cloud/notebooks/python_sdk/deployments/foundation_models/prompt_tuning/Use watsonx to tune IBM 'granite-13b-instruct-v2' model with cars4u document.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "![image](https://raw.githubusercontent.com/IBM/watson-machine-learning-samples/master/cloud/notebooks/headers/watsonx-Prompt_Lab-Notebook.png)\n",
        "# Use watsonx to tune IBM `granite-13b-instruct-v2` model with Car Rental Company customer satisfaction document"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "#### Disclaimers\n",
        "\n",
        "- Use only Projects and Spaces that are available in watsonx context.\n",
        "\n",
        "\n",
        "## Notebook content\n",
        "\n",
        "This notebook contains the steps and code to demonstrate support of prompt tuning in watsonx.\n",
        "\n",
        "Some familiarity with Python is helpful. This notebook uses Python 3.11.\n",
        "\n",
        "\n",
        "## Learning goal\n",
        "\n",
        "The goal of this notebook is to demonstrate how to:\n",
        "- upload of dataset with prompts to container (project)\n",
        "- trigger prompt tuning process\n",
        "- review of prompt tuning process details/results\n",
        "- deploy prompt-tuned model/asset\n",
        "- evaluation (via inference) of prompt-tuned model/asset\n",
        "\n",
        "\n",
        "## Contents\n",
        "\n",
        "This notebook contains the following parts:\n",
        "\n",
        "- [Setup](#setup)\n",
        "- [Data loading](#data_loading)\n",
        "- [Initialize experiment](#initialize)\n",
        "- [Run Prompt Tuning](#run_tuning)\n",
        "- [Prompt Tuning details](#run_details)\n",
        "- [Deploy](#deploy)\n",
        "- [Foundation Models Inference](#models_inference)\n",
        "- [Analyze the satisfaction](#predict)\n",
        "- [Summary and next steps](#summary)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "<a id=\"setup\"></a>\n",
        "## Set up the environment\n",
        "\n",
        "Before you use the sample code in this notebook, you must perform the following setup tasks:\n",
        "\n",
        "-  Create a <a href=\"https://cloud.ibm.com/catalog/services/watsonxai-runtime\" target=\"_blank\" rel=\"noopener no referrer\">watsonx.ai Runtime Service</a> instance (a free plan is offered and information about how to create the instance can be found <a href=\"https://dataplatform.cloud.ibm.com/docs/content/wsj/getting-started/wml-plans.html?context=wx&audience=wdp\" target=\"_blank\" rel=\"noopener no referrer\">here</a>)."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "### Install and import the `datasets` and dependencies"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "!pip install -U ibm-watsonx-ai | tail -n 1\n",
        "!pip install \"scikit-learn==1.3.2\" | tail -n 1\n",
        "!pip install matplotlib | tail -n 1\n",
        "!pip install wget | tail -n 1"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "### Watsonx API connection\n",
        "This cell defines the credentials required to work with watsonx API for Foundation\n",
        "Model inferencing.\n",
        "\n",
        "**Action:** Provide the IBM Cloud user API key. For details, see <a href=\"https://cloud.ibm.com/docs/account?topic=account-userapikey&interface=ui\" target=\"_blank\" rel=\"noopener no referrer\">documentation</a>."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "import getpass\n",
        "from ibm_watsonx_ai import Credentials\n",
        "\n",
        "credentials = Credentials(\n",
        "    url=\"https://us-south.ml.cloud.ibm.com\",\n",
        "    api_key=getpass.getpass(\"Please enter your watsonx.ai api key (hit enter): \"),\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "### Defining the project id\n",
        "The Foundation Model requires project id that provides the context for the call. We will obtain the id from the project in which this notebook runs. Otherwise, please provide the project id."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "import os\n",
        "\n",
        "try:\n",
        "    project_id = os.environ[\"PROJECT_ID\"]\n",
        "except KeyError:\n",
        "    project_id = input(\"Please enter your project_id (hit enter): \")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Create an instance of APIClient with authentication details."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "from ibm_watsonx_ai import APIClient\n",
        "\n",
        "client = APIClient(credentials)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "To be able to interact with all resources available in watsonx.ai Runtime, you need to set **project_id** which you will be using."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "'SUCCESS'"
            ]
          },
          "execution_count": 5,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "client.set.default_project(project_id)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "<a id=\"data_loading\"></a>\n",
        "## Data loading"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "This example uses the Consumer Financial Protection Bureau training dataset. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "filename = 'car_rental_prompt_tuning_training_data.json'"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Download training data from git repository and create data assets."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Creating data asset...\n",
            "SUCCESS\n"
          ]
        }
      ],
      "source": [
        "import wget\n",
        "import os\n",
        "\n",
        "url = \"https://raw.github.com/IBM/watson-machine-learning-samples/master/cloud/data/prompt_tuning/car_rental_prompt_tuning_training_data.json\"\n",
        "if not os.path.isfile(filename):\n",
        "    wget.download(url)\n",
        "\n",
        "asset_details = client.data_assets.create(name=filename, file_path=filename)\n",
        "asset_id = client.data_assets.get_id(asset_details)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "collapsed": false,
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "8e29dfdf-3940-41d1-832a-3b4d72064c2a\n"
          ]
        }
      ],
      "source": [
        "print(asset_id)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Define connection information to training data."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "from ibm_watsonx_ai.helpers import DataConnection\n",
        "\n",
        "data_conn = DataConnection(data_asset_id=asset_id)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "<a id=\"initialize\"></a>\n",
        "## Initialize experiment "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "from ibm_watsonx_ai.experiment import TuneExperiment\n",
        "\n",
        "experiment = TuneExperiment(credentials, project_id=project_id)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "All avaliable tasks are presented under `Tasks` Enum"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "{'QUESTION_ANSWERING': 'question_answering',\n",
              " 'SUMMARIZATION': 'summarization',\n",
              " 'RETRIEVAL_AUGMENTED_GENERATION': 'retrieval_augmented_generation',\n",
              " 'CLASSIFICATION': 'classification',\n",
              " 'GENERATION': 'generation',\n",
              " 'CODE_GENERATION_AND_CONVERSION': 'code',\n",
              " 'EXTRACTION': 'extraction',\n",
              " 'TRANSLATION': 'translation'}"
            ]
          },
          "execution_count": 11,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "{task.name: task.value for task in experiment.Tasks}"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Define `prompt_tuner` parameters. \n",
        "\n",
        "**Action:** For parameters details, see <a href=\"https://ibm.github.io/watsonx-ai-python-sdk/tune_experiment.html#ibm_watsonx_ai.experiment.fm_tune.TuneExperiment.prompt_tuner\" target=\"_blank\" rel=\"noopener no referrer\">documentation</a>."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 36,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "prompt_tuner = experiment.prompt_tuner(name=\"granite-13b-instruct-v2 prompt tuning - python SDK\",\n",
        "                                       task_id=experiment.Tasks.CLASSIFICATION,\n",
        "                                       base_model='ibm/granite-13b-instruct-v2',\n",
        "                                       accumulate_steps=16,\n",
        "                                       batch_size=8,\n",
        "                                       learning_rate=0.0002,\n",
        "                                       max_input_tokens=128,\n",
        "                                       max_output_tokens=2,\n",
        "                                       num_epochs=6,\n",
        "                                       tuning_type=experiment.PromptTuningTypes.PT,\n",
        "                                       init_text = \"Extract the satisfaction from the comment. Return simple '1' for satisfied customer or '0' for unsatisfied. Comment:\",\n",
        "                                       init_method = \"text\",\n",
        "                                       verbalizer=\"{{input}}\",\n",
        "                                       auto_update_model=True\n",
        "                                       )"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "You can review previous set parameters"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 37,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "{'base_model': {'model_id': 'ibm/granite-13b-instruct-v2'},\n",
              " 'accumulate_steps': 16,\n",
              " 'batch_size': 8,\n",
              " 'init_method': 'text',\n",
              " 'init_text': \"Extract the satisfaction from the comment. Return simple '1' for satisfied customer or '0' for unsatisfied. Comment:\",\n",
              " 'learning_rate': 0.0002,\n",
              " 'max_input_tokens': 128,\n",
              " 'max_output_tokens': 2,\n",
              " 'num_epochs': 6,\n",
              " 'task_id': 'classification',\n",
              " 'tuning_type': 'prompt_tuning',\n",
              " 'verbalizer': '{{input}}',\n",
              " 'name': 'granite-13b-instruct-v2 prompt tuning - python SDK',\n",
              " 'description': 'Prompt tuning with SDK',\n",
              " 'auto_update_model': True,\n",
              " 'group_by_name': False}"
            ]
          },
          "execution_count": 37,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "prompt_tuner.get_params()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "<a id=\"run_tuning\"></a>\n",
        "## Run Prompt Tuning"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Run a prompt tuning process of foundation model on top of the training data referenced by DataConnection (tuning may take some time).\n",
        "\n",
        "\n",
        "By changing the `background_mode` parameter to `True`, the prompt tuning process will run in the background."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 38,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "\n",
            "##############################################\n",
            "\n",
            "Running 'd4adb955-b2c3-4ce2-a348-87e806c14559'\n",
            "\n",
            "##############################################\n",
            "\n",
            "\n",
            "pending...............\n",
            "running.............................................................................................................................................................................................................................................................................................\n",
            "completed\n",
            "Training of 'd4adb955-b2c3-4ce2-a348-87e806c14559' finished successfully.\n"
          ]
        }
      ],
      "source": [
        "tuning_details = prompt_tuner.run(training_data_references=[data_conn], background_mode=False)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "<a id=\"run_details\"></a>\n",
        "## Prompt Tuning details"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Check status/state of initialized Prompt Tuning run if ran in background mode or when process finish if background mode is off."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 39,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "'completed'"
            ]
          },
          "execution_count": 39,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "prompt_tuner.get_run_status()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Get prompt tuning run details."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 40,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "{'metadata': {'created_at': '2024-02-05T13:41:07.947Z',\n",
              "  'description': 'Prompt tuning with SDK',\n",
              "  'id': 'd4adb955-b2c3-4ce2-a348-87e806c14559',\n",
              "  'modified_at': '2024-02-05T14:09:45.830Z',\n",
              "  'name': 'granite-13b-instruct-v2 prompt tuning - python SDK',\n",
              "  'project_id': '4b865c15-f0e8-4054-9b1b-12a17b62d2f6',\n",
              "  'tags': ['prompt_tuning',\n",
              "   'wx_prompt_tune.08749bd2-b5d4-47ac-9965-240e9dc5d72f']},\n",
              " 'entity': {'auto_update_model': True,\n",
              "  'description': 'Prompt tuning with SDK',\n",
              "  'model_id': '23854716-30a3-4d89-8931-61e3a5f1c08b',\n",
              "  'name': 'granite-13b-instruct-v2 prompt tuning - python SDK',\n",
              "  'project_id': '4b865c15-f0e8-4054-9b1b-12a17b62d2f6',\n",
              "  'prompt_tuning': {'accumulate_steps': 16,\n",
              "   'base_model': {'model_id': 'ibm/granite-13b-instruct-v2'},\n",
              "   'batch_size': 8,\n",
              "   'init_method': 'text',\n",
              "   'init_text': \"Extract the satisfaction from the comment. Return simple '1' for satisfied customer or '0' for unsatisfied. Comment:\",\n",
              "   'learning_rate': 0.0002,\n",
              "   'max_input_tokens': 128,\n",
              "   'max_output_tokens': 2,\n",
              "   'num_epochs': 6,\n",
              "   'num_virtual_tokens': 100,\n",
              "   'task_id': 'classification',\n",
              "   'tuning_type': 'prompt_tuning',\n",
              "   'verbalizer': '{{input}}'},\n",
              "  'results_reference': {'connection': {},\n",
              "   'location': {'path': 'default_tuning_output',\n",
              "    'training': 'default_tuning_output/d4adb955-b2c3-4ce2-a348-87e806c14559',\n",
              "    'training_status': 'default_tuning_output/d4adb955-b2c3-4ce2-a348-87e806c14559/training-status.json',\n",
              "    'model_request_path': 'default_tuning_output/d4adb955-b2c3-4ce2-a348-87e806c14559/assets/d4adb955-b2c3-4ce2-a348-87e806c14559/resources/wml_model/request.json',\n",
              "    'assets_path': 'default_tuning_output/d4adb955-b2c3-4ce2-a348-87e806c14559/assets'},\n",
              "   'type': 'container'},\n",
              "  'status': {'completed_at': '2024-02-05T14:09:45.710Z',\n",
              "   'running_at': '2024-02-05T13:42:38.256Z',\n",
              "   'state': 'completed'},\n",
              "  'tags': ['prompt_tuning',\n",
              "   'wx_prompt_tune.08749bd2-b5d4-47ac-9965-240e9dc5d72f'],\n",
              "  'training_data_references': [{'connection': {},\n",
              "    'location': {'href': '/v2/assets/8e29dfdf-3940-41d1-832a-3b4d72064c2a',\n",
              "     'id': '8e29dfdf-3940-41d1-832a-3b4d72064c2a'},\n",
              "    'type': 'data_asset'}]}}"
            ]
          },
          "execution_count": 40,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "prompt_tuner.get_run_details()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Let's summarize the run."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 41,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Enhancements</th>\n",
              "      <th>Base model</th>\n",
              "      <th>Auto store</th>\n",
              "      <th>Epochs</th>\n",
              "      <th>loss</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Model Name</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>model_d4adb955-b2c3-4ce2-a348-87e806c14559</th>\n",
              "      <td>[prompt_tuning]</td>\n",
              "      <td>ibm/granite-13b-instruct-v2</td>\n",
              "      <td>True</td>\n",
              "      <td>6</td>\n",
              "      <td>0.1123</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                               Enhancements  \\\n",
              "Model Name                                                    \n",
              "model_d4adb955-b2c3-4ce2-a348-87e806c14559  [prompt_tuning]   \n",
              "\n",
              "                                                             Base model  \\\n",
              "Model Name                                                                \n",
              "model_d4adb955-b2c3-4ce2-a348-87e806c14559  ibm/granite-13b-instruct-v2   \n",
              "\n",
              "                                            Auto store  Epochs    loss  \n",
              "Model Name                                                              \n",
              "model_d4adb955-b2c3-4ce2-a348-87e806c14559        True       6  0.1123  "
            ]
          },
          "execution_count": 41,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "prompt_tuner.summary()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "collapsed": false,
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Plot learning curves."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 42,
      "metadata": {
        "collapsed": false,
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABdEAAAGJCAYAAABlx3qDAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdeVxUVf8H8M/sw77JJiLgioArKuFeomBWWrZolkZmPZWW2vI89pSm2WOZmWaWWWmL+dNsMdtQ0rRScofct8CNXZZhZ5i5vz+uMzoywAAzDMvn/XrNC+bOueeeOeLce75z7vdIBEEQQERERERERERERERE1Ujt3QAiIiIiIiIiIiIiouaKQXQiIiIiIiIiIiIiohowiE5EREREREREREREVAMG0YmIiIiIiIiIiIiIasAgOhERERERERERERFRDRhEJyIiIiIiIiIiIiKqAYPoREREREREREREREQ1YBCdiIiIiIiIiIiIiKgGDKITEREREREREREREdWAQXQiapSEhAT06dMHarUaEokEBQUF9m6SWRKJBK+++qq9m0FERM3UI488guDg4AbtK5FIMGPGjDrLffrpp5BIJEhLSzNuGzFiBEaMGGF8npaWBolEgk8//bRBbbGVt956C506dYJMJkOfPn3qvf+uXbsgkUjw9ddfW79xzUxwcDAeeeSRBu3L6xUiaosae46hhrHGWN7Sa6CWrjHXiTdf61HLxSA6UTNkGGQfPHjQ3k2p1dWrV3H//ffDwcEBq1atwhdffAEnJye7tefnn3/mwJOIqBU4evQo7r33XgQFBUGtViMgIACjRo3CypUr7d00u7D3+W379u148cUXMXjwYKxbtw7/+9//aiy7YcMGLF++vOkaR0REVmcYjxoearUa3bp1w4wZM5CVlWXVY9XnHEPWU5+x/N69e/Hqq6822wlzRE1Fbu8GEFHLdeDAARQVFeG1115DTEyMvZuDn3/+GatWrTIbaCgrK4Nczo88IqLmbu/evbj11lvRsWNHTJ8+HX5+frh06RL++usvrFixAjNnzrR3Exvs4YcfxsSJE6FSqWosExQUhLKyMigUCuO22s5vTWHnzp2QSqX45JNPoFQqay27YcMGHDt2DLNmzWqaxhERkc0sXLgQISEhKC8vx59//okPPvgAP//8M44dOwZHR0erHKM+5xiynvqM5ffu3YsFCxbgkUcegbu7e9M0kKgZYkSJiBosOzsbAFrEiVStVtu7CUREZIHXX38dbm5uOHDgQLXzi+G801LJZDLIZLJayxhm/DUn2dnZcHBwYHCDiKiNGTNmDPr37w8AeOyxx+Dl5YVly5bh+++/x6RJkxpVd2lpKRwdHa1+jhEEAeXl5XBwcLBKfa1VSxrLEzUXTOdC1IIdOXIEY8aMgaurK5ydnTFy5Ej89ddfJmW0Wi0WLFiArl27Qq1Ww8vLC0OGDEFiYqKxTGZmJuLj49GhQweoVCr4+/tj3LhxJjlbbzZixAhMnToVADBgwABIJBJj/s+acoHenAvMkB/1q6++wuuvv44OHTpArVZj5MiROHfuXLX99+3bh9tvvx0eHh5wcnJCr169sGLFCgBijrJVq1YBgMmthwbmcoxa0n+GWxn37NmDOXPmwNvbG05OTrj77ruRk5NTY/8QEVHDnD9/HuHh4WYHdT4+PtW2rV+/HgMHDoSjoyM8PDwwbNgwbN++3fj6999/j7Fjx6J9+/ZQqVTo3LkzXnvtNeh0ujrbsnTpUgwaNAheXl5wcHBAZGRkrTm9v/zyS3Tv3h1qtRqRkZH4/fffTV43lxP9ZjfnRK/p/CYIAoKDgzFu3LhqdZSXl8PNzQ1PPPFEre+vqqoKr732Gjp37gyVSoXg4GC89NJLqKioMJaRSCRYt24dSkpKjMeuKV/7iBEj8NNPP+HChQvGsjfnD9Xr9Raf8+Pi4uDm5gZHR0cMHz4ce/bsqfX9AKbXFgsWLEBAQABcXFxw7733orCwEBUVFZg1axZ8fHzg7OyM+Ph4k/drab8AYqBm0aJF6NChAxwdHXHrrbfi+PHjZttVUFCAWbNmITAwECqVCl26dMGbb74JvV5f53siImoubrvtNgBAamqqcdv69esRGRkJBwcHeHp6YuLEibh06ZLJfiNGjEBERAQOHTqEYcOGwdHRES+99FKt5xhLP4uDg4Nxxx13YNu2bejfvz8cHBzw4YcfWuV8sG7dOtx2223w8fGBSqVCWFgYPvjgg2r9YmjDn3/+iYEDB0KtVqNTp074/PPPq5UtKCjA7NmzERwcDJVKhQ4dOmDKlCnIzc01lqmoqMD8+fPRpUsXqFQqBAYG4sUXX6zWvpps3rzZ+G/Srl07PPTQQ7hy5YrJv0dNY/mbvfrqq3jhhRcAACEhIcZ/p5uvZbZs2YKIiAioVCqEh4cjISGhWl1XrlzBo48+Cl9fX2O5tWvXWvSeDLnXN2/ejLCwMDg4OCA6OhpHjx4FAHz44Yfo0qUL1Go1RowYYfZaq65+ufm9qNVqRERE4LvvvjPbJr1ej+XLlyM8PBxqtRq+vr544oknkJ+fb9F7opaHM9GJWqjjx49j6NChcHV1xYsvvgiFQoEPP/wQI0aMwO7duxEVFQVAPOktXrwYjz32GAYOHAiNRoODBw/i8OHDGDVqFABgwoQJOH78OGbOnIng4GBkZ2cjMTERFy9erHHxjP/+97/o3r071qxZY7zNr3Pnzg16L2+88QakUimef/55FBYWYsmSJZg8eTL27dtnLJOYmIg77rgD/v7+ePbZZ+Hn54eTJ0/ixx9/xLPPPosnnngC6enpSExMxBdffGG1/jOYOXMmPDw8MH/+fKSlpWH58uWYMWMGNm3a1KD3TERE5gUFBSEpKQnHjh1DRERErWUXLFiAV199FYMGDcLChQuhVCqxb98+7Ny5E6NHjwYgBq6dnZ0xZ84cODs7Y+fOnZg3bx40Gg3eeuutWutfsWIF7rrrLkyePBmVlZXYuHEj7rvvPvz4448YO3asSdndu3dj06ZNeOaZZ6BSqfD+++8jLi4O+/fvr/N91Kam85tEIsFDDz2EJUuWIC8vD56ensbXfvjhB2g0Gjz00EO11v3YY4/hs88+w7333ovnnnsO+/btw+LFi3Hy5EnjgPGLL77AmjVrsH//fnz88ccAgEGDBpmt77///S8KCwtx+fJlvPPOOwAAZ2dnkzKWnPN37tyJMWPGIDIyEvPnz4dUKjUGMv744w8MHDiwzn5bvHgxHBwc8J///Afnzp3DypUroVAoIJVKkZ+fj1dffRV//fUXPv30U4SEhGDevHn16hcAmDdvHhYtWoTbb78dt99+Ow4fPozRo0ejsrLSpC2lpaUYPnw4rly5gieeeAIdO3bE3r17MXfuXGRkZDCHPBG1GOfPnwcAeHl5ARDvHnvllVdw//3347HHHkNOTg5WrlyJYcOG4ciRIyZfiF+9ehVjxozBxIkT8dBDD8HX1xf9+/ev8Rxj6WcxAJw+fRqTJk3CE088genTp6N79+7G1xpzPvjggw8QHh6Ou+66C3K5HD/88AOeeuop6PV6PP300yZtOHfuHO69915MmzYNU6dOxdq1a/HII48gMjIS4eHhAIDi4mIMHToUJ0+exKOPPop+/fohNzcXW7duxeXLl9GuXTvo9Xrcdddd+PPPP/H444+jR48eOHr0KN555x2cOXMGW7ZsqfXf6NNPP0V8fDwGDBiAxYsXIysrCytWrMCePXuM/yb1Gcvfc889OHPmDP7v//4P77zzDtq1awcA8Pb2Npb5888/8e233+Kpp56Ci4sL3n33XUyYMAEXL140/q1kZWXhlltuMQbDvb298csvv2DatGnQaDQWpYH7448/sHXrVmPfL168GHfccQdefPFFvP/++3jqqaeQn5+PJUuW4NFHH8XOnTvr1S+AmKN/woQJCAsLw+LFi3H16lXjhMObPfHEE8Z6n3nmGaSmpuK9997DkSNHsGfPHpPUfNRKCETU7Kxbt04AIBw4cKDGMuPHjxeUSqVw/vx547b09HTBxcVFGDZsmHFb7969hbFjx9ZYT35+vgBAeOutt6zWzqCgIGHq1KnVyg8fPlwYPny48flvv/0mABB69OghVFRUGLevWLFCACAcPXpUEARBqKqqEkJCQoSgoCAhPz/fpE69Xm/8/emnnxZq+lgDIMyfP9/43NL+M7zHmJgYk2PNnj1bkMlkQkFBgdnjERFRw2zfvl2QyWSCTCYToqOjhRdffFHYtm2bUFlZaVLu7NmzglQqFe6++25Bp9OZvHbj53VpaWm1YzzxxBOCo6OjUF5ebtw2depUISgoyKTczftWVlYKERERwm233WayHYAAQDh48KBx24ULFwS1Wi3cfffdxm2Gc0pqaqpx283nxtTUVAGAsG7dOuO2ms5vp0+fFgAIH3zwgcn2u+66SwgODjbph5slJycLAITHHnvMZPvzzz8vABB27txp3DZ16lTBycmpxrpuNHbs2Gr9KAiWn/P1er3QtWtXITY2ttq/Y0hIiDBq1Khaj284TkREhMnfzKRJkwSJRCKMGTPGpHx0dLRJey3tl+zsbEGpVApjx441aedLL70kADC5DnrttdcEJycn4cyZMyZ1/uc//xFkMplw8eJF47abr1eIiOzBcL769ddfhZycHOHSpUvCxo0bBS8vL8HBwUG4fPmykJaWJshkMuH111832ffo0aOCXC432T58+HABgLB69epqxzJ3jqnPOSooKEgAICQkJJiUbez5QBDMX0PExsYKnTp1MtlmaMPvv/9u3JadnS2oVCrhueeeM26bN2+eAED49ttvq9VrOJd88cUXglQqFf744w+T11evXi0AEPbs2VNtX4PKykrBx8dHiIiIEMrKyozbf/zxRwGAMG/ePOM2S2IOBm+99Va16xcDAIJSqRTOnTtn3JaSkiIAEFauXGncNm3aNMHf31/Izc012X/ixImCm5ub2b6++TgqlcqkDR9++KEAQPDz8xM0Go1x+9y5c03aW59+6dOnj+Dv728yzt++fbsAwOTv448//hAACF9++aVJOxMSEqptv/laj1oupnMhaoF0Oh22b9+O8ePHo1OnTsbt/v7+ePDBB/Hnn39Co9EAEHOcHT9+HGfPnjVblyH/3K5du+x221F8fLxJDryhQ4cCAP755x8AYtqV1NRUzJo1q9rt/TembLFUffrP4PHHHzc51tChQ6HT6XDhwoV6H5+IiGo2atQoJCUl4a677kJKSgqWLFmC2NhYBAQEYOvWrcZyW7ZsgV6vx7x58yCVml7S3vh5fWNO1KKiIuTm5mLo0KEoLS3FqVOnam3Ljfvm5+ejsLAQQ4cOxeHDh6uVjY6ORmRkpPF5x44dMW7cOGzbts2i1DEN0a1bN0RFReHLL780bsvLy8Mvv/yCyZMn13qO/PnnnwEAc+bMMdn+3HPPAQB++uknG7S47nN+cnIyzp49iwcffBBXr15Fbm4ucnNzUVJSgpEjR+L333+3KAXKlClTTGaARUVFQRAEPProoybloqKicOnSJVRVVQGwvF9+/fVXVFZWYubMmSb9bG4m3ebNmzF06FB4eHgY309ubi5iYmKg0+mqpf0hImouYmJi4O3tjcDAQEycOBHOzs747rvvEBAQgG+//RZ6vR7333+/yWebn58funbtit9++82kLpVKhfj4eIuOW99zVEhICGJjY83W1dDzAWB6HVBYWIjc3FwMHz4c//zzDwoLC032DwsLM57TAHGmdvfu3Y3nNwD45ptv0Lt3b9x9993V2mk4l2zevBk9evRAaGioSb8aUunc3K83OnjwILKzs/HUU0+ZrK8yduxYhIaG2uzcHhMTYzKTvVevXnB1dTW+d0EQ8M033+DOO++EIAgm7ys2NhaFhYVmr61uNnLkSJM75Q13j0+YMAEuLi7VthuOb2m/ZGRkIDk5GVOnToWbm5ux3KhRoxAWFmbSls2bN8PNzQ2jRo0yeT+RkZFwdnau9d+JWi6mcyFqgXJyclBaWmpym5pBjx49oNfrcenSJYSHh2PhwoUYN24cunXrhoiICMTFxeHhhx9Gr169AIgXM2+++Saee+45+Pr64pZbbsEdd9yBKVOmwM/Pr0neT8eOHU2ee3h4AIAxqG+4bbAxt8PfqD79Z2kbiYjIegYMGIBvv/0WlZWVSElJwXfffYd33nkH9957L5KTkxEWFobz589DKpVWG9Tc7Pjx43j55Zexc+fOal+Q3jwAvtmPP/6IRYsWITk5uVqe8Jt17dq12rZu3bqhtLQUOTk5NjunTpkyBTNmzMCFCxcQFBSEzZs3Q6vV4uGHH651vwsXLkAqlaJLly4m2/38/ODu7m6zL4nrOp8avvQ35Go1p7Cw0LifpccxDIYDAwOrbdfr9SgsLISXl5fF/WL4efO/u7e3d7W2nT17Fn///bfJre83aukL5hJR67Vq1Sp069YNcrkcvr6+6N69u/GL67Nnz0IQBLPnPwDVUlkEBARYvHhofc9RISEhNdbV0PMBAOzZswfz589HUlISSktLTcoXFhaaBFpvPg4gnuNuHC+eP38eEyZMqLGtgNivJ0+ebNA5w9Av5sa5oaGh+PPPP2s9dkPV9d5zcnJQUFCANWvWYM2aNWbrsORcWJ9/S+D6tYWl/VLTud2w742B/rNnz6KwsNDsej2Wvh9qeRhEJ2rlhg0bhvPnz+P777/H9u3b8fHHH+Odd97B6tWr8dhjjwEQZ03deeed2LJlC7Zt24ZXXnkFixcvxs6dO9G3b996H7OmmW86nQ4ymazadnPbAPEb6+aiJbSRiKi1USqVGDBgAAYMGIBu3bohPj4emzdvxvz58y3av6CgAMOHD4erqysWLlyIzp07Q61W4/Dhw/j3v/9d64zmP/74A3fddReGDRuG999/H/7+/lAoFFi3bh02bNhgrbfYaBMnTsTs2bPx5Zdf4qWXXsL69evRv39/swNFcxpyR1dj1HU+NfybvPXWW+jTp4/ZsjfnWa/PcSw9n1uzX/R6PUaNGoUXX3zR7OvdunWz2rGIiKxp4MCB6N+/v9nX9Ho9JBIJfvnlF7OfrTd/Vt84q9tSln4W11Z3Q88H58+fx8iRIxEaGoply5YhMDAQSqUSP//8M955551q1xDWGi/q9Xr07NkTy5YtM/v6zQHj5sDSc/tDDz1U45fkhkl+DTmOPcbqer0ePj4+JncD3qimL0GoZWMQnagF8vb2hqOjI06fPl3ttVOnTkEqlZqcXD09PREfH4/4+HgUFxdj2LBhePXVV41BdADo3LkznnvuOTz33HM4e/Ys+vTpg7fffhvr16+vd/s8PDxQUFBQbfuFCxdM0qdYynBr2LFjxxATE1NjOUsvsurbf0REZH+GQXxGRgYA8dyg1+tx4sSJGoOtu3btwtWrV/Htt99i2LBhxu2pqal1Hu+bb76BWq3Gtm3boFKpjNvXrVtntry5tGlnzpyBo6NjowdStZ3fPD09MXbsWHz55ZeYPHky9uzZY9FClUFBQdDr9Th79ix69Ohh3J6VlYWCggIEBQVZva2WMJzzXV1daz3n24ql/WL4efbsWZNrm5ycnGp3qXXu3BnFxcV2eT9ERLbSuXNnCIKAkJAQq38ZaKtzVH388MMPqKiowNatW01mQDcmTUfnzp1x7NixOsukpKRg5MiR9T6nGvrl9OnTxvQvBqdPn7bbud3b2xsuLi7Q6XR2O7cDdffLjef2m90cO+jcuTN+/fVXDB48uEFfEFHLxJzoRC2QTCbD6NGj8f333yMtLc24PSsrCxs2bMCQIUPg6uoKQFwF/UbOzs7o0qWL8bb00tJSlJeXm5Tp3LkzXFxcTG5dr4/OnTvjr7/+QmVlpXHbjz/+iEuXLjWovn79+iEkJATLly+vFpy/8dtlJycnADAbwL9RffqPiIia1m+//WZ25pAhP6phhvX48eMhlUqxcOHCarPBDPsbZibdWF9lZSXef//9Otshk8kgkUhM8pmnpaVhy5YtZssnJSWZ3OZ76dIlfP/99xg9enSNM6QsVdf57eGHH8aJEyfwwgsvQCaTYeLEiXXWefvttwNAtYC7Yebb2LFjG9zWutLk1CYyMhKdO3fG0qVLUVxcXO31nJycBtdtCUv7JSYmBgqFAitXrjT5+zL3Bcb999+PpKQkbNu2rdprBQUFJvl3iYhainvuuQcymQwLFiyodt4WBKHaOLQ+bHWOqg9z1xCFhYU1fpluiQkTJhjT1N3McJz7778fV65cwUcffVStTFlZGUpKSmqsv3///vDx8cHq1atNxvK//PILTp482ahzO1D3OLsmMpkMEyZMwDfffGP2SwRbn9st7Rd/f3/06dMHn332mcm1TGJiIk6cOGFS5/333w+dTofXXnut2vGqqqoa3FfUvHEmOlEztnbtWiQkJFTb/uyzz2LRokVITEzEkCFD8NRTT0Eul+PDDz9ERUUFlixZYiwbFhaGESNGIDIyEp6enjh48CC+/vprzJgxA4A4S27kyJG4//77ERYWBrlcju+++w5ZWVkWDcLNeeyxx/D1118jLi4O999/P86fP4/169ebLDZSH1KpFB988AHuvPNO9OnTB/Hx8fD398epU6dw/Phx46DUsKDbM888g9jY2FoDCZb2HxERNa2ZM2eitLQUd999N0JDQ1FZWYm9e/di06ZNCA4ONi5K1qVLF/z3v//Fa6+9hqFDh+Kee+6BSqXCgQMH0L59eyxevBiDBg2Ch4cHpk6dimeeeQYSiQRffPGFRbf3jh07FsuWLUNcXBwefPBBZGdnY9WqVejSpQv+/vvvauUjIiIQGxuLZ555BiqVyhioX7BgQaP7pK7z29ixY+Hl5YXNmzdjzJgxNebnvFHv3r0xdepUrFmzxpj2Zv/+/fjss88wfvx43HrrrQ1u66ZNmzBnzhwMGDAAzs7OuPPOOy3eXyqV4uOPP8aYMWMQHh6O+Ph4BAQE4MqVK/jtt9/g6uqKH374oUFts4Sl/eLt7Y3nn38eixcvxh133IHbb78dR44cwS+//IJ27dqZ1PnCCy9g69atuOOOO/DII48gMjISJSUlOHr0KL7++mukpaVV24eIqLnr3LkzFi1ahLlz5yItLQ3jx4+Hi4sLUlNT8d133+Hxxx/H888/36C6bXWOqo/Ro0dDqVTizjvvxBNPPIHi4mJ89NFH8PHxMd4VV18vvPACvv76a9x333149NFHERkZiby8PGzduhWrV69G79698fDDD+Orr77Cv/71L/z2228YPHgwdDodTp06ha+++grbtm2rMcWOQqHAm2++ifj4eAwfPhyTJk1CVlYWVqxYgeDgYMyePbtB7TZch/z3v//FxIkToVAocOeddxqD65Z444038NtvvyEqKgrTp09HWFgY8vLycPjwYfz666/Iy8trUNssUZ9+Wbx4McaOHYshQ4bg0UcfRV5eHlauXInw8HCTL/eHDx+OJ554AosXL0ZycjJGjx4NhUKBs2fPYvPmzVixYgXuvfdem70nshOBiJqddevWCQBqfFy6dEkQBEE4fPiwEBsbKzg7OwuOjo7CrbfeKuzdu9ekrkWLFgkDBw4U3N3dBQcHByE0NFR4/fXXhcrKSkEQBCE3N1d4+umnhdDQUMHJyUlwc3MToqKihK+++sridh44cKDaa2+//bYQEBAgqFQqYfDgwcLBgweF4cOHC8OHDzeW+e233wQAwubNm032TU1NFQAI69atM9n+559/CqNGjRJcXFwEJycnoVevXsLKlSuNr1dVVQkzZ84UvL29BYlEItz4EQdAmD9/vkl9lvRfTe/R0Pbffvutzn4iIiLL/fLLL8Kjjz4qhIaGCs7OzoJSqRS6dOkizJw5U8jKyqpWfu3atULfvn0FlUoleHh4CMOHDxcSExONr+/Zs0e45ZZbBAcHB6F9+/bCiy++KGzbtq3aZ/jUqVOFoKAgk7o/+eQToWvXroJKpRJCQ0OFdevWCfPnzxduvoQGIDz99NPC+vXrjeX79u1b7RxhOKekpqYat918bjR3Dqzt/Gbw1FNPCQCEDRs21Ny5N9FqtcKCBQuEkJAQQaFQCIGBgcLcuXOF8vJyk3JTp04VnJycLKqzuLhYePDBBwV3d3cBgLFP63vOP3LkiHDPPfcIXl5egkqlEoKCgoT7779f2LFjR63Hr+k4NZ3PDf+eOTk5xm2W9otOpxMWLFgg+Pv7Cw4ODsKIESOEY8eOCUFBQcLUqVNNyhYVFQlz584VunTpIiiVSqFdu3bCoEGDhKVLlxqvyQTB/PUKEVFTq22cd7NvvvlGGDJkiODk5CQ4OTkJoaGhwtNPPy2cPn3aWGb48OFCeHi42f1rOsdY+lkcFBQkjB07ttr+1jgfbN26VejVq5egVquF4OBg4c033xTWrl1b7VxeUxtuPscLgiBcvXpVmDFjhhAQECAolUqhQ4cOwtSpU4Xc3FxjmcrKSuHNN98UwsPDjdc3kZGRwoIFC4TCwsLqnXiTTZs2Ga+NPD09hcmTJwuXL1+2qB9q8tprrwkBAQGCVCo1ef+Ga6CbmTsXZmVlCU8//bQQGBgoKBQKwc/PTxg5cqSwZs2aOo9v7jiGa4i33nrLZHtN//aW9IsgiH/TPXr0EFQqlRAWFiZ8++23Zq8TBUEQ1qxZI0RGRgoODg6Ci4uL0LNnT+HFF18U0tPTjWXM/R1QyyQRBK6KR0RERETUks2ePRuffPIJMjMz4ejoaO/mEBERERG1KsyJTkRERETUgpWXl2P9+vWYMGECA+hERERERDbAnOhERERERC1QdnY2fv31V3z99de4evUqnn32WXs3iYiIiIioVWIQnYiIiIioBTpx4gQmT54MHx8fvPvuu+jTp4+9m0RERERE1CoxJzoRERERERERERERUQ2YE52IiIiIiIiIiIiIqAYMohMRERERERERERER1YA50RtIr9cjPT0dLi4ukEgk9m4OERG1MoIgoKioCO3bt4dUyu+8G4rnayIisjWes62D52wiIrKlxp6vGURvoPT0dAQGBtq7GURE1MpdunQJHTp0sHczWiyer4mIqKnwnN04PGcTEVFTaOj5mkH0BnJxcQEgdryrq2uj6tJqtdi+fTtGjx4NhUJhjea1Suwny7GvLMN+shz7ynLW6iuNRoPAwEDj+YYahudr+2BfWYb9ZDn2lWXYT5azZl/xnG0d1jpn8/+B5dhXlmE/WY59ZTn2lWWa0/maQfQGMtxe5urqapVBuaOjI1xdXfkfpxbsJ8uxryzDfrIc+8py1u4r3s7cODxf2wf7yjLsJ8uxryzDfrKcLfqK5+zGsdY5m/8PLMe+sgz7yXLsK8uxryzTnM7XTNhGRERERERERERERFQDBtGJiIiIiIiIiIiIiGrAIDoRERERERERERERUQ0YRCciIiIiIiIiIiIiqgGD6ERERERERERERERENWAQnYiIiIiIiIiIiIioBgyi25lOL2Bfah4O5UqwLzUPOr1g7yYREREREVELw3EFEf8fEBGR7TCIbkcJxzIw5M2deGjtQXx+VoaH1h7EkDd3IuFYhr2bRkREbcyqVasQHBwMtVqNqKgo7N+/v9bymzdvRmhoKNRqNXr27Imff/7Z5HVBEDBv3jz4+/vDwcEBMTExOHv2rEmZ119/HYMGDYKjoyPc3d3NHufixYsYO3YsHB0d4ePjgxdeeAFVVVWNeq9ERK0NxxVE/H9ARES2xSC6nSQcy8CT6w8jo7DcZHtmYTmeXH+YJ3oiImoymzZtwpw5czB//nwcPnwYvXv3RmxsLLKzs82W37t3LyZNmoRp06bhyJEjGD9+PMaPH49jx44ZyyxZsgTvvvsuVq9ejX379sHJyQmxsbEoL79+3qusrMR9992HJ5980uxxdDodxo4di8rKSuzduxefffYZPv30U8ybN8+6HUBE1IJxXEHE/wdERGR7zSKIXp/Zb8ePH8eECRMQHBwMiUSC5cuXVyvz6quvQiKRmDxCQ0NNypSXl+Ppp5+Gl5cXnJ2dMWHCBGRlZVn7rZml0wtY8MMJmLuxzLBtwQ8neOsZERE1iWXLlmH69OmIj49HWFgYVq9eDUdHR6xdu9Zs+RUrViAuLg4vvPACevTogddeew39+vXDe++9B0Cchb58+XK8/PLLGDduHHr16oXPP/8c6enp2LJli7GeBQsWYPbs2ejZs6fZ42zfvh0nTpzA+vXr0adPH4wZMwavvfYaVq1ahcrKSqv3AxFRS8NxBZnD8fV1/H9ARETWIrd3Awyz31avXo2oqCgsX74csbGxOH36NHx8fKqVLy0tRadOnXDfffdh9uzZNdYbHh6OX3/91fhcLjd9q7Nnz8ZPP/2EzZs3w83NDTNmzMA999yDPXv2WO/N1WB/al61b8hvJADIKCzH7hN5GBjsZfP2tBSGu/dLSwGFApBIxIdUav53IiKqW2VlJQ4dOoS5c+cat0mlUsTExCApKcnsPklJSZgzZ47JttjYWGOAPDU1FZmZmYiJiTG+7ubmhqioKCQlJWHixIkWtS0pKQk9e/aEr6+vyXGefPJJHD9+HH379q22T0VFBSoqKozPNRoNAECr1UKr1Vp03JoY9m9sPW0B+8oy7CfLsa/M22fhuCLpXDaiQjybrmEtgDX/pprT3yXH19UZ/h/sT81DdGeOr4mIqGHsHkS/cfYbAKxevRo//fQT1q5di//85z/Vyg8YMAADBgwAALOvG8jlcvj5+Zl9rbCwEJ988gk2bNiA2267DQCwbt069OjRA3/99RduueWWxr6tWmUX1XyCv9HJtHL4MhhspNeLPy9evB4svzFobngOXP9dLhd/Gh5yefXy5n4395yBeSJqjXJzc6HT6UwC1QDg6+uLU6dOmd0nMzPTbPnMzEzj64ZtNZWxRE3HufEYN1u8eDEWLFhQbfv27dvh6Oho8bFrk5iYaJV62gL2lWXYT5ZjX5k6lCsBIKuz3PY/9uHqSc7CNccaf1OlpaVWaIl1cHzd+HJERETm2DWI3pDZb5Y6e/Ys2rdvD7VajejoaCxevBgdO3YEABw6dAhardZkhlxoaCg6duyIpKQksyd5a85s83K0rNtdnfRwdW0+sxrsTacT+8LFRQuZDBAEMbAuCNd/NzwHTF8zvH4jQbgeGDf8XlMQHQBksuvB+Bt/ry2gf3OdN79mK5ytZRn2k+XYV5azVl+xrxtm7ty5JrPkNRoNAgMDMXr0aLi6ujaqbq1Wi8TERIwaNQoKhaKxTW3V2FeWYT9Zjn1lnldqHj4/e7DOcqOHRnEm+k2s+TdlGBvaW0saXwPWG2NbOr72cpTz+uomvMa3DPvJcuwry7GvLNOc7hyzaxC9IbPfLBEVFYVPP/0U3bt3R0ZGBhYsWIChQ4fi2LFjcHFxQWZmJpRKJdzd3asdtylmtukFwF0pQ0ElANQcSV24PQX7Th7BCH8BimaRvb55SEnhDCRLcbaWZdhPlmNfWa6xfdVUs9ratWsHmUxWLW9pVlZWjTPO/Pz8ai1v+JmVlQV/f3+TMn369LG4bX5+ftXyuBqOW1PbVCoVVCpVte0KhcJqgTdr1tXasa8sw36yHPvKVHQXH/i7qZFZWG42H7QEgJ+bGtFdfCCT8rZKc6zxN9Vc/iZb0vgasN4Yu+7xtQB3JZBz4i/8fNLiatsUXuNbhv1kOfaV5dhXlmkOd47ZPZ2LLYwZM8b4e69evRAVFYWgoCB89dVXmDZtWoPqtPbMNkVwFmZuTAEAkwteybXnHd0dcbGgFD9elOFgvgNmDO6GYZ18IGnDOUV0Oi1SUhLRu/coyGTN40K1LjfPlr/xubnXa5stb1DbrHbDLHlB0OLo0URERY2Co6MCcrmYR55McVab5dhXlrNWXzXVrDalUonIyEjs2LED48ePBwDo9Xrs2LEDM2bMMLtPdHQ0duzYgVmzZhm3JSYmIjo6GgAQEhICPz8/7Nixwxg012g02LdvH5588kmL2xYdHY3XX38d2dnZxjyuiYmJcHV1RVhYWP3fLBFRKyOTSjD/zjA8uf5wjWXm3xnGADo1ii3G14B1x9i1ja8BCRbd0xux4b5m9mzbeI1vGfaT5dhXlmNfWaY53Tlm1yB6Q2a/NYS7uzu6deuGc+fOARBnr1VWVqKgoMDk2/LajmvtmW139OkAuVyGBT+cMFkExc9Njfl3hmF0mB++PZyONxNOIrOoDC8npKBvgBdmDw9Dl3aNux29pZPJFC0miG5tNwbjzQXoBUFcgNWwCGtOjgJSqQJKpRhEd3QEVCoYn8tb5ddo9cdZbZZjX1musX3VlP08Z84cTJ06Ff3798fAgQOxfPlylJSUGPOpTpkyBQEBAVi8eDEA4Nlnn8Xw4cPx9ttvY+zYsdi4cSMOHjyINWvWAAAkEglmzZqFRYsWoWvXrggJCcErr7yC9u3bGwP1AHDx4kXk5eXh4sWL0Ol0SE5OBgB06dIFzs7OGD16NMLCwvDwww9jyZIlyMzMxMsvv4ynn37a7DmZiKgtiovwx2NDQ/DRH6km2yUS4L1JfREX4V/DntTatKTxNWDdMXZd42v+P6gdr/Etw36yHPvKcuwryzSHO8fsmiTkxtlvBobZb4bZbNZQXFyM8+fPG28pj4yMhEKhMDnu6dOncfHiRasety5xEf7489+3Yf2j/TGlqw7rH+2PP/99G+Ii/CGVSnBv/wDsemEEnhzWBUqZFEeuXMUj//cHluw8ivzSiroPQK2OYea5TCYGwJVKMSiuVgMODmKQ3MkJcHYWy7u7iw+FAtBqgdxc4PJlIDUVSEsTF2nNzQU0GqC8HNDp7PjmiMhuHnjgASxduhTz5s1Dnz59kJycjISEBOPt4BcvXkRGRoax/KBBg7BhwwasWbMGvXv3xtdff40tW7YgIiLCWObFF1/EzJkz8fjjj2PAgAEoLi5GQkIC1Gq1scy8efPQt29fzJ8/H8XFxejbty/69u2LgwfF/L4ymQw//vgjZDIZoqOj8dBDD2HKlClYuHBhE/UMEVHLkKkRxwa3R/jioc46OCllEATA04lfOLYlHF+L4+tO7cRUMLNv8TWOr4mIiBrL7vNQ6zv7rbKyEidOnDD+fuXKFSQnJ8PZ2RldunQBADz//PO48847ERQUhPT0dMyfPx8ymQyTJk0CALi5uWHatGmYM2cOPD094erqipkzZyI6OtrmK4ffTCaVICrEE1dPCogK8ax2q6WTSo5/394dD94SiEU/nMK2kxnYcuwiEs+k47GobpjQKwhyGROmU80kEjHYrlRe36bXi0H18nKgqEicwS6TXQ/KOzpen62uUIiBeyJq3WbMmFFj+pZdu3ZV23bffffhvvvuq7E+iUSChQsX1hrw/vTTT/Hpp5/W2q6goCD8/PPPtZYhImrLyrU67DwpzjyOHxSE9KNXUObmi28Op2Pb8UxEd/aycwupKXF8LYGzWgxzdPdyYCojIiKyGrsH0R944AHk5ORg3rx5yMzMRJ8+farNfpPeEMFLT09H3759jc+XLl2KpUuXYvjw4cZB/uXLlzFp0iRcvXoV3t7eGDJkCP766y94e3sb93vnnXcglUoxYcIEVFRUIDY2Fu+//37TvOkGCPR0xIdT+yHp3FW8uvUETmdrsOKPE/ju6AXMGhaGW4J97N1EakGkUjFYfuPdkzqdGFgvKQEKC8XAuiHli4OD+LgxsN6G0/MTERERNRt7zuWipFIHP1c1egW4If0oMDpMDKInHMvEvDvCIGUgsc3g+JqIiMg27B5EB+o3+y04OBiCYG7t+es2btxY5zHVajVWrVqFVatWWdzO5iC6ixd+njUEG/66hGWJp3GxoARzth7ALUHeeHZYGII8nO3dRGqhZDLxcUOmBVRViYF1jQbIzxe3GQLrTk6m+dWZwouIiIio6SUcywQAxIb7GoPlgzt5wkkpQ6amHCmXC9C3o4c9m0hNjONrIiIi62OShhZIJpXg4UEdsfvfIxAfHQKZVIK/LuTgofW/493fT6CoQmvvJlIrYZiF7uoKeHiI+dVVKjEdTF6eaX71tDQgJ0cMuJeVXV/clIiIiIhso0qnR+K1VC6xEdcXcFQpZLg1VLxTNeF4pl3aRkRERNSaMIjegrmqFZg/LgzbZg3DsC4+0AkCNian4v7PdmHLsYvQ6WufUUBUXxKJOOPc0RFwcwM8PU0XLs3JERcrvXnh0qIiLlxKREREZG37U/NQUKqFp5MSA4M9TV4bc20xxW3HMuucaUxEREREtWsW6Vyocbr4OOPzxwZg58lsLPzhBNLySrBk51F8k3IBs4eHoV8HLiZEtlPXwqXFxeJzuVwMtnPhUiIiIiLr+OVaKpdRPXwhl0mh1V+fsTCiuzeUcinSrpbiVGYRevi72quZRERERC0eg+ityG09fDC0Wzt8tucCVuw4g/NXNZjx7V8Y0dkPM4f2gL+ro72bSG1EfRcudXQUc7Fz4VIiIiIiy+j1ArZdS9USd0MqFwMnlRzDunrj15NZSDiWySA6ERERUSNw/mcro5BJ8diwEOx+8VZM7N8RUgmw63wmJn6xGx/uPY3SSiaqJvswLFrq4iLmV/f0FPOtA2JQPT39em711FQgK0vcXloqBt+JiIiI6LojlwqQXVQBF5Ucg7qYv/PUEFzfxrzoRERERI3CIHor5emkxBv39sSPM4diYJAXtDo9Pjt4Dg98vgsJpy5Dz7yI1AzUtXDppUs1L1zK/OpERETUlhkC47f18IFKLjNbJqaHD+RSCU5lFiE1t6Qpm0dERETUqjCI3sqFtXfFpn9F4YMHIxHg5oirpRVYuD0Fj2/ai+OZ+fZuHpEJSxcuNcxWv3nhUr3e3u+AiIiIyPYEQcAvxzIAAHHh1VO5GLg7KhHdWZylnnCMs9GJiIiIGopB9DZAIpFgTC8/7Hh+GJ6L6Q4HhQwnsgsw/au9WLAtGTnF5fZuIlGNDAuXOjmJAXVPTzEljEwmBs6zs4ELF64H1q9cEWexFxcDFRUMrBMREVHrcyJDg0t5ZVDJpRje3bvWsrHXguwJTOlCRERE1GAMorchaoUMM2O6YNfzIzC+dwcAwLbTV/DA57vw6f5zqKhifgxqGQwLlzo7Xw+sOzmJrxUXAxkZpvnVMzKA/HxxUdPKSnFRUyIiIqKWatu1WeXDu3nDUSmvtezocF9IJEDKpQKkF5Q1RfOIiIiIWh0G0dsgXzc1lk/qje+eGozeAe4or9JhzV+nMemL3fjtXAYERhipBZLJxPzqNS1ceuXK9cB6Wpq4cKlGI75eWckc60RERNRy/HItiD6mZ82pXAx8XNToH+QBgAuMUhsjQLw1tbJSzA2p1QJVVeKFv04n3rIqCJxhQ0REFql92gK1an07umPLjEH49lA63kw4hcyiMvz358Po094Ts4eHo6u3q72bSNQocvn1xUsB8fq4qkq8fs7LE38CYm51pVIMxCuV4kMuF58bHnK5OANeIrHf+yEiIiI6l12Ms9nFkEsluC3U16J9YsP9cCAtHwnHMhE/OMTGLSRqJkpLxdkzwPWLeInE9HcDw3aJRLzoB65f/Bu21fR6bfXX53VLyxIRkV0wiN7GSSQSTOgfgDG9fPHejn/w8Z/nkZyeh/iNf+DOsI54PLobPBxV9m4mkVUYFi5VKMTnOp2Y7kV17U+8qkqcrGKYmAKI18aGQLpUej3IrlCYBtgNv/PaloiIiGzJMJt8UJd2cHNQWLRPbLgfFv10EgfS8pBbXIF2zry+pzbixtk0hhnnN/688XfDAKCm1839XhNBqHlg0JCA+o2/Wxrcv3FbTfUbbsctKbk+SLJG8J+IqBViEJ0AAI5KOV4c0w2Tojpg0Y+nsO1EBr4/fhG/nk3HtKiumNArGAoZs/9Q62QIiJsjCKZ3fJaWimlg9Prr14c3BtkVCjEor1BUn81uKENERETUUIYgelx43alcDAI9HdEzwA1HrxTi1xNZmDiwo62aR9R8yKTiBXlzY0lAv7bXq6osL2v4vabAtuGLg0uXxIGKoWxDZ8jfGOg3/Kwr4H9joL+2+usb0K9rPyKiemqGZxSyp0BPR3w4pR/+Op+HV7cex6ksDd794yS+O3oRs4aFITrYx95NJGpSEsn1tDA1MQTZdTqgvFycyGG4HgVMZ7PL5eJMdpWqeoDd8CAiIiIy53J+Kf6+XAiJRFwwtD7iIvxw9EohfjmWySA6kT01p9Qshpno7u6mA5H6Bulv/v3mQH9d+5lzY/C/pi8CGjpDvqb0PTX9buinggJxQGerVD1M30PUrDGITmbd0tkTPz07BBv3XcLbiadxqaAEz209gFs6euOZYWEI9nS2dxOJmo26gt96/fU1jCorgbIy04VMJRLTtDDMy05ERETmbDueBQAYEOxZ75QsseF+eGvbaew9n4vCMq3FqWCIqA1qCTO2GxLcryl9T111GQZvmZnmA9y1BfwbOkP+xln9taXqufn1uo7ZmIA+0/dQG8cgOtVIJpVgcnRH3NnXH8u3ncPn+1Lx18UcHPjyd0zoHYxpUV3houLFN1FdDLnUa6LXX5/JfmNe9hvvvjQE1JmXnYiIqO3adqz+qVwMuvg4o4uPM85lF+O3U9kY3zfA2s0jImo6TRnErWnGviXqk6rH3O96ff2+HLj52HXN5gcaPkPeXPoeQxsyM6/P2rckT39dx29s8J/IChhEpzq5qhWYN64HHhrUEQu3nsCus9n4KjkV205dwRPR3XBneEfIpPxQImoow+QBRQ3fSTUmL/vNAXbmZSciImqZcooqcOBCHgAgNqL+QXQAGBPhh5U7zyHhWCaD6ERETaElpGZp6Gx+c+l7qqrEn0VF1wee9cnTbyhz8+sNCajfuG998vTfmO6ntvob+/rNixlTs8cgOlmsk7cTPp02AL+dysHCH04g9Woxlvx2DN/8fQGzh4ejXwcvezeRqFUyzES3dl52cyljmJediIioedp+IhOCAPTu4IYAd4cG1REbLgbRd53JRmllFRyVHA4SEbV51pyxbZi17+Ji3YFlQ2fzG34agvv1+aLA8Lsls/kN6hNQNwzY09JMc+3XN+BvybGsOdu/DeNVE9XbraHeGNJ1KD7bcwErdpzB+atFmPHtXxje2Q8zh/RAezdHezeRqM2xdl52w2x2w3PDdcSNgXkiIiJqOgnXUrk0dBY6AIS3d0UHDwdczi/D72dyEBfhb63mERER2U5LSM3SkOD+jfvWJ09/XbPXG5u+x1x/35i+p67gfn3y9Nf1uiFw0Qxm7DOITg2ikEnx2LAQTIgMwFsJZ7Dx4AXsPp+JvWnZmNQ3BFP6d+HMFqJmpCF52QsLr5+fDeerCxcABwfxYZjJrlBc/+KciIiIrK+wVIuk81cBNCwfuoFEIkFcuB8+/jMVCccyGUQnIiKylvoG+g23mjs4NO3t4A3JzX9j2RvT91hS183HtiSVz419afhyoays9qBGE2CUkxrFw0mJ/02IwEPRHbFg6wnsS7uKzw+ex08nLuOpwaGIDQ2AlJE1omavrrzshrvfAPHcpdGI5zep9HqqGQcHQK02DazXloKGiIiILLPjVBaq9AK6+7qgk7dzo+oa01MMou84mY3KKj2Uci6WQkRE1Ga0hNQsNaXisTOGN8gqwtq7YuMTUUg4loVFP57ElcJSvJaYgq9TLmD28DBE+HvYu4lE1AiGc+zNX5ILAqDVijPYCwuBPHG9M2NAXi4HHB3FL4wNzxUKLm5KRERUH79YIZWLQd9AD3i7qJBTVIE953Nxa3efRtdJREREZDU3zkRvRou2MYxBViORSDCmpx92vjAMz8WEwlEhw8nsAjy+eS8WbEtGTnG5vZtIRFYmkYgBcgcHwNUV8PAQH87O4rlOqwVyc4ErV8T1UtLSgNRU4PJlcbtGA5SWiuWawRfLREREzU5JRRV+P5MDoHGpXAykUgliw30BANuuBeeJiIiIqHYMopPVqeQyzIzpjN+eH4G7e3cAAGw7fQUPfL4L6/afRUWVro4aiKilk8nEhUmdnAB3dzGw7u4uBtsBMXCelQVcvHg9sJ6aCmRkAPn5QFERUF5uuvgpERFRW7T7TA4qqvTo6OmIHv4uVqkzLlzMhb79RBZ0en6LTURERFQXpnMhm/F1U+OdSb0xdXAQXv3+BJKv5OOjv85g67FLmDm0B27t4gdJc8/DRERWI5GYz5NuSAlTVSWmhLl69fri3gqF+DAsZGp4LpczJQwREbUNCddmi4+JsN61c1QnT7g7KpBXUokDaXm4pZOXVeolIiIiaq0YRCeb69PRHd/NiMZ3h9PxZsIpZBWV4eVfDqNPe0/MGh6Gbt5u9m4iEdmRISXMzQtt63RiYL2yEigpuZ7uxRCIV6nEhUxvzLculzf/NVKIiIgsVVGlw85T2QCskw/dQCGTIqaHL74+dBkJxzIZRCciIiKqQ7OYx7dq1SoEBwdDrVYjKioK+/fvr7Hs8ePHMWHCBAQHB0MikWD58uW11v3GG29AIpFg1qxZJttHjBgBiURi8vjXv/5lhXdD5kgkEtwTGYDfXhiOp4d3hVImRXJ6HuL/70+8seMo8ksr7N1EImpm6koJU1ICZGeLKWFSU6+nhTGkhCkuZkoYIiJq2facy0VxRRV8XVXo08HdqnUb8qsnHMuEnildWhWOr4mIiKzP7kH0TZs2Yc6cOZg/fz4OHz6M3r17IzY2FtnZ2WbLl5aWolOnTnjjjTfg51f7bIwDBw7gww8/RK9evcy+Pn36dGRkZBgfS5YsafT7odo5KuV4YUw37Hx+BMaEtYcAYOvxi7j/8134v8P/QKvT27uJRNSMGVLCODgALi5iUN3TU1zUVKEA9HoxJUx6OnDhwvXA+oULYsC9sFAMvldWimWJiIiaM0Mql9hwP0il1r3VakjXdnBSypCpKcffVwqtWjfZD8fXREREtmH3IPqyZcswffp0xMfHIywsDKtXr4ajoyPWrl1rtvyAAQPw1ltvYeLEiVCpVDXWW1xcjMmTJ+Ojjz6Ch4eH2TKOjo7w8/MzPlxdXa3ynqhuHTwc8MGUvtg4PRo9fF1RUlmFlX+exEPrf8feNPMXeERENZFKxbQujo5iQN3DQ3w4OYmvVVYCubnApUumC5levizmYNdogLIyMTe7wMl4RETUDFTp9Eg8kQUAiLNiKhcDtUKGW0N9AFwP1lPLx/E1ERGRbdg1J3plZSUOHTqEuXPnGrdJpVLExMQgKSmpUXU//fTTGDt2LGJiYrBo0SKzZb788kusX78efn5+uPPOO/HKK6/A0dHRbNmKigpUVFxPOaLRaAAAWq0WWq22UW017N/YelqiyI4u+O6pKGw+mI5lv57FpcISPL/1AAYGemHmkO4I9nQGAOj0Ao5czsGRXAn0F7PRt4M3ZFaejdOa6HRak59kHvvJci25rwyLkarV4nNBEHOtV1WJwfP8fNOycrlYVqUyzbUuk1l2PGt9prfFcwIREV23PzUP+aVaeDgqMDDY0ybHiIvww49/ZyDhWAb+HdfdaguXkn20pPE1YJsxtk4voKhM3PdkbjlGaKs4bqyF9lreQy3zH9aK/WQ59pXl2FeWMfZTVZU4660xdTVyf7sG0XNzc6HT6eDr62uy3dfXF6dOnWpwvRs3bsThw4dx4MCBGss8+OCDCAoKQvv27fH333/j3//+N06fPo1vv/3WbPnFixdjwYIF1bZv37691guD+khMTLRKPS2RK4D/RADbrkixO0OC/ZeuYurGPRjqJ6CDo4CfLklRUCkBIMPnZ5PhrhRwT7Aevb04ZbQ2KSlt92+qPthPlmNfWa6xn+mlpaVWagkREbVECcfF2eGjwnwhl9nmBuIR3X2glEuRdrUUp7OKEOrHmcMtWUsaXwPWH2OnXJXg2zTDuBFYkVKAz07mc9xogcSUFHs3oUVgP1mOfWU59pVlEvfsaXQdjR1j2zWIbguXLl3Cs88+i8TERKgN0w7NePzxx42/9+zZE/7+/hg5ciTOnz+Pzp07Vys/d+5czJkzx/hco9EgMDAQo0ePbvRtalqtFomJiRg1ahQUCkWj6mrp7gGQmluC1386g93ncrA7w/ysgYJKCdaekWFRXG8M7+xrtkxbptNpkZKSiN69R0Ema9t/U7VhP1muLfeVTid+4W34KQhibnaZ7PqsdbVa/F38CNdi587Gf6YbZmMREVHbo9cL2HYtiG6LVC4Gzio5hnVth19PZiPhWCaD6FSNrcbXgHXH2NuOZ2FdUgpuDpUXVkqw7owMK+OCENvZrV51tgVanQ6JKSkY1bs3FJbedtkGsZ8sx76yHPvKMsZ+GjwYCrfGfY43doxt1yB6u3btIJPJkJWVZbI9KyurzkVNanLo0CFkZ2ejX79+xm06nQ6///473nvvPVRUVEBm5o8zKioKAHDu3DmzJ3mVSmU2R5xCobBa4NuadbVk3fzd8dljA/HbyWw89sVB6PQ1zxp498/TGN4lgLfo1UAmU7S5gGdDsJ8s1xb7SiYT863f6MaUMGVlQFGRuF0iEXOwA43/TOf5gIio7Uq+XIAsTQWcVXIM7tLOpseKi/A3BtFnxXSz6bHItlrS+Bqw3hhbpxfw+i+nqwXQAUAAIAHw+p/pGNPFg+PGGihkMgbxLMB+shz7ynLsK8so5PJGj5Ebu79dFxZVKpWIjIzEjh07jNv0ej127NiB6OjoBtU5cuRIHD16FMnJycZH//79MXnyZCQnJ5s9wQNAcnIyAMDf379BxyXrUytltQbQASC7uBwp6XlN1CIiIpFEIs46d3AAXFyuL2Tq6iouYtoSrVq1CsHBwVCr1YiKisL+/ftrLb9582aEhoZCrVajZ8+e+Pnnn01eFwQB8+bNg7+/PxwcHBATE4OzZ8+alMnLy8PkyZPh6uoKd3d3TJs2DcXFxSZltm3bhltuuQUuLi7w9vbGhAkTkJaWZpX3TETUXBkW+rwt1AcquW0H1jE9fCCTSnAqswhpuSU2PRbZVlsdX+9PzUNGYXmNrwsAMoq12J9eXGMZIiKiutg1iA4Ac+bMwUcffYTPPvsMJ0+exJNPPomSkhLEx8cDAKZMmWKyMEplZaXx5F1ZWYkrV64gOTkZ586dAwC4uLggIiLC5OHk5AQvLy9EREQAAM6fP4/XXnsNhw4dQlpaGrZu3YopU6Zg2LBh6NWrV9N3ApmVXVTzhdCNckssK0dEZGtSqRhgb2k2bdqEOXPmYP78+Th8+DB69+6N2NhYZGdnmy2/d+9eTJo0CdOmTcORI0cwfvx4jB8/HseOHTOWWbJkCd59912sXr0a+/btg5OTE2JjY1Fefv0ze/LkyTh+/DgSExPx448/4vfffze5HTw1NRXjxo3DbbfdhuTkZGzbtg25ubm45557bNcZRER2JgiCMYhuy1QuBu6OSkR38gJwPQ87tVxtcXxt6bgxu4SLthMRUcPZPYj+wAMPYOnSpZg3bx769OmD5ORkJCQkGBdDuXjxIjIyMozl09PT0bdvX/Tt2xcZGRlYunQp+vbti8cee8ziYyqVSvz6668YPXo0QkND8dxzz2HChAn44YcfrP7+qOF8XGrOuXejdk6WlSMiIvOWLVuG6dOnIz4+HmFhYVi9ejUcHR2xdu1as+VXrFiBuLg4vPDCC+jRowdee+019OvXD++99x4AMQC0fPlyvPzyyxg3bhx69eqFzz//HOnp6diyZQsA4OTJk0hISMDHH3+MqKgoDBkyBCtXrsTGjRuRnp4OQLyFXKfTYdGiRejcuTP69euH559/HsnJyY1eWZ2IqLk6mVGEi3mlUMmlGN7Nu0mOGXstWG8I3lPL1RbH15aOG32cmCqPiIgarlksLDpjxgzMmDHD7Gu7du0yeR4cHAxBqN/K2jfXERgYiN27d9erDmp6A0M84e+mRmZhudn8dgDQzkmF3u09m7RdREStSWVlJQ4dOmQyK00qlSImJgZJSUlm90lKSjJZCAwAYmNjjQHy1NRUZGZmIiYmxvi6m5sboqKikJSUhIkTJyIpKQnu7u7o37+/sUxMTAykUin27duHu+++G5GRkZBKpVi3bh0eeeQRFBcX44svvkBMTEyN+ewqKipQUVFhfG5YPEar1TY68G7YnwH8urGvLMN+slxb6quf/74CABjWtR2UUqFe77mh/XRbNy/MkwDJlwpwMbcI/m6tf5KKNf+mmtvfZVsbX9c1bpQA8HNWYGB756ZuGhERtSLNIohOZI5MKsH8O8Pw5PrDkABmL4icFAqUlwtwcmyB+ROIiJqB3Nxc6HQ64ww1A19fX5w6dcrsPpmZmWbLZ2ZmGl83bKutjI+Pj8nrcrkcnp6exjIhISHYvn077r//fjzxxBPQ6XSIjo6uln/9RosXL8aCBQuqbd++fTscHR1r3K8+EhMTrVJPW8C+sgz7yXJtoa++TpYBkMBXm4Gff05vUB0N6adgZxlSiyRY8fVvGOZfv6BqS2aNv6nS0lIrtIQaqrZxo2GUOH9YBy4qSkREjcIgOjVrcRH++OChfljwwwmTxWLaOSuhKa/ChYJivLXrGF4c0ROODKQTEbUqmZmZmD59OqZOnYpJkyahqKgI8+bNw7333ovExERIzCSgnzt3rskseY1Gg8DAQIwePRqurq6Nao9Wq0ViYiJGjRrV6JXdWzv2lWXYT5ZrK331T04JMpP2QC6VYNb9MXBzqN97bUw/ZbqlYXHCGVyWtMPttw+o174tkTX/pgx3PZH91DRu9HNWYP6wDojr4m6/xhERUavAIDo1e3ER/hgV5oekc9nY/sc+jB4aheguPth9JhuPfXYQ289dQoCrIyb36wIrTTIkImoz2rVrB5lMhqysLJPtWVlZ8PMzv6Cdn59freUNP7OysuDv729Spk+fPsYyNy9cWlVVhby8POP+q1atgpubG5YsWWIss379egQGBmLfvn245ZZbqrVNpVJBpVJV265QKKwWeLNmXa0d+8oy7CfLtfa+2nEmFwAQ3dkL7VwbfmHbkH66vVcAFiecwYG0fGgq9PByrv5Z2hpZ42+qNf9NtiSGceOoZbvwT24pZvfzwIxBQZyBTkREVmH3hUWJLCGTShAV4onIdgKiQjwhk0pwW6gvXr0rHACw7vBp7DyXjuJiOzeUiKiFUSqViIyMxI4dO4zb9Ho9duzYgejoaLP7REdHm5QHxNvhDeVDQkLg5+dnUkaj0WDfvn3GMtHR0SgoKMChQ4eMZXbu3Am9Xo+oqCgA4u3xUqnppYpMJjO2kYiotdl2bWHPMRH+dZS0vkBPR0QEuEIvAL+ezKp7B6JmSCaVwFktzhXs7qFiAJ2IiKyGQXRq0aZEB+PRwSEAgKV/pOBkdj54NyURUf3MmTMHH330ET777DOcPHkSTz75JEpKShAfHw8AmDJlisnCo88++ywSEhLw9ttv49SpU3j11Vdx8OBB4yJmEokEs2bNwqJFi7B161YcPXoUU6ZMQfv27TF+/HgAQI8ePRAXF4fp06dj//792LNnD2bMmIGJEyeiffv2AICxY8fiwIEDWLhwIc6ePYvDhw8jPj4eQUFB6Nu3b9N2EhGRjV0pKEPK5UJIJMCoMN+6d7CBuHDxTqBfrgXziYiIiEjEIDq1eP8d2wMxPXxRqdPj1Z0HkVNWgoICoJ6LzBMRtVkPPPAAli5dinnz5qFPnz5ITk5GQkKCcWHQixcvIiMjw1h+0KBB2LBhA9asWYPevXvj66+/xpYtWxAREWEs8+KLL2LmzJl4/PHHMWDAABQXFyMhIQFqtdpY5ssvv0RoaChGjhyJ22+/HUOGDMGaNWuMr992223YsGEDtmzZgr59+yIuLg4qlQoJCQlwcHBogp4hImo6hlnoA4I84e1in1QqcRFiEH3PuVxoyrV2aQMRERFRc8Sc6NTiyaQSvDupD+7/MAnHrmgw79cDWH7HIBQWKuHmBphZd46IiG4yY8YM40zym+3atavatvvuuw/33XdfjfVJJBIsXLgQCxcurLGMp6cnNmzYUGu7Jk6ciIkTJ9ZahoioNUg4LgbRYyPMr0fRFLr4uKCztxPO55Tgt1PZGNcnwG5tISIiImpOOBOdWgVHpRyfTB2A9m5qpF4tweu7D0Gh0iM/H2DaXCIiIiJqznKKKnAgLQ/A9dng9mLIx57AlC5ERERERgyiU6vh66rGJ48MgLNKjv1peVh14G+4ugooKAB0Onu3joiIiIjIvMQTWRAEoFcHNwS42zddlSGIv+t0DsoqeRFNREREBDCITq1MD39XrJrcDzKpBFuSr+Db0+fg4QEUFgJVVfZuHRERERFRdcZULuH2nYUOAOHtXRHg7oAyrQ67z+TYuzlEREREzQKD6NTqDO/mjdfGiYvbLd9xBvsyr8DLC9BoAC3XRyIiIiKiZqSwTIu953IB2D+VCyCuaWFox7bjTOlCREREBDCITq3Ug1Ed8cSwTgCA/3z7Ny6U5sHbGygqAior7dw4IiIiIqJrdp7KQpVeQDdfZ3T2drZ3cwAAY64F0X89mYXKKi4wRERERMQgOrVa/44LxZgIP1Tq9PjXlwdRLCmGnx9QUgKUl9u7dUREREREwC9Hxdnecc0glYtBv44e8HZRoai8CnvP59q7OURERER2xyA6tVpSqQTL7u+D3oHuKCjVYtpnByBRV8LfXwyil5XZu4VERERE1JaVVlYZ847HNoNULgZSqQSjw3wBMKULEREREcAgOrVyDkoZPp7SHwHuDki7WoonvjgIRxcdAgLE/OglJfZuIRERERG1VbtP56CiSo9ATweE+bvauzkmDHnRtx/Pgk4v2Lk1RERERPbFIDq1et4uKnwaPwAuajkOXsjHi1//DVdXAQEBgCCIedKJiIiIiJraL8fEWd5jIvwhkUjs3BpTt3TygpuDAldLKnEwLc/ezSEiIiKyKwbRqU3o6uuC1Q9FQi6VYGtKOt5JPAMXFyAgAJDJgMJCe7eQiIiIiNqSiioddp7KBgDENqN86AYKmRQxPcSULoZgPxEREVFbxSA6tRmDu7TD/+7uCQB4d+c5fH3oMpycxEC6UgkUFIgz04mIiIiIbG3vuasorqiCj4sKfQPd7d0cswwpXbYdz4TAC2UiIiJqwxhEpzbl/gGBePrWzgCAud/+jb3nc+HgIAbSHR2B/HxAr7dzI4mIiIio1Uu4Nrs7NtwPUmnzSuViMLRrOzgqZcgoLMffl3nrJhEREbVdDKJTm/PcqO64o5c/tDoB//riEM5lF0GtBtq3B1xdxRnpDKQTERERka1U6fTYfsKQD735pXIxUCtkuDXUBwBTuhAREVHbxiA6tTlSqQRL7+uNyCAPaMqrEP/pAeQWV0CpBPz9AQ8PMZCu09m7pURERETUGu1Py0N+qRbujgoMDPG0d3NqFXctX3vCsQymdCEiIqI2i0F0apPUChnWPByJjp6OuJRXhumfH0S5VgeFAvDzA7y8xMVGtVp7t5SIiIiIWptt12Z1j+rhC7mseQ/Jbg31gVIuRdrVUpzJKrZ3c4iIiIjsonlfsRHZkJezCuviB8DNQYEjFwvw3Fcp0OsFyGSAry/g7Q0UFQGVlfZuKRERERG1Fnq9gG3HswBcX7izOXNWyTGsazsA1/O4ExEREbU1DKJTm9bZ2xkfPhwJhUyCn45m4K3tpwEAUqkYRPf1BUpKgPJyOzeUiIiIiFqF5MsFyNSUw1klx+Au7ezdHIvEXkvp8suxDDu3hIiIiMg+GESnNu+WTl54c0IvAMAHu85j4/6LAMRAert2YnqX8nKgrMyerSQiIiKi1sCQyuXWUB+oFTI7t8YyMT18IZNKcCqzCGm5JfZuDhEREVGTYxCdCMA9/Trg2ZFdAQD/3XIMf5zNAQBIJGJ+9PbtxfzopaX2bCURERERtWSCICDhuBhENyzY2RJ4OClxSydxAdRtx5nShYiIiNqeZhFEX7VqFYKDg6FWqxEVFYX9+/fXWPb48eOYMGECgoODIZFIsHz58lrrfuONNyCRSDBr1iyT7eXl5Xj66afh5eUFZ2dnTJgwAVlZWVZ4N9RSzYrpirv7BkCnF/DU+sM4nVlkfM3dXQyk6/VAMddTIiIiIqIGOJVZhAtXS6GSSzGiu7e9m1MvcRH+AGD8EoCaL46viYiIrM/uQfRNmzZhzpw5mD9/Pg4fPozevXsjNjYW2dnZZsuXlpaiU6dOeOONN+DnV/vsjQMHDuDDDz9Er169qr02e/Zs/PDDD9i8eTN2796N9PR03HPPPVZ5T9QySSQSvDGhJwYGe6KoogqPfnoA2UXXk6G7uoqBdKkU0Gjs2FAiIiIiapF+uZbKZVg3bzip5HZuTf3EhvlCIgGOXCxARiHzHDZXHF8TERHZht2v3JYtW4bp06cjPj4eALB69Wr89NNPWLt2Lf7zn/9UKz9gwAAMGDAAAMy+blBcXIzJkyfjo48+wqJFi0xeKywsxCeffIINGzbgtttuAwCsW7cOPXr0wF9//YVbbrmlWn0VFRWoqKgwPtdci6JqtVpotdp6vmtThv0bW09r1xT9JAXw3qReuH/NfqRdLcW0Tw/gy0cHwEEp5qtUqQAfHyAzE7h6FXBzE1O+NDc6ndbkJ5nHfrIc+8pyer11Pqt4TiAian0M+dBbUioXAx9XNfp19MChC/nYfjwLUwcF27tJZEZLGV8DthtjC3oBAKDT6aHV6RpcT1tg6B/2U+3YT5ZjX1mOfWUZYz9VVYl5lhtTVyP3t2sQvbKyEocOHcLcuXON26RSKWJiYpCUlNSoup9++mmMHTsWMTEx1U7yhw4dglarRUxMjHFbaGgoOnbsiKSkJLMn+cWLF2PBggXVtm/fvh2Ojo6NaqtBYmKiVepp7Zqinx4KBN4plOHoFQ0eWpWI+G56SJthsLwuKSn8m7IE+8ly7CvLNfazqpSLMBARtSr/5BTjdFYR5FIJRvbwsXdzGiQu3A+HLuQj4Vgmg+jNUEsaXwO2G2NrNDIAEiRnZKCqIr3B9bQliSkp9m5Ci8B+shz7ynLsK8sk7tnT6DoaO8a2axA9NzcXOp0Ovr6+Jtt9fX1x6tSpBte7ceNGHD58GAcOHDD7emZmJpRKJdzd3asdNzPTfI6/uXPnYs6cOcbnGo0GgYGBGD16NFxdXRvcVkD8JiQxMRGjRo2CQqFoVF2tWVP3U0T/fDy87iD+zpPimCwE/4nrbvJ6ZSWQlSXmSHdzE9O8NBc6nRYpKYno3XsUZDL+TdWE/WQ59pXl8vK0SEtr/GeVhnmjiIhalW3HxfzQ0Z294O6otHNrGiYuwg+v/3wS+1KvIq+kEp5OLfN9tFYtaXwN2G6M/fGFJKC4CH38/TG6Bd710ZS0Oh0SU1IwqndvKGQyezen2WI/WY59ZTn2lWWM/TR4MBRubo2qq7FjbLunc7G2S5cu4dlnn0ViYiLUarXV6lWpVFCpVNW2KxQKqwV0rVlXa9ZU/XRLFx8sva83nt2YjE/2XECwtwseviXohnYAgYFiapfCQjGQ3tw+92QyBQOeFmA/WY59VTfDF2qN/azi+YCIqHVJOJYBQAxEt1SBno4Ib++K4+kaJJ7IxAMDOtq7SWRjthpfA7YbY0uu3UIsk0kZmLKQQiZjX1mA/WQ59pXl2FeWUcjljR4jN3Z/u86dbdeuHWQyWbVVu7Oysupc1KQmhw4dQnZ2Nvr16we5XA65XI7du3fj3XffhVwuh06ng5+fHyorK1FQUGC141LrNK5PAJ4f3Q0AMP/7Y/jttOmCPAoF4O8PeHqKgfSqKnu0koiIiIiasysFZUi5XAiJBBgV5lv3Ds2YIZ97wrGaZxiTfXB8TUREZDt2DaIrlUpERkZix44dxm16vR47duxAdHR0g+ocOXIkjh49iuTkZOOjf//+mDx5MpKTkyGTyRAZGQmFQmFy3NOnT+PixYsNPi61Xk/f2gX3RnaAXgBmfHkYJ9JNb/+QywFfX6BdO0CjEdO8EBEREREZbD8uBpz7B3nAx8W6s3mbmmEm/Z5zV6Ep5yLYzQnH10RERLZj93Quc+bMwdSpU9G/f38MHDgQy5cvR0lJiXE18SlTpiAgIACLFy8GIC6WcuLECePvV65cQXJyMpydndGlSxe4uLggIiLC5BhOTk7w8vIybndzc8O0adMwZ84ceHp6wtXVFTNnzkR0dHSNi55Q2yWRSPC/u3viSn4Zkv65ikc/PYAtTw+Gn9v1AZBMBvj4iD+zsgAnJ8DMnYlERERE1AYZZm3HtoL8zF19XdDZ2wnnc0rw26lsjOsTYO8m0Q04viYiIrINuwfRH3jgAeTk5GDevHnIzMxEnz59kJCQYFwM5eLFi5DesGJjeno6+vbta3y+dOlSLF26FMOHD8euXbssPu4777wDqVSKCRMmoKKiArGxsXj//fet9r6odVHKpVj9UCTu+WAPzueUYNpnB/DVE9FwUl3/LySVirPRJRIgOxvQ6wEHBzs2moiIiIjsLre4AgfS8gC07HzoN4qL8MOq384j4Vgmg+jNDMfXREREtmH3IDoAzJgxAzNmzDD72s0n7uDgYAiCUK/6zZ381Wo1Vq1ahVWrVtWrLmq73BwV+DR+IMav2oPj6Ro8839HsGZKf8iuLVwDiAF0Ly9xRnpmJiAIgKOjHRtNRERERHaVeCILegHoGeCGDh6t48IwLtwfq347j12nc1BWqYODkguiNSccXxMREVmfXXOiE7U0gZ6O+Ghqf6jkUuw4lY3XfjxRrYxEAnh4AO3bAzodUFxsh4YSERERUbNgSOXSWmahA0BEgCsC3B1QptXh97M59m4OERERkc0xiE5UT/06euCdB/oAAD7dm4Z1e1LNlnNzEwPpEom44CgRERERtS2FZVrsPZ8LoHUF0SUSifH9bLv2JQERERFRa8YgOlED3N7TH/8ZEwoAWPjjCSSeyDJbzsVFDKQrFEBBgZjehYiIiIjahp2nsqDVCejq44zO3s72bo5VGYLoiSezUFmlt3NriIiIiGyLQXSiBnpiWCdMGhgIQQCe+b8jOHq50Gw5JycxkO7gwEA6ERERUVvSGlO5GPTr6IF2zioUlVch6Z+r9m4OERERkU0xiE7UQBKJBAvHRWBo13Yo0+ow7bMDSC8oM1vWwUEMpDs7A/n5gJ6TdYiIiIhatdLKKuw+I+YLjw1vfUF0mVSC0eG+AK5/WUBERETUWjGITtQICpkUqyb3Q3dfF2QXVeDRTw+gqFxrtqxKBfj7i7nSCwrERUeJiIiIqHX6/UwOyrV6BHo6ILy9q72bYxNjDCldTmRCp+ftlkRERNR6MYhO1EiuagXWxg+At4sKpzKLMGPDEVTpzE81VyrFQLqHB1BYCFRVNXFjiYhqsGrVKgQHB0OtViMqKgr79++vtfzmzZsRGhoKtVqNnj174ueffzZ5XRAEzJs3D/7+/nBwcEBMTAzOnj1rUiYvLw+TJ0+Gq6sr3N3dMW3aNBQXF1erZ+nSpejWrRtUKhUCAgLw+uuvW+dNExHZ0C+GVC7hfpBIJHZujW3c0skLrmo5cosrcTAtz97NISIiIrIZBtGJrCDA3QGfTO0PtUKK3WdyMH/rcQg1JD+XywE/P8DLC9BoAK35ietERE1m06ZNmDNnDubPn4/Dhw+jd+/eiI2NRXZ2ttnye/fuxaRJkzBt2jQcOXIE48ePx/jx43Hs2DFjmSVLluDdd9/F6tWrsW/fPjg5OSE2Nhbl5eXGMpMnT8bx48eRmJiIH3/8Eb///jsef/xxk2M9++yz+Pjjj7F06VKcOnUKW7duxcCBA23TEUREVlJRpcPOk+JnaGvMh26gkEkRE3YtpctxpnQhIiKi1otBdCIr6dXBHSsm9oVEAny57yI+/iO1xrIyGeDrC3h7A0VFQEVFEzaUiOgmy5Ytw/Tp0xEfH4+wsDCsXr0ajo6OWLt2rdnyK1asQFxcHF544QX06NEDr732Gvr164f33nsPgDh7fPny5Xj55Zcxbtw49OrVC59//jnS09OxZcsWAMDJkyeRkJCAjz/+GFFRURgyZAhWrlyJjRs3Ij093Vjmgw8+wPfff4+77roLISEhiIyMxKhRo5qkX4iIGmrv+asoqqiCj4sKfQM97N0cm4q7lu9927HMGieREBEREbV0cns3gKg1iQ33w39v74FFP53E/345iUBPB8RF+JstK5UCPj5iQD07GxAEQK1u4gYTUZtXWVmJQ4cOYe7cucZtUqkUMTExSEpKMrtPUlIS5syZY7ItNjbWGCBPTU1FZmYmYmJijK+7ubkhKioKSUlJmDhxIpKSkuDu7o7+/fsby8TExEAqlWLfvn24++678cMPP6BTp0748ccfERcXB0EQEBMTgyVLlsDT09Ns2yoqKlBxwzeTGo0GAKDVaqFt5K0/hv0bW09bwL6yDPvJci2tr375W/wycFQPH+h0VU22Fo6hfwpKC+CgdIBcKodMKrPpMaND3OGolCG9sBxHLlxFzwA3mx7PWqz5N9VS/i6JiIio4RhEJ7KyaUNCcOFqKb746wJmbUrGRjcH9Al0N1tWIhHTushkQEaGGEh3cGja9hJR25abmwudTgdfX1+T7b6+vjh16pTZfTIzM82Wz8zMNL5u2FZbGR8fH5PX5XI5PD09jWX++ecfXLhwAZs3b8bnn38OnU6H2bNn495778XOnTvNtm3x4sVYsGBBte3bt2+Ho6Oj2X3qKzEx0Sr1tAXsK8uwnyzXEvpKJwA/pcgASOBWnIaff6757kRb2btrb5Mer5uLFMlXpXh/axLuDDK/NlBzZY2/qdLSUiu0hIiIiJozBtGJrEwikWD+nWG4nF+K307n4LHPDuC7pwYj0NN88EYiERcalUrFQLpeDzg5NXGjiYiaIb1ej4qKCnz++efo1q0bAOCTTz5BZGQkTp8+je7du1fbZ+7cuSaz5DUaDQIDAzF69Gi4uro2qj1arRaJiYkYNWoUFApFo+pq7dhXlmE/Wa4l9dW+1DyU/HUQ7g4KzLg/BgpZ02XQzNJk4dCfh9B7UG9ACmj1WlTpq1Clq4IAARJIoJApoJAqoFaooZKroJAoIJfJoZAqGrwAqr5DBpI3H8W5CmeMGTO4RSykas2/KcNdT0RERNR6MYhOZANymRQrH+yH+1Yn4WSGBo9+egBfPzkIbg41X6C7uV0PpBcVAS4uTdhgImqz2rVrB5lMhqysLJPtWVlZ8PMzvxien59freUNP7OysuDv729Spk+fPsYyNy9cWlVVhby8POP+/v7+kMvlxgA6APTo0QMAcPHiRbNBdJVKBZVKVW27QqGwWuDNmnW1duwry7CfLNcS+urXU7kAgJgwXziqq38e2UpFVQWKtEUAAJlcBplcBiWUJmX0gh5anRY6QQeNVgN9pThrXCEVA+lquRoOcgdjoF0hU0AqqftLgJhwfyi/PY60q6VIy69AN9+WcyFrjb+p5v43SURERI3HhUWJbMRZJcfaR/rD11WFs9nFePrLw9Dqar+91cUFCAgQ07sUFjZRQ4moTVMqlYiMjMSOHTuM2/R6PXbs2IHo6Giz+0RHR5uUB8Tb4Q3lQ0JC4OfnZ1JGo9Fg3759xjLR0dEoKCjAoUOHjGV27twJvV6PqKgoAMDgwYNRVVWF8+fPG8ucOXMGABAUFNSYt01EZBN6vYCEY2JKqjER5r+ItJWC8gJo9bXn5pZKpFDJVXBUOMJN7QYPBw+4q92hkqsgCAI0FRpcKbqCtII0pOanIjU/FZcKLyG3JBeaCg3KtGWo0ldVq9dFrcDQru0AwPj+iYiIiFoTBtGJbMjfzQGfTB0AR6UMf57LxcvfHYMgCLXu4+QkBtKVSqCgQMyTTkRkS3PmzMFHH32Ezz77DCdPnsSTTz6JkpISxMfHAwCmTJlisvDos88+i4SEBLz99ts4deoUXn31VRw8eBAzZswAIKa1mjVrFhYtWoStW7fi6NGjmDJlCtq3b4/x48cDEGeUx8XFYfr06di/fz/27NmDGTNmYOLEiWjfvj0AcaHRfv364dFHH8WRI0dw6NAhPPHEExg1apTJ7HQiouYi5XIBMjXlcFLKMLhLuyY7bpm2DPll+XBS1j8noEQipnhxUDjAVeUKTwdPeDp4wlHpCIlEgrKqMmSXZONS4SWkFaQhLV98ZJdko7C8EKXaUmh1WsSGi+tg/MIgOhEREbVCTOdCZGMRAW5YOakvpn9+EJsOXkJQO0c8NaJLrfs4OIiB9MxMID8fcHcXU70QEdnCAw88gJycHMybNw+ZmZno06cPEhISjAuDXrx4EdIbPoQGDRqEDRs24OWXX8ZLL72Erl27YsuWLYiIiDCWefHFF1FSUoLHH38cBQUFGDJkCBISEqBWq41lvvzyS8yYMQMjR46EVCrFhAkT8O677xpfl0ql+OGHHzBz5kwMGzYMTk5OGDNmDN5+++0m6BUiovpLOC4GkG8N9YFaIWuSYwqCgPyyfOgFPZRyZd07WEgulUMuNR0uGtLBaPVa5JbmQi/oxTzrUgW6tgdkEuBkhgYnMnPQuZ2rxelgiIiIiJo7BtGJmsDIHr6Yf2c45m89jiUJp9HR0xF39Gpf6z5qNdC+vRhILyhgIJ2IbGvGjBnGmeQ327VrV7Vt9913H+67774a65NIJFi4cCEWLlxYYxlPT09s2LCh1na1b98e33zzTa1liIiaA0EQsO3aLOy4JkzlUqotRUF5AZxVzkDtmQMbzZAORoXrud4FQYBWr4VUWoWegQ5IvliGrw+fxQMDvSCXXsuzrnAw5lhXSBWQSZvmCwYiIiIia2FIjqiJTB0UjPjBwQCAOV+l4NCF/Dr3USoBf3/Aw0MMpOt0tm0jERERETXM6awipF0thVIuxa3dfZrkmIIgIK8sDxKJpNqs8aYikUiglCnhqHDEyB5eAID95yvhoHAAABRXFiOjKAMXCi8Y86xfLLyI3JJck3QwdaU8JCIiIrInBtGJmtDLY8MQ08MXlVV6TP/8IC5eLa1zH4UC8PMDvLzExUa1ta8XRURERER28MtRcRb6sK7ecFI1TUC7uLIYRZVFcFY6N8nx6jKsuzsA4NiVEuQX6+GgcICLygUeDh7wdPCEs9IZUqkU5VXlyC7NxmXNZTHHekEaLhRcQFZxFgrKC1BSWYJKXSUD60RERNRsMIhO1IRkUgnendQHEQGuyCupxCOf7kdhad1RcZkM8PUFvL2BoiKgsrIJGktEREREFtt2vGlTuegFPfLK8iCTyJpNehRvFwUiAsTFTX8/U1jtdZlUBrVcDWelM9zV7vBw8ICb2g0KmQJVQhXyyvLEwHpBGlLzU/FP/j9IL0pHflk+iiqKUF5VDp2et2YSERFR02MQnaiJOSrl+GTqALR3U+OfnBI8sf4gKqvqTmAplYpBdF9foKQEKC9vgsYSERERUZ1Sc0twKrMIcqkEMT2aJpVLUUURiiqL4KR0apLjWWpEqBsAYNfpAovK35gOxk3tBk8HT7ir3aGWiwtRF1UUIb0oHRcLLxpnrV8qvITcklxoKjQo05ahSl9lq7dDREREBIBBdCK78HVV45NHBsBZJcdf/+ThP9/+bdHtqlIp0K6dmN6lvBwoK2uCxhIRERFRrRKuLSga3dkL7o5Kmx9Pp9chrywPKpkKUknzGtINv5bSJflCMQpKGxbclkgkUMgUJulgPBw84KR0gkQiQVlVGbJKsnCx8KIxz3pafhqyS7JRWF7IdDBERERkdc3riouoDenh74pVk/tBJpXg28NXsHLnOYv2k0jE/Ojt24v50UvrTqtORERERDaUcC2VS2x406Ry0VRoUKItgaPCsUmOVx8BHip09XWATgD+NJPSpTFuTAdjyLPuqnaFXCqHVq9FbmmuMR1MWn4aUgtSkVGUgfyyfBRXFqOiqgJ6oe47QImIiIhuxiA6kR0N7+aNhePCAQDLEs9gy5ErFu/r7i4G0vV6oLjYRg0kIiIiolqlF5Qh5VIBJBJgdJivzY+n1WmRV5YHtVwNiURi8+M1xIhrs9EtTenSGFKJFCq5Ck5KJ2OedXe1O1RyFQRBQGFFIdKL0o151lPzU3G58DLyyvJs3jYiIiJqPRhEJ7KzyVFBeHxYJwDAi1//jf2pll/Qu7qKgXSpFNBobNVCIiIiIqrJ9muz0CM7esDHVW3z42kqNCirKoOD3MHmx2ooQ170A6lFKKlo+oVAb0wH46pyNc5ad1SKM/dLtCXIKs4CAM5MJyIiIos0iyD6qlWrEBwcDLVajaioKOzfv7/GssePH8eECRMQHBwMiUSC5cuXVyvzwQcfoFevXnB1dYWrqyuio6Pxyy+/mJQZMWIEJBKJyeNf//qXtd8akUX+ExeKuHA/VOr0ePyLg0jNLbF4X2dnMZCuUAAFBQBTPxIRERE1nV+u5UOPi7B9KpdKXSXyyvLgIHdotrPQASC4nRodvVTQ6gTsPdd8ZnrIpXJjnnUHRfP9EqKxOL4mIiKyPrsH0Tdt2oQ5c+Zg/vz5OHz4MHr37o3Y2FhkZ2ebLV9aWopOnTrhjTfegJ+f+QvVDh064I033sChQ4dw8OBB3HbbbRg3bhyOHz9uUm769OnIyMgwPpYsWWL190dkCalUgnce6IPege4oKNUift1+5JdUWry/oyMQEAA4ODCQTkRERNRUcosrcCBNvIuwKfKhF5QXoEJX0ewDwBKJ5HpKl1MFdm1LW8PxNRERkW3YPYi+bNkyTJ8+HfHx8QgLC8Pq1avh6OiItWvXmi0/YMAAvPXWW5g4cSJUKpXZMnfeeSduv/12dO3aFd26dcPrr78OZ2dn/PXXXyblHB0d4efnZ3y4urpa/f0RWcpBKcPHU/ojwN0BaVdL8fgXB1Gutfz2V7VanJHu7CwG0omIiIjItn49kQW9AEQEuCLQ07aLfJZXlSO/LB9OCiebHsdahl8Loied16BCy5QpTYXjayIiItuQ2/PglZWVOHToEObOnWvcJpVKERMTg6SkJKscQ6fTYfPmzSgpKUF0dLTJa19++SXWr18PPz8/3HnnnXjllVfg6Gj+4reiogIVFRXG55prCai1Wi20Wm2j2mjYv7H1tHZtoZ/c1VJ89HBfPPDRfhxIy8fzXyVj2X09Lb5dVyoFvL2BqiqxjzQaLZydxe1UnU6nNflJNWNfWU6vt85nVWv+rCMiai0SruVDj2uCWej5Zfmo0lfBReVi82NZQ6i/A3xdFcjSaLHvHw2GXQuqk+20pPE1YLsxtqAXrrVVD62u6XPytySG/mE/1Y79ZDn2leXYV5Yx9lNVFWDnMbZdg+i5ubnQ6XTw9TVdxd7X1xenTp1qVN1Hjx5FdHQ0ysvL4ezsjO+++w5hYWHG1x988EEEBQWhffv2+Pvvv/Hvf/8bp0+fxrfffmu2vsWLF2PBggXVtm/fvr3WC4P6SExMtEo9rV1b6KeHQyRYfUqKH49mQpuXjts7Nmz2zrlzrb+vrCElhf1kKfaV5Rr7WVVaWmqllhARkS0Ulmmx51wuACAuwt+mxyrVlqKwvBDOSmebHseaJBIJRoS6Y9P+HOw6XcAgehNoSeNrwHZjbI1GBkCC5IwMVFWkN7ietiQxJcXeTWgR2E+WY19Zjn1lmcQ9expdR2PH2HYNottS9+7dkZycjMLCQnz99deYOnUqdu/ebTzRP/7448ayPXv2hL+/P0aOHInz58+jc+fO1eqbO3cu5syZY3yu0WgQGBiI0aNHN/o2Na1Wi8TERIwaNQoKhaJRdbVmbamfbgfQ4dBlvLTlBLZdkeK2qJ64p2+Axfsb+iomZhS0WgWKioCiIqCqClCpxNzpzXgtqiaj02mRkpKI3r1HQSZr3X9TjcW+slxenhZpaY3/rDLMxiIioubpt1PZ0OoEdPFxRhcf2wW3BUFAflk+9NBD0cLOwcO7i0H0PWc10Or0UMh4e2RLZe3xNWC7MfbHF5KA4iL08ffH6Ca4S6Ql0+p0SExJwajevaGQyezdnGaL/WQ59pXl2FeWMfbT4MFQuLk1qq7GjrHtGkRv164dZDIZsrKyTLZnZWXVuKiJpZRKJbp06QIAiIyMxIEDB7BixQp8+OGHZstHRUUBAM6dO2f2JK9SqczmiFMoFFYL6FqzrtasrfTTg7eE4HJBBd7fdR4vf38CgV7OGNS5Xb3qUCoVcHJSwN0dqKgAiovFfOkaDSCXi8H0NtCVdZLJFAwMW4h9VTdD+qTGfla1hc85IqKWLOFY06RyKdGWiLPQVS1nFrpBzw5O8HCSI7+kCocvFCOqE3Nk21JLGl8DthtjS6TibCGZTMrAlIUUMhn7ygLsJ8uxryzHvrKMQi5v9Bi5sfvbdSqAUqlEZGQkduzYYdym1+uxY8eOavnVGkuv15vkW7tZcnIyAMDf37a3YhLVx/Oju+OOXv7Q6gT864tDOJdd3OC6VCrAywsIDgaCggAXF6CsDMjPF38KgvXaTUQNU1VVhV9//RUffvghioqKAAAZGRkoLm74/30iIrK+skoddp3JBgDERdguiK4X9Mgvy4dUKoVc2vJuIpZJJRjeTZw1tvtUgX0b0wZwfE1ERGQ7dr8SmzNnDqZOnYr+/ftj4MCBWL58OUpKShAfHw8AmDJlCgICArB48WIA4mIpJ06cMP5+5coVJCcnw9nZ2fjN+Ny5czFmzBh07NgRRUVF2LBhA3bt2oVt27YBAM6fP48NGzbg9ttvh5eXF/7++2/Mnj0bw4YNQ69evezQC0TmSaUSLL2vNzIKy3HoQj7iP92PLU8Nhpdz9RkblpLJxAC6s/P12emFhWIwXaEQZ6fL7f7JQNT2XLhwAXFxcbh48SIqKipwyy23AACWL18OQRCwevVqO7eQiIgMdp/JRrlWjw4eDghvb7vZ1cWVxdBUaOCmbtzty/Y0PNQdW45cxe4zhXguToBMypyCtsTxNRERkW3YPVT2wAMPICcnB/PmzUNmZib69OmDhIQE42IoFy9ehFR6fcJ8eno6+vbta3y+dOlSLF26FMOHD8euXbsAANnZ2ZgyZQoyMjLg5uaGXr16Ydu2bRg1ahQA8Rv6X3/91XhBERgYiAkTJuDll19uujdOZCG1QoY1D0fi7vf34mJeKR77/CD+b/otUCsad7uPRAKo1eLD3R0oLRWD6cXFgF4vBtPVauu8ByKq27PPPov+/fsjJSUFXl5exu133HEHZs2aZb+GERFRNTemcpHYaKEZnV6HvNI8KGQKSCUtN5d4vyBnuKhlyC+pwtHLJejTseWlpWlJOL4mIiKyDbsH0QFgxowZmDFjhtnXDCdug+DgYAh15J345JNPan09MDAQu3fvrlcbiezJy1mFdfEDcM/7e3HkYgGe+yoFKyf1hdRKM3nkcsDV9XqKl+JiMW96Xh6gVIoBdaboIrKtP/74A3v37oVSqTTZ3rFjR1y5csVOrSIioptVVumx46TtU7kUVRahWFsMd7W7zY7RFBQyKQZ3dUXC0XzsPl3AIHoT4PiaiIjI+lrulAaiNqaztzNWPxQJhUyCn45m4K3tp61+DIkEcHQEfHzEvOmBgWIQvahIXJC0lrSHRNRIer0eOp2u2vb09HS4uLjYoUVERGTO3vO5KKqogreLCv06etjkGFX6KlwtvQqVTNWiZ6EbjOjuDgDYdaqgzoAtERERUXPU8q/IiNqQ6M5eeOMeMa/gB7vOY+P+izY7lkIBuLmJwfSgIHFR0qoqcXZ6SYmY8oWIrGf06NFYvny58bkhPcD//vc/3H777XZqFRER3cyQyiU23NdqdwXerLC8EGVVZXBUONqk/qYW1ckVaoUUWRotTmWU2bs5RERERPXGIDpRCzMhsgOeGdkVAPDfLcfwx9kcmx5PIgGcnABfXzGYHhAgpn/RaMTZ6ZWVNj08UZvx9ttvY8+ePQgLC0N5eTkee+wxAEBGRgbefPNNO7eOiIgAQKcXsP1EFgAgLtzfJseo1FUivywfDnIHm+Vbb2oqhRTRncUFWHefLrBvY4iIiIgagEF0ohZodkxX3N03ADq9gKfWH8aZrKImOa5SCXh4iMH0jh3F3ysqxNnppaUA784largOHTogJSUFL730EmbPno1evcS7Tv744w/4+PjYuXVERAQAB9LykFdSCTcHBaI6edrkGJpyDcp15XBQONikfnsZHuoGgCldiIiIqGVqVBC9srISp0+fRlVVlbXaQ0QWkEgkeGNCTwwM9kRRRRXi1x1AdlF5kx1fKgWcnQF/fyA4WPwpkYgz0wsLAa22yZpC1KrI5XI89NBDWLJkCZYtWwYAcHBoXUEUIqKWzJDKJaaHLxQy689HqqiqQF5ZXqtJ43KjwV3coJBJcDGvAmm5TXfd2tJwjE1ERNQ8NejKr7S0FNOmTYOjoyPCw8Nx8aKYl3nmzJl44403rNpAIjJPJZfhw4cjEdLOCVcKyjD9s4Moq9RBpxewLzUPh3Il2JeaB53etjN9VCoxX3pwsDg73dUVKCsD8vPFn5xoRGS59PR0fPXVV3jvvfewevVqAMDq1avx7rvv2rllRESk1wvYdlwMoo+J8LPJMQrKC1Cpr4RarrZJ/fbkpJJhQIi4UPauU4V2bk3zwzG2dej0AorLxS8gTudX2HwsREREbUeDguhz585FSkoKdu3aBbX6+gVeTEwMNm3aZLXGEVHtPJyUWPfIAHg4KpByuRCTPkrC4Dd24qG1B/H5WRkeWnsQQ97ciYRjGTZvi0wGuLiIOdODg8Uc6oAYTC8qEhclJaKaffrppwgJCcG0adOwdOlSrFq1CgCwatUqkwVHiYjIPv6+UoiMwnI4KmUY0rWd1esv05ahoLwAzkpnq9fdXIzo7g4A2MW86NVwjN14CccyMOTNnfgntxQA8M7hfAz59DgSzhXYt2FERNQqNCiIvmXLFrz33nsYMmSIyWI34eHhOH/+vNUaR0R1C27nhDVT+kMulSD5UiEyNaa3x2YWluPJ9YebJJBuoFYD7dpdz53u6AiUlHB2OlFtXnnlFcybNw+FhYVIS0vD0aNHAQBHjx7FP//8Y+fWERGRIZXLraE+UCtkVq1bEATkl+VDp9dBKVNate7mZEg3N8gkwNmsMlzJr7B3c5oVjrEbJ+FYBp5cfxgZhTeNhYq1ePLnVAbSiYio0RoURM/JyTG7yFlJSUmrWUGeqCXp19EDziq52dcM8eoFP5xo8tsZ5XIxvUuHDuLsdG9vMYCenw8UFwM6XZM2h6hZKy0txcSJEyGVcs1vIqLmRhAE44SEuHDrp3Ip1ZaisKIQzqrWOwsdANwd5egTJL7H3ZyNboJj7IbT6QUs+OEEzI10jGOh3y8ztQsRETVKg0bq/fv3x08//WR8bjipf/zxx4iOjrZOy4jIYvtT81BQVvNqngKAjMJy7E/Na7pG3UAiARwcAB8fcXZ6YKCYS72oSFyMtIITkYgwbdo0bN682d7NICIiM05nFSHtaimUciluDa0e6GwMQRCQX54PAJBLzU+KaE2MKV2YF90Ex9gNtz81r9oM9BsJADKKtdifXtx0jSIiolanQVdp//vf/zBmzBicOHECVVVVWLFiBU6cOIG9e/di9+7d1m4jEdUhu6jmi8YbXbhagujOXjZuTe0UCsDN7foCpEVFgEYjpntRqcRgOyfiUlu0ePFi3HHHHUhISEDPnj0hXMt79NJLL0GpVGLZsmV2biERUdtlSOUyrGu7Gu/+a6jiymJoKjRwUbpYtd7malg3N7y97TKOXSlBTpEW3i4KezepWeAYu+EsHQtll9Q86YiIiKguDQpVDRkyBCkpKaiqqkLPnj2xfft2+Pj4ICkpCZGRkdZuIxHVwcdFXXchAHO/PYrxq/Zg6bbTSDp/FRVV9sunIpGIudJ9fcXZ6R06iAF2jUacnV5ZabemEdnF4sWLsW3bNmRlZeHo0aP4+++/AQB///03kpOT7ds4IqI2zhBEj7VyKhe9oEdeWR5kEhlkUuvmWW+uvF2VCA9wBAD8zpQuRhxjN5ylYyEfJ35hQ0REDVfvaRRarRZPPPEEXnnlFXz00Ue2aBMR1dPAEE/4u6mRWVhuNhcgAMikEuj0ApIvFSD5UgHe++0cHBQyDAzxxJAu7TC4SzuE+rlAKm36nItKpfhwdQVKS8XZ6UVFYt50tVp8cHY6tXZvv/021q5di0ceeQQAoNFo4Obmhh9//BGurq72bRwRURuWlluCU5lFkEkliOnha9W6iyqKUFRZBHe1u1Xrbe5uDXXH8Sul2H26ABP6e9u7OXbHMXbj1DUWkgDwc1ZgYPvWveYAERHZVr3DUgqFAt98840t2kJEDSSTSjD/zjAA4kXijSTXHqse7Iu9/7kNb93bC+P7tEc7ZxXKtDrsPpOD138+idvf/QMD//crnvm/I/jqwCVcKShr6rcBqRRwdgb8/cXZ6QEB4rbCQvGh5R2Y1IqpVCoMHjzY3s0gIqKbJBwXZ6FHd/KCh5PSavXq9DrkleVBKVNCKmlbswWGXcuLfuRCMQpLq+zbmGaAY+zGqWssBADzh3WAzA6ThYiIqPVo0NXa+PHjsWXLFis3hYgaIy7CHx881A9+bqa3M/q5qfHBQ/0QF+GP9u4OuK9/IJZP7IsD/x2JhFlD8fLYHhjR3RsOChlyiyuxNSUdL37zNwa/sRO3Ld2FV7YcQ8KxTBTWsnCpLahUgIeHGEzv2FGcpV5eDuTlibnUhZqm3BO1UM8++yxWrlxp72YQEdFNjKlcIqybykVToUGJtgROCier1tsSdPBQoauPA3QC8MdZLjAKcIzdWDWOhZwV+OD2EMR1cbdPw4iIqNVo0Ko4Xbt2xcKFC7Fnzx5ERkbCycn0wu+ZZ56xSuOIqH7iIvwxKswPSeeysf2PfRg9NArRXXzMzrqQSCQI9XNFqJ8rHhvaCZVVehy5mI8953Lxx7lcpFwqwD+5JfgntwRf/HUBUgnQq4O7MfVLvyB3qOS2z90pkwEuLuKjvFxcgLSgQHzIZGJedbl11/cisov9+/dj586d+PHHHxEeHm7cPnnyZCgUCnz77bd2bB0RUduUUViG5EsFkEiA2DDrpXLR6rTIK8uDWq6GRNI2Z8cOD3XD2ewy7D5VgDt623fh++aAY+zGM4yFRi3bhX9ySzG7nwdmDAriDHQiIrKKBoWePvnkE7i7u+PQoUM4dOiQyWsSiYQneCI7kkkliArxxNWTAqJCPC2+aFTKpYjq5IWoTl6YM7o7NOVa/HX+qjGo/k9Oid3zqRvyo7u7i8H0wkLxZ1UV4OAgvtZGx6HUCri7u+Oee+4xPtdey1/k5uYGhcL2C2GtWrUKb731FjIzM9G7d2+sXLkSAwcOrLH85s2b8corryAtLQ1du3bFm2++idtvv934uiAImD9/Pj766CMUFBRg8ODB+OCDD9C1a1djmby8PMycORM//PADpFIpJkyYgBUrVsDZuXrO0nPnzqFv376QyWQoKCiw6nsnIqrJ9uNZAIB+HT3g42rZ4oWW0FRoUFZVBg+1h9XqbGlGhLrj498zsT+1CCUVOjip2sbCqjXhGNs6ZFIJnNVimKO7h4oBdCIispoGBdFTU1Ot3Q4iamZc1QqMDvfD6HDx1uX0gjLsOZeLPedy8ee5q8gtrsDuMznYfSYHAODlpMTgLu3EoHrXdghwd7BZ22QyMb2LYXZ6cbEYUM/PBxQKcXa6rG2Pw6gFWrdunclzjUaDL7/8Eu+//77NFxbdtGkT5syZg9WrVyMqKgrLly9HbGwsTp8+DR8fn2rl9+7di0mTJmHx4sW44447sGHDBowfPx6HDx9GREQEAGDJkiV499138dlnnyEkJASvvPIKYmNjceLECajVYiBq8uTJyMjIQGJiIrRaLeLj4/H4449jw4YNJsfTarWYNGkShg4dir1799q0L4iIbvTLsQwAwBgrpnKp1FUirywPDnKHNjsLHQBC2qnR0VOFi3kV2HtOg1HhbfcLBYBjbCIiouau0SvYCIIAgcmJiVq9uvKpXy0xzad+axPkU5dIxBno3t5AcDAQGCg+LyoS072Ul9vksEStzrJlyzB9+nTEx8cjLCwMq1evhqOjI9auXWu2/IoVKxAXF4cXXngBPXr0wGuvvYZ+/frhvffeAyBeGyxfvhwvv/wyxo0bh169euHzzz9Henq6Md/ryZMnkZCQgI8//hhRUVEYMmQIVq5ciY0bNyI9Pd3keC+//DJCQ0Nx//3327QfiIhudLW4AvtT8wAAseHWC6IXlBegQlcBB4XtJhy0BBKJBMND3QEAu08X2LUtzQ3H2ERERM1PgzMJf/7553jrrbdw9uxZAEC3bt3wwgsv4OGHH7Za44ioebIkn3pqbglSb8in3rODO4baMJ+6XA64uYkz1MvKxEC6RiMuRKpUirPTpY3+2pDIuvr164cdO3bAw8MDffv2NZmRqNPpAABDhw6FTCbD4cOHbdKGyspKHDp0CHPnzjVuk0qliImJQVJSktl9kpKSMGfOHJNtsbGxxgB5amoqMjMzERMTY3zdzc0NUVFRSEpKwsSJE5GUlAR3d3f079/fWCYmJgZSqRT79u3D3XffDQDYuXMnNm/ejOTkZIvywldUVKCiosL4XKPRABBnsxtS5DSUYf/G1tMWsK8sw36ynD36atuxdOgFIMzfBX4uCqscu6KqAleLrkItU0NXpbNCK00Z6rRF3bYwtIsLvtibhaRzGpSWaaFSNN3FmqGPqrRVkEoad1xr/V1yjE1ERNR8NSiIvmzZMrzyyiuYMWMGBg8eDAD4888/8a9//Qu5ubmYPXu2VRtJRM2bJfnUUy4VIOVaPnW1QoqBIV7GoLo186lLJGLA3NER8PS8vhBpYeH1mesqlVUORdRo48aNg+raH+S4ceNMgugVFRX4+++/MXbsWGMZW8jNzYVOp4Ovr+mCeb6+vjh16pTZfTIzM82Wz8zMNL5u2FZbmZtTxcjlcnh6ehrLXL16FY888gjWr19vcUqbxYsXY8GCBdW2b9++HY6OjhbVUZfExESr1NMWsK8sw36yXFP21fqTUgBSBMsL8PPPPzfZca0hZW+KvZtgEUEA3JUyFFTqsfGnFPT0bPrZ17/++muj6ygtLW10HRxjExERNW8NCqKvXLkSH3zwAaZMmWLcdtdddyE8PByvvvoqT/BEbZwl+dR/P5OD322cT12hEBchNcxO12jEGeolJeIipE2wViNRrebPn2/8/dVXXzV5TaPRYPHixfjPf/5j85zozdX06dPx4IMPYtiwYRbvM3fuXJNZ8hqNBoGBgRg9enSj+1Gr1SIxMRGjRo1qksVeWzL2lWXYT5Zr6r4qKtfi+f27AAiYOX4ouvhUX/C4vsq0ZbhUeAkOCgcoZLZ5D7oqHVL2pqD3oN6QWfmuP1sZpU3H5kO5uKxoh6nDApvsuCXlJTi9/zRiYmKgVCobVZfhrqfG4BibiIioeWtQED0jIwODBg2qtn3QoEHIyMhodKOIqHUx5FO/r38gBEHA6awi/Hk2F3+ey8W+f/KM+dS3poh5kEPaOYkB9S7tEN3ZC24OjRtoSqWAk5P48PK6PjvdMN6pquJCpGR/nTp1woEDB+Dl5WWyvaCgAH369ME///xjk+O2a9cOMpkMWVlZJtuzsrLg52c+B7Cfn1+t5Q0/s7Ky4O/vb1KmT58+xjLZ2dkmdVRVVSEvL8+4/86dO7F161YsXboUgJgjVq/XQy6XY82aNXj00UertU2lUpmdua9QKKwWeLNmXa0d+8oy7CfLNVVf/XE8G1qdgM7eTugR0PgFLwVBQFFZESRyCdQqtRVaWDuZXNZigugjwjyw+VAu9pzTQJBIIZc1zWKrhv6RK+SN/puyxt8kx9g2oNNdv9Bvw4v4EhGRdTQo+VuXLl3w1VdfVdu+adMmdO3atdGNIqLWy5BP/bGhnfBp/ECkzB+NTY/fgpm3dUHfju6QSmDMpf6v9YfQd+F2jFu1B0u3nUbS+auoaGSOT6US8PAAgoLEhUgBoLhYnKlOZE9paWnGPOg3qqysxOXLl212XKVSicjISOzYscO4Ta/XY8eOHYiOjja7T3R0tEl5QEyxYCgfEhICPz8/kzIajQb79u0zlomOjkZBQQEOHTpkLLNz507o9XpERUUBEHOvJycnGx8LFy6Ei4sLkpOTjTnTiYhsIeGYmFYqLsI6C4qWaEtQWF4IZ2XjZ7S3Nr06OMHDSY6ich0OXyiyd3PshmNsG5DLxYv8ggJxoaT8fHEWTWkpUFEhBtiJiIgs1KCZ6AsWLMADDzyA33//3Zivbc+ePdixY4fZEz8RUU1uzKf+3OjuKCzTYt8/V/HnOXGmuq3yqRtmpwOAv794XV1QIC5Oyokq1JS2bt1q/H3btm1wc3MDcD2/6pIlSxASEmLTNsyZMwdTp05F//79MXDgQCxfvhwlJSWIj48HAEyZMgUBAQFYvHgxAODZZ5/F8OHD8fbbb2Ps2LHYuHEjDh48iDVr1gAQvyybNWsWFi1ahK5duyIkJASvvPIK2rdvj/HjxwMAevTogbi4OEyfPh2rV6+GVqvFjBkzMHHiRLRv395Y5kYHDx6EVCpFRESETfuDiNq2skoddp0WU86NifCvo3Td9IIe+WX5kEqlkEsbNPxq1WRSCYZ1c8P3R65i9+lCDOzUNlOYcYxtAz4+QIiPGCw3PCoqrgfQKyrE2eoGcrn4kErFn7xVlYiIbtCgq7gJEyZg3759eOedd7BlyxYA4kB3//796Nu3rzXbR0RtjJuD+Xzqf17LqZ5bXGmVfOo6vYB9qXk4lCuBV14e+gT44GquBPn5gIsL86VT0zEElSUSCaZOnVrt9d9++w3vvPOOTdvwwAMPICcnB/PmzUNmZib69OmDhIQE48KgFy9ehFR6/ea1QYMGYcOGDXj55Zfx0ksvoWvXrtiyZYtJcPvFF19ESUkJHn/8cRQUFGDIkCFISEiAWn09jcGXX36JGTNmYOTIkZBKpZgwYQLeffddm75XIqK67D6TgzKtDgHuDghv3/iAbnFlMTQVGriq2mZw2BLDu7tfC6IXYE5sB8istOB8S8IxtnXo9AKKy8UZ5qezizEqIgAycznvDaleDA+tFqisBMrLxefl5YBeL5aVSK4H1WWy64F2IiJqUxr8yR8ZGYn169fj0KFDOHToENavX9/gk/uqVasQHBwMtVqNqKgo7N+/v8ayx48fx4QJExAcHAyJRILly5dXK/PBBx+gV69ecHV1haurK6Kjo/HLL7+YlCkvL8fTTz8NLy8vODs7Y8KECdXyuxKR/Rnyqa+Y2BcH/huDhFlD8fLYHv/P3r3HRV3l/wN/feYODDPDcEdAwPv9goqkeSm8ZLlZ1lZraW6rW6tt5X5rs63M2v1ZrZVdTKtNrS272KbdVVLRMgTF+w0v4J2LMMxwh2Hm8/vj44wioAMMzACv5+PBA2fmzGfOvAXO57znfN4HY3oEw0cpd9ZTf+p/+zHi5c0YuzgFz607iPUHc2GpsNZ7zPUHczDylc24f8UufHxcjvtX7MK4tzbjoDkHwcFSzfRLi4CJWpzdbofdbkd0dDTy8/Odt81mMwAgIyMDt912W4v3Y+7cuTh9+jSqqqqQlpbmLKkCACkpKVi1alWt9nfffTcyMzNRVVWFgwcPYtKkSbUeFwQBL774InJzc1FZWYmff/4Z3bt3r9XGaDRi9erVKCkpgcViwYoVK6DVNlzq4MEHH3TGhYiopWw4dLmUi9DMy9NsdhtM5SYoZArIZVzV2pD4GC20ajlMZTU4eL7M093xGHfNsTvq/Npxjp9VIJ3Iv/HzSYx8ZTPWH6ynprxcDqjV0qWpej0QFARERACxsZe/OncGOnWSVrT7+krPq66WSsIUFUlfFos0eaiokBLxotiK75iIiFpTk5LoP/74IzZs2FDn/g0bNtQZTK/niy++wLx587BgwQLs3r0bAwYMwIQJE+psNuZQXl6OuLg4vPzyyw1ueBYZGYmXX34ZGRkZ2LVrF2666SbcfvvtOHTokLPNE088ge+++w5r1qzB1q1bceHCBdx5552N6jsRtS531FNffzAHj3yyGzmWylrHzrVUYu5nu7E7PwedOknnv2bz5QUoRC0tOzsbQUFBnu4GEVGHVl1jx89HpMSfO+qhl1aXotRaCj+VX7OP1Z4p5TKM7C6t1E85avZsZzzEXXPsjjq/vtY5/iOf7K4/kV4fx6pzjQbQagGDQUqwR0YCcXGXE+zR0VLSPTBQSsYD0up1i+Vygr24WEqwO1a3M8FORNSmNSmJ/vTTT9e7+Zkoinj66acbdazXX38ds2bNwsyZM9G7d28sX74cvr6+WLFiRb3thw4din//+9+49957oXYMVleZPHkyJk2ahG7duqF79+7417/+Ba1Wix07dgAALBYLPvzwQ7z++uu46aabEB8fj5UrV+K3335ztiEi7+eop/638T2w9i8jsOf58Xj/gXhMT+yMuGA/2EU4a6nf98EO9H9hA/76+V7Ud/rquO/F7w9D6y8iKko6bzabpQUnRERE1P6lZhWipLIGwf5qxEcHNOtYNfYaFJQXQC1XQyaw9MP1jO5hAABszbRA7IDJRnfNsTvi/NpmF7Hwu8PXPMdf+N1h2OzN/LkSBKnmo4+PVP8xIEBapR4dfTm5HhMDREVJCXaDQWpvt1/e4NSRYC8pubzBaT3/70RE5H2aVBP9+PHj6N27d537e/bsiRMnTrh8nOrqamRkZGD+/PnO+2QyGZKSkpCamtqUrtVhs9mwZs0alJWVITExEYB0abzVakVSUlKtvkdHRyM1NRXDhw+vc5yqqipUVVU5bxcXFwMArFYrrNb6S0a4yvH85h6nvWOcXNdRY+WrAMZ2D8TY7oEAeiDHUonfThZi+8lC/HbShMKyaqDe02uJCCDHUonUE/lIiDUiJERaiFJYKJ3/+nXgRWQ2m7XWd2qY3e6e37+O9vtLROQNHKtVx/cObdLG5VcqqSpBRU0FAjTNS8Z3FAlxOmiUMuRaqpGZW4Ge4b6e7lKrcsccuy3NrwH3zbHTsk11VqBf6epz/BbjSLI7NldylKiz2S7XYHd8r6qSVqhXVUnJ9Csvf3XUXW+F+uvWSwl8KxP518Q4uY6xch1j5RpnnBz7VzTnWM18fpOS6Hq9HllZWYiJial1/4kTJ+DXiCxTQUEBbDabc+Myh9DQUBw9erQpXXM6cOAAEhMTUVlZCa1Wi7Vr1zpPSnJzc6FSqWAwGOq8bm5ubr3HW7RoERYuXFjn/o0bN8LX1z0neMnJyW45TnvHOLmOsQJ8ACT5ATf3AzadF/Dd2evXI934SxoKj3S8FVCu2LePP1Ouau7vXzkL8xMRtSqbXcTGQ+4p5WK1WVFYXggfhU+z66p3FBqlDMO76JBy1IyUo+YOl0R3xxy7Lc2vAffNsTMKBADXP8d/5/t0HI6wI0Qj5bvpsuR9+zzdhTaBcXIdY+U6xso1ydu3N/sYzZ1jNymJfvvtt+Pxxx/H2rVr0aVLFwDS4P63v/0Nv/vd75rVIXfp0aMH9u7dC4vFgq+++gozZszA1q1b6/103xXz58/HvHnznLeLi4sRFRWF8ePHQ6fTNauvVqsVycnJGDduHJSOT62pDsbJdYxV/YKyTfhuxa7rtht/Y0KdVSrV1cDFi1JpQ1/fy6UPOwqbzYp9+5IxYMA4yOX8mboWk8mKU6ea//vnWI1FREStY9cp6ao1vY8Sw+MCm3UsS6UFlbZKGH1acNVrOzSmpx4pR83YmmnGw2MjPN2dVuXtc2x3z68B982xA7NN+Pj49c/xd+TLsCNfhiCtCkM7B2BYbACGdg5AtxBts688cTtRvLxy3bF6vbpaWrlutV6+z8FRy/3KVewufFJgtdmQvG8fxg0YAKWcmx83hHFyHWPlOsbKNc44jRgBpV7frGM1d47dpCT6q6++iokTJ6Jnz56IjIwEAJw9exajRo3C4sWLXT5OUFAQ5HJ5nV278/LyGtzUxFUqlQpdu3YFIO1yvnPnTrz55pt47733EBYWhurqapjN5lqfll/rddVqdb014pRKpduSlO48VnvGOLmOsaotsWsIwvUa5FoqGyzqEq7XILFrCORXnUgrldL+QkVFUjLdbr98hWZHIpcrmUS/DscVt439/ZPL5cjJyUFISIjz+QBgMpkQEBBQb51UIiJyn58OSitmk3qFQilvevmEqpoqmCpM8FV2rJXU7jCiqx5KuYDThVU4VVCJmCCNp7vUatwxx25L82vAfXNsV87xtWoF+kT4Y+9ZCwpKq/HToTz8dOnKkwBfJYbGGDEs1ojhcYHoFa6rMxfwCJWq/vvt9tqlYa4sD2OzScn2KxPsV5eHqSdZp5TLmcRzAePkOsbKdYyVa5QKRbPzW819fpPLufz2229ITk7Gvn374OPjgwEDBuDGG29s1HFUKhXi4+OxadMmTJkyBQBgt9uxadMmzJ07tylda5DdbnfWW4uPj4dSqcSmTZswdepUAEBmZibOnDnjrOtGRO2PXCZgweTeeOST3RBQf3X0e4ZGNXjSLJcDQUHSKvT8fMBkAvT6es9DiRqtoU3UqqqqoGpoEkVERG4hiiI2HJKS6M0t5WKuNKPaXg2tugN+2t5Mfmo5hsT4I/VkMbYcNWPmyOb9X7Ql7phjd9T59bXO8R1n9Yvv7o+JfcNRVWPDvrMWpGcXIi3bhF2nilBUbsXGw3nYeFhKqvtrFM6kekKsEX076Zv1wZrbyWQNJ9ivrr1eUyMl1x2JdcdGpqIorVR3rFavrpYmOS1Yf52IqK1rVBI9NTUVhYWFuO222yAIAsaPH4+cnBwsWLAA5eXlmDJlCt5+++0Gd/Wuz7x58zBjxgwMGTIEw4YNw5IlS1BWVoaZM2cCAKZPn45OnTph0aJFAKTNUg4fPuz89/nz57F3715otVrnJ+Pz58/HLbfcgujoaJSUlGD16tVISUnBhg0bAEgnKA899BDmzZsHo9EInU6HRx99FImJiQ1uekJE7cPEvuFYdv9gLPzucK0NiHyUMlRY7fjot1OYOjgSUcaGV4/5+0vnmBcvSivTfX2lVepETfHWW28BAARBwH/+8x9oL13iUFkp/Xz+3//9H3r27Omx/hERdQT7z1mQY6mEr0qOG7sFNfk4FdYKmCvN0KqYQG+qMT0NSD1ZjK0dJInu7jl2R51fN3SOH6bXYMHk3pjYNxwAoFbIMSxWSpDPBWC12XHgvAXp2SakZRVi16kilFTWYPPRfGw+mg8A8FXJEd85AAmxRiTEBaJ/pB5qhZeuopHLG17hc3Vy3WqVNjUFpH9XV0ur3EVRSqZfuXpdoWAheSLq8BqVRH/xxRcxZswY3HbbbQCkzUVmzZqFGTNmoFevXvj3v/+NiIgIvPDCCy4f85577sHFixfx/PPPIzc3FwMHDsT69eudm6GcOXMGsis+Db1w4QIGDRrkvL148WIsXrwYo0ePRkpKCgAgPz8f06dPR05ODvR6Pfr3748NGzZg3Lhxzue98cYbkMlkmDp1KqqqqjBhwgS8++67jQkHEbVRE/uGY1zvMKSeyMfGX9Iw/sYEDOociHvf34ED5y2Y9fEu/O+RG+CnbvhPpEoFhIdLyfOLF6XzTq2W55bUeG+88QYAaRXk8uXLIb808bHb7QCkzU+WL1/usf4REXUE6y+tQh/bIwQaZdOSY6Iowlxphs1ug0rOK4ia6sbuerzyI3AsrwIXzFWIMLTvjWjcPcfuyPPr+s7x6yvTeCWlXIbB0QEYHB2Ah0d3gc0u4vCFYqRdWqmenm2CpcKKX44X4JfjBQAAtUJ6zrBYIxLijBgcHdDkvxutypEMv/IDGasV2LMH6NxZmsg4EuyO+uuO72VlUoIdqL88DCdBRNQBNCqJvnfvXrz00kvO259//jmGDRuGDz74AAAQFRWFBQsWNCqJDgBz585t8PIyx8DtEBMT0+Al7w4ffvjhdV9To9Fg6dKlWLp0qcv9JKL2Qy4TkBBrROEREQmxRiiVCrw/PR6T396Oo7klmPflXiybFn/NTYZkMiAw8PKqdLMZ0OlY3oUaJzs7GwAwduxYfP311wgICAAgbXqi1+uxdu3aZm9gTUREDRNFEesPNr+US7m1HOZKM/xUfu7qWodk8FVgYLQWu0+XIuWoGX8YHurpLrWolphjd+T59dXn+I2tbS6XCegXqUe/SD3+dGMc7HYRx/JLkJZlQlp2IdKzTSgorUZqViFSswqBTYBSLmBApAEJcUYkxAYivnPANRfjeCWFQtoE6mp2e+3V644Ee2Wl9O+KCum7ozyMC/XXiYjaqkb9ZS8qKnJ+gg0AW7duxS233OK8PXToUJw9e9Z9vSMiakXheh+898Bg3Pd+GjYcysObm47jiXHdr/s8rbZ2eReNBvDxaYUOU7uyZcsWT3eBiKhDOpZXiuyCMqjkMoztGdKkY4iiiKLKIgCAkhtwN9uYHgbsPl2KrZmWdp9E5xzbu8lkAnqG6dAzTIcZN0gfOJy8WCatVL+UWM8rrsKu00XYdboIS7echFwmoG8nPYZfWqke39kIvU8b/bsgk0lfDSXYr0yuO2quV1XVX3/dsRLekWBn/XUiamMalUQPDQ1FdnY2oqKiUF1djd27d2PhwoXOx0tKSpq90ykRkSfFdzbin1P64qn/7cebm46jV7i/s4bitSiVQFiYlDzPzweKi6Xa6byyka5l3rx5eOmll+Dn54d58+bVeqy6uhoA8Mwzz0ClUuH111/3RBeJiNo9xyr0G7sFQdvE1aOl1aUoriqGv8rfnV3rsEb30OP1jedw4FwZCkqsCPJvv3NMzrHbFkEQ0DVEi64hWkxL6AxRFHHGVI60LBN2XFqpfq6oAvvOmrHvrBnvbcuCIAC9w3VIiA1EQpwRw2KMCPBrByWfHBuc1rfJ6dWr163Wywl2q1VawW6zXT7O1avXmWAnIi/UqLPESZMm4emnn8Yrr7yCdevWwdfXt9Zu4fv370eXLl3c3kkiotb0+6FROJxTjFW/ncK8L/chJsgPPcOuX05DJgMCAqTzSMeqdJ1OOhckqs+ePXtgtVqd/76S7dLEYv/+/VDwh4iIqMU46qFPaGIpF7toh6nCBJkgg1zG0gXuEKxToU8nXxw6X45tx8y4Mz7Y011qMZxjt22CIKBzoB86B/rh90OjAADnzRVIy5IS6mnZJmQXlOHQhWIculCMFdulMn49Qv2lhHqsVAIm2L+d1f53bHB69Ya4olh/gr2y8nL99ZoaaZW7INSfYOcqJSLykEbNyl966SXceeedGD16NLRaLT766COorvjUccWKFRg/frzbO0lE1NqevbUXjueXYPuJQvzpo134du5IGF1cMeLnJyXSCwoAk4nlXahhW7ZsQVZWFvR6fZ1yLo6a6N9//z1rohMRtZDThWU4klMMuUzAuF5NKxtSWl2KkuoSGDQG93augxvTw4BD58uRcrR9J9E5x25/Ohl8cOfgSNw5OBIAkFdceWmTUqkEzPH8UmTmlSAzrwQfp54GAMQF+0kr1S+VgAnXt9PJw5VlXa4minXLw7haf93xbyKiFtSovzJBQUHYtm0bLBYLtFot5FdtErFmzRpotVq3dpCIyBMUchneuW8wbl+6HWdM5fjLpxn470MJUMpdu7Tw6vIuZjOg13PhBNXVrVs35OTkICREqsN7zz334K233oIPP3khImpxjlIuw+OaVl7BZrehsLwQKrkKMoHlB9xpdA8Dlm6+gD2nS2Epr4Het30myDjHbv9CdRr8bkAEfjcgAgBQWFrlXKWelm3C0dxiZF0sQ9bFMnyWfgYAEG30RUKstFJ9eFwgIgN8ILT3iYQgSJOo5tRfd3Ak6q9cyU5E1ExNOhPR6/X13m80GpvVGSIibxLgp8J/ZgzBHUu3Y0eWCS99fxgv3t7X5ecLAmAwSFcx5udL5V38/es/L6SOSxTFWrd//PFHLFq0iEl0IqJW4CjlMrFP00q5lFSXoMxahgBNgDu7RQAijWp0DdHgRH4lfj1uwa0DAj3dpRbFOXbHEahV45Z+4biln7TvkqXcivRTl1aqZ5tw8LwFZ0zlOGMqx5qMcwCACL1GKv0SJ61Wjw3ya/9J9Ss1tv76lSvYKyulJDxweSX8lavXWX+diFzUPj/OJyJyk+6h/njjnoGY/d8MfJx6Gr3CdbhvWHSjjuHjA3TqBBQWSl8qFeDr20IdJiIiIpfkWiqx54wZADC+CUn0GnsNCssLoVFoOlYyqxWN6WnAifxcbM00t/skOnVcel8lxvUOxbjeUkmpkkorMk4XSSvVswqx/5wFFyyVWLf3AtbtvQAACPZXS6vULyXWuwZrIZN10L9Djam/7kiqV1VJifbycqkdcHnVuuO7QsHLiImoFibRiYiuY3yfMPxtXHe8lnwMz39zEF1DtBga07hVQQoFEBIi1Ud3lHfR6bjwgaQNqa5OvjAZQ0TU8jZcWoUe3zkAoTpNo59vqbSgwlqBAB+uQm8po3sY8J9tuUjPKkFZlQ1+apZkoPbPX6PEmB4hGNNDKvVXXl2DPWfMSMsqxI5sE/aeNeNiSRV+2J+DH/bnAACMfioMjQmQ6qrHGdEzTAd5R02qOzS2/rpjg1PHd5vtcoLdobJSStZzg1OiDolJdCIiF8y9qSuO5pbghwM5ePi/Gfj20ZHoZGhcuQ1BkOqiO8q7mM2AVlv/VYnUcYiiiAcffBDqS6tnKisr8fDDDzs3FZs2bRqUSiW+/vprT3aTiKjdcdRDb0opl2pbNUwVJvgoO0CdYg+KC9YgyqjGWVMVUk8WI6k3P7CgjsdXpcCIrkEY0TUIAFBptWHfWfOlmuqFyDhdBFNZNTYcysOGQ3kAAJ1GgaEx0ialw2ID0TdCB4WLezt1CI2pv26zAWVllx9zbHDqcHV5GNZfJ2q3mEQnInKBIAj49939kV1QhsM5xZj98S589fAN8FE1/iRJo5HKu2g0QEGBtNjBz68FOk1twowZM2rdvv/++wEAVqsVgFQjVclC+kREbmUqq0ZadiEAYEITkuiWSguqbFUw+rBedUsSBAGje+jxSWo+th41M4lOBECjlEu10eMCAXRDdY0dB85bkJZdiLQsEzJOF6G4sgabjuZj09F8AICfSo74GCMSYqWv/pEGqBRMqtervvrrjs19O3eWEvCO5LqjPIyj/vqVG5wKQu3EOuuvE7V5TKITEbnIV6XA+9Pjcfs723HoQjGe/Gof3r5vUJNWoMnlQHDw5fIuRUXSKnWeV3U8K1eurPf+4uJifPrpp3j33Xeh0+lauVdERO3bz4fzYBeB3uE6RAc2bqOSyppKFFUUwU/JT8Bbw5ieBnySmo/fThSjqsYONRN/RLWoFDLEdw5AfOcA/GUMUGOz43BOMdKyTEjLNmHnKRMsFVZsO3YR245dBABolDIMjg7AkM4G2C0Cqqw2LtpwhVwurV6/Xv11m+1yWZiqKunfFRX1b3DK+utEbQaT6EREjRAZ4Itl98fjDx/swPf7c9ArXIc5Y7s26ViCINVFv7K8i59f3XMyIiIicq+fDkp1hG/p2/hV6OZKM6x2K7Rqrbu7RfXoFe6LEH8l8kusSM8qwY3d9Z7uEpFXU8hl6B9pQP9IA2aNioPdLuJobgnSsguRnm1CerYJhWXV+O1kIX47WQhAjvf+tRmDogIulX8xIr5zAHxVTBe5zNX6645Ee3X15frrVVVSuZirNzi9chU7E+xEXoF/FYmIGmlYrBELb++Df6w9iMUbM9Ej1B9JvUObfDy1GoiIAHx8gIsXpXMpLeflRERELaK40ortJ6RSLhMbmUQvt5bDXGGGVsWBurUIgoDRPQ1Ys/MitmaamUQnaiSZTEDvCB16R+gwc0QsRFHEifxSpGWbkHqyAL8czUGxFUg/ZUL6KRMAQCET0C9Sj2GxRgyPDcSQmAD4a7hSvUlcqb9+5Sr2qqrLG5teXX/96vIwrL9O1KqYRCciaoJpCZ1xJKcYn+w4g8e/2Iu1f7kB3UL9m3w8uRwICrq8Kt1kksq78LyIiIjIvbYczUe1zY64YD90DXE9GS6KIooqiiBChFLOZFJrGtNDjzU7L+KXYxbU2EQo5FyVSdRUgiCgW6g/uoX64574CPzwwzn0GT4au89eLgFz3lyBPWfM2HPGjPe2ZkEmAH0ipKR6Qqy0Wt3gq7r+i9G1Oeqv18eRWL8ywe4oD2OzXf4uirVXwrP+OlGLYRKdiKiJFkzug+N50iqOP328C9/MGdHsk0l/fymRfvGiVCfd11eqm05ERETuseFQLgBgYp+wRu1rUmYtg6XSwjIuHtA/SguDrwLm8hrsOVOCobHcK4TIXQQBiAn0Q7cwA+4ZGg0AOGsqR3q2SdqsNNuE04XlOHDeggPnLfjw12wAQM8wf2mj0rhADIs1IkjLmpRuJZc3vKLq6uS6oyzM1fXXRbFueRjWXydqMibRiYiaSCmX4d1pg/G7d7bjdGE55q7eg1Uzh0Ihb96n/ioVEB4uJc+vLO/Ccx0iIqLmqai2YctRaWO9W/qGu/w8u2hHUUURBEGAQsYpVGuTywSM6q7Ht3sLkXLUwiQ6UQuLMvoiyuiLqfGRAIBcS6UzoZ6WVYiTF8twNLcER3NL8FHqaQBA1xCtc6X68LhAhOq4EqjFOJLh19vg1FF/varq8veysssbnNZXHoaTTqIG8QyQiKgZArVqfDB9CKYu+w2/nijA//vxKJ6f3LvZx5XJgMDAy6vSzWZpE1KWdyEiImq6bccvosJqQyeDD/p2cj0RW1ZdhuKqYujUTN56ypieBny7txBbM82YNyESchkTPUStJUyvwe0DO+H2gZ0AAAWlVdJK9SwpsX40twQn8ktxIr8Uq9POAAA6B/pKK9VjA5EQZ0RkgK8n30LHcK0NTu32+hPslZXSvx311x3lYVh/nagOJtGJiJqpd4QOr/1+AP7y6W6s2J6NXuH+uHtIlFuOrdVKK9MLCqTyLhqNtAEpERERNd6Gg1IplwmNKOVis9tQWF4IhUwBuYxJBE+Jj9FCq5bDVFaDQ+fL0D+KZXWIPCVIq8akfuGY1E+6osdcXn2p/IsJ6dkmHLpgwenCcpwuLMeXu84BADoZfC6VfzFiWGwgYgJ9G1VSi5pJJpO+rrXB6ZVfjvIwjn+z/joRk+hERO4wqV84/npTV7y1+QT+sfYg4oK1iO8c4JZjq1RAWJiUPM/PB4qLpdrpPOckIiJyXXWNHT8fyQMATOwb5vLzSqtLUWothUFjaKGekSuUchlGdNNhw8EipGSamUQn8iIGXxXG9wnD+D7S39biSisyThVhR3Yh0rNNOHDOgvPmCny95zy+3nMeABDir0ZCXOCl1epGdA3RMqnuKY4NTuvb5PTq1ev11V+32S4fx/F/aLVK/2aCndoRJtGJiNzk8aTuOJpbgo2H8/DwJxn4bu5IhOndUwtQJgMCAqTzGsemozpd/VfqERERUV2pWYUorqxBkFbt8gfdNfYaFJQXQC1XQyYwEeBpY3oYpCT6UQsevbkTE25EXkqnUWJszxCM7RkCACirqsHuM0WXSsCYsPesGfklVfhu3wV8t+8CACDQT4VhscZLddUD0TPMHzKWbfI8xwanrtRft1qlmuuAlGSvrJRWuTuS6ay/Tm0c0y9ERG4ikwl4/Z6BuPPd7TiWV4rZ/92FL/+cCI3SfZd++/ldLu9iMrG8CxERkavWXyrlMr5PqMv1tEuqSlBRU4EAjXuuLqPmSeiig1ohINdSjWO5FegRzhrLRG2Bn1qBG7sF48ZuwQCASqsNe86YL5WAKcTuM0UoLKvGTwdz8dOlv9V6HyWGxhidJWB6h+ugkPPDTK/RUP11gwHYtw/o3Flq09j6645/E3kh/mQSEbmRVq3Af6YPxe+W/or95yx4+n/78cY9A926UkqprF3exWKRVqXzg3wiIqL62ewikg9LiZmJfVwr5WK1WVFYXggfhQ9XPHsJjVKGxC46pGRakJJpZhKdqI3SKOVI7BKIxC6BALqhusaO/efMSLtUVz3jlAmWCit+PpLnLMOlVSswJCbAuVK9f6QeSibVvZdS2bz66w6ORP2VK9mJPIRJdCIiN4sO9MW7fxiMB1akY93eC+gVrsOfR3dx62sIgvQhv1otJdKLiqQ66fWdpxAREXV0GaeLUFBaDZ1GgeFxgS49x1JpQaWtEkYfYwv3jhpjTE8DUjIt2JppwZ/HRHi6O0TkBiqFDENijBgSY8ScsUCNzY6DF4qRnl2ItCwT0k+ZUFJZg5TMi0jJvAgA8FHKMbizAQmxUl31AVEGt14BTC2ksfXXr1zB7igPA1xeCX/l6nXWX6cWxiQ6EVELuKFrEJ67tRde+O4wXl5/FN3D/DG2R4jbX8fHB+jUCSgslL5UKsCXi7KIiIhq+elgDgAgqXcoVIrrT7KraqpQVFkEXyUHVW9zQ1c9FDIBpwoqcaqgEjFB7tl/hoi8h0Iuw8AoAwZGGTB7VBfY7CKO5hYjLUsq/5KebUJRuRXbTxRi+4lCAFIifmCUAcNjjUiIC8SgaAN8VUx5tSmNqb/uSKpXVUmJ9vJyqR1wedW647tCwcu2yS34F4WIqIXMuCEGR3JK8MWus/jrZ3uwbs4IdAnWuv11FAogJESqj56fD5jNUnkXfhBPREQEiKKIDQcbV8rFXGlGta0aASrWQvc2Wo0cQ2P9kXqyGClHzXhwpGv/p0TUdsllAvpE6NEnQo8/joyF3S7ixMVSpGUVYselzUoLSquQnm1CerYJ2HwCCpmA/pF6JMRJK9XjOwfAX8PLdtukhuqvA1Li/OrkutUqJdgd3222ywl2R6KeG5xSEzCJTkTUQgRBwItT+uDExVJknC7CrI92Ye2cEdD7uP/kTRAAvf5yeRezGdBq679KjoiIqCM5cN6CC5ZK+CjlGNU9+LrtK6wVMFea4afya4XeUVOM7qFH6slibM1kEp2oI5LJBHQP9Uf3UH88kBgDURSRXVAm1VTPKkRatgk5lkrsPmPG7jNmLEs5CZkA9O2kR0KsEcNiAzEsxgi9L5PqbZ4guF5/3WaTVq5fvcGpw9XlYVh/na7iFesUly5dipiYGGg0GiQkJCA9Pb3BtocOHcLUqVMRExMDQRCwZMmSOm0WLVqEoUOHwt/fHyEhIZgyZQoyMzNrtRkzZgwEQaj19fDDD7v7rRFRB6dWyLH8/niE6zXIKijDXz/bA5tdbLHX02ik8i4hIdIVbWVlLfZS1M40ZiwGgDVr1qBnz57QaDTo168ffvzxx1qPi6KI559/HuHh4fDx8UFSUhKOHz9eq43JZMK0adOg0+lgMBjw0EMPobS01Pl4SkoKbr/9doSHh8PPzw8DBw7Ep59+6r43TUQdwvpLq9DH9gy+br1cURRhrjTDZrdBJecn0d7qxu56yAQgM7cCOeYqT3fH63B+TR2NIAiIC9bivmHRWHLvIPz29E345amx+Pdd/XFXfCSijb6wi8D+cxZ88Es2Zn28CwNf2ohb3vwFL3x7CD8dyEFhKf+WtDuO+uu+vtKl2gEBQFgYEBMDxMZK32NigMhIIDxc2mRMJpMS62Vl0sZjRUXSCrWSEinpXl19uS47dTgeT6J/8cUXmDdvHhYsWIDdu3djwIABmDBhAvLz8+ttX15ejri4OLz88ssIC6t/1cHWrVsxZ84c7NixA8nJybBarRg/fjzKrsomzZo1Czk5Oc6vV1991e3vj4go2F+N9x8YArVChq3HLuLV9Udb9PXkciA4WDoXEARp3Oc4T9fS2LH4t99+w3333YeHHnoIe/bswZQpUzBlyhQcPHjQ2ebVV1/FW2+9heXLlyMtLQ1+fn6YMGECKisrnW2mTZuGQ4cOITk5Gd9//z22bduG2bNn13qd/v3743//+x/279+PmTNnYvr06fj+++9bLhhE1K6IouhMok/sG37d9uXWcq5CbwMC/JQYGC2VyEvJtHi4N96F82siKakeZfTF3UOisPjuAdj21Fj89vRNWHLPQNw3LBpxwX4QReBITjFW/XYKj3y6G/H//BnjXt+KZ9cdwLf7LiC/uPL6L0Rtl6P2up+fdEm30QhEREjJdUeCvXNnaVIdEiK1A6QSMSUltRPspaVSgt1qvVw2htolj5dzef311zFr1izMnDkTALB8+XL88MMPWLFiBZ5++uk67YcOHYqhQ4cCQL2PA8D69etr3V61ahVCQkKQkZGBUaNGOe/39fVt8ESBiMid+kXq8epd/fHY53vx3rYs9ArXYcqgTi32eoIgfdh+ZXkXP7+6e7QQAY0fi998801MnDgRTz75JADgpZdeQnJyMt555x0sX74coihiyZIlePbZZ3H77bcDAD7++GOEhoZi3bp1uPfee3HkyBGsX78eO3fuxJAhQwAAb7/9NiZNmoTFixcjIiICzzzzTK3Xfeyxx7Bx40Z8/fXXuO222+p9L1VVVaiqurySqLi4GABgtVphtVqbFSfH85t7nI6AsXIN4+S6psbqeF4psgrKoJQLGBkXcM3ni6KIgtIC1NTUQKaQwVZja1afPcHR57bY98a6sZsOu0+XIuVIEX4fH9jo5ztiVGOtgUxo3toyb/od5vyaqH4RBh9MGdTJOQfLL6l01lBPyzIhM68Ex/NLcTy/FJ/sOAMAiA3yk+qpR+vBheodRGPqr9ts0sp0R/31qippBfvVG5yy/nq74dEkenV1NTIyMjB//nznfTKZDElJSUhNTXXb61gs0uoEo9FY6/5PP/0Un3zyCcLCwjB58mQ899xz8PX1rfcYnJR7HuPkOsbKNa0dp0l9QnD4xli890s2/v6//YgOUKNfJ32LvqZMJq1KVyqBggJpfNc2YW9Tm81a6zs1zG53z89Va/1cNmUsTk1Nxbx582rdN2HCBKxbtw4AkJ2djdzcXCQlJTkf1+v1SEhIQGpqKu69916kpqbCYDA4E+gAkJSUBJlMhrS0NNxxxx31vrbFYkGvXr0afD+LFi3CwoUL69y/cePGBsf4xkpOTnbLcToCxso1jJPrGhurDecEAHJ087fhl80bW6ZTXmjfb/s83YUWZ6wCAAUOni/Dlp93Q9/E6js///xzs/tSXl7e7GO4Q1uaXwMtN8fmXMh1HTlWARo5JvQKxoRe0l4ZprJqZJw2I/2UCemninAktwTZBWXILijD5zvPAlDgP1nbMCwmAENjjBgWG4DoAB8ITIrW0iF+pq5MsjtWqDvqr9ts0ldNjZRUr6qS/u3Y4NRBJoP10s+Otbqam5ldg/VS3KyOTWObc6xmPt+jSfSCggLYbDaEhobWuj80NBRHj7qn3IHdbsfjjz+OESNGoG/fvs77//CHP6Bz586IiIjA/v378fe//x2ZmZn4+uuv6z0OJ+Xeg3FyHWPlmtaMU08R6G2Q4bAZmLliB/7Wz9bkSZ8n7NvHnylXNffnqrUm5E0Zi3Nzc+ttn5ub63zccd+12oSEhNR6XKFQwGg0Ottc7csvv8TOnTvx3nvvNfh+5s+fXyvBX1xcjKioKIwfPx46na7B57nCarUiOTkZ48aNg7K+zYvIibFyDePkuqbGavnSVAAluH9MP0yKb/gKMLtox/ni86ioqYBO3by/FZ5kq7Fh32/7MOCGAZAr2v+GaF/knMDhnHJYAjtj7KDGrUYvqyxDZnomkpKSoGpm8sKR/PW0tjS/Blp+js25kOsYq8sGARgUA5RHAtklAk4UCzhZLOBsKXDeXIm1e3Owdm8OAECvEtHFX0RXvYiuOhEhGi40duDPlOuSryiJSQ1L3r692cdo7hzb4+VcWtqcOXNw8OBB/Prrr7Xuv7Lmar9+/RAeHo6bb74ZJ0+eRJcuXeoch5Nyz2OcXMdYucZTcRp9sxV3vZeOrIIyrM0PxH//OBRqRetsUVFdDRQWSuVdfH1dL+9is1mxb18yBgwYB7mcP1PXYjJZcepU83+uvGVC7i22bNmCmTNn4oMPPkCfPn0abKdWq6Gu5wdbqVS67ffcncdq7xgr1zBOrmtMrM4UluNIbgnkMgET+0Vc83nFVcUot5fD4GtodmkPbyBXyDtEEn1MLwMO55Rj2/FiTB0acv0nXMERH4VS0ezfv470++uu+TXQcnNszoVcx1i5xmq14vv1yQjoFo/dZ4uRfqoI+89bYKkGdhcK2F0otQv0U2FoTACGXfrqFqKFTNaxsur8mXKdM1ajR0MpCNKKdUdZmOrqyyvaRVH6cpSacZSHUSg6xKc2VpsNyfv2YdyIEVDqm3clf3Pn2B5NogcFBUEulyMvL6/W/Xl5eW6ppTZ37lznRmWRkZHXbJuQkAAAOHHiRL2DPCfl3oNxch1j5ZrWjpNRqcSHDw7F7e/8ij1nLVj4/VG8elf/VrkUUKkEfHykq84uXpTGZa3W9bFXLlcyiX4dsku5l+b+XLXWz2RTxuKwsLBrtnd8z8vLQ3h4eK02AwcOdLa5epOzmpoamEymOq+7detWTJ48GW+88QamT5/e+DdJRB3S+kPSSsGEWCOMfg2vNLbZbSgsL4RKrmoXCfSOZEwPA97dfAG7T5WguKIGOp92v0bsmtrS/Bpo+Tk250KuY6yuTy0HxvQMxbh+0s9+RbUNe84WIS3LhLTsQuw5Y0ZhWTXWH8rD+kPS76DBV4mhMUYkxBoxPC4QvcJ1kHeQpDp/plyn9POrGytH/XVHIr2mpnb9dUeJGLtdan9l3fV2Wn9dqfD8h94ePUtUqVSIj4/Hpk2bnPfZ7XZs2rQJiYmJTT6uKIqYO3cu1q5di82bNyM2Nva6z9m7dy8A1JrsExG1lNggP7zzh8GQCcCajHNYuf1Uq722TAYEBkobjSuV0qp0W/vff4wa0JSxODExsVZ7QLpk09E+NjYWYWFhtdoUFxcjLS3N2SYxMRFmsxkZGRnONps3b4bdbndOvAEgJSUFt956K1555ZVaq9yIiK5n/UGpNNTEvtdOHpZUl6DMWgY/pV9rdIvcKNKoRtcQDWwi8Otxi6e743GcXxO1Hh+VHDd0CcIT47rj89mJ2P/CeKx5OBH/N747buwWBF+VHOZyK5IP5+GfPxzBbW//ioELN2LmynQs33oSe84UwWqze/ptkLcSBGmyrtFIq94MBiAkBIiOBmJjgZgY6XvnzkCnTkBAgFRXXRSBigppkm8yAUVFQHExUF4urXDnxL9ZPP5R/bx58zBjxgwMGTIEw4YNw5IlS1BWVubcTXz69Ono1KkTFi1aBEDaLOXw4cPOf58/fx579+6FVqtF165dAUiXmK1evRrffPMN/P39nbVV9Xo9fHx8cPLkSaxevRqTJk1CYGAg9u/fjyeeeAKjRo1C//79PRAFIuqIRnUPxjOTeuGfPxzBv348gu6h/hjZLajVXl+rlcbZggJpbNVopFXq1PE0dix+7LHHMHr0aLz22mu49dZb8fnnn2PXrl14//33AQCCIODxxx/HP//5T3Tr1g2xsbF47rnnEBERgSlTpgAAevXqhYkTJ2LWrFlYvnw5rFYr5s6di3vvvRcREREApBIut912Gx577DFMnTrVOZ6rVKo6m5kREV0pr7gSu8+YAQDjezecRK+x16CwvBAahYabw7VRo3sYcCI/FylHLZjUv3F10dsjzq+JPEOtkGNojBFDY4yYC8Bqs+PgeQvSsk1IzzZhZ7YJJVU12JJ5EVsyLwIAfFVyxHcOQEKsEcNiAzEgSg91ByjFRc0kkzW8EemVG5s6vq7c4NSRSL+yPMyVK9hlvCLvWjyeRL/nnntw8eJFPP/888jNzcXAgQOxfv1652YoZ86cgeyK/8QLFy5g0KBBztuLFy/G4sWLMXr0aKSkpAAAli1bBgAYM2ZMrddauXIlHnzwQahUKvz888/OE4qoqChMnToVzz77bMu+WSKiqzw0MhaHc4rx9e7zmLN6N76ZMwIxQa23Ek6lAsLCpOR5fr70IbW/f7u78ouuo7Fj8Q033IDVq1fj2WefxTPPPINu3bph3bp1tTYYe+qpp1BWVobZs2fDbDZj5MiRWL9+PTQajbPNp59+irlz5+Lmm2+GTCbD1KlT8dZbbzkf/+ijj1BeXo5FixY5J/sAao35RET12XBISvINjjYgTK9psJ2l0oIKawUCfAJaq2vkZmN6GvDhL7lIzypGWZUNfuqOnYDi/JrIOyjlMgyKDsCg6AA8PLoLbHYRR3KKsSOrEOnZJqSfMsFcbsUvxwvwy/ECAIBaIcOgaAMSYgOREGvEoOgA+Kg69t80aiS5XPqqL8l+dXLdUX+9qkr6d0XF5ZXqMlnd8jBMsHs+iQ5ItdXmzp1b72NXT5JjYmIgiuI1j3e9x6OiorB169ZG9ZGIqCUIgoD/d0c/ZF0sw96zZsz6eBe+/ssN8Ne0Xv04mezy1V8XL0qr0nU6aaykjqMxYzEA3H333bj77rsbPJ4gCHjxxRfx4osvNtjGaDRi9erVDT6+atUqrFq1qsHHiYga4kopl2pbNYoqiuCj9OEq9DYsLliDyAA1zhVVYcfJYtzcmx+IcH5N5H3kMgF9O+nRt5Mef7oxDna7iGP5JUjPNjnrqheUVmNHlgk7skwAAKVcwIBIA4bFGpEQF4j4zgHQqjlJoyZyJNiv3otCFOtPsFdWSnXYHavY7XZptV19CfYOch7F3z4iIg/TKOV474F4TH77VxzPL8UTX+zF+w8MafWd3P38Lpd3MZlY3oWIiNomU1k10rKlBMSEPg0n0S2VFlTaKmH0YXmotkwQBIzpqccnqflIyTQziU5EbYJMJqBnmA49w3SYnih9mHXyYpmUVM8uRFqWCbnFldh1ugi7Thfh3ZSTzkR8Qqy0WemQGCP0Pty8k5rpyrIuV3NscHrll2OD05oaafV6Tc3l8jBXJtYd/25H2te7ISJqo0J1Grw/fQh+/14qfj6Sj9eTj+H/JvRo9X4olbXLu1gs0qp0IiKituLnw3mw2UX0Ctehc2D9JdIqaypRVFHEzUTbiTE9DPgkNR+pJ4pRVWOHWsFLzomobREEAV1DtOgaosUfEqIhiiLOmMqRdsVK9XNFFdh31ox9Z814f1sWBAHoHa6TVqrHBmJYrBFGvwZqZRM1hWODU2U9H9bY7XUT7K7UX79yJXsbwyQ6EZGXGBhlwMt39sO8L/fhnS0n0DPcH7f1j2j1fgiCtPm3Wi0l0ouKAF/fVu8GERFRk6y/VA/9lmuUcjFXmmG1W6FVa1urW9SCekb4IsRfifwSK3ZmlWBkd72nu0RE1CyCIKBzoB86B/rh90OiAADnzRVIv7RKPS3bhOyCMhy6UIxDF4qxcvspAED3UK1UUz3OiGGxRoT4N7wvCFGzODY4bWz99ZoaaSW7zSYlHxwJ9itXr3tp/XUm0YmIvMidgyNxJKcYH/ySjf9bsw8xgX7o28kzE0EfH6BTJ6CwUEqmExERebuSSit+vbRBW0P10Mut5TBXmOGn4ir09kImCBjVw4Cvdl1ESqaZSXQiapc6GXxwx6BI3DEoEgCQX1wprVTPljYrPZZX6vz6747TAIC4ID8kxF1eqR5hYL1OagWNrb/uSLBXVwPl5VI7QEqmO+qtX2d/jtbAJDoRkZd5+pZeyMwrxbZjFzH741349tGRCNKqr//EFqBQACEh0vesLMBsBvz96/+wmYiIyNO2ZF5Etc2OuCA/dAupu8pcFEUUVRTBLtqhknMwa0/G9tTjq10X8etxC2psIhTyjrHJGRF1XCE6DSYPiMDkAdLVy4WlVdh5StqYND3bhCO5xcgqKENWQRk+Sz8LAIgy+kgr1S+VgIkycnNtakWNrb9utQJlZdLj5eVAgGf3PWESnYjIy8hlAt6+dxCmvLsd2QVleOSTDHz6p+FQeai+pyBcrotuNAIlJdL45e/fJsuYERFRO7b+YA4AaRV6fUmBMmsZLJUWlnFph/pHaWHwVcBcXoO9Z0oxJNbf010iImpVgVo1JvYNx8S+4QAAS7kVO09dXql+8EIxzpoqcNZ0Dl9lnAMAhOs1SIg1YtilEjBxQX5MqpNnNFR/3WoF9u/3TJ+uwiQ6EZEX0vsq8cH0Ibhj6XbsPFWEBd8ewv+7o6/HT2hCQqQPf00madNRpRLw87t8hRUREZGnVFpt2HL0IoD6S7nYRTuKKoogCAIUMk6D2hu5TMCN3fX4bm8hthw1M4lORB2e3leJpN6hSOodCgAorarBrlNSPfX0bBP2nzMjx1KJdXsvYN3eCwCAYH/1pY1KpZXq3UK0kMk42SMCmEQnIvJaXUO0eOu+QfjjRzvxWfoZ9A73xwOJMZ7uFnx9pXrpOp1UL72oSLrtw/J6RETkQduOXUSF1YYIvQb96tlPpKy6DMVVxdCpdR7oHbWGMT2kJPq2TDP+NjESMn7KT0TkpFUrMKZHCMb0CAEAVFTbsPtMkVRXPasQe86acbGkCj/sz8EP+6UruwJ8lRjmWKkea0SvcB3kTKpTB8UkOhGRFxvbMwRPTeiJV9YfxcLvDqNriD8SuwR6ulvOEi9+ftKKdJNJSqb7+bFeOhERecb6Q7kAgAn1lHKxi3YUlhdCIVNALmMtsvZqSKw//NQyFJbV4ND5MvSLZNkeIqKG+KjkGNE1CCO6BgGQrujad9aM9GxptXrG6SIUlVux4VAeNhzKAwD4axQYFmOUVqvHBaJvhA4KuWfKjhK1NibRiYi83MOj43Akpxjf7ruAv3yagW/njkSU0dfT3QIg1UQ3GgGtVkqiFxWxXjoREbW+6ho7fj4sTfBvuVQL9kolVSUos5ZBr6m7Qp3aD6VchhHd9Nh4sAhbjpqZRCciagSNUo6EuEAkxAXiUUhj68ELFqRlSXXVd50qQkllDTYdzcemo/kAAD+VHIM7B2B43KWV6qF+nn0TRC2ISXQiIi8nCAJemdofWQWlOHi+GLM+3oX/PXID/NTe8ydcpQJCQ6XkOeulExFRa9uRVYjiyhoEaVWI7xxQ67Eaew0KKwqhkqsgE7harr0b08OAjQeLsPWoBY/e3Mnj+8kQEbVVKoUMg6MDMDg6AI+M6YIamx1HckqQll2IHVkm7DxlgqXCil+OF+CX4wUAALVChmhfGU5qTiKxazAGRRugUXJ1FbUP3pOBISKiBvmo5Hj/gSH43TvbcTS3BH/7ch/enTbY6zZ5Yb10IiLyBEcpl3G9w+rUai2pKkG5tRwBmoD6nkrtTEKcP9QKATmWahzLq0CPMO+4eo+IqK1TyGXoF6lHv0g9/nRjHOx2EZl5JUjLKnRuVlpYVo3jxTIc33ISb205CZVchgFReiTEBiIhzojB0QFetRiMqDH4k0tE1EZEGHyw/P7BuO+DHVh/KBdvbT6Ox5O6e7pbdVxZL724WEqmm0xSyRfWSyciInez2UVsvFSrdWLfsFqPWW1WmCpM8FH4cEVyB+GjkiOxiw4pmRZsPWpmEp2IqIXIZAJ6hevQK1yHB0fEQhRFHL1gxorvf0G5thN2nipCfkkVdp4qws5TRXhnC6CQCejbSY+EOCOGxwYiPiYAOo3S02+FyCVMohMRtSFDYoz455S++Pv/DmDJz8fRM8wfE+up/eoN5HIgIEBKprNeOhERtZSM00UoKK2CTqNAYlztzbctlRZU1FTA6GP0UO/IE0b3NCAl04KUTAtmj4nwdHeIiDoEQRDQNUSLkWEiJk3qD4VCgdOF5UjLLrxUV92E8+YK7D1rxt6zZry3NQsyAegdoZNWqscaMTTGiAA/rrwi78QkOhFRG3PP0GgcySnBqt9OYd6X+xAT5IeeYTpPd6tBrJdOREQtaf1BqZRLUq9QqBSXa55X1VShqLIIvkquRO5obuiqg0Im4FRBJU4VVCImSOPpLhERdTiCICAmyA8xQX64Z2g0AOBcUTnSsqTSL2nZhThVWI6D54tx8HwxPvw1GwDQM8wfCbFGDIsNxLBYI4L91Z58G0ROTKITEbVB/7i1F47lleC3k4X400e78O3ckTB6+Sf2jnrpej3rpRMRkXuIoogNl+qhT7iqlIu50oxqWzUCVKyF3tH4axQYEqPFjqwSbM00IyYo7PpPIiKiFhcZ4IvIeF9MjY8EAORaKpGWXXgpqW7CifxSHM0twdHcEnyUehoA0CXYDwlx0kr1hNhAhOn5wSh5BpPoRERtkFIuw9I/DMbtS7fjjKkcf/k0A/99KAFKuez6T/YgQZBWpPv6sl46ERE138HzxThvroCPUo5R3YKd91dYK2CuNMNP5efB3pEnjelpkJLoR82YMYJJdCIibxSm1+D2gZ1w+8BOAICC0irsvJRQ35FViMy8Epy8WIaTF8uwOu0MAKBzoC+GxRidifUoI684o9bBJDoRURsV4KfCB9OH4M53t2NHlgkvfX8YL97e19PdcgnrpRMRkTv8dDAHADC2ZzB8VNIAIooizJVm1Nhr4C/392T3yINGdtdD9tNZHM2tQI65CuEGlgMgIvJ2QVo1bukXjlv6Sft+mcursfNUEdKyCpGWbcKhCxacLizH6cJyrMk4BwDoZPDBsFijtFI9LhAxgb7cTJxaBJPoRERtWI8wf7xxz0DM/m8GPk49jV7hOtw3LNrT3XLZlfXSi4qkeulyubQynec9RER0LaIoOuuhT+hzeaVxubUc5koztCqtp7pGXsDop8SAKC32nCnF1kwL7k0I8XSXiIiokQy+KozrHYpxvUMBAMWVVmScLrq0UWkhDpyz4Ly5Amv3nMfaPecBACH+aimpHheI4bFGdA3RMqlObsEkOhFRGze+TxjmjeuO15OP4flvDqJriBZDY4ye7lajOOql63Ssl05ERK45kV+KrIIyqOQy3NRTSpA6VqGLEKGUKz3cQ/K00T31l5LoZibRiYjaAZ1GibE9QjC2h/Q3vby6BrtPm5GWXYi0LBP2njUjv6QK3+/Pwff7pavVjH6qS+VfpJrqPcP8IZMxqU6NxyQ6EVE78OhNXXE0txg/HsjFI59k4Ju5I9HJ0LYy0KyXTkREjeFYhT6iayD8NVLCvMxaBkuVBf4qlnEhYHQPA5ZsPI/9Z8tQWGpFoJYfrBARtSe+KgVGdgvCyG5BAIBKqw17z5qRlmVC+qlCZJwugqmsGusP5WL9pY3IdRrFpfIvgRgWa0SfCB0UXr63GHkHJtGJiNoBQRCw+O4ByC4ox5GcYsz+eBe+evgGZ33YtqS+eukVFVIynfXSiYjI4adLSfRb+kp1U+2iHaYKE2SCDHIZBwwCQnUq9I7wxeEL5diWacEd8UGe7hIREbUgjVKO4XGBGB4XCKAbqmvsOHDejB1ZJqRnm7DrlAnFlTX4+Ug+fj6SDwDQqhWI7xxwaaW6Ef06GaBSMKlOdTGJTkTUTviqFPhgejx+9852HLpQjCe/2oe37xvUZuu/Oeql63TSinTWSyciIoczheU4nFMMmQAkXaqTWlpdiuKqYhg0Bs92jrzK6B4GHL5QjpRMM5PoREQdjEohQ3xnI+I7GzFnLFBjs+PQhWKkZRciPVtKrBdX1mDrsYvYeuwiAECjlCG+cwCGxQQiIc6IgVEGaJT8cJ6YRCcialciA3yxbNpgTPtPGr7fn4Ne4TrMGdvV091qFh8fICKC9dKJiOiyDZcuyU6IDYTRTwWb3YbC8kKo5CrIBK4eo8vG9NRj2ZYL2H26BMUVNdD5cApMRNRRKeQyDIgyYECUAbNHdYHNLuJobrFU/iXbhPRTJpjKqrH9RCG2nygEICXiB0YZkHCpBMzgzgb4qjiWdET8XyciamcS4gLxwu/64Nl1B7F4YyZ6hPo7V+m1VayXTkREV3LUNZ3YNwwAUFJdgjJrGQI0AZ7sFnmhKKMGXUI0OJlfie3HLbilf6Cnu0RERF5CLhPQJ0KPPhF6/HFkLOx2ESculiIt24S0rEKkZZtwsaTKuWr9bZyAQiagf6Qew2KllepDOgc492ah9o1JdCKiduj+4Z1xJKcYn6adweNf7MXav9yAbqFtf5O1K+ulm81SIr28XEqws146EVHHkFdciYzTRQCACX3CUGOvQWF5ITQKTZstYUYta3QPA07m52LLUSbRiYioYTKZgO6h/uge6o8HhneGKIrILihDerbJmVi/YKnE7jNm7D5jxvKtJyETgL6d9BgWY0RCXCCGxRih92VSvT3yimsdly5dipiYGGg0GiQkJCA9Pb3BtocOHcLUqVMRExMDQRCwZMmSOm0WLVqEoUOHwt/fHyEhIZgyZQoyMzNrtamsrMScOXMQGBgIrVaLqVOnIi8vz91vjYjIYxZM7oNhsUaUVtVg1se7YCm3erpLbqNSASEhQOfOgF4PlJRIX6Lo6Z4REVFL23hpFfqgaAPC9BpYKi2osFbAR8E6X1S/MT0MAID0rGKUV9s825lWwPk1EZF7CIKAuGAt7h0WjTfuGYjtT9+EX54ai8V3D8Dd8ZGINvrCLgL7z1nwn1+zMevjXRj40kZMXLINL3x7CD8dyEFhaZWn3wa5iceT6F988QXmzZuHBQsWYPfu3RgwYAAmTJiA/Pz8etuXl5cjLi4OL7/8MsLCwupts3XrVsyZMwc7duxAcnIyrFYrxo8fj7KyMmebJ554At999x3WrFmDrVu34sKFC7jzzjtb5D0SEXmCSiHDsmmD0cngg1OF5Zj72W7U2Oye7pZbOeqlR0UBarVUL72iwtO9IiKiluQs5dInDNW2ahRVFMFH6cNV6NSgLiEadApQodomYsfJYk93p0Vxfk1E1HIEQUCU0Rd3xUfi33cPwLanxiJ1/k14896BuG9YNLoE+0EUgaO5JVj12yk88uluxP/zZyS9vhX/WHsA3+67gLziSk+/DWoij5dzef311zFr1izMnDkTALB8+XL88MMPWLFiBZ5++uk67YcOHYqhQ4cCQL2PA8D69etr3V61ahVCQkKQkZGBUaNGwWKx4MMPP8Tq1atx0003AQBWrlyJXr16YceOHRg+fLg73yIRkccEatX4YPoQTF32G345XoBFPx3Fc7f19nS33Kqheul+flJinYiI2o+i8mrsyDIBkOqhWyotqLRVwuhj9HDPyJsJgoAxPQ34NDUfKUfNuKlX+62dz/k1EVHrCtf74PaBnXD7wE4A4KyhnpZdiLQsEzLzSnAivxQn8kvxadoZAEBMoC+GxgRAbREwwFyBmGCWf2kLPJpEr66uRkZGBubPn++8TyaTISkpCampqW57HYvFAgAwGqWT64yMDFitViQlJTnb9OzZE9HR0UhNTa13kK+qqkJV1eVLMIqLpRUMVqsVVmvzSiQ4nt/c47R3jJPrGCvXdJQ4dQv2wSt39sFfv9iPD3/NRvcQX9w5qFOjjtFWYuXYaNRikWqml5a2fr10u909sfL2WBMRecKmoxdhs4voFa5DqF6OM+Yi+Cn9PN0tagPG9JCS6L+dKEZVTfu6Ms+hLc2vgZabY7eV81ZvwFi5hnFyHWMFGDQyjO8VhPG9ggBICwB2nTIj/ZQJO08X4UhOCU4VluNUYTkAOT557Rd0MmgwLCYAQ2OMGBYTgGgjr7BzcP5M2WyAh+fYHk2iFxQUwGazITQ0tNb9oaGhOHr0qFtew2634/HHH8eIESPQt29fAEBubi5UKhUMBkOd183Nza33OIsWLcLChQvr3L9x40b4+vq6pa/JycluOU57xzi5jrFyTUeJ04ROMmw4L8Mzaw8i99g+xDRhn9GOEit3aG6sysvL3dQTIqL2Y+NhqcbyxD7SKvRqezW0aq2He0VtQa8IXwT7K3GxxIpd2SUYGOXxi7Ldri3Nr4GWn2PzvNV1jJVrGCfXMVZ1DQIwqDNQ0QnIKhFwoljAyWIBZ0uB8+ZKrN2bg7V7cwAAeqWILjrpq6tORKiPdAV2R5a8Zw+wZ0+zjtHcOXb7O3O4ypw5c3Dw4EH8+uuvzTrO/PnzMW/ePOft4uJiREVFYfz48dDpdM06ttVqRXJyMsaNGwelkpdwNIRxch1j5ZqOFqeJdhFzP9+H5CP5+OSUL75+ZDjCdBqXnttWYyWKQFmZVN6lrAzQaKQ66i3JZLLi1Knmx8qxGouIiCSVNuDXE4UAgDE9DSiqKIRWxQQ6uUYmCBjdQ4+vdhUg5agZA6OCPN2lNsld82ug5ebYbfW81RMYK9cwTq5jrFzniNUNo8fiQE4Zdp4qws5TRdh3zgKLFdhdKGC3dNqDQD8VhsYEYGhMAIbFBKB7iBYyWcfIqjt/pgYNgjI8vFnHau4c26NJ9KCgIMjl8jq7dufl5TW4qUljzJ07F99//z22bduGyMhI5/1hYWGorq6G2Wyu9Wn5tV5XrVZDXU9xXaVS6bY/DO48VnvGOLmOsXJNR4rTG/cOwp3vbsexvFLM/WwfvvhzIjRK12udtMVYqVSATne5XrrF0rL10mWXtuxubqzaWpyJiFra4SIBVpuImEBfBPlXw1Jlh0qu8nS3qA0Z09OAr3YV4JfjFjx6c6Cnu+N2bWl+DbT8HLstnrd6CmPlGsbJdYyV6wx+Priplw439ZISxJVWG3afKZLqqmeZsPtMEQrLqrH+UB7WH5L+vut9lBgaY8TwOCMSYgPRO0IHeTtPqivl8mb/TDX3+bJmPbuZVCoV4uPjsWnTJud9drsdmzZtQmJiYpOPK4oi5s6di7Vr12Lz5s2IjY2t9Xh8fDyUSmWt183MzMSZM2ea9bpERN5Oq1bgP9OHwuCrxL5zFsz/+gBEUfR0t1qcXA4EBADR0UBICFBdLdVMr6nxdM+IiMhV+0zS5DCpdxAsVRaWcaFG6x+lhcFXgeIKG/adrfB0d9yO82siorZPo5Tjhi5BeDypOz6bPRz7XxiPrx5OxJMTeuDGbkHwVclhqbDi5yN5+OcPRzD5nV8xcOFGPLgyHctSTmL3mSJYbe1z7w9P83g5l3nz5mHGjBkYMmQIhg0bhiVLlqCsrMy5m/j06dPRqVMnLFq0CIC0Wcrhw4ed/z5//jz27t0LrVaLrl27ApAuMVu9ejW++eYb+Pv7O+uw6fV6+Pj4QK/X46GHHsK8efNgNBqh0+nw6KOPIjExkTuHE1G7Fx3oi3f/MBgPrEjH2j3n0SvcH7NHdfF0t1qFSiUl0f39pRIvFouUYPfzu7yCnKg+NruItGwTMgoEBGabkNg1pN2v9mgqxso1jJPrbHYRvxwvwIFLSfQhMUoIggiFzONTGWpjFDIBN3bT47t9hVi3pwjdZAICsk0Y0S2s3fz+cX5NRNS+qBVyDIkxYkiMEXPGdoXVZsehC8VIyypEWrYJO7NNKKmqQUrmRaRkXgQA+CjliO8cgIRYIxLiAjEgSg+1wvUr0Kl+Hj/zvOeee3Dx4kU8//zzyM3NxcCBA7F+/XrnZihnzpyB7IrMxoULFzBo0CDn7cWLF2Px4sUYPXo0UlJSAADLli0DAIwZM6bWa61cuRIPPvggAOCNN96ATCbD1KlTUVVVhQkTJuDdd99tuTdKRORFbugahOdu7YUXvjuMl386iu6h/hjTI8TT3Wo1Pj5ARIRU5qWwUFqV7uPT8vXSvdnSpUvx73//G7m5uRgwYADefvttDBs2rMH2a9aswXPPPYdTp06hW7dueOWVVzBp0iTn46IoYsGCBfjggw9gNpsxYsQILFu2DN26dXO2MZlMePTRR/Hdd985x+Q333wTWu3l1aX79+/HnDlzsHPnTgQHB+PRRx/FU0891TJBaMD6gzlY+N1h5FgqAcjx8fFdCNdrsGByb0zs27y6fO0NY+Uaxsl1tWMlJTmfXXsKj4+PxE29PNs3apuMWmkKnHqyDKmQ4+Pju9vV7x/n10RE7ZtSLsPAKAMGRhnw59FdYLOLOJJTjLRsE9KyCpF+ygRzuRW/nijArycKAAAqhQyDogxIiAvE8FgjBkUHwEfFpHpjeTyJDki11ebOnVvvY46B2yEmJua6pQdcKU2g0WiwdOlSLF261OV+EhG1JzNuiMGRnBJ8sessHv1sD9bNGYEuwR3n0nhBkFak+/oCJSVSMt1katl66d7qiy++wLx587B8+XIkJCRgyZIlmDBhAjIzMxESUvfDld9++w333XcfFi1ahNtuuw2rV6/GlClTsHv3bvTt2xcA8Oqrr+Ktt97CRx99hNjYWDz33HOYMGECDh8+DI1G2tB22rRpyMnJQXJyMqxWK2bOnInZs2dj9erVAKSNX8aPH4+kpCQsX74cBw4cwB//+EcYDAbMnj27VWKz/mAOHvlkN64+s8i1VOKRT3Zj2f2D20XSxR0YK9cwTq5rKFYFpTY8+/Vp/L+pMozpafBE16iNSjlqxkfb8+rc395+/zi/JiLqOOQyAX076dG3kx4PjYyF3S7ieH4p0rILkZZlQlq2CQWlVVKSPduEtwAo5QL6RxqcK9XjOwdAq/aKFLFXY4SIiDooQRDw4pQ+OHGxFBmnizDr411YN2cEdJqOtQGMXA4YDFLy3GyWEukVFYBWCyg6yCj5+uuvY9asWc5LvZcvX44ffvgBK1aswNNPP12n/ZtvvomJEyfiySefBAC89NJLSE5OxjvvvIPly5dDFEUsWbIEzz77LG6//XYAwMcff4zQ0FCsW7cO9957L44cOYL169dj586dGDJkCADg7bffxqRJk7B48WJERETg008/RXV1NVasWAGVSoU+ffpg7969eP3111sliW6zi1j43eE6CTwAzvue++YQugRr200ZgKay2UU8t+4gY3UdjJPrrhUrhyXJ53Bjd32HjxW5xmYXsWTjuXofEyFd57Dwu8MY17v9lHYhIqKORyYT0CPMHz3C/DE9UfqgNKugDGlZJqRnSyVgciyVyDhdhIzTRXg35aSUiI/QISEuEMNijBgaa4Tep2PlBVzRQdIDRERUH7VCjuX3x+N37/yKrItl+Otne/DhjKEdcvKoVALBwVLyvCPVS6+urkZGRgbmz5/vvE8mkyEpKQmpqan1Pic1NRXz5s2rdd+ECROwbt06AEB2djZyc3ORlJTkfFyv1yMhIQGpqam49957kZqaCoPB4EygA0BSUhJkMhnS0tJwxx13IDU1FaNGjYJKpar1Oq+88gqKiooQEBBQp29VVVWoqqpy3i4uLgYAWK1WWK3WRkQGzhPMa7lYUoVxb2xr1HE7KsbKNYyT6/KLrdhzqhiDojvOVVTXY6ux1fpOl+05U4r8kobHARFAjqUSqSfykRBrbNSxGzu+EBERtRZBENAlWIsuwVr8ISEaoijirKkCO7ILkZ5tQlp2Ic6aKrDvnAX7zlnw/rYsCALQK0yHhDgjEmKNGBYbCKOf6vov1s4xiU5E1MEF+6vx/gNDcNfy35CSeRGvbjiK+bd03EKz9dVL12iksi/tUUFBAWw2m7NWqkNoaCiOHj1a73Nyc3Prbe/YaMzx/Xptri4Vo1AoYDQaa7WJjY2tcwzHY/Ul0RctWoSFCxfWuX/jxo3wbeR/YkaBAOD6tQKVgghFO/6gxRU1dsAqXv/Dt44eK8bJda7Gateu4xBPXb/UREez77d9nu6C13H1b/rGX9JQeKRxP1Pl5eVN7BUREVHrEgQB0YG+iA70xe+HRAEALpgrkOZIqmeZkFVQhsM5xTicU4yV208BALqHajEs1oiE2EAkxBkR4q/x4LvwDCbRiYgI/SL1ePWu/njs8714b2sWeoXpMGVQJ093y2NYL73tmj9/fq1V8sXFxYiKisL48eOh0+kadazAbBM+Pr7ruu3+M2MwhsXWTeh3JOnZRZixas9123X0WDFOrnM1VkOGdONK9CvYamzY99s+DLhhAOQKbhh2JeFMKT4+nnXdduNvTGj0SnTHVU9ERERtUYTBB3cMisQdgyIBAPnFlUjLNjlXqh/LK3V+fbLjDAAgLshPSqrHSYn1CIOPJ99Cq2ASnYiIAAC3D+yEIzklWL71JP7+v/2IC/ZD/0iDp7vlUR2hXnpQUBDkcjny8mpvtJaXl4ewsLB6nxMWFnbN9o7veXl5CA8Pr9Vm4MCBzjb5+fm1jlFTUwOTyVTrOPW9zpWvcTW1Wg11PZ90KJVKKJWNq+uX2DUE4XoNci2V9dZlFgCE6TUY2T28Q5ZAutLI7hqE648wVtfBOLnuerECgCB/OfpF+zJZXA+5Qs64XGVQjA4h/soGS7o4fv8Su4Y0+vevseMLERGRNwvRaTB5QAQmD4gAABSWVmHnqSLnZqVHcouRVVCGrIIyfL7zLAAgyuiDYTHSKvXhsYGIMvpAENrX+WwHv1CUiIiu9OSEHripZwiqauyY/XEG8ouvXQ+6o3DUS4+OlpLqpaXSCnW73dM9az6VSoX4+Hhs2rTJeZ/dbsemTZuQmJhY73MSExNrtQeA5ORkZ/vY2FiEhYXValNcXIy0tDRnm8TERJjNZmRkZDjbbN68GXa7HQkJCc4227Ztq1VrNjk5GT169Ki3lIu7yWUCFkzuDUBKrlzJcXvB5N4dPtkJMFauYpxcd71YCQD+Nj4aFTVlKK0uhSiypAtdm1wm4PHxkfU+xt8/IiKihgVq1ZjYNwwLJvfBj4/diL3PjceHM4Zg9qg4DIiUNnk/a6rA/3afw1Nf7ceof2/BDS9vxmOf78HqtDM4ebF9nKsxiU5ERE5ymYAl9w5El2A/5BZX4uFPMlBeXYO0bBMyCgSkZZtgs7f9wa+pfHyA8HApma7RSKvT20MZ1Hnz5uGDDz7ARx99hCNHjuCRRx5BWVkZZs6cCQCYPn16rY1HH3vsMaxfvx6vvfYajh49ihdeeAG7du3C3LlzAUh19h5//HH885//xLfffosDBw5g+vTpiIiIwJQpUwAAvXr1wsSJEzFr1iykp6dj+/btmDt3Lu69915EREgrHv7whz9ApVLhoYcewqFDh/DFF1/gzTffrLOpaUua2Dccy+4fjDB97Zp/YXoNlt0/GBP7hjfwzI6HsXIN4+S668Xq3iG9EKWLglKmRFFlEapqqho4EpFkTE8D/t/UWIT41145zt8/IiIi1+l9lbi5VyiemdQL38wdiX0LxuOjPw7DX8Z0QXznACjlAnIslfhm7wU8s/YAbn5tK4b+axPmfLobH6eeQmZuCewu5BVsdvFyLuJcicdzEe3kYnQiInIXnUaJD6YPwe1Lt2P3GTPiX/oZFVYbADk+Pr4L4XoNFkzu3WEnmoIglXPx8Wk/9dLvueceXLx4Ec8//zxyc3MxcOBArF+/3rmJ55kzZyCTXf7c/YYbbsDq1avx7LPP4plnnkG3bt2wbt069O3b19nmqaeeQllZGWbPng2z2YyRI0di/fr10GguJ8M+/fRTzJ07FzfffDNkMhmmTp2Kt956y/m4Xq/Hxo0bMWfOHMTHxyMoKAjPP/88Zs+e3QpRuWxi33CM6x2G1BP52PhLGsbfmNCky/07AsbKNYyT664XK3+1P3yUPiiqKIKpwoSKmgro1DrIBK4VovqN6WnAjd31SDtZgP17TiNp5FCM6BbG3z8iIqIm0qoVGN09GKO7BwMAKqpt2HOmCDuyTUjPLsTuM2YUlFbhhwM5+OFADgAgwFeJoTFGJMQFIiHWiF7hulpj8fqDOVj43WHkWCoh5SJOInzzeY/mIphEJyKiOuKCtXjwhhi8vfnEpQT6ZbmWSjzyye4Ov2KroXrpbfUqtblz5zpXkl8tJSWlzn1333037r777gaPJwgCXnzxRbz44osNtjEajVi9evU1+9W/f3/88ssv12zTGuQyAQmxRhQeEZEQa2Sy5RoYK9cwTq67XqwUMgWC/YLhp/JDYXkhzJVm+Ch84KNs/xtcUdPIZQIGRPlCeZa/f0RERO7mo5Ljhq5BuKFrEACgqsaGfWctSMsqRPopE3adKkJRuRUbD+dh42Fpzyt/jUJKqscaYbOL+PeGzDr74ng6F8EkOhER1WGzi/gq41y9j4mQaocu/O4wxvXmyi1HvXStFigqklamExFR6/NV+kKj08C/yh8FZQUwVZigU+ugkHHKQ0REROQpaoUcw2KNGBZrBABYbXYcOG9BWpa0Un3XqSKUVNZg89F8bD6a3+BxPJ2L4BklERHVkZ5tunTZVP1EADmWSqRnm5DYJbD1OubFfHykOum+vkBWlqd7Q0TUMckEGQwaA3wUPjBVmFBUUQS5TA6tSgtB6Ngf+hIRERF5A6VchsHRARgcHYBHxnSBzS7i8IVipGUX4qeDucg4XdTgcz2Zi2ASnYiI6sgvaTiB3pR2HYUgSOVdiIjIs9QKNcK0YdCqtCgoL0BRRRH8VH5QK9ro5hVERERE7ZRcJqBfpB79IvUI9ldfM4nu4IlcBHfcISKiOkL8NddvBGDbsYsoqbS2cG+IiIgaTxAE+Kv9EaWPQpg2DNW2apgrzbDZbdd/MhERERG1OldzEa62cycm0YmIqI5hsUaE6zW43oXv/9t9HqNe3YL3t51EpZVJCSIi8j4KmQJBfkGI1kfDX+UPS5UF5dZyT3eLiIiIiK5yvVyEACBcr3HWV29NTKITEVEdcpmABZN7A0CdwUu49DXrxljEBfuhqNyK//fjUYx6dQv+u+M0qmvsrd1dIiKi6/JR+qCTrhMidZGACBRVFMFq49VURERERN7ierkIAFgwuXerbyoKMIlOREQNmNg3HMvuH4wwfe3LpML0Giy7fzD+cWtvbHx8FP59V390Mvggv6QKz607iJtfT8H/Ms7BZhc91HMiIqL6OTYejTZEI9A3EGXVZSipKoEocswiIiIi8gbXy0VM7BvukX5xY1EiImrQxL7hGNc7DKkn8rHxlzSMvzEBiV1DnJ/6KuQy3D0kCr8bGIEvdp7FW5tO4KypAn9bsw/Lt57E38Z3x4Q+YRCE1v+UmIiIqCEquQqhfqHSxqNlBSiqLIKv0hcaRevX1yQiIiKi2hy5iHGvpyCroBx/S+qKv9zU3SMr0B24Ep2IiK5JLhOQEGtEfJCIhFhjvYOWWiHH9MQYbHtqDP4+sSf0Pkoczy/Fw5/sxu/e2Y6txy5ylR8REXkVQRCgVWmljUf9wlBjq4G5ghuPEhEREXkDuUyAVi2t/+4R5u/RBDrAJDoREbmRr0qBR8Z0wbanxuKvN3WFn0qOA+ctmLEiHfe8vwM7T5k83UUiIqJa5DI5gvyCEKWPgk6j48ajRERERFQHk+hEROR2eh8l5o3vgW1PjcWfRsZCpZAhPduEu5en4sGV6Th43uLpLhIREdXio/RBhH8EonRRECDAVGHixqNEREREBIBJdCIiakGBWjWeva03tj45BvcNi4ZCJiAl8yJue/tX/OXTDJzIL/F0F4mIiJxkggx6jR5R+igE+waj3FqO4qpiliQjIiIi6uCYRCciohYXrvfBojv74ed5ozFlYAQEAfjxQC7Gv7EN/7dmH86aeNk8ERF5D5VchVBtKKL0UdDINSiqLEJlTaWnu0VEREREHsIkOhERtZqYID8suXcQ1j82CuN7h8IuAl9lnMNNr6Xg+W8OIr+YCQoiIvIejo1Hw7Xh3HiUiIiIqANjEp2IiFpdjzB/vD99CNbNGYEbuwXBahPxceppjPr3Fiz66QiKyqo93UUiIiIA0sajgb6BiDZEQ6/Ro7iqGGXVZSzxQkRERNSBMIlOREQeMzDKgP8+lIDVsxIwONqASqsd723NwqhXt+DNn4+jtKrG010kIiICAGgUGkT4RyBSFwmZIENRZRGqbfzQl4iIiKgjYBKdiIg87oYuQfjfIzdgxYND0Ctch5KqGrzx8zGMenULPtiWhUorL50nIiLPEwQBeo0e0fpoBPsGo8JageKqYthFu6e7RkREREQtyONJ9KVLlyImJgYajQYJCQlIT09vsO2hQ4cwdepUxMTEQBAELFmypE6bbdu2YfLkyYiIiIAgCFi3bl2dNg8++CAEQaj1NXHiRDe+KyIiaixBEHBTz1D88OhIvPOHQYgL8oOprBr/+vEIRv97Cz5NOw2rjUkKIiLyPKVciVBtKKL10fBR+MBcaUaFtcLT3SICwDk2ERFRS/BoEv2LL77AvHnzsGDBAuzevRsDBgzAhAkTkJ+fX2/78vJyxMXF4eWXX0ZYWFi9bcrKyjBgwAAsXbr0mq89ceJE5OTkOL8+++yzZr8fIiJqPplMwG39I7DxiVF4dWp/ROg1yCuuwj/WHsTNr23F2j3nYLOzDi0REXmen8oPkbpIhGvDYRftKKooQo2dpcjIczjHJiIiahkKT77466+/jlmzZmHmzJkAgOXLl+OHH37AihUr8PTTT9dpP3ToUAwdOhQA6n0cAG655Rbccsst131ttVrd4EkCERF5nkIuw++HRuH2QRH4LO0M3tlyAmdM5Xjii31YlnIS88b1wIQ+oRAEwdNdJSKiDsyx8aifyg+mchOKKougkqvgq/TlGEWtjnNsIiKiluGxJHp1dTUyMjIwf/58530ymQxJSUlITU1t8ddPSUlBSEgIAgICcNNNN+Gf//wnAgMDG2xfVVWFqqoq5+3i4mIAgNVqhdVqbVZfHM9v7nHaO8bJdYyVaxgn13kyVjIA04ZF4o6BYfjvjrN4/5dsHMsrxcOfZKB/Jx0eT+qKkV0CvSZR4a5Y8eeSiKht0Sg0CPcPh1atxcWyiyiqLIJWpYVKrvJ016iD4Bwbzudf+Z0axli5hnFyHWPlOsbKNaIoXYVuq6nx+BzbY0n0goIC2Gw2hIaG1ro/NDQUR48ebdHXnjhxIu68807Exsbi5MmTeOaZZ3DLLbcgNTUVcrm83ucsWrQICxcurHP/xo0b4evr65Z+JScnu+U47R3j5DrGyjWMk+s8HasoAPP7AVsuyJCSI2D/+WL88aPd6KoTcWuUDXE6j3avlubGqry83E09ISKi1iIIAnRqHXwUPiiqLIKp3IQKawX81f6QCR7fjoraOc6xa/P0eWtbwli5hnFyHWPlOsbq2iwWOQABe/buRfXpPc06VnPn2B4t5+Ip9957r/Pf/fr1Q//+/dGlSxekpKTg5ptvrvc58+fPx7x585y3i4uLERUVhfHjx0Ona17Wxmq1Ijk5GePGjYNSqWzWsdozxsl1jJVrGCfXeVus7gJQWFqF5duy8Wn6WZwoBt48pMDo7kF44uau6BPhuWy6u2LlWI1FRERtj1KuRIhfCPyUfiisKIS50gwfhQ98lD6e7hpRi/CmOba3nbd6M8bKNYyT6xgr1zFWrvnP6VSgrASDBg5EUp/wZh2ruXNsjyXRg4KCIJfLkZeXV+v+vLy8Vq+jFhcXh6CgIJw4caLBAV6tVkOtVte5X6lUuu2H3Z3Has8YJ9cxVq5hnFznTbEKC1Dihdv7Yfbornh783F8uescth4rwNZjBbi1fzjmjeuOLsFaj/WvubHyljgTEVHT+an84KP0gaXSgoLyAhRVFMFf7Q+FrEOuZaIWxjl2yxynI2CsXMM4uY6xch1jdW2Osq1yhaLZcWru8z12TaFKpUJ8fDw2bdrkvM9ut2PTpk1ITExs1b6cO3cOhYWFCA9v3icaRETU+iIMPlh0Z3/8PG80fjcgAoIA/LA/B+Ne34on1+zDuSKWRSEiIs+RCTIE+AQgWh+NAE0ASqtLUVpd6qzxSeQunGMTERG1HI8ugZg3bx5mzJiBIUOGYNiwYViyZAnKysqcO4lPnz4dnTp1wqJFiwBIG6UcPnzY+e/z589j79690Gq16Nq1KwCgtLQUJ06ccL5GdnY29u7dC6PRiOjoaJSWlmLhwoWYOnUqwsLCcPLkSTz11FPo2rUrJkyY0MoRICIid4kN8sNb9w3CI2O64LWNx/DzkTysyTiHdXvP4w/DojHnpq4I8dd4uptERNRBqRVq58ajBeUFKKosgp/SD2pF3ZW4RE3FOTYREVHL8GgS/Z577sHFixfx/PPPIzc3FwMHDsT69eudG6GcOXMGMtnlxfIXLlzAoEGDnLcXL16MxYsXY/To0UhJSQEA7Nq1C2PHjnW2cdRYmzFjBlatWgW5XI79+/fjo48+gtlsRkREBMaPH4+XXnqp3kvJiIiobekVrsN/ZgzB7jNFeG1jJrafKMRHqafxxa6zmDkiFn8eFQeDr8rT3SQiog7IsfGor9IXRRVFKCwvRGVNJTceJbfhHJuIiKhleLwY39y5czF37tx6H3MM2g4xMTHXvexxzJgx12zj4+ODDRs2NLqfRETUtgyODsCnfxqO304U4N8bM7HnjBnLUk7ik9TTmD0qDjNHxkKr9vgwSEREHZBCpkCwXzD8VH4oLOfGo+RenGMTERG5H5c7EBFRu3ZD1yB8/cgN+M/0IegZ5o+Sqhq8lnwMo17dgv/8koVKq83TXSQiog7KV+mLTrpO6OTfCaIowlRhQo29xtPdIiIiIqKrMIlORETtniAISOodih//eiPeum8QYoP8YCqrxj9/OIKxi1PwWfoZWG12T3eTiIg6IMfGo1H6KAT6BHLjUSIiIiIvxCQ6ERF1GDKZgN8NiEDyE6PwytR+iNBrkGOpxPyvDyDp9a34Zu952O1MWhARUetTK9QI04YhShcFpUyJosoiVNVUebpbRERERAQm0YmIqANSyGW4Z2g0Nv/fGDx/W28E+qlwurAcj32+F7e8+Qs2HsrlCkAiImp1giDAX+2PKH0UwvzCUG2rhrnSDLvIq6WIiIiIPIlJdCIi6rA0Sjn+ODIW254aiycn9IC/RoHMvBLM/m8Gprz7G349XsBkOhERtTqFTIEgvyBE6aPgr/KHpdKCcmu5p7tFRERE1GExiU5ERB2en1qBOWO74tenbsKcsV3go5Rj31kz7v8wDX/4IA0Zp4s83UUiIuqAnBuP6joBIlBUUQSrzerpbhERERF1OEyiExERXaL3VeLJCT2x7amxePCGGKjkMqRmFWLqst/w0KqdOHyh2NNdJCKiDkYmyGDQGBBtiIbRx4iy6jKUVJXwSikiIiKiVsQkOhER0VWC/dV44Xd9sOXJMbhnSBTkMgGbjuZj0lu/YO7q3ci6WOrpLrqNyWTCtGnToNPpYDAY8NBDD6G09Nrvr7KyEnPmzEFgYCC0Wi2mTp2KvLy8Wm3OnDmDW2+9Fb6+vggJCcGTTz6JmpqaWm1SUlIwePBgqNVqdO3aFatWrar1+KJFizB06FD4+/sjJCQEU6ZMQWZmplveNxFRW6OSq6SNR/VRUMvVKKooQmVNpae7RURERNQhMIlORETUgE4GH7xyV38kPzEKkwdEAAC+35+DcW9sw9+/2o/z5opa7W12EWnZJmQUCEjLNsFm9/5VgtOmTcOhQ4eQnJyM77//Htu2bcPs2bOv+ZwnnngC3333HdasWYOtW7fiwoULuPPOO52P22w23HrrraiursZvv/2Gjz76CKtWrcLzzz/vbJOdnY1bb70VY8eOxd69e/H444/jT3/6EzZs2OBss3XrVsyZMwc7duxAcnIyrFYrxo8fj7KyMvcHgoioDXBsPBqpj0SYNgxWmxWWSgtsdpunu0ZERETkVja7iNIqaSFWZm6Jx+fXCo++OhERURsQF6zF2/cNwsOj4/D6xmPYdDQfX+w6i7V7zuMPCdGYM7YrMk6bsPC7w8ixVAKQ4+PjuxCu12DB5N6Y2Dfc02+hXkeOHMH69euxc+dODBkyBADw9ttvY9KkSVi8eDEiIiLqPMdiseDDDz/E6tWrcdNNNwEAVq5ciV69emHHjh0YPnw4Nm7ciMOHD+Pnn39GaGgoBg4ciJdeegl///vf8cILL0ClUmH58uWIjY3Fa6+9BgDo1asXfv31V7zxxhuYMGECAGD9+vW1XnvVqlUICQlBRkYGRo0a1ZKhISLyao6NR/1UfiisKIS50gyNQgO1oPZ014iIiIiabf3BnCvm18BrP5/A6p3nPDq/ZhKdiIjIRX0i9PjwwaHIOF2Ef284ih1ZJqz67RRWp51Btc1ep32upRKPfLIby+4f7JWJ9NTUVBgMBmcCHQCSkpIgk8mQlpaGO+64o85zMjIyYLVakZSU5LyvZ8+eiI6ORmpqKoYPH47U1FT069cPoaGhzjYTJkzAI488gkOHDmHQoEFITU2tdQxHm8cff7zB/losFgCA0WhssE1VVRWqqqqct4uLpTr2VqsVVmvzNuNzPL+5x+kIGCvXME6uY6zqp4ACwZpg+Mh8UFheiMKqQgCArYYr06/HEaMaaw1kQvMu0ObPJRERkfusP5iDRz7ZjavXnXt6fs0kOhERUSPFdw7AZ7OGY/uJQry64Sj2n7PU204EIABY+N1hjOsdBrlMaNV+Xk9ubi5CQkJq3adQKGA0GpGbm9vgc1QqFQwGQ637Q0NDnc/Jzc2tlUB3PO547FptiouLUVFRAR8fn1qP2e12PP744xgxYgT69u3b4HtatGgRFi5cWOf+jRs3wtfXt8HnNUZycrJbjtMRMFauYZxcx1i5Zt9v+zzdhTbj559/bvYxysvL3dATIiIistlFLPzucJ0EOuD5+TWT6ERERE0gCAJGdguCXNYT932Q1mA7EUCOpRLp2SYkdglslb49/fTTeOWVV67Z5siRI63SF3eZM2cODh48iF9//fWa7ebPn4958+Y5bxcXFyMqKgrjx4+HTqdrVh+sViuSk5Mxbtw4KJXKZh2rvWOsXMM4uY6xck11dTV+/vlnJI5JhEKhgChKU1Dx0lT06tt2u3QVlR32WrddefzqYzpc7zUbun01ESIECC5/r0VwfJP+IQhCndvV1mqcyjiFpKQkqFSqevvgKsdVT0RERNQ86dkmZwmX+nhifu3AJDoREVEz5JdUXb8RgPyShk8E3O1vf/sbHnzwwWu2iYuLQ1hYGPLz82vdX1NTA5PJhLCwsHqfFxYWhurqapjN5lqr0fPy8pzPCQsLQ3p6eq3n5eXlOR9zfHfcd2UbnU5XZxX63LlznZueRkZGXvN9qdVqqNV1awIrlUq3Jd7ceaz2jrFyDePkOsbKNQG+Aa0Sp6sT4vXd54nbouhI84sN3q5RSBuVueNnij+TRERE7uHqvLk159cOTKITERE1Q4i/xq3t3CE4OBjBwcHXbZeYmAiz2YyMjAzEx8cDADZv3gy73Y6EhIR6nxMfHw+lUolNmzZh6tSpAIDMzEycOXMGiYmJzuP+61//Qn5+vrNcTHJyMnQ6HXr37u1s8+OPP9Y6dnJysvMYgJQIefTRR7F27VqkpKQgNja2kZEgIqKWdPUK70s32gSr1Yp92Od8D0REROR53ji/dmjeDipEREQd3LBYI8L1mgZzBgKAcL0Gw2Ib3gzTU3r16oWJEydi1qxZSE9Px/bt2zF37lzce++9iIiIAACcP38ePXv2dK4s1+v1eOihhzBv3jxs2bIFGRkZmDlzJhITEzF8+HAAwPjx49G7d2888MAD2LdvHzZs2IBnn30Wc+bMca4Sf/jhh5GVlYWnnnoKR48exbvvvosvv/wSTzzxhLN/c+bMwSeffILVq1fD398fubm5yM3NRUVFRStHioiIiIiIiFqaN8+vmUQnIiJqBrlMwILJ0urqqwd6x+0Fk3t73aaiDp9++il69uyJm2++GZMmTcLIkSPx/vvvOx+3Wq3IzMystWnaG2+8gdtuuw1Tp07FqFGjEBYWhq+//tr5uFwux/fffw+5XI7ExETcf//9mD59Ol588UVnm9jYWPzwww9ITk7GgAED8Nprr+E///kPJkyY4GyzbNkyWCwWjBkzBuHh4c6vL774ooWjQkRERERERK3Nm+fXLOdCRETUTBP7hmPZ/YOx8LvDtTZBCdNrsGByb0zsG+7B3l2b0WjE6tWrG3w8JiamzqZvGo0GS5cuxdKlSxt8XufOneuUa7namDFjsGfPngYfb2izOSIiIiIiImqfvHV+zSQ6ERGRG0zsG45xvcOQeiIfG39Jw/gbE5DYNcRrV6ATEREREREReSNvnF8ziU5EROQmcpmAhFgjCo+ISIg1MoFORERERERE1ATeNr9mTXQiIiIiIiIiIiIiogYwiU5ERERERERERERE1AAm0YmIiIiIiIiIiIiIGsAkOhERERERERERERFRA5hEJyIiIiIiIiIiIiJqAJPoREREREREREREREQNUHi6A22VKIoAgOLi4mYfy2q1ory8HMXFxVAqlc0+XnvFOLmOsXIN4+Q6xsp17oqVY3xxjDfUNByvPYOxcg3j5DrGyjWMk+vcGSuO2e7hrjGbvweuY6xcwzi5jrFyHWPlGm8ar5lEb6KSkhIAQFRUlId7QkRE7VlJSQn0er2nu9FmcbwmIqLWwjG7eThmExFRa2jqeC2I/Li8Sex2Oy5cuAB/f38IgtCsYxUXFyMqKgpnz56FTqdzUw/bH8bJdYyVaxgn1zFWrnNXrERRRElJCSIiIiCTsfpaU3G89gzGyjWMk+sYK9cwTq5zZ6w4ZruHu8Zs/h64jrFyDePkOsbKdYyVa7xpvOZK9CaSyWSIjIx06zF1Oh1/cVzAOLmOsXIN4+Q6xsp17ogVV7M1H8drz2KsXMM4uY6xcg3j5Dp3xYpjdvO5e8zm74HrGCvXME6uY6xcx1i5xhvGa35MTkRERERERERERETUACbRiYiIiIiIiIiIiIgawCS6F1Cr1ViwYAHUarWnu+LVGCfXMVauYZxcx1i5jrFqv/h/6zrGyjWMk+sYK9cwTq5jrNov/t+6jrFyDePkOsbKdYyVa7wpTtxYlIiIiIiIiIiIiIioAVyJTkRERERERERERETUACbRiYiIiIiIiIiIiIgawCQ6EREREREREREREVEDmEQnIiIiIiIiIiIiImoAk+getnTpUsTExECj0SAhIQHp6eme7pLX2bZtGyZPnoyIiAgIgoB169Z5ukteadGiRRg6dCj8/f0REhKCKVOmIDMz09Pd8krLli1D//79odPpoNPpkJiYiJ9++snT3fJ6L7/8MgRBwOOPP+7prnidF154AYIg1Prq2bOnp7tFbsTx2jUcs13DMds1HK+bjmN2wzhme5/GjrFr1qxBz549odFo0K9fP/z444+1HhdFEc8//zzCw8Ph4+ODpKQkHD9+vFYbk8mEadOmQafTwWAw4KGHHkJpaWmtNvv378eNN94IjUaDqKgovPrqq+55w83gjbFKSUnB7bffjvDwcPj5+WHgwIH49NNP3femm8Ab43SlEydOwN/fHwaDoVnv0x28NVaiKGLx4sXo3r071Go1OnXqhH/961/uedNN4K1x2rBhA4YPHw5/f38EBwdj6tSpOHXqlFvec1N5Ilb/+te/cMMNN8DX17fB36szZ87g1ltvha+vL0JCQvDkk0+ipqamcW9OJI/5/PPPRZVKJa5YsUI8dOiQOGvWLNFgMIh5eXme7ppX+fHHH8V//OMf4tdffy0CENeuXevpLnmlCRMmiCtXrhQPHjwo7t27V5w0aZIYHR0tlpaWerprXufbb78Vf/jhB/HYsWNiZmam+Mwzz4hKpVI8ePCgp7vmtdLT08WYmBixf//+4mOPPebp7nidBQsWiH369BFzcnKcXxcvXvR0t8hNOF67jmO2azhmu4bjddNwzL42jtnepbFj7Pbt20W5XC6++uqr4uHDh8Vnn31WVCqV4oEDB5xtXn75ZVGv14vr1q0T9+3bJ/7ud78TY2NjxYqKCmebiRMnigMGDBB37Ngh/vLLL2LXrl3F++67z/m4xWIRQ0NDxWnTpokHDx4UP/vsM9HHx0d87733Wi4Y1+GtsfrXv/4lPvvss+L27dvFEydOiEuWLBFlMpn43XfftVwwrsFb4+RQXV0tDhkyRLzllltEvV7v9vffGN4cq0cffVTs0aOH+M0334hZWVnirl27xI0bN7ZMIK7DW+OUlZUlqtVqcf78+eKJEyfEjIwMcdSoUeKgQYNaLhjX4alYPf/88+Lrr78uzps3r97fq5qaGrFv375iUlKSuGfPHvHHH38Ug4KCxPnz5zfq/TGJ7kHDhg0T58yZ47xts9nEiIgIcdGiRR7slXfjhNx1+fn5IgBx69atnu5KmxAQECD+5z//8XQ3vFJJSYnYrVs3MTk5WRw9z5+ymAAAEaVJREFUejQn5PVYsGCBOGDAAE93g1oIx+um4ZjtOo7ZruN4fW0cs6+PY7Z3aewY+/vf/1689dZba92XkJAg/vnPfxZFURTtdrsYFhYm/vvf/3Y+bjabRbVaLX722WeiKIri4cOHRQDizp07nW1++uknURAE8fz586IoiuK7774rBgQEiFVVVc42f//738UePXo08x03nbfGqj6TJk0SZ86c2fg36QbeHqennnpKvP/++8WVK1d6PInurbE6fPiwqFAoxKNHj7rnjTaTt8ZpzZo1okKhEG02m7PNt99+KwqCIFZXVzfzXTeNJ2J1pYZ+r3788UdRJpOJubm5zvuWLVsm6nS6Wn/nr4flXDykuroaGRkZSEpKct4nk8mQlJSE1NRUD/aM2guLxQIAMBqNHu6Jd7PZbPj8889RVlaGxMRET3fHK82ZMwe33nprrb9XVNfx48cRERGBuLg4TJs2DWfOnPF0l8gNOF5Ta+CYfX0cr13DMds1HLO9Q1PG2NTU1Do/3xMmTHC2z87ORm5ubq02er0eCQkJzjapqakwGAwYMmSIs01SUhJkMhnS0tKcbUaNGgWVSlXrdTIzM1FUVNTMd9543hyr+lgsFo+Mad4ep82bN2PNmjVYunRp899sM3lzrL777jvExcXh+++/R2xsLGJiYvCnP/0JJpPJPW++Ebw5TvHx8ZDJZFi5ciVsNhssFgv++9//IikpCUql0j0BaARPxcoVqamp6NevH0JDQ2u9TnFxMQ4dOuTycZhE95CCggLYbLZa/4EAEBoaitzcXA/1itoLu92Oxx9/HCNGjEDfvn093R2vdODAAWi1WqjVajz88MNYu3Ytevfu7elueZ3PP/8cu3fvxqJFizzdFa+WkJCAVatWYf369Vi2bBmys7Nx4403oqSkxNNdo2bieE0tjWP2tXG8dh3HbNdwzPYeTRljc3Nzr9ne8f16bUJCQmo9rlAoYDQaa7Wp7xhXvkZr8uZYXe3LL7/Ezp07MXPmTBffnft4c5wKCwvx4IMPYtWqVdDpdE18h+7jzbHKysrC6dOnsWbNGnz88cdYtWoVMjIycNdddzXx3TadN8cpNjYWGzduxDPPPAO1Wg2DwYBz587hyy+/bOK7bR5PxcoV7vqbrnC5JRG1GXPmzMHBgwfx66+/erorXqtHjx7Yu3cvLBYLvvrqK8yYMQNbt27lxPwKZ8+exWOPPYbk5GRoNBpPd8er3XLLLc5/9+/fHwkJCejcuTO+/PJLPPTQQx7sGRF5O47Z18bx2jUcs13HMZuo5WzZsgUzZ87EBx98gD59+ni6O15l1qxZ+MMf/oBRo0Z5uitez263o6qqCh9//DG6d+8OAPjwww8RHx+PzMxM9OjRw8M99A65ubmYNWsWZsyYgfvuuw8lJSV4/vnncddddyE5ORmCIHi6i+0OV6J7SFBQEORyOfLy8mrdn5eXh7CwMA/1itqDuXPn4vvvv8eWLVsQGRnp6e54LZVKha5duyI+Ph6LFi3CgAED8Oabb3q6W14lIyMD+fn5GDx4MBQKBRQKBbZu3Yq33noLCoUCNpvN0130WgaDAd27d8eJEyc83RVqJo7X1JI4Zl8fx2vXcMxuOo7ZntOUMTYsLOya7R3fr9cmPz+/1uM1NTUwmUy12tR3jCtfozV5c6wctm7dismTJ+ONN97A9OnTG/kO3cOb47R582YsXrzY+Tf6oYcegsVigUKhwIoVK5r4jpvOm2MVHh4OhULhTKADQK9evQCg1ctveXOcli5dCr1ej1dffRWDBg3CqFGj8Mknn2DTpk3XLLfUUjwVK1e46286k+geolKpEB8fj02bNjnvs9vt2LRpE+s8UpOIooi5c+di7dq12Lx5M2JjYz3dpTbF8Wk3XXbzzTfjwIED2Lt3r/NryJAhmDZtGvbu3Qu5XO7pLnqt0tJSnDx5EuHh4Z7uCjUTx2tqCRyzm47jdf04Zjcdx2zPacoYm5iYWKs9ACQnJzvbx8bGIiwsrFab4uJipKWlOdskJibCbDYjIyPD2Wbz5s2w2+1ISEhwttm2bRusVmut1+nRowcCAgKa+c4bz5tjBQApKSm49dZb8corr2D27NnNf8NN5M1xSk1NrfU3+sUXX4S/vz/27t2LO+64wz0BaARvjtWIESNQU1ODkydPOtscO3YMANC5c+fmvO1G8+Y4lZeXQyarndZ1jPd2u72pb7nJPBUrVyQmJuLAgQO1PphITk6GTqdr3NWNLm9BSm73+eefi2q1Wly1apV4+PBhcfbs2aLBYKi1WyyJYklJibhnzx5xz549IgDx9ddfF/fs2SOePn3a013zKo888oio1+vFlJQUMScnx/lVXl7u6a55naefflrcunWrmJ2dLe7fv198+umnRUEQxI0bN3q6a15v9OjR4mOPPebpbnidv/3tb2JKSoqYnZ0tbt++XUxKShKDgoLE/Px8T3eN3IDjtes4ZruGY7ZrOF43D8fs+nHM9i7XG2MfeOAB8emnn3a23759u6hQKMTFixeLR44cERcsWCAqlUrxwIEDzjYvv/yyaDAYxG+++Ubcv3+/ePvtt4uxsbFiRUWFs83EiRPFQYMGiWlpaeKvv/4qduvWTbzvvvucj5vNZjE0NFR84IEHxIMHD4qff/656OvrK7733nutEJX6eWusNm/eLPr6+orz58+vNaYVFha2QlTq8tY4XW3lypWiXq93fwAawVtjZbPZxMGDB4ujRo0Sd+/eLe7atUtMSEgQx40b1wpRqctb47Rp0yZREARx4cKF4rFjx8SMjAxxwoQJYufOnT12TumpWJ0+fVrcs2ePuHDhQlGr1TrnIyUlJaIoimJNTY3Yt29fcfz48eLevXvF9evXi8HBweL8+fMb9f6YRPewt99+W4yOjhZVKpU4bNgwcceOHZ7uktfZsmWLCKDO14wZMzzdNa9SX4wAiCtXrvR017zOH//4R7Fz586iSqUSg4ODxZtvvpkTchdxQl6/e+65RwwPDxdVKpXYqVMn8Z577hFPnDjh6W6RG3G8dg3HbNdwzHYNx+vm4ZhdP47Z3udaY+zo0aPrjCFffvml2L17d1GlUol9+vQRf/jhh1qP2+128bnnnhNDQ0NFtVot3nzzzWJmZmatNoWFheJ9990narVaUafTiTNnznQmWxz27dsnjhw5UlSr1WKnTp3El19+2b1vvAm8MVYzZsyod0wbPXq029+/q7wxTlfzhiS6KHpvrM6fPy/eeeedolarFUNDQ8UHH3zQYx/MiKL3xumzzz4TBw0aJPr5+YnBwcHi7373O/HIkSPuffON5IlYNfR3aMuWLc42p06dEm+55RbRx8dHDAoKEv/2t7+JVqu1Ue9NEEVRdH3dOhERERERERERERFRx8Ga6EREREREREREREREDWASnYiIiIiIiIiIiIioAUyiExERERERERERERE1gEl0IiIiIiIiIiIiIqIGMIlORERERERERERERNQAJtGJiIiIiIiIiIiIiBrAJDoRERERERERERERUQOYRCciIiIiIiIiIiIiagCT6ETkdQRBwLp16zzdDSIiIroGjtdERERNM2bMGDz++ONuOdYLL7yAgQMHuuVYRNQwJtGJqJYHH3wQgiDU+Zo4caKnu0ZERESXcLwmIiLybg2N1SdOnMDXX3+Nl156ydk2JiYGS5Ysue4x6/sA+//+7/+wadMmN/eeiK6m8HQHiMj7TJw4EStXrqx1n1qt9lBviIiIqD4cr4mIiLxbfWN1cHAw5HK5215Dq9VCq9W67XhEVD+uRCeiOtRqNcLCwmp9BQQEAJA++V62bBluueUW+Pj4IC4uDl999VWt5x84cAA33XQTfHx8EBgYiNmzZ6O0tLRWmxUrVqBPnz5Qq9UIDw/H3Llzaz1eUFCAO+64A76+vujWrRu+/fZb52NFRUWYNm0agoOD4ePjg27dutU5MSEiImrvOF4TERF5t/rGarlcXqucy5gxY3D69Gk88cQTztXq9YmJiQEA3HHHHRAEwXn76nIuDz74IKZMmYL/9//+H0JDQ2EwGPDiiy+ipqYGTz75JIxGIyIjI+uMyWfPnsXvf/97GAwGGI1G3H777Th16pSbI0LUdjGJTkSN9txzz2Hq1KnYt28fpk2bhnvvvRdHjhwBAJSVlWHChAkICAjAzp07sWbNGvz888+1Jt3Lli3DnDlzMHv2bBw4cADffvstunbtWus1Fi5ciN///vfYv38/Jk2ahGnTpsFkMjlf//Dhw/jpp59w5MgRLFu2DEFBQa0XACIiojaA4zUREZH3+/rrrxEZGYkXX3wROTk5yMnJqbfdzp07AQArV65ETk6O83Z9Nm/ejAsXLmDbtm14/fXXsWDBAtx2220ICAhAWloaHn74Yfz5z3/GuXPnAABWqxUTJkyAv78/fvnlF2zfvh1arRYTJ05EdXW1+980UVskEhFdYcaMGaJcLhf9/Pxqff3rX/8SRVEUAYgPP/xwreckJCSIjzzyiCiKovj++++LAQEBYmlpqfPxH374QZTJZGJubq4oiqIYEREh/uMf/2iwDwDEZ5991nm7tLRUBCD+9NNPoiiK4uTJk8WZM2e65w0TERG1QRyviYiIvFt9Y/Vdd90liqIojh49WnzsscecbTt37iy+8cYb1z0mAHHt2rW17luwYIE4YMCAWq/buXNn0WazOe/r0aOHeOONNzpv19TUiH5+fuJnn30miqIo/ve//xV79Ogh2u12Z5uqqirRx8dH3LBhQyPeNVH7xZroRFTH2LFjsWzZslr3GY1G578TExNrPZaYmIi9e/cCAI4cOYIBAwbAz8/P+fiIESNgt9uRmZkJQRBw4cIF3HzzzdfsQ//+/Z3/9vPzg06nQ35+PgDgkUcewdSpU7F7926MHz8eU6ZMwQ033NCk90pERNRWcbwmIiLybleP1VeOuy2pT58+kMkuF58IDQ1F3759nbflcjkCAwOdY/a+fftw4sQJ+Pv71zpOZWUlTp482Sp9JvJ2TKITUR1+fn51Ltd2Fx8fH5faKZXKWrcFQYDdbgcA3HLLLTh9+jR+/PFHJCcn4+abb8acOXOwePFit/eXiIjIW3G8JiIi8m4tOVZfS33j87XG7NLSUsTHx+PTTz+tc6zg4OCW6yhRG8Ka6ETUaDt27Khzu1evXgCAXr16Yd++fSgrK3M+vn37dshkMvTo0QP+/v6IiYnBpk2bmtWH4OBgzJgxA5988gmWLFmC999/v1nHIyIiam84XhMREbUNKpUKNpvtuu2USqVL7Rpr8ODBOH78OEJCQtC1a9daX3q93u2vR9QWMYlORHVUVVUhNze31ldBQYHz8TVr1mDFihU4duwYFixYgPT0dOdGZNOmTYNGo8GMGTNw8OBBbNmyBY8++igeeOABhIaGApB2D3/ttdfw1ltv4fjx49i9ezfefvttl/v3/PPP45tvvsGJEydw6NAhfP/9986kABERUUfB8ZqIiKh9iImJwbZt23D+/PlaY3l97TZt2oTc3FwUFRW57fWnTZuGoKAg3H777fjll1+QnZ2NlJQU/PWvf3VuPkrU0TGJTkR1rF+/HuHh4bW+Ro4c6Xx84cKF+Pzzz9G/f398/PHH+Oyzz9C7d28AgK+vLzZs2ACTyYShQ4firrvuws0334x33nnH+fwZM2ZgyZIlePfdd9GnTx/cdtttOH78uMv9U6lUmD9/Pvr3749Ro0ZBLpfj888/d18AiIiI2gCO10RERO3Diy++iFOnTqFLl//fzh3bUAjDABR0Wgo2gOZXlGyCWIF5mIUB6DIVXZjAv0JCSHcL2F2kJyu/v9+n7Pse53nGMAwxz/Nj87uui1prjOMY67rGNE2xbVtc1xV93z82B76stNba20sA31FKieM4YlmWt1cBABLeawAAeI5LdAAAAAAASIjoAAAAAACQ8J0LAAAAAAAkXKIDAAAAAEBCRAcAAAAAgISIDgAAAAAACREdAAAAAAASIjoAAAAAACREdAAAAAAASIjoAAAAAACQENEBAAAAACBxA70rloNnYq+WAAAAAElFTkSuQmCC",
            "text/plain": [
              "<Figure size 1800x400 with 3 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "prompt_tuner.plot_learning_curve()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "<a id=\"deploy\"></a>\n",
        "## Deploy"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "You can specify `model_id` from tuning details."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 43,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "'23854716-30a3-4d89-8931-61e3a5f1c08b'"
            ]
          },
          "execution_count": 43,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "model_id = None\n",
        "\n",
        "if 'model_id' in tuning_details.get('entity', {}):\n",
        "    model_id = tuning_details['entity']['model_id']\n",
        "model_id"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Create online deployment for published model."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 44,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "\n",
            "#######################################################################################\n",
            "\n",
            "Synchronous deployment creation for uid: '23854716-30a3-4d89-8931-61e3a5f1c08b' started\n",
            "\n",
            "#######################################################################################\n",
            "\n",
            "\n",
            "initializing\n",
            "ready\n",
            "\n",
            "\n",
            "------------------------------------------------------------------------------------------------\n",
            "Successfully finished deployment creation, deployment_uid='92cbc6b8-7202-41df-9143-aced05bb23aa'\n",
            "------------------------------------------------------------------------------------------------\n",
            "\n",
            "\n"
          ]
        }
      ],
      "source": [
        "from datetime import datetime\n",
        "meta_props = {\n",
        "    client.deployments.ConfigurationMetaNames.NAME: \"PT DEPLOYMENT SDK - project\",\n",
        "    client.deployments.ConfigurationMetaNames.ONLINE: {},\n",
        "    client.deployments.ConfigurationMetaNames.SERVING_NAME : f\"pt_sdk_deployment_{datetime.utcnow().strftime('%Y_%m_%d_%H%M%S')}\"\n",
        "}\n",
        "deployment_details = client.deployments.create(model_id, meta_props)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Additionally you can get deployment details by printing `deployment_details`"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "deployment_details"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "You can specify `deployment_id` from deployment details."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 46,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "'92cbc6b8-7202-41df-9143-aced05bb23aa'"
            ]
          },
          "execution_count": 46,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "deployment_id = deployment_details['metadata']['id']\n",
        "deployment_id"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "<a id=\"models_inference\"></a>\n",
        "## Foundation Models Inference on `watsonx.ai`"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Provide a set of model parameters."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 47,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "from ibm_watsonx_ai.metanames import GenTextParamsMetaNames as GenParams\n",
        "\n",
        "generate_params = {\n",
        "    GenParams.MAX_NEW_TOKENS: 3,\n",
        "    GenParams.STOP_SEQUENCES: [\"\\n\\n\"]\n",
        "}"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Initialize the `ModelInference` class."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 48,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "from ibm_watsonx_ai.foundation_models import ModelInference\n",
        "\n",
        "tuned_model = ModelInference(\n",
        "    deployment_id=deployment_id,\n",
        "    params=generate_params,\n",
        "    api_client=client\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Alternatively you can use `credentials` and `project_id` to initialize the `ModelInference` class.\n",
        "\n",
        "```\n",
        "tuned_model = ModelInference(\n",
        "    deployment_id=deployment_id,\n",
        "    params=generate_params,\n",
        "    credentials=credentials,\n",
        "    project_id=project_id\n",
        ")\n",
        "\n",
        "```"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Get deployment model inference details."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "tuned_model.get_details()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Analyze the product class for a sample prompt."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 50,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "'0'"
            ]
          },
          "execution_count": 50,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "response = tuned_model.generate_text(prompt=\"Extract the satisfaction from the comment. Return simple 1 for satisfied customer or 0 for unsatisfied.\\nComment: I was penalty charged because vehicle was not returned with full tank.\\nSatisfaction:\\n\")\n",
        "response"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "<a id=\"predict\"></a>\n",
        "## Analyze the product classes."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Download and prepare the `cfpb_test` dataset."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 51,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "\n",
        "filename = 'car_rental_prompt_tuning_testing_data.json'\n",
        "\n",
        "url = \"https://raw.github.com/IBM/watson-machine-learning-samples/master/cloud/data/prompt_tuning/car_rental_prompt_tuning_testing_data.json\"\n",
        "if not os.path.isfile(filename):\n",
        "    wget.download(url)\n",
        "\n",
        "data = pd.read_json(filename)\n",
        "prompts = list(data.input)\n",
        "satisfaction = list(data.output)\n",
        "prompts_batch = [\"\\n\".join([prompt]) for prompt in prompts]"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "### Calculate the accuracy of tuned model"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 52,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "tuned_model_results = tuned_model.generate_text(prompt=prompts_batch)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 53,
      "metadata": {
        "collapsed": false,
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "tuned_model_results = [int(x) for x in tuned_model_results]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 54,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "accuracy_score: 0.916083916083916\n"
          ]
        }
      ],
      "source": [
        "from sklearn.metrics import accuracy_score\n",
        "\n",
        "print(f'accuracy_score: {accuracy_score(satisfaction, tuned_model_results)}')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "### Calculate the accuracy of base model"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Initialize the `ModelInference` class with base model."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 55,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "base_model = ModelInference(\n",
        "    model_id='ibm/granite-13b-instruct-v2',\n",
        "    params=generate_params,\n",
        "    api_client=client\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 56,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "base_model_results = base_model.generate_text(prompt=prompts_batch)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 57,
      "metadata": {
        "collapsed": false,
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [],
      "source": [
        "base_model_results = [int(x) for x in base_model_results]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 58,
      "metadata": {
        "pycharm": {
          "name": "#%%\n"
        }
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "base model accuracy_score: 0.8181818181818182\n"
          ]
        }
      ],
      "source": [
        "from sklearn.metrics import accuracy_score\n",
        "\n",
        "print(f'base model accuracy_score: {accuracy_score(satisfaction, base_model_results)}')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "<a id=\"summary\"></a>\n",
        "## Summary and next steps\n",
        "\n",
        " You successfully completed this notebook!\n",
        " \n",
        " You learned how to use prompt tuning in watsonx for analyze the product classes for a sample prompts.\n",
        " \n",
        "Check out our _<a href=\"https://ibm.github.io/watson-machine-learning-sdk/samples.html\" target=\"_blank\" rel=\"noopener no referrer\">Online Documentation</a>_ for more samples, tutorials, documentation, how-tos, and blog posts. "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "### Authors: \n",
        "\n",
        "**Lukasz Cmielowski, PhD**, Senior Technical Staff Member watsonx.ai.\n",
        "\n",
        "**Mateusz Szewczyk**, Software Engineer at watsonx.ai."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "pycharm": {
          "name": "#%% md\n"
        }
      },
      "source": [
        "Copyright © 2024-2025 IBM. This notebook and its source code are released under the terms of the MIT License."
      ]
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "notebook-samples",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.11.9"
    },
    "orig_nbformat": 4
  },
  "nbformat": 4,
  "nbformat_minor": 2
}