochs, loss_val_arr[0:len(loss_arr)], 'b', label='Val Loss', color='green')\n",
        "plt.title('Loss')\n",
        "plt.xlabel('Steps')\n",
        "plt.ylabel('Loss')\n",
        "plt.legend()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 563
        },
        "id": "upRF9YF9uakc",
        "outputId": "b1c06d38-eecd-483c-c581-3780ffc0904a"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "<ipython-input-45-e0ae6e1b77d8>:2: UserWarning: color is redundantly defined by the 'color' keyword argument and the fmt string \"b\" (-> color=(0.0, 0.0, 1.0, 1)). The keyword argument will take precedence.\n",
            "  plt.plot(epochs, loss_arr_by_epoch, 'b', label='Training Loss', color='pink')\n",
            "<ipython-input-45-e0ae6e1b77d8>:3: UserWarning: color is redundantly defined by the 'color' keyword argument and the fmt string \"b\" (-> color=(0.0, 0.0, 1.0, 1)). The keyword argument will take precedence.\n",
            "  plt.plot(epochs, loss_by_epochs, 'b', label='Val Loss', color='green')\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "epochs = range(1, args.epochs + 1)\n",
        "plt.plot(epochs, loss_arr_by_epoch, 'b', label='Training Loss', color='pink')\n",
        "plt.plot(epochs, loss_by_epochs, 'b', label='Val Loss', color='green')\n",
        "plt.title('Loss')\n",
        "plt.xlabel('Epochs')\n",
        "plt.ylabel('Loss')\n",
        "plt.legend()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JEgct23xcHYA"
      },
      "source": [
        "# Classification report\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2B7CnDExcG5G",
        "outputId": "d5791088-c532-45c8-a112-44418caff0c4"
      },
      "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 seqeval\n",
            "  Downloading seqeval-1.2.2.tar.gz (43 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m43.6/43.6 kB\u001b[0m \u001b[31m1.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25h  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "Requirement already satisfied: numpy>=1.14.0 in /usr/local/lib/python3.10/dist-packages (from seqeval) (1.22.4)\n",
            "Requirement already satisfied: scikit-learn>=0.21.3 in /usr/local/lib/python3.10/dist-packages (from seqeval) (1.2.2)\n",
            "Requirement already satisfied: scipy>=1.3.2 in /usr/local/lib/python3.10/dist-packages (from scikit-learn>=0.21.3->seqeval) (1.10.1)\n",
            "Requirement already satisfied: joblib>=1.1.1 in /usr/local/lib/python3.10/dist-packages (from scikit-learn>=0.21.3->seqeval) (1.2.0)\n",
            "Requirement already satisfied: threadpoolctl>=2.0.0 in /usr/local/lib/python3.10/dist-packages (from scikit-learn>=0.21.3->seqeval) (3.1.0)\n",
            "Building wheels for collected packages: seqeval\n",
            "  Building wheel for seqeval (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
            "  Created wheel for seqeval: filename=seqeval-1.2.2-py3-none-any.whl size=16165 sha256=8f799517f97d07fa0a04acf6c88c6a38d5be93ae57959e369fd1154c1f4167ac\n",
            "  Stored in directory: /root/.cache/pip/wheels/1a/67/4a/ad4082dd7dfc30f2abfe4d80a2ed5926a506eb8a972b4767fa\n",
            "Successfully built seqeval\n",
            "Installing collected packages: seqeval\n",
            "Successfully installed seqeval-1.2.2\n"
          ]
        }
      ],
      "source": [
        "!pip install seqeval"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "e5kPVmKYcPgD"
      },
      "outputs": [],
      "source": [
        "#from sklearn.metrics import f1_score\n",
        "from seqeval.metrics import classification_report, f1_score\n",
        "import json"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "tMSSIsxOcRN7"
      },
      "outputs": [],
      "source": [
        "f1 = open('/content/score_eval/test_data.json')\n",
        "f2 = open('/content/output_result/CMeEE_test.json')\n",
        "\n",
        "fullTestDataset = json.load(f1)\n",
        "predictedTestDataset = json.load(f2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Lr7w-5Ikrlic",
        "outputId": "2b64f124-d8f9-4969-868c-4c07d201e304"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "9868"
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ],
      "source": [
        "len(predictedTestDataset)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WTbQlB7ErigR",
        "outputId": "3dda3aad-667b-4693-ba93-f99f1c3ce659"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "9868"
            ]
          },
          "metadata": {},
          "execution_count": 28
        }
      ],
      "source": [
        "len(fullTestDataset)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "jydrSFL7c-sA"
      },
      "outputs": [],
      "source": [
        "trueLabels = []\n",
        "predictedLabels = []\n",
        "labelWrongOrNull = '-100'\n",
        "\n",
        "for i in range(len(fullTestDataset)):\n",
        "  tempTrue = []\n",
        "  tempPredicted = []\n",
        "\n",
        "  for j in range(len(fullTestDataset[i]['entities'])):\n",
        "    itemTrue = fullTestDataset[i]['entities'][j]\n",
        "    predictedItem = next((predictedItem['type'] for predictedItem in predictedTestDataset[i]['entities'] if  itemTrue['start_idx'] == predictedItem['start_idx'] and itemTrue['end_idx'] == predictedItem['end_idx']), labelWrongOrNull)\n",
        "\n",
        "    tempTrue.append(itemTrue['type'])\n",
        "    tempPredicted.append(predictedItem)\n",
        "\n",
        "  predictedLabels.append(tempPredicted)\n",
        "  trueLabels.append(tempTrue)\n",
        ""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uNvODBDoKCgm",
        "outputId": "332c373f-5e80-403e-fc34-d18143f94695"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "9868\n",
            "9868\n"
          ]
        }
      ],
      "source": [
        "print(len(predictedLabels))\n",
        "print(len(trueLabels))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "DhROcGq1kP14",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "b1e78ea2-3dd8-462c-8463-9ee3b5711279"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/seqeval/metrics/sequence_labeling.py:171: UserWarning: bod seems not to be NE tag.\n",
            "  warnings.warn('{} seems not to be NE tag.'.format(chunk))\n",
            "/usr/local/lib/python3.10/dist-packages/seqeval/metrics/sequence_labeling.py:171: UserWarning: dis seems not to be NE tag.\n",
            "  warnings.warn('{} seems not to be NE tag.'.format(chunk))\n",
            "/usr/local/lib/python3.10/dist-packages/seqeval/metrics/sequence_labeling.py:171: UserWarning: ite seems not to be NE tag.\n",
            "  warnings.warn('{} seems not to be NE tag.'.format(chunk))\n",
            "/usr/local/lib/python3.10/dist-packages/seqeval/metrics/sequence_labeling.py:171: UserWarning: dep seems not to be NE tag.\n",
            "  warnings.warn('{} seems not to be NE tag.'.format(chunk))\n",
            "/usr/local/lib/python3.10/dist-packages/seqeval/metrics/sequence_labeling.py:171: UserWarning: sym seems not to be NE tag.\n",
            "  warnings.warn('{} seems not to be NE tag.'.format(chunk))\n",
            "/usr/local/lib/python3.10/dist-packages/seqeval/metrics/sequence_labeling.py:171: UserWarning: pro seems not to be NE tag.\n",
            "  warnings.warn('{} seems not to be NE tag.'.format(chunk))\n",
            "/usr/local/lib/python3.10/dist-packages/seqeval/metrics/sequence_labeling.py:171: UserWarning: dru seems not to be NE tag.\n",
            "  warnings.warn('{} seems not to be NE tag.'.format(chunk))\n",
            "/usr/local/lib/python3.10/dist-packages/seqeval/metrics/sequence_labeling.py:171: UserWarning: equ seems not to be NE tag.\n",
            "  warnings.warn('{} seems not to be NE tag.'.format(chunk))\n",
            "/usr/local/lib/python3.10/dist-packages/seqeval/metrics/sequence_labeling.py:171: UserWarning: mic seems not to be NE tag.\n",
            "  warnings.warn('{} seems not to be NE tag.'.format(chunk))\n",
            "/usr/local/lib/python3.10/dist-packages/seqeval/metrics/sequence_labeling.py:171: UserWarning: -100 seems not to be NE tag.\n",
            "  warnings.warn('{} seems not to be NE tag.'.format(chunk))\n",
            "/usr/local/lib/python3.10/dist-packages/seqeval/metrics/v1.py:57: UndefinedMetricWarning: Recall and F-score are ill-defined and being set to 0.0 in labels with no true samples. Use `zero_division` parameter to control this behavior.\n",
            "  _warn_prf(average, modifier, msg_start, len(result))\n"
          ]
        }
      ],
      "source": [
        "report = classification_report(trueLabels, predictedLabels)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Z_i677Xzf0Ay"
      },
      "outputs": [],
      "source": [
        "micro_f1 = f1_score(trueLabels, predictedLabels)\n",
        "macro_f1 = f1_score(trueLabels, predictedLabels, average='macro')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LYKBTGR5kYBD",
        "outputId": "430024f1-c049-4f62-c950-ace4fcf8911d"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "         100       0.00      0.00      0.00         0\n",
            "          ep       1.00      0.94      0.97      1421\n",
            "          ic       0.33      0.07      0.12       362\n",
            "          is       0.37      0.09      0.15      3523\n",
            "          od       0.43      0.13      0.20      3793\n",
            "          qu       0.99      0.86      0.92      1229\n",
            "          ro       0.31      0.05      0.09      1310\n",
            "          ru       0.37      0.07      0.11      1275\n",
            "          te       0.94      0.57      0.71      1358\n",
            "          ym       0.15      0.04      0.06      2086\n",
            "\n",
            "   micro avg       0.32      0.26      0.29     16357\n",
            "   macro avg       0.49      0.28      0.33     16357\n",
            "weighted avg       0.50      0.26      0.32     16357\n",
            "\n"
          ]
        }
      ],
      "source": [
        "print(report)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8wdSe-wykFhN",
        "outputId": "14b1137a-f0b5-4d12-e969-0aa3f8107e47"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.2860207844451894"
            ]
          },
          "metadata": {},
          "execution_count": 34
        }
      ],
      "source": [
        "micro_f1"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "u8Y4SAwWkHOF",
        "outputId": "96ad726a-e463-47cf-98b8-dfe73af32291"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.3337625221709553"
            ]
          },
          "metadata": {},
          "execution_count": 35
        }
      ],
      "source": [
        "macro_f1"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}