notebooks/getting-started.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Copyright 2022 Google LLC\n",
    "#\n",
    "# Licensed under the Apache License, Version 2.0 (the \"License\");\n",
    "# you may not use this file except in compliance with the License.\n",
    "# You may obtain a copy of the License at\n",
    "#\n",
    "#      http://www.apache.org/licenses/LICENSE-2.0\n",
    "#\n",
    "# Unless required by applicable law or agreed to in writing, software\n",
    "# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
    "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
    "# See the License for the specific language governing permissions and\n",
    "# limitations under the License."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Template to run a T5X job on Vertex AI\n",
    "\n",
    "This notebook serves as a template for configuring and running a T5X job on\n",
    "Google Vertex AI. The job can be a fine-tuning job or a training job from\n",
    "scratch, or it can be an evaluation job or an inference job. The template lets\n",
    "you clone the notebook for a specific job, task, dataset, or configuration and\n",
    "get a quick start on running a T5X job on Vertex AI.\n",
    "\n",
    "## Objective\n",
    "\n",
    "This template notebook performs three tasks. First, it runs through a\n",
    "fine-tuning job from a pre-trained model for an English to German translation\n",
    "task using the [T5X framework](https://github.com/google-research/t5x) on\n",
    "[Vertex AI](https://cloud.google.com/vertex-ai) with the\n",
    "[`wmt_t2t_translate`](https://www.tensorflow.org/datasets/catalog/wmt_t2t_translate)\n",
    "dataset. It then performs the additional tasks of evaluation and of batch\n",
    "inference on the fine-tuned model. \n",
    "\n",
    "The notebook is organized as following:\n",
    "\n",
    "-  [Preparation: Imports and initialization](#preparation):\n",
    "    Initialize and configure an experiment for any of the tasks to run.\n",
    "-  [Task 1: Fine-tune a pre-trained model on Vertex AI](#task1):\n",
    "    Configure and fine-tune a pretrained T5 base model for translation task\n",
    "    with T5X on Vertex AI.\n",
    "-  [Task 2: Run model evaluation on Vertex AI](#task2):\n",
    "    Configure and evaluate a trained model with T5X on Vertex AI.\n",
    "-  [Task 3: Run batch inference on Vertex AI](#task3):\n",
    "    Configure and run batch inference using the trained model with T5X on Vertex AI.\n",
    "\n",
    "## Before you begin\n",
    "\n",
    "Before you run this notebook, review the prerequisites:\n",
    "\n",
    "1. For background about how to run and monitor T5X jobs with Vertex AI,\n",
    "    including environment setup and data preparation tasks, see the [README\n",
    "    file](README.md) for this repository.\n",
    "1. Ensure that you have set up the development environment and installed the\n",
    "    required libraries as described in the environment setup of the [README\n",
    "    file](README.md). "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "toc-hr-collapsed": true
   },
   "source": [
    "## <a name=\"preparation\"></a>Preparation: Imports and initialization\n",
    "\n",
    "In this section, you import required libraries, configure environment settings,\n",
    "and initialize an experiment in Vertex AI."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Call the\n",
    "[`autoreload`](https://ipython.org/ipython-doc/3/config/extensions/autoreload.html)\n",
    "extension to reload modules automatically before executing code."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Reloads modules automatically before executing any code/script\n",
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import libraries\n",
    "\n",
    "Import the required libraries."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import time\n",
    "from datetime import datetime\n",
    "import pandas as pd\n",
    "\n",
    "# Import the Vertex AI SDK for Python\n",
    "from google.cloud import aiplatform as vertex_ai"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Import [`utils.py`](utils.py), which is a\n",
    "custom script that has utility functions to streamline configuration and\n",
    "submission of a Vertex AI custom job and to track the generated artifacts and\n",
    "metrics from the custom job."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import utils"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Configure environment settings\n",
    "\n",
    "Based on the [environment setup](README.md) you did previously, configure the\n",
    "following environment settings:\n",
    "\n",
    "-  `PROJECT_ID`: Configure the Google Cloud project ID.\n",
    "-  `REGION`: Configure the\n",
    "    [region](https://cloud.google.com/vertex-ai/docs/general/locations) to use\n",
    "    for Vertex AI operations throughout this notebook.\n",
    "-  `BUCKET`: Configure the Google Cloud Storage bucket name that you created\n",
    "    during environment setup. Vertex AI uses this bucket for operations such as\n",
    "    staging the code and saving generated artifacts.\n",
    "-  `TENSORBOARD_NAME`: Configure the managed TensorBoard instance name that\n",
    "    you created during the environment setup."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Project definitions\n",
    "PROJECT_ID = '<YOUR PROJECT ID>' # Change to your project id.\n",
    "REGION = '<YOUR REGION>'  # Change to your region.\n",
    "\n",
    "# Bucket definitions\n",
    "BUCKET = '<YOUR BUCKET NAME>' # Change to your bucket."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Tensorboard definitions\n",
    "TENSORBOARD_NAME = '<YOUR TENSORBOARD NAME>' # Change to your Tensorboard instance name"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get Vertex AI TensorBoard ID based on name\n",
    "TENSORBOARD_ID = ! gcloud ai tensorboards list --filter=\"displayName={TENSORBOARD_NAME}\" --format=\"value(name)\" --region={REGION} 2>/dev/null \n",
    "TENSORBOARD_ID = TENSORBOARD_ID[0]\n",
    "\n",
    "print(f\"TENSORBOARD_ID = {TENSORBOARD_ID}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Configure a custom container image\n",
    "\n",
    "The notebook uses\n",
    "[Vertex AI custom training jobs](https://cloud.google.com/vertex-ai/docs/training/create-custom-job)\n",
    "with a\n",
    "[custom training container](https://cloud.google.com/vertex-ai/docs/training/containers-overview)\n",
    "to run T5X training, evaluation, and inference tasks. The\n",
    "[base custom training container image](Dockerfile)\n",
    "that you created as part of the environment setup\n",
    "packages the T5X framework and its dependencies and sets the default entrypoint\n",
    "to the T5X launch script.\n",
    "\n",
    "Most samples in the repository use the base custom training container image\n",
    "`t5x-base` directly. When you are deriving an image from the base container\n",
    "image, configure your custom container image name (`IMAGE_NAME`) as shown in the\n",
    "following code:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Configure the custom container image name\n",
    "IMAGE_NAME = 't5x-base' # Change to your image name\n",
    "IMAGE_URI = f'gcr.io/{PROJECT_ID}/{IMAGE_NAME}'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Verify that the image exists in Container Registry:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "! gcloud container images describe $IMAGE_URI"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If the results show that the image exists, then the step is completed correctly.\n",
    "If the results show that image does not exist, review the preceding steps and\n",
    "try them again. Don't proceed in the notebook until this verification step shows\n",
    "success."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Configure experiment settings\n",
    "\n",
    "The notebook uses Vertex AI Experiments to track the run, including\n",
    "configuration, dataset location, artifacts, and metrics for reproducibility.\n",
    "Configure the experiment name, and set the workspace and run settings."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "EXPERIMENT_NAME = '<YOUR EXPERIMENT NAME>' # Change to your experiment name\n",
    "\n",
    "EXPERIMENT_WORKSPACE = f'gs://{BUCKET}/experiments/{EXPERIMENT_NAME}'\n",
    "EXPERIMENT_RUNS = f'{EXPERIMENT_WORKSPACE}/runs'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Initialize the Vertex AI SDK for Python\n",
    "\n",
    "Initialize the Vertex AI SDK for Python for your project, bucket, and\n",
    "experiment."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "vertex_ai.init(\n",
    "    project=PROJECT_ID,\n",
    "    location=REGION,\n",
    "    staging_bucket=EXPERIMENT_WORKSPACE,\n",
    "    experiment=EXPERIMENT_NAME\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Configure the dataset location\n",
    "\n",
    "Configure the location where training, evaluation, and test datasets are staged,\n",
    "as prepared during the environment setup."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "TFDS_DATA_DIR = f'gs://{BUCKET}/datasets'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": [],
    "toc-hr-collapsed": true
   },
   "source": [
    "## <a name=\"task1\"></a>Task 1: Fine-tune a pre-trained model on Vertex AI\n",
    "\n",
    "This section details how to run a T5X job on Vertex AI, starting with\n",
    "fine-tuning a pretrained model. You perform the following steps:\n",
    "\n",
    "-  Define parameters required for a T5X job by using Gin files and by\n",
    "    overwriting parameters.\n",
    "-  Configure experiment settings to track with\n",
    "    [Vertex AI Experiments](https://cloud.google.com/vertex-ai/docs/experiments/intro-vertex-ai-experiments).\n",
    "-  Create and submit\n",
    "    [custom job](https://cloud.google.com/vertex-ai/docs/training/create-custom-job)s\n",
    "    to run on Vertex AI using convenient utility functions.\n",
    "-  Integrate with\n",
    "    [Vertex AI TensorBoard](https://cloud.google.com/vertex-ai/docs/experiments/tensorboard-overview).\n",
    "-  Monitor the progress of run through Cloud Logging and Vertex AI TensorBoard.\n",
    "-  Track experiments metadata, artifacts and metrics from Vertex AI Experiments."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Configure the T5X job with the Gin framework\n",
    "\n",
    "T5X uses the [Gin framework](https://github.com/google/gin-config) to run\n",
    "training, evaluation, and inference jobs. To understand how T5X employs Gin, see\n",
    "the\n",
    "[T5X Gin primer](https://github.com/google-research/t5x/blob/main/docs/usage/gin.md).\n",
    "\n",
    "Typically, you create a Gin file to define the parameters for your task. This\n",
    "Gin file can refer to other Gin files such as a model Gin file or a base run Gin\n",
    "file. \n",
    "\n",
    "You can also create multiple Gin files for a task and configure a T5X job. When\n",
    "multiple Gin files are configured, they are imported in the order that you\n",
    "specify and they overwrite any earlier parameters. You can find sample Gin files\n",
    "for some predefined and custom tasks in the [`config` directory](config/)\n",
    "of this repo.\n",
    "\n",
    "In this notebook, you configure Gin files for the job through the `GIN_FILES`\n",
    "parameter and pass them as an argument to the custom job spec.\n",
    "\n",
    "Gin makes it easy to change (overwrite) specific parameters in the Gin files by\n",
    "passing them as command-line arguments. You can overwrite default parameters by\n",
    "adding them to the `GIN_OVERWRITES` variable as shown in the example that\n",
    "follows.\n",
    "\n",
    "Run the following code, which configures the base Gin file for the fine-tuning\n",
    "job using the `GIN_FILES` parameter and which specifies overwrites in\n",
    "`GIN_OVERWRITES`. (If you copy the notebook and use it for a different task or\n",
    "for a different dataset, you update this code with references to your\n",
    "job-specific Gin files and overwrites.)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The example below configures the base Gin file for the fine-tuning job\n",
    "via `GIN_FILES` parameter and specifies any overwrites in `GIN_OVERWRITES`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Specify Gin file path(s) for the task\n",
    "GIN_FILES = ['../configs/finetune_t511_base_wmt.gin']  # Change to your Gin file path"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The base Gin configuration for fine-tuning task runs for 3000 steps (configured as `TRAIN_STEPS`)  with evaluation period (`EVAL_PERIOD`) set to 500 steps between each evaluation. Please note that `TRAIN_STEPS` is set to a lower number of steps for demonstration purposes. For better results, you should increase the number of training steps.\n",
    "\n",
    "```\n",
    "TRAIN_STEPS = 1_003_000  # 1000000 pre-trained steps + 3000 fine-tuning steps.\n",
    "EVAL_PERIOD = 500\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "from __gin__ import dynamic_registration\n",
      "\n",
      "import __main__ as train_script\n",
      "from t5.data import mixtures\n",
      "from t5x import models\n",
      "from t5x import partitioning\n",
      "from t5x import utils\n",
      "\n",
      "include \"t5x/examples/t5/t5_1_1/base.gin\"\n",
      "include \"t5x/configs/runs/finetune.gin\"\n",
      "\n",
      "MIXTURE_OR_TASK_NAME = \"wmt_t2t_ende_v003\"\n",
      "TASK_FEATURE_LENGTHS = {\"inputs\": 256, \"targets\": 256}\n",
      "TRAIN_STEPS = 1_003_000  # 1000000 pre-trained steps + 3000 fine-tuning steps.\n",
      "DROPOUT_RATE = 0.0\n",
      "INITIAL_CHECKPOINT_PATH = \"gs://t5-data/pretrained_models/t5x/t5_1_1_base/checkpoint_1000000\"\n",
      "LOSS_NORMALIZING_FACTOR = 233472\n",
      "EVAL_PERIOD = 500\n",
      "\n",
      "BATCH_SIZE = 128\n"
     ]
    }
   ],
   "source": [
    "! cat ../configs/finetune_t511_base_wmt.gin"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Specify parameters to be overwritten in Gin files\n",
    "GIN_OVERWRITES = [\n",
    "        'USE_CACHED_TASKS=False'\n",
    "    ]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For example, when you run an inference task using the same base Gin file, you\n",
    "specify the other required parameters as `GIN_OVERWRITES` as shown in the\n",
    "snippet. You do the same thing for an evaluation task.\n",
    "\n",
    "``` python\n",
    "GIN_OVERWRITES = [\n",
    "        f'CHECKPOINT_PATH'=\"{CHECKPOINT_PATH}\"',\n",
    "        f'INFER_OUTPUT_DIR'=\"{INFER_OUTPUT_DIR}\"',\n",
    "        'USE_CACHED_TASKS=False'\n",
    "    ]\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Configure the Vertex AI custom job\n",
    "\n",
    "Configure the custom job run ID, name, and base output directory for the run.\n",
    "The `RUN_NAME` value can be arbitrary. We recommend setting it to a specific\n",
    "task or topology so that you can identify the run when you're exploring the\n",
    "experiment. For example, use `t5-finetune-v2-8`, where `t5` is the model,\n",
    "`finetune` is the task, and `v2-8` is the TPU topology."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "RUN_NAME = f'<YOUR RUN NAME>' # Change to your run name for the custom job\n",
    "RUN_ID = f'{EXPERIMENT_NAME}-{RUN_NAME}-{datetime.now().strftime(\"%Y%m%d%H%M\")}'\n",
    "RUN_DIR = f'{EXPERIMENT_RUNS}/{RUN_ID}'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The T5X launch script allows to specify the function to run: `train`, `eval`,\n",
    "and `infer`. Depending on the job, you configure the `RUN_MODE` setting. For the\n",
    "fine-tuning example, you set `RUN_MODE` to `train`. The Gin file and the\n",
    "overwrites that you configured earlier parameterize the invoked function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "RUN_MODE = 'train'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To help with troubleshooting, display the values of local variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for key in [\n",
    "    \"PROJECT_ID\", \"REGION\", \"BUCKET\", \"TENSORBOARD_NAME\", \"TENSORBOARD_ID\", \n",
    "    \"IMAGE_NAME\", \"IMAGE_URI\", \n",
    "    \"EXPERIMENT_NAME\", \"EXPERIMENT_WORKSPACE\", \"EXPERIMENT_RUNS\", \n",
    "    \"TFDS_DATA_DIR\", \"GIN_FILES\", \"GIN_OVERWRITES\", \n",
    "    \"RUN_NAME\", \"RUN_ID\", \"RUN_DIR\", \"RUN_MODE\"\n",
    "]:\n",
    "    print(f\"{key}={eval(key)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create the Vertex AI custom job\n",
    "\n",
    "The utility function\n",
    "[`utils.create_t5x_custom_job`](https://github.com/GoogleCloudPlatform/t5x-on-vertex-ai/blob/master/notebooks/utils.py#L37)\n",
    "creates a Vertex AI custom job spec to run the T5X job (training, evaluation,\n",
    "and inference) and to return the job spec. The function stages the required Gin\n",
    "configuration files in the Cloud Storage bucket before creating the custom job\n",
    "spec.\n",
    "\n",
    "The function requires you to specify the resource spec to run the script, a\n",
    "custom container image, and other information such as Gin configuration file\n",
    "paths. For parameter descriptions and usage, see the\n",
    "[function's doc string](https://github.com/GoogleCloudPlatform/t5x-on-vertex-ai/blob/master/notebooks/utils.py#L37).\n",
    "\n",
    "For TPUs, define the worker pool specification (`ACCELERATOR_TYPE` and\n",
    "`ACCELERATOR_COUNT`) based on TPU availability in the region. See\n",
    "[Vertex AI locations](https://cloud.google.com/vertex-ai/docs/general/locations#accelerators)\n",
    "for details."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define worker pool resource specification\n",
    "# Machine to run the custom job on. For TPUs, use `cloud-tpu`.\n",
    "MACHINE_TYPE = 'cloud-tpu'\n",
    "# Accelerator type to attach to the machine. For TPUs, use `TPU_V2`, `TPU_V3`.\n",
    "ACCELERATOR_TYPE = 'TPU_V2'\n",
    "# Number of accelerators to attach to the machine.\n",
    "# For TPUs, specify the number of cores to be provisioned.\n",
    "ACCELERATOR_COUNT = 8"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create the custom job spec."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "job = utils.create_t5x_custom_job(\n",
    "    display_name=RUN_ID,\n",
    "    machine_type=MACHINE_TYPE,\n",
    "    accelerator_type=ACCELERATOR_TYPE,\n",
    "    accelerator_count=ACCELERATOR_COUNT,\n",
    "    image_uri=IMAGE_URI,\n",
    "    run_mode=RUN_MODE,\n",
    "    gin_files=GIN_FILES,\n",
    "    model_dir=RUN_DIR,\n",
    "    tfds_data_dir=TFDS_DATA_DIR,\n",
    "    gin_overwrites=GIN_OVERWRITES\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The function returns the custom job spec that you can submit to Vertex AI to run\n",
    "the script, as shown in the next section."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "job.job_spec"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Submit the custom job to Vertex AI and track the experiment\n",
    "\n",
    "The utility function\n",
    "[`utils.submit_and_track_t5x_vertex_job`](https://github.com/GoogleCloudPlatform/t5x-on-vertex-ai/blob/master/notebooks/utils.py#L253)\n",
    "provides a way to submit the custom job to Vertex AI. The function tracks the\n",
    "experiment and logs the metadata and artifacts generated by the experiment, such\n",
    "as Gin configuration files, dataset location, and model architecture (including\n",
    "the lineage).\n",
    "\n",
    "The job may take > 30 minutes to completion and be patient for it to finish."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "utils.submit_and_track_t5x_vertex_job(\n",
    "    custom_job=job,\n",
    "    job_display_name=RUN_ID,\n",
    "    run_name=RUN_ID,\n",
    "    experiment_name=EXPERIMENT_NAME,\n",
    "    execution_name=RUN_ID,\n",
    "    tfds_data_dir=TFDS_DATA_DIR,\n",
    "    model_dir=RUN_DIR,\n",
    "    vertex_ai=vertex_ai,\n",
    "    run_mode=RUN_MODE\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Monitor the job with Vertex AI TensorBoard\n",
    "\n",
    "Vertex AI Training does not support built-in integration with Vertex AI\n",
    "Tensorboard for TPU-based training jobs. However, you can use the\n",
    "`tb-gcp-uploader` command-line utility to manually\n",
    "[upload Vertex AI TensorBoard logs](https://cloud.google.com/vertex-ai/docs/experiments/tensorboard-overview#uploading_logs)\n",
    "to Vertex AI TensorBoard. This step allows you to monitor the training in near\n",
    "real time as Vertex AI TensorBoard streams in Vertex AI TensorBoard logs while\n",
    "they are written to the Cloud Storage bucket.\n",
    "\n",
    "Run the following command from the terminal window to sync logs to Vertex AI\n",
    "TensorBoard."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "cmd = f\"\"\"\n",
    "tb-gcp-uploader --tensorboard_resource_name {TENSORBOARD_ID} \\\n",
    "--logdir {EXPERIMENT_RUNS} \\\n",
    "--experiment_name {EXPERIMENT_NAME}\n",
    "\"\"\"\n",
    "\n",
    "print(cmd)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To access the TensorBoard instance for the experiment, run the following command\n",
    "to generate and display a URL. Then click or copy the URL to your browser."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "TENSORBOARD_URL = f\"https://{REGION}.tensorboard.googleusercontent.com/experiment/{TENSORBOARD_ID.replace('/', '+')}+experiments+{EXPERIMENT_NAME}/\"\n",
    "print(f\"TensorBoard URL for the experiment is located at {TENSORBOARD_URL}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Alternatively, you can access the Vertex AI TensorBoard experiment from the [console](https://console.cloud.google.com/vertex-ai/experiments/)."
   ]
  },
  {
   "attachments": {
    "0a2e3dc9-cbdb-4f21-be20-d60b69c5cff3.png": {
     "image/png": "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"
    },
    "be2862b0-0e40-47f5-95f9-e07e451a4d8d.png": {
     "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgAAAASCCAIAAABLssNEAAAAAXNSR0IArs4c6QAAAERlWElmTU0AKgAAAAgAAYdpAAQAAAABAAAAGgAAAAAAA6ABAAMAAAABAAEAAKACAAQAAAABAAACAKADAAQAAAABAAAEggAAAAAWh7VqAABAAElEQVR4AexdB3xUxda/2zebHtJ7CKH3bqPaUEDEBvgp9t7L8wnP9vTZUN+z4UNRHyoKItJsFKX3HkpCEtJ7L9vr999cuCyb7GZzN9ndJOfqL8zOnTPnzH/mnjN3Zu45goqKisjISIYuQoAQIAQIgR6GgNBkMvWwJlNzCQFCgBAgBKwIkAGgcUAIEAKEQA9FgAxAD+14ajYhQAgQAkKz2UwoEAKEACFACPRABIQWi6UHNpuaTAgQAoQAISAkCAgBQoAQIAR6JgJkAHpmv1OrCQFCgBBgaAmIBgEhQAgQAj0UAXoD6KEdT80mBAgBQoAMAI0BQoAQIAR6KAJkAHpox1OzCQFCgBAgA0BjgBAgBAiBHooAGYAe2vHUbEKAECAEyADQGCAECAFCoIciQAagh3Y8NZsQIAQIATIANAYIAUKAEOihCJAB6KEdT80mBAgBQoAMAI0BQoAQIAR6KAJkAHpox1OzCQFCgBDgaQAaGxtfe+01QfMVHh7+119/cVBWV1cvXLiQvfXmm2+qVCr2VmFh4cMPP8zmL1682Gg0Iv+LL75YsGABF5by0KFD48ePb2howK3+/fv/8ccfs2bNAsmuXbuQs3v37lGjRrE1PPvss7W1tWzN+AsB2PzQ0FAUQ86yZcvmzp1rG+3g7bff/uCDDzgSShAChAAh0NMRyMnJQUiAdl16vR6oPf/885mZmWfPnv3vf/+Ln6dPn0YlrO6+7777Tp48eeTIkaSkpFtvvRX5JSUlKPPyyy+jGKvNH3vsMeS/8847d999t8FgYAXYvn07itXU1OAnErg+/PDDpUuX1tfX//rrr/i5cePGvLw82Ik+ffrceeedLNXKlStxa+3atVlZWd999x3SJ06cOHPmDBJc61AnfkIqloT+EgKEACFACFxQka5jUVZWhqk0lDJHMnTo0B9++AE/WV3M3Tp16hQUPfT7+++/P2PGDI1Gw5Ls3bv3iSeeQPq9996DtcDbAJu/Y8cOqGnOADz66KNsPv7+9ttvv/zyC/eTtSL4iXeRtLS0b775hruFV5MNGzbgZ3Jy8ueff87mr1q1auzYsZyl4QpTghAgBAiBHouAGAq3vVd0dPTTTz9dVFSEeT0m2tDa6enpbCX5+fnz588PDg5mfw4cOPDjjz9G+vjx4xMmTJDL5Ww+1nlwIQ3c2ZxW/06ePJnLnzZtGuwH5vhY+Tl8+DAUOnsLK07Z2dmonCuJ9ww2/Z///AcrSHhREIvF69ate+SRR5DgilGCECAECIEejgCfPQAsAWFxPzEx8e9//zs0MpRv3759WRwxxfb392+JKUikUmnLfLscO3vAGQwUgwkZN25cv3798JIhkUhmz57N0rKr/H5+fnZV4SdrY2CiysvLsTRka05aFqYcQoAQIAR6GgJ8ZsTYccU2LzYAUlJSoNah9J977jlWEQcGBpaWlnIgYgcY27x4J8BLg+2eLRZ5vvzyyyeffBJTcq1Wy5XHeg6Xtk2gHiwZYTth3759CoUCt7htZ1b1g2lkZCRLsmfPnuLiYhSOiop64IEHsNwEEiw0wWLZ1klpQoAQIAR6OAJ83gAwoQZqcXFx7KR+zZo1ubm5rAGYOHEiNmOPHTvGwopNXSwWYSKPDQAszWP/ls3Hgv4LL7wgk8mwS4y5eVVVFfKxc7BixYqIiAi2jO1ftVqNhSZwZLU/tpqxBc0WiI2NffDBB1EJe5QIy0Q333wz1oXYu9D72B5444035syZY1shpQkBQoAQIAT4vAGwaym33XYbVtih66H6Bw0ahEk30Bw2bNinn346YsQI7O5iaxdrRND1mKRjmeiVV17p3bs31uUx/X/99dcxT0f5Sy+9FMc9Bw8e/Mwzz2Dn9qabblq+fDm3EIR3C7aHYBVQ4T333FNRUYHafvzxx2uvvRanjKDocQgVXK6++mocM50yZQrMT2pqKkqyhAMGDICVwgvEyJEj2Rz6SwgQAoQAIcAiIMBBSWjM9sKBg0A4i4k5O87/QOkXFBRgko4FeqHQ+kqBjQEcuIRhGDNmDOb4XOU4A4oDmpjF4zg/Zu5sPpaGsL5fV1c3fPjw+Pj4jIwMVmsfPXoUZbCMwxbDBB852G3G2R7YDNzCT5Rn3xiwdoQTR6DFtjPMD94tOKZYO8ICFL424HIoQQgQAoQAIQAEeBqAroIdXhFgIWB4YFS6iswkJyFACBACnkGAzx6AZyRzkwsWizD3h/bHX7yauFkbkRMChAAh0P0Q4LMH0CVQCAsLw9IWzoxio4JdmOoSYpOQhAAhQAh4DIFuvgTkMRyJESFACBACXQ6BbrsE1OV6ggQmBAgBQsDDCJAB8DDgxI4QIAQIAV9BgAyAr/QEyUEIEAKEgIcRIAPgYcCJHSFACBACvoIAGQBf6QmSgxAgBAgBDyNABsDDgBM7QoAQIAR8BQEyAL7SEyQHIUAIEAIeRkCIULoeZknsCAFCgBAgBHwBAXoD8IVeIBkIAUKAEPACAmQAvAA6sSQECAFCwBcQoCUgX+gFkoEQIAQIAS8gQAbAC6ATS0KAECAEfAEBoUgk8gU5SAZCgBAgBAgBDyMgQGwvvV6PKIxcIEYPS0DsCAFCgBAgBLyCgBhvAAiZm5iYSOdBvdIBxJQQIAQIAW8hYDUA4I04vWQAvNUHxJcQIAQIAa8gcG4PgNZ/vII+MSUECAFCwIsIIFoifQrgRfyJNSFACBACXkOAjoF6DXpiTAgQAoSAdxEgA+Bd/Ik7IUAIEAJeQ4AMgNegJ8aEACFACHgXAdoA8C7+xJ0QIAQIAa8hQAbAa9ATY0KAECAEvIsAGQDv4k/cCQFCgBDwGgJkALwGPTEmBAgBQsC7CJAB8C7+xJ0QIAQIAa8hQAbAa9ATY0KAECAEvIsAGQDv4k/cCQFCgBDwGgJkALwGPTEmBAgBQsC7CJAB8C7+xJ0QIAQIAa8hQAbAa9ATY0KAECAEvIsAGQDv4k/cCQFCgBDwGgJkALwGPTEmBAgBQsC7CJAB8C7+xJ0QIAQIAa8hQAbAa9ATY0KAECAEvIsAGQDv4k/cCQFCgBDwGgJkALwGPTEmBAgBQsC7CJAB8C7+xJ0QIAQIAa8hQAbAa9ATY0KAECAEvIsAGQDv4k/cCQFCgBDwGgJkALwGPTEmBAgBQsC7CHSYATCZTN99993QoUMFAsGzzz7b2NjINWzDhg19+vRh8ysrK9l8i8WyaNEiZOL67LPP9Ho98tetW/fggw9yhAsXLly1ahV+njlz5o477vjhhx9QePfu3cjZuXPnVVddhZ+zZs3CXY4kKyvryiuvRP7kyZNPnz6N/Pfee++tt97iCrz//vv/+te/uJ+UIAQIAUKgxyLQYQYA2nzx4sXffPNNZmZmeXk59DWL6R9//HH33XfDNhQXF4eHh8+ePVutVuPW22+/DSWek5MD9Q29D72MzKamJlZrs7Rnz55lCxuNRtSwadMmlExLSztw4MCECRP+/ve/o84rrriif//+9fX1ICksLOzXr98DDzxQVVX12GOPDRo0qKioaOrUqQsWLEAOChgMhueee27ixIls/fSXECAECIEejYBKpYLOxfwdU3J3rr1795aVlbE1QEEDU0zq8RMJ2AA2H/r9xhtvhNKvqalBPlceOTfffLNSqfz++++nTJnCiTFv3rxly5bh58mTJ1G+pKSEvYVp/vHjx9l0XV0dbqEJ+Pnxxx8//vjjbD7+wkKsXbsWJgQ2A8YDOQcPHkRh5HBlKEEIEAKEQI9FQAyF2CHX+PHj9+zZ89///hd6n51uI8FqZ6wLsSwCAgKg4uVy+a5du5ATHR3N5vfu3RuvDn5+fmazuVVh0D3IByF7Fwo9Nzf3zTffhEXR6XTI1Gq1oMUrBd42uBr++c9/glAqleKdYP369VgygmV66qmnwIgrQwlCgBAgBHosAh22BPTRRx89//zz119//UMPPYR1HgCKhXjoZSSggjl8WSXO5nOZKMkqZVbR2+ZzaSS4u9DjqampI0eOBK8777yTLYOXGLxScEYCmRKJhGU9c+bMTz75pKCg4N1337399ttt66Q0IUAIEAI9FoGOMQBY0nnyyScx/R8zZkxiYmJsbCwAhb6OiYlBIjs7m8UXS/lr1qyprq5OSEhADru+jwRW8FevXo0FeplMZmsbUJ4ltPuLzYYlS5Zce+21SUlJycnJ7F2o+0mTJmEliiu8devWY8eO4SfYzZgxA8tBWEQaPHgwV4AShAAhQAj0ZAQ6xgBg3j169Ggs7+DwDzZ1X3rpJWCKJSAoZUy9X3nlFXbRH5sBeDkQCoXYqsUu7uuvv45pO4qhzObNm1EY+VhHOnHiBIwH1PfKlSvxctCyezD9h73BfB9nij788EMUYBeCsMGA/d5Tp04hB/sE7HYC0ni9wB4DFn9efvll21eEljVTDiFACBACPQiBjtoEPnz4MIfaTz/9hHR6ejr0OPQ7DmKyt6C4sXaPTFyY9U+fPp3Nx0oOdoDZfJwmYjNvu+02TNu/+OIL5GPLF5nYWmDLYGth7NixbLEvv/wSewxLly5lb2Gzl83H3+3bt7OZ+IuDScjBZjKXQwlCgBAgBHo4AgIYACyOY+qNiTmnOnkn8AYQFBTUKrlGo2m5+4plH5FIZMcaKz+42pyqYwc4MDCwVV4NDQ3BwcG2t3DQCO8c+fn5thsStgUoTQgQAoRAT0Ogw04BscA50v6421L7IxPLPi0RFzdfLfPtchxpfxSz1f6wMfgaAKtSb7zxBml/OwzpJyFACPRkBDpg1u/j8GElCktPOHI6d+5cHxeVxCMECAFCwJMIdPAbgCdFd5HX8OHDscyFBSW7hSYXyakYIUAIEALdFYHubwCwx6BQKLpr/1G7CAFCgBDgjUD3XwLiDQ0REgKEACHQvREgA9C9+5daRwgQAoSAQwTIADiEhm4QAoQAIdC9ESAD0L37l1pHCBAChIBDBMgAOISGbhAChAAh0L0RIAPQvfuXWkcIEAKEgEMEyAA4hIZuEAKEACHQvREgA9C9+5daRwgQAoSAQwTIADiEhm4QAoQAIdC9EehuXwIj2gw8krJRBOD7Ad6nEQ0Gfujg9LW9HQlyeA+NjIzkTY6Y9fBYBwd5PLijCYh2ALd4YWFh/Mhra2vhCC8qKoofOWCEv9X4+HhHcTqd4Anh4fwV8iNcDw9y1Ax3sHBSi9if/MhBhb5DqFEnQjq5BcRAjtBG+Iy89WLWEWaNVGE26ExajUmjZvRai6U5oGnzSKuobYhKSJCHhDFCBzW0Xq81F3XDfSH63d/fn1/fIVQGaNs78IRiccWuvxq2bWZEYgaPi8AcMvW6iNGXWBwEam21BRAez+CQIUPI80qr+PhaZnczAOHNF4dyZmZmXFxcq45IuTJOEnArjehmvD1JQAn26tUrJCTECQsnt6DF4L4UGtxJGSe3YDwQjAHNd1LGyS1oEHBng7s5KeboFsK9wfzwJocBQMQI3uSADkEjYL0ciddmPkvu0ACYjEZVk0HZZDLrzAIT4ye1+EmEIolQLpcEBOF/XeaZhL79ZLILwVDb5GhbAOEx0HbE0LbNdD2NfodDXJgQ10lQsnz/ruL7b2eSGUHzuoDFxDSt+Spu/YGI4WPaVU+rLn7bVQMV9hgC3c0A2AGH+QiPORRXCWj5zUDZGtwnd1N4d8jR8K5LDsnZyzpTb/91nvqit0YMBZNOY1QpjcpGk15naY5XivqFUpnITyEJChXJ5MLz7s0tAgGKt5/zOQpWAE+Sm3TaE288KugTKJCdm68AOLOgKGPJ+2H/+Vok83NdGHceGde5UMkOQaCbG4AOwYgq6YkIWNd4rMYDit6oURmg9NUq6H2EukamQCy2Kv3AYIl/oEjux5bs0ijp8dLxS7r0Mmuwbu4SKmIbN6zUvfahIrIdBoAjp4TvI0AGwPf7iCT0NALWCbhBr6ut1mmURo3GYjJaJcBkXySGa1kpJvt+fkKJzNNidRo/dXlJ0V+/i9JaMhAwRmwHXPQm1LIQ5XRdBMgAdN2+I8k7GgGLxTrZb2rA4r6hJE8rFUDhQ/GLpDJJUIg4IFCs8MfPjubqtfp01ZUVRw8Wbfih5vPlZgsjTLWXxKKvU0y9WsZ3K8K+OvrtewiQAfC9PiGJPIuAGYs8ykaDsgHr+xaTCRreAnXIMGJFgF94JJS+UMJzL9ez7XCVm0nVVHHkQMHqZVUffstuU8Cmxb38lCK575l7HpFeEiMQWtWCxajXH1KP+mghcHC1airX1RAgA9DVeozk7SAEcHbT0NRoaKrH1i43r8eOLlb2Rf4BMpPIPyVNLOw+H8qY1MqajJN5K5eWL/qy+byq9V0m+rn7Em+YFzN6nFBuDZoUmJh8/MrrzP6MRcWIY5jLDu2NGjW+g/CmanwRATIAvtgrJFMnIWAxGoxqlb4Jk/0m62Sf3dHF0r4iQBxoPb7JTvaRjV0ABuffu6wB4I4/Qe/XZp8p+vXH8p/e1R89t5wf+di8hFl3xIwdLwm86Ixy4tRpiRZLxs5tfgq/5FHjOqkXqFrfQYAMgO/0BUnSwQicU4IWC844GlSNxqZGJCw4ndm8qYlTm5jp4xgP9H6LYzxde9sTZsykUdVllhZtWl/22wrtvhPmBut8P+yBm+NmzI2/dIIsLNwOa43epDeamjQGlc5c4R83MOEiw2BXmH52GwTIAHSbrqSG2CCA7z9MRj00vgqL+02Y+LOfswpEIjE0vn8gdnRFUnkLvW9TQ9dMWvS6xpLC8p9/aNy9SXdgt7HEqvdD7roh+pob4ydMCYiKsX7l23wZTRZofOj9BrVebzAbzUDIavZgNU1GE66uCQBJ3T4EyAC0Dy8q7fsImPU6TVWFobRA7dc8vAU4xiMV+QWwx3iEYonvN6G9EuLQqqqyomzXn6VbNzb+vsKiZizVTMBtU6OvmR0/8cqg+CSB1HpoVW80G7TGRrVerTMaTBaDCd/6WVlB6UtEKCLyl0n85ZJApglmg66egAAZgJ7Qyz2ljVjh0dVWGXGOU6/DJ+DWb7WaHTMIZTKc4e+GKJiMKhzlPLAber9hw5f6HOt8P2DWOOmwy5Nn3BI/cLDQz99gZhq02O1W6QxmncEEnc/iIBIKpPiaDZbRqvfFSItF59R+/flEN0SMmnQxAmQALsaDfnVNBOCRTVdTiSP8WMiAUzNpcJg42hiQ3JeXJwifh8Bi0dbXVh7aU/LXptrVn+izrXpfce3gmHtnxl93U1BiSlGDSiuQlKiF6tpGrPOcn+YzUPrQ+AqZWCGz/pWI4LGN5vo+392dKSAZgM5El+rufASMGrWuugKuGnBwBx9uSUPD5L0iLGKJoKax85l7moNBpaw6ur9k+5bKpW8b8hmhnJGMjoiedSO2dv1S+gpCo1Rafa3alFdcD0dyIRYZtDuUvkxineP7y8VISMXd52Crp9Hvjvy6vwHgzsPx6D7QuuPV1k1ysHaHO2jdabvvk5s1am11hVHViDVskUSCWb/1cEvzJifWts/78uHR7dY1cfeb34F9Zzboq48fKt6+pWLpK/pMRpjACCOYkGtuC7n+jqA+A4RR8VjfacJMv0YJjS8UCTG1D/aXxvXyx0wfGr+9s3z3xy0f0InGGwh0NwNQU1MDL/YskvDlC7+4cMrPz686HuCKigo4VXaHXKfTwTGy9VB5Oy9wh1d3eNbFZ6o8yPEMV1VVofmQn4d3RpDDGzNcIsMVNj9ylUoF9OAKmwc5oIIraYQTAAL25JAMjuq1Kn1NZbPqh5ETCgNDZL2ChcCpopI92o9DLCAvKSlpJ+rnigNwCA9/+kCPXw0glzdfvPpOWFFVjc3qAGWTviivds+2siULdRmMUMHowxnTNaOFk+cEDR4hSEqrMxhq1GbL2TzofTFm+mIBND48UvtbGiR6i75JoG08t83reivYgYeG8x545eXlcMruOkcq6UUEeI5vL0rsnDU7ebFqCbz8ikTsT4xpHs8hqLiLH7kXuYM1x905Yq3e5WiBQKsFnGe6Tw4NDta27vih59EL+voaY101Dr1AAJzjlIVHiQJDrLese5uY9Z934WmxsH3nXE5Hd8GI63pHZZznc+TtGDksajA5BoM+++TZ1V8b1y7V5TH6KEYnYEQzJ4becF9Y737i8GizyYSv2Mw6bfPCjkwhxTRfYN3CbXbbhmqq0Wtw6m+FAwbSuaT2dznJkWiH8OerYRtx/hf96+sIdDcDgCAYtnEwTp48icAavAPCIKhWTEwM74Aw8LDrTkAYzN9lMhlCkvEbRHiA8f7BO54M5q94/+BNjvceyM+bHNNPgG/bdmNjPRZ8EH3FEhoCT5xY6JeE9HKEDN4b8C6IvnNUoM18kGPk2FqgNklsC+DlCdzRfbaZbaaVeVklu7fnfrtIWZ6tsjBqESObOSXimjnJw8cGJKRAoWMB30+C/VtRgFyM7VxHFQJ8vHuFhoY6KuA8H+DjDYB332HgOa+f7voOAt3NALRElscshqsEtPZLENw9FxLeJYfk7rTdV8gtFl19ra6mwmwwwEkbIpPIe0VKg0OhDZ30ABrOXtZJcPsv0LrffNdHTkNhfvG+nWfXrag+9ptGx5gljCR+XPg1c9JGjg/q3dfPTybHLq5UCL2PXVxXWsPK70rJVsuw0LV6y5VM1xvuSm1UplMR6P4GoFPho8o7CYHmVR2TsaFWXVWOLVCs7Iv8/GVQ/YHBzlV/J8nT4dViy7a+vDR/7678bZtrDmzR1RRYBIw85ZKoK2+KGTlG7ReQkpIUFR4qx9qOW3v5HS44VditECAD0K26s3s0BkG4tM2f8moCZFjvhkNmrPXDf0M3UP1wsFBTWpF7+FD+ti21uzcaKjMsRkbee1TiDfcljL8iZtDgkLBQP6nwTOaZ6GBpgIzWUrrHiPbdVpAB8N2+6YGSYbKvr6vRN9bp1SqB2ShSBMhDwyUBgV0dCuxQNzY0FWVknt2xrWzzOl3uboGR8U9IS75rQfKlE+OHj/IPC8OBfZtmYhnG5hclCYHOQYAMQOfgSrW2EwGzXq+rq9I31GH6LxAJRfDXFpWgwM5n17ys52+a90K1Wn15cVnOjq2FW35THlrFmBl5QFCfGx7sO+XK2HET/CN47vB3TVRIap9DgAyAz3VJTxMIs37EJsSsH0cbsfSPeCzy8CgzPuVt0nZdKEwWfERSW3r8WM6mX+v/WmY2asWMMWroVWnXzug95dqQlD7dYDmr6/YOSc4hQAaAg4ISnkYAbju1cOCDWb/ZLBAKEHcXql8k94McJuuZny65CAKvawXp6cfWr1P+udLYmI+FrICwpN5X3dD/pjsj+vYVKrr8cpanRwnx60wEyAB0JrpUtwME4LYTh/oNjfX4egv/4VinPDwaPjsvFO9q2l9nZkpy8rJ27ihc/T9NxjZxWLKwPj95xOWD7ns+YfQ4aXjUhaZRihDwGQTIAPhMV/QMQRCJV1NVbvXi0Dy/l4b2kkdEd10f/TjVU1VZm7V3f+7q7xq3f29R+AnrNFjYj5x564hbbu81cGjP6FVqZVdFwJsGQKvVHjt2TCqVDhkyBB+dtoSwurr61KlT+CR1zJgx3OeF2dnZcO8THBw8cuTIliSU47MIIAwvVD9CFUL1Y8FHGhaOWT9CdPmswE4Eg/Gqb9Lmnzx1Zs2PVRv+bWFEAr02UMv0ve/xvtNvihg26nR2jn9KV93BdtJwutXNEPDaQWO4KvvPf/6TnJwMVf7555/ju387ZI8fP75ly5axY8fik/R33nlHo9GgwIoVK/CR/WWXXQar8P7779uR0E/fRABu+ptyzygLc0043CkSY8oflDbYLyquK2p/tcFy5uSZ3/67dNVt1+164KrK9R/Jigz9JsyY8dn628/UXPbSOxEjxsIhJ75btgadp4sQ8G0EvPYG8NFHH73++ussOPPnz9+wYcPcuXNtsULOP/7xD+TASMybNy8jIyMlJQXWAiYBmQMHDmxoaDhz5ky/fv1sqSjtIwjgPA92cU3KRnVlGVb8kRbCY3NYhCy0V1cMzqUzMRUlZXmHDxdt+bXh8DZTXZakzhw/emja3MeSJk4NjE/GqU8fQZ7EIARcR8A7BgCLP6NGjeKkRPAK+J9qbGyEA142Mycn58orr4Rj59LSUnjUgg1ISEiAf64RI0ZwVPD3m5iYyP2khA8hYHXbWasryVeL8SWvSSiVwVO/NASqv8toSTgRgrTWD3er6ouzz+Zt+b1q71/6/N0itT402r/3/a8mXzEpcvgYoUzuQ7B3kCj4JO3ir9I6qF6qxvcQ8I4BgLd0O4/h8JoJD/KcAYAnd/ijZzV+UVHRzp078RIQERHBArhv376DBw/ihSAuLq5NSLnNgzZLtiwAWiJvCYuzHIsFxzrhtBkGQGgyiv0UfojOCN9t7bzghtN95Pl5goOkZoFApTNlnMop3ruzeNsm7amtIos5UCGNm3l/7ylXxYy5DBsYzhvUPHD4Wzsvklc3aA5n18qk4pGy4F7BfMybOx3nHFW62+EIeMcAINyH3SjBA2/78TsKYJPgvvvuQ4PhlRevCLt27Zo4cSLbfkz84Wv3l19+gbtjO4+7sBzYOmYrx1+8amRmZsKfs23lLoIIkfLy8lADP3JwLygogA3DshUP7iDH2w/2xrHSxcO9IsgBIGDE3gkPcqhO8MWF+Xub5FY9i/9RUtVkbKxFeF7kaI2mao3BT2NgdDVMRVW7EAA5JM/Pz0coHgjgYn9xxUAOKpBjbCDdJmvIzqAB+Eco1Bss2LCoyc4q3fGX4fhfpvocqYYJ7zskdPzU0PETEW7XGBRcUFtvqa51VC3LEdwhAwRwVIyTtmUCNWDgYT7EOxIR5kzWkJAhIa5zt4rNCLafbHjwj2JGb7SezQ3IXjot/pL+gc1eZV39JgMDjwLCtOxTn83xjgGAn/ezZ8/agoJNYFtVjvSgQYO4AsnJyenp6VDueFHA9i9MAm5df/3127Ztu+aaa7hiSMAHOvS19WFuDghTWFgYHR3N+0HChjO2oPEsuf4gccLgScAzDHkQn4AfOZQgjkhBAB7kQAAXPPLDUrapwTmZuQRo0QUwP22TCwTY8NTXVhnU9YzZxAQGiMIjsM2rQXSWkpLYuAQzMtt/oe0AH33HQ3hwg/LFiAI5GuKcOQpA8RlMUOmNxadOFu3Yqj681VyZIRXLQgMDk+c8nTLzVkVSb6EiADFn4CXaxc/T8P6KjsNYdc691bsQCYufeEbwQsyv62H1QYuHxXVyvHCt2pr14Oq8fiFixJeBYEaz6c6VZ5ff3veGK1Jh3FsVtWVmm4C3JKEcLyLgHQOA0bl//35b3Y0HHuMVQGDIYgwNGDBgz549HC7QpNDC2AdOTU3Fyg+bj6erpXaA9sfFEWIWjwk474AwYApRYT+4CtuVCAwMBDn+touKKwxCGADe5GxIFjSBq7BdCezKoFPaIDeZrA586mpkJqNMoRDJ5FD94gDrRo5ErQ6oR2hyntCxrNvg7rg9GBigRdc7LnLuToPaUFZQnLd3V/mW9epTW/Hu4M/UR4+e6T/q8mE33hoYn9RmDa0WAHd0PT8DgApZct5dz4NcrTPev6mkbzACi50zmXBDnRIg/mZ3+cxJg4L826EoMPBaxYQyfRAB/suUbjZm+vTpf/75J1vJDz/8wJ7twT7w4sWLkYkdArwiYAkFaazzrFmz5vLLL0cZJNjzoFBPX3755fjx49sUw/VJUMuqQNuTyVsCciHHYtFUlDTknLYe7TcasM0Lx20BKX1Z7Y9i8O7gdeicCIDTnAWltdvWb1z39KObb+id8/6jxj0/RfvpLn/0Hzeuy7zmf2uDp17vF5Nwob3tTFnHjRsfM7tDy0ra3hoalTpjpRpK37ahcqFgYwlmX+1zytRe1rYcKe1hBNph2DtWMnzGdfr06VWrViH43KWXXpqUZJ1qYdkBZ/xZRk8//TT2fnFhHjRnzhx2Nnf//ffDbLCLjI8++ihWOTtWKqrNFQSMaqW6tLA5TgsjVijkkbFiBc/3DFfYdWAZvYWpa9LnHztWuH1LxcrXTGaxFNETC5jUp2f3mfdQxJDhguZTPVBhbAAyhvHaDKkDW91mVXkVTb/sOisNstcG1oV/hBo+/07QZj1UoMshYN/lnmwAzvLjsuUIAzB8+HAu54rmi/uJBIwBXh1scyjtYQQw8dfVVmOKj5jsithEkd+FBTcPS9ImO05xYReiUWMuzMop2L2jbPXn2oKDEgHjn8PEz+yf9tBr8ZdNFAe1+5xSm9x9v0BFvXb/yZJ/bz67LavOquglQrMUUeQvCF5hMD82uFd4iO928QVZKcULAW8aAF4CE5HXEMD3XOqSAutXXXhXCw3Hp7y+PDcUCgVmRqAyMOWFJYXp6cW/rmw6tk2gKpJVmFMGMH0e+Ax6P6jPAK+h6T3GTVpjenbF8u15nx2tYPCFm8nChMo/nZqYmhB67WeHQySCKAmOTjH5erPeInhs9lD6JsB7fdXpnMkAdDrE3YOBrq5GW1GCpRE4blPEJIp9O0qXziQor9Wodx4q3vJr5Y7fLFUnpQ3qKBmTdM+CpCunRQwb3S0/4HI+0vRGS25p3bo9+T8eLD1SpWGMFkz5X5oUf9XYpDH9o+VSq1OmnPjQJetOLD1dqxAJXhwdcde0gckx577NdF453e2iCJAB6KId5zmxEaJLXVpgUClxQgsu+xUxCb7sw0dtZMpKKs7u3JH78zfmgmPChuLgCib2jnnJ190Ye+kkWRhOmtmscXgORa9xwjp+UWXTjuMl6/cVrcpvYjRGiHL38PDrxyVMHpEQGiS3hSM1Nvjdhy+//2yBWCJOSWz7K0uvtYoYdxACZAA6CMjuVw1W0PFFlapJVVpkNurhwMcvOh6O+32zoVBzTTpLWVFp3q7txb/+qD/9R4BZF9F3bMLDzydOvS4wPkEgsQk24Jtt6DipsGiDFbAGlX7fqdKNh4q/Olnd0KDHt11XpgTePCb26nHJcRGBUrHD/W2FTCSWdEkvrR0HYU+piQxAT+npdrfTbNJXlqpNGrPVo4M/9ntx1rPdlXQ+AVR/vcZUVliat+Ov0t9WGjN+D9AwyVfd5H/5VaNm3yYN6nHnxDR606ki5Z5NRb9mNR0rszrf7hchf/rKxOvGJ/VPDg/0a8Xvul0vAVI3jrDaVUY/fRoBMgA+3T3eEs4atqU434SIXb1CZb2i/CJjvCWJE774OLVOaSgrLs/964/yDd+a8nYG6Zi4a27rd8dD4SNGn84t6FHa32i2FFU07kwvWbGn6PcCJYNvd/3E940Iv/mKlJH9oyOCrYE26SIE7BAgA2AHCP2E855KLT7v0mqEUql/Qm+h7x30NJih+vVlRWUFWzeW/vgRU30qWMXETbu57z1PRY0cKxBL4MDIYjTgOH9P8ExQ06jdc7J07d7Cr45XW/W+RDi+l+T/Lku4/tI+ydG0hUtPtDMEyAA4Q6en3YPSVBXnGzUquELGN72yXjG+pv31cNrTpC8vLC788/eyFe+LlAUhFZa4G2b0f/yliOGjL2zw9oAlDJ3RfORMxfo9+W8fKGfUBmzYBAWInxifOHdiqthUHxPRKzCYtH9Pe4Lb3V4yAO2GrLsSwIGzprzYOmsWivzjks1Gk7oaH3z5yqU1mKtUxoq8/MJN68t+eEOmawjLZGLvvHrQ394NGzTMV6T0iBwZhXWb9uc/9WcB06CzfrhlYW4bFvHINX2G94sJ8rM+0bm5DXojHx98HhGfmPgQAmQAfKgzvCUK/PaoS/INTY0QQBwY5B+bhIOelhZBOr0lHlzzV6pMVTk5+b+tqlz5qsLERGUz8U/f1X/d8yF9L/qS3FsSeoZvcY1q59Gilzbmni1sYsQCxmAekRr8j2t6XzEiKYKX437PiE1cfBmBbm4AsAQMt8y8OwC08CfqLXKwdoc7aF1pu1mtUpcVWnDURyKRR8WJzx+bcZHcEThukoutXimFKj1T3qirysnKX/d97e9LAk11iTom8YGFaXPv9k9KdcQa+Sx33hsAIHRTfpDjciKh81vN1OfGba1SdzSz/Is/z648XWNd4xIKUuMDX7oy+bLhCX1iWnc06/Vx67x1dNd3EOhuBgCxJBFHhX3y8RgolcpDhw7BQTQPD4Ugz83NRXgZ3uSIywGfwLh4cEcT4PMOHvnb5dWdG1ggh0N8xANA4MzWuaMEYvbWV5saai2MRSCWiiPjhEo93gZQCW7CMysuxIRp6XOb4+IoAXL4BEbct/aGo8FHSQK4cDDjZKfxVG5x9pmipt0bjfvWK6oLY/IY/8fuDLzqBnNCSla90lx92BF35ENm9B18Sjsp4+QWEAM5msBbiSOiC0ICwJu3Ey6t3sLXF7DbpSVFxWWVVUrRtjNNf+So8hoNjBkhWoR39ZFf1j9wcILCT6ZTVeQcLm3FTT/AR+QM1ht2613fKuPzmSDHmEHDeQ889Lutp/fzFdO/vohAdzMAKSkprGNRgA0/o5s2bRo6dCjvgDCoAbGI4YGOx4ME+wHLgWgwvAPCwBs2NAiimvDgjscY9gOetJOTk1tqcLwWWT16lhSaZRFMVDgOesrZg57n905BDvuBcGZ9+/ZtSd7mQAY5TC8cuCKqj4vkVtUvEGBjs05lqFPqxDlZ0m0bGrL2KcoyA2uZhFf/lTp7riIuEaVYE+VcBgSEAf7oetTpvGSrdyEzyOGpEAOg1QJtZoKwd+/e7Zo64NMtMNXqTUUVDbv21205U32gwcToTZjyPzI68rqxiZcNjglQSOC0zepoG2f1HVxoMqv94VPdWTkH5JABYZQgP2IB8SAH97KyMgd1U7bPIcBzfPtcO84L1PzufOHVG8MRQ5n3PA6EmIO7Q+4md7vmnG+lS/+CNasH8UjbEeiqK7Q1lUKzWYxYOXDso2glbAvIcYG2Jbldba3+BC2Ed50cqr9WaaxvVNfmZhdvXKva+kNoQx7ESnjyrdSb5gVYVf/5q0Vzzt+48C/6ne27C1ntSUHxNbfe+qc9dBfKgrC9I6e8VrXvVNmv+4uWZtVjfR9qfjo+3B0XP2lEQlxkoK2n/gvj+wLDi1I8uNvSW5vtxlODhtvWRmlfRoDn+PblJnWsbDwmQbYCuEluW1V7062yNuv16vIiIxz7MAz8OvhFxQpErY+BVsldl8F1cny5Wq00NEL152SWbF6r3rxEUdIYncT43/rc6Aee8I91KyoLLIHrMtuWdF1+WyouDXIXa2hS649mV245VPTF0cryWh0W+UfH+l8eL739qv4DU2PhlYGrs10JF7m3q04XC3uRtYsSUjEOgdYffu42JboTAvqGOnjzx34vPHpizUcaHObd1ql0xhqlsbFRVZt9uuT31dqNHwZUMrEDIxP+9XrsVdMLG1XuaH/vNq1N7jqDOae4bsuhgu8Olh0qUaF8SLD06ctipo1NHD0guqa8ICYmlLf2b5M7FSAEWATIAPSIkYB1c015ib6+GnusiN5ldewjaff+ZEchhfXrJq2hRmloUmrqz5ws+mWlYf1ncOAT2j888fnXUq670S8qxogw5OnpHcXRp+rR6I2bDxR8/ufZX8/UWz2TykU39g2+9fLky4fGxUecC6xWYcQ2hONlfp9qDwnTlREgA9CVe8812eHYR1WUZzbhY1EhJv7yXpGu0XV8KSwOYI+3VmVQq7W1p48Vr/3WtHFZoIoJS2ZSnv485bpZsrAIlqvZ1A2/Y8Ji18b9ec/8nJlXqUYErkGR8vmXxM+8JDk1IQzH+ukiBDyPABkAz2PuKY7Ny9/66gp1ZSlUP6b8/ompIi959MR0tqZJW6sy6nX66uMHi378QrB9dUAjIwtn+iz+X+/rZokDgz2Fixf4QPX/uif3/1Zn6Gq1mPVfkhT45i2DLx0aL8WZHroIAe8hQAbAe9h3MmeLQa8pyNZGYE4tsEZwjPZ0fA9WtxnhvadRV6s2mgz6isN7Sr75QHpwa1AZI+vD9PtsVfLV1wtl3dNRpbDZAOMz5nU7s2//McPqrsdkmZoW8uotQy4fEtvJnU/VEwIuIUAGwCWYulwhXW0VtD9jMOAYrH90gsTjERxxql1rNJfW6erUBrNOU3ZgV9lX/5KfORxcyCiuEPf7enXi1Glw29nlgHVRYKlEWKfWb9yed9eqDGvcXaP55iHhz84aOH6gLzrWdrFRVKz7IUAGoLv1qdlosH7hpVFhwV0UFByYOgBfWHm4kWqdqaBKU9JoEdfUQ/VXfveBLGNPcD4TOGNY2ldvxk+40otb0B6AAotdP++r/veB7MomPU703zUq8pHpA0f3j/J0N3igqcSiiyNABqCLd+DF4usb6jXlRRazCTH9pBGxFkWAh7V/k8ZQ1ajXmQUNdfUNe7acPrBJfOCPgCIm+LaJaV+/GHvZ5O6t+qsaNBt2nf1gc96pRj2jMz84OvL+6wYM7xtJS/0Xj1P65SsIeNMA7Nu3Dx+d47zbkCFD4DPADhLk//XXX/AoAI80l1xySVpaGldg79698NAQHx/P5VCi+aBnMU76I5qf2D8wID5JXV0DVxCeQQbOCRCBFic79WZGW1dTtX9Hwy/fKU5tEmYxwXfNSLvrsZhLJ3Vv1V9Rp96wO3fxX3lHKzXw5nNrqt9Ts4eNHRRPqt8zI5C48EPAawZg27ZtcHRz8803G43GrVu3wmkBPLfYtuH7778fN24c9D4KLF++HD4LoPHhmwzO3VauXPnEE0+QAeDgMsKjZ2mB2aCHB00c9LQeprS6evPEQXKTCSc7dbVKg8HC6GqrK/dsbfplmejQDrmOUcycOfCrJ6NHXyqUyTlRu18Cqn/drrOfbM0/AdUvET45PuaOq/tKdFUD+kaQ9u9+3d3NWuQ1A7Bnz54FCxYATZgB+A5ctmyZrQGAH0fofXbWD7cks2fPho9PFK6qqurfv/8jjzwCg9HNeoJncywWTVW5vrbKYjGLpHJ84SXyVARHg8lc26SrVxsNJou+vqZix0bV2iWi9GNyExP6f3fFzLhVHxkXO+Qio86zjb5KVlmv/nl7zpt/5hfhcKdUBNU//+p+Q9OsCz6nTlUZTeZuu8ftqz1CcrUXAe8YAHjKxezeVla5XA7lHmE9s2i90tPThw8ffuTIEXjEhFevCRMm9OvXD/l9+vTBX7gb9JjDKd7OZKzN6OTLrNepSwvZCI7w66CISejAFX/nDceCT2m9DqF39bXVFTs3qb5/RVJQJRcyoXPn97njoajhozVGU15uLm8AnHN3sdoOqaRVXpX1mrU7sh/8PZdRGqD6Hxsfc9+0/kNTI7htXrx9dR73VkXyncwe23Df6QLXJfGOAYCuh6dcWynhM9nWezvW/Y8dOzZz5swbb7yxqanp66+/vuOOO0JDQ21JWk1j3YNb+mDdWFq/qbfJbJWq1UyQg85NclYYVNIqCyeZLPfWyfGEwXMyDnpWlGLFH994yWMS4NnN2srzjJyRO+F6/hbIITMrNvuXvYPzRPivuklXVq/T1VZWbf1N9c1T0hpGLmWCpt/U/+EXIoaORERJFDaqNawwtuTnq2/jX7QOVK23vQ1S622OHJUgzdbjAt25IrbkVkibV9LYhqNqLPhg1v/Ib2cZnRELPncND3/6hkFD+1g/rkZJCM1ytKb5Djy2Bt7kbN+xYqMS1xvOluwQ8vYypfLeQsA7BgDzfazw2LYZ25UYeVwOVnjgiT41NRU5cMc/d+7czZs333rrrVwBR4ns7OyioiK2KvxVqVSHDx/G/oGj8k7yQZ6Xl4eAMPzI8RgXFxe7ExAGgTXwooNK2If5nKhQRSajsarMorV6EBP6BYjCY5ims4zlokcd3OHQHzbVYUAYJy1v1qHWcDDNAWFsuaPaGo2lvkGj3r9Vsuq14CZG2sTIpk8NuuV+WWr/QpM5/9ARaELUjVAwYN3egDCcUBgeAB/6i4cKQyUYPyC3nVJwNbuSQJNzc3MhPN4+UV4khHNpYUWD/o9j9f85qrS2TyCYnuw3/5KwvjFyY0PhgQP5dtWCO/areASEQT0AGYcj0Pt2kyQ7Fo5+ghzrpYiBgXgMtn3nqLxdPsjZgDD2A8+unIOfIIfk1113nYP7lO1bCHjHAERFReEkjy0SiD2CKBZcDpS+7U+MZhcPtMBs4OLqgdnAWhPsDZfTrgTESExM5GcAwAgxlXDhOWwXU64w4jpBg3DLYmy+RaNSFedbwqyxAPFxrzjI4VsRXrMQESwujucHwNBf6BRENeHkQaK0wajZvbNs2SLF6d/9spiIZ+4e8PDfQvr0ty3DpqE9sXyHDZuWt1zJgQZHpw8bxj/aO8htd5VcYWpbBuQICMPOJKqa9N9vynhqYyFjQIQWy73jop+bPaR/okPkUQ9GDqYv/AwAyENCQmJjYyGDrUiup/Pz81EDLtdJbEvCfmDmYTfwbAs4T2PgOS9Ad30HAe8YADwYGRkZHAp1dXUYcND4mPdB6cA8QGsvWrRo/PjxbBlMiJKTk7nyricwA+I3hWRZQA3hcp2dXUmwdoe8mflF3KH6DY1WF5IiucI/LlnoNOJgS3I78Zz/BLkddOnHs45/+m7D7q/D88xR144dsmJx+NBRjippSe6oZKv5HUKO3seEtNX6nWeCEAMHHzOX1Kg//+XkP//ItwZhN1uenRD/0PRBfeLatugdDr5zge3uujnw3CRH2+3koZ8+i4B3DADgeO655957773JkyfjPX3//v0PPvggMnHq/8033/zwww+RxtGgr776asyYMXghxa7vnDlzOBBhJ/g92FwNXS5hPehZkm/GupmAkYdHyyOiPdkEvUbz59fLcj59TiTyi9OYh6/8MXnarG7syAGjq7zesHrZwdf+KrQG4xUJXp6SMP/aAb1j2lb9nuwX4kUIuImA1wwAXm+feuoprLBj7j927Fj2XRurJbAKbJPw/o6TP1D9OO9vt/2L8m42u8uQN09gsdmLLV/ILJRI/ONTRHI+Wxo8msxu5+bt2r7tg7ebsg/LKlQjX3h28J0PIo4Yj9q6Ckl+eeN3m8+8srPIrMVMVvDGVUnzruqfEhPUVeQnOQkB1xHwmgGAiDjgj4jntrJi5gV7wOXIZLJWV35Ya8EV664JHO+xuvIvyDGorRuP0pAOPujpHDeRWKwpztuxfEn6D5+YDNqYsddN+f7V8CEjnVN16bu5pQ0//JX11q5ilcqIOC3/nJp4xzUDkqNJ9XfpXiXhnSHgTQPgTC66h5OUDbXm+hpTr16I4OgXFScJ4rmnxwNLo7KpaO2K9I+fr1GLRGFpox55dtzNtwi9FEuAh/ztJcktrV/xZ/Zru4v1TQYmQPKvq5NGxAuumjBC7CBgcnvrp/KEgG8iQAbAN/uFwWavvrIULt3EAYHyqDgPOtKxlO/bdfrTN4v2/KGKSIuaPWfsHXcl973oLJCPQsZLLOus/88z/9hZwqgMTKDkzWkpc6b2TYkJPnkivfk4K69KiYgQ6CIIkAHwxY4yaTWa8mJIJg2LUMSneExEVVlxxpefFL3yTt0AxjhiRvzs+RNvmBnsf2FRzmOSeIBRXlnDD1uyFmKb12ASBkj+eW3KvCutqh+scQbIRCF5PdAHxMLbCJAB8HYPtOCPg6s47okPuwRyf6mn4vdajIYzP35z9sX7dAam/oqh0bc/HTzikpAgSbfU/mcx699y5qU/ccLHzMjEr01OmX9Nv6QoWutvMRYpo7sjQAbA53pYVZQLv55imVwWHecJf54MgyC9R1+Yo96Taw5hTPOeHjLngZCEJH9GpVHW+xw67gmUU9rwzR8Zr28tZgTWw50LJiXhXH9CxIUvEN2rnqgJgS6GABkA3+owbWWZUa2EV2f/hOTa8kosRnSqfNqq8hOLFxW9+oEYBzuHBEX8Y23YsLFSoSU5wk/ZoFV3LvNObZl95dklDf/dcOoDrPXD4YiZeeXqpPuuHxQfzvNTW/va6Tch0DURIAPgQ/1mVCmhkfGplyIuUSCRda72NxkyVyzL+r/7zVGMLJAJf22J/1U3MyJJgIRJbJ4Rw6+ZD0HDSxT2a8EzRXVv/5T+v/3l8N2GWf8701LuuHZgTKiHvqXgJTgREQIeQoAMgIeAbpMNln2UBTmMUIBwLp194rPi0N4Trzyg3H0SfhJi730k6s5nlYFRJoMhXCGKDbvIS2ubYvtmAXhxkIiFUP1vrUpfdqQSNpXxE384PfXWKf2iSfX7Zp+RVN5AgAyAN1BvydNiURbmwg8kwrngyH/L+x2Voy4rxppPxecfmSqZ4FuuGPj8W4K+o6sadWa9HpoxMpin17yOEq+j6skra3x5ddHPBdlW920y8afTU2+a3DcqhGb9HQUw1dNNECAD4KwjsYbALiM4K+T0novkiOtiMeiEInFgotUDNns1M+fjy6xVckSPyV69PGfRvcZKRhLDDPpsefzVM6oM0voGLfxNJEb4h/hLz3O2/uui5LYktumOFd62ZudpndG8emvW/NWZRoM5yE/8zg1pN03qG9F+w+b15jtvZpt33Ze/TRaOCrjJ2lG1lN8ZCHQ3A6DVauEDmR2CcOYOx4RwMNfs3LHdK9ogV6vVIAfuqKG96MNfBcjhiRoOLezcatpWBX8PCOioRWgXgcAvPkWl1eIYKAqAHMEM4CkPlTght63KNs2SAwqNVmcxmxozT2R+9Eb9sl+EUUzgLXcMfOhZ/5R+2ZVNKk0TnN0n9PKTCgxKpZ6rgSUHa37cgT9wAy2cQvPwDQlySN4uciFohIKiKs2bK44vO1WLtf5706R/mzsmISoYB2qBpOs9iJoAOLijCXBM4johix7IQcKSo+3tJUclqAHkkBm9wIMcVCCH5Bh7/EYOyOGmBX/5kQM33q6kuRFICc8g0N0MAFyHwsEcHiHAhycBegTxYRBYg9+DVF5ejnrcIcdThIhmjrhD+xuVjcbKEnCRhEeLaust1TVsx0N4uGWH32w0gd9zWF1TYzSZ8G1X9U//03673NLEBN46OeKOx/3SBpQbmdJDGYjlK5MIY0OkddX1NRd/9wQAEQ0GkTshAA/urAoDenD5x48clg9xReAGvE37AV5ikbBRrd+wv+y1g42MznJNUsB9EyLl+lKLti4/v84R+CzOLf+iQsgM7hAeepAHOUjQdqhg2P72kkMeCAByNBwC8CDHyIELRWhhXPzIwR2zH94DD+QpKSktgaUcH0SguxkAxG/BxQGNx3jQoEG8A8KgnqSkJN4BYaA9EerSWVwOk1GZn2MO9MOurx8i+l58IUACaoiMtIYb5HFVlhQX/L627qfFupMZ4iQm5dlP+916p0gRoDFYiqqUSYFmP6koMTJAfCEO20VMWg0Ic1EJpz8w94f+cicgDJQXGwjaKR/rzYKKpk/XH/76dBOU56LrU+6+bnCvIFl6uqRf/wFt0joqABuAkQNl6qhAm/mIX43ua7NYqwWgfxHJBwC2erfNTAx4ONa186HbJhVXAGHsYPl4DzwEQeKqooSPI9DdDEBLuHnMQLlKQNvmDJQr3DIBclwt88/lYOO3KM+o14nk8pbaH2XaIHdYr/VG1fFDpz58o3HDOji0ibz7gUGPPBeUkob8OqWutFZjNluCFBKcghc51m9ouDPhnXLHTc+QY2Huf7+dumdVJmMyT4wPeHPesEsHW11Vs9xhQjCbbkvSVu6DkAWftwEAeSeOnFZEvijLnZGDitwkd6fhFzWDfnQ+At3fAHQ+hjw5qMtLjBqVQCRWxCXzrKI1Ml1dzYnP3itZ+LYgkZGPSBv6z6Ux4yewBSsbtOV1GqTDg2Td4LgnnPk8/vn+X7MRIk3w9nW9H7phaLCie7otaq2fKY8Q6AAEyAB0AIg8qtDVVevrsdwvUMQkijrKzbLFnLNmReb9txvFjCicCXtoQeItd8X0sU78cZXWqqsbdUhEh/hFhnT5455Lfzl5/8pMq0eHEPnBx8eP7sdzicIuMgAAQABJREFUoawZG/pDCPRQBMgAeKHjTTqtprwER4vkETGSwI7xQVaXeeLg/UPVRxizmol6bN6IBW+r/QKxAY3m4QRTQaWyUWNAOjFcERIg80KbO45lZmHdA0v27yxoZIyWt2f1eermkTIE7KWLECAE2o8AGYD2Y+YeBVaXlfnZqAMbv/KIKPcqs1JrqytPfPJ28Wv/FoQz8v7M8M92Ro+9HPmN5eUCHIG0MDlljTq9CQclU2MCsfHrPkdv1WA0WxavTX9yXTZW/C9JCPz03tEj+kR4SxjiSwh0AwTIAHi6E+HvATZAKJX6x104rcRPCHzbdXbDT9lv/p++iIFe7/Pup/3mzBf5nTs6Ao2v0Ztyypqg/eEXoXd0oNTRiR9+7D1LdSynasG3R37Pa8S62We39L/z2kEKWRc2Zp4Fj7gRAq0jQAagdVw6KVdTWWrSqq3OPhHmReD4CI4L7KtPHjv57sKGX36z1DGRz9439KmFAfHJtnQqnamoRhcVbZZLRSlRATgsb3u3C6V1BvOS9elP/pHLqIw3DQx9ae7wYak08e9CHUii+i4CZAA81zf6hjpdDTw8M4r4JJGM/zYsjg9lfvt5ycf/NBUzAbMu6f/UP+Mvn4LP3mxbUqfSF+G4p8US5CeJ6+WHz31t73aJNCsyJv6vLD+6Hkd9pKIvbx9429R+/nI66tMlOpCE7AIIkAHwUCdhuUZTVgRmsl4RkkBr3EEeFyrJ+31t9gdPao9UiOOZvl99mjprjiwkzK6qinpNdSO+H7b0CpAmRnRJ7554X1FqTV/9dvrenzIYrWnukF5/v23Y0N7hdi2ln4QAIeAOAt40APjcpq6uDt/atPqtrNFoxLfsXNsCAwPxeSR+wkcK6+qkVSquvE8lLHDJUJSH9or9/Hk7+6zNPHli0Uu1X60VBjMRD9w7+NEXgnufO+Jp29iSGnWtUicUCHsFSEL8ut7En23L0ZzqhT+XbC/VMTLR53MHwIO/XEIr/rb9TGlCoAMQ8JoBwNeGixcvHjx4MKvQZ82aBd8ptg3av38/vPpwH9NPnDgRPnmOHDly/PjxkSNHIgF/I5MmTbIl8dm0uqzIpNcJRCJ/G2efrkjLfsiqranM+ObzwmdesgQw/tOGDHh+UcKEqYzIvu+w4FNUpWpUG+D+Pq6XQqfU6nTWg/9d62rSGL7fnPHQmmy04uYBYS/NGT40lSb+XasPSdoug4C9EvGY4B9++OEDDzzAejvJy8vbuHHj9OnTbblnZmbee++9tjnwTXbixIm7774bmcOGDfvpp59KSkrgMsW2jK+lrc4+66oNjfVI+Mcn46/rEsJgwGwUbfntxHPX6xHKMJxJefmtAXc/Iglo5dMBo8mSV9GkxXFPoSApwj/AT1Le1G4Ppq7L1kklj2ZXLvzu6O85DYy/+NWxoQvvmdyVDy51EkhULSHQYQh4xwBg+g9PVZyvK8zld+zYAU/Otl7b2NkrvEJybwYZGRmXXXYZ1/Qrr7xy+/btPm0A4Na4oU6kbYKvAnlkrFjRvuDj2rys3A8WatbsZaRM6Jxrh7/07+A+/bnm2yZ0RtPZMiV892DpHIf9u+JxT5xY/XzDiaesE3/LjH6hr80ZLDdUkfa37WVKEwIdjoB3DAAcxvbu3du2MVjQxwQ/OjqayywuLl6+fHlMTMzZs2cHDhwI1Q8nhbAHXAGsHeHifjpK8HMHxtYGWt7uwFAD3ArrK8sVQUGy4BBZWPvWMc6uXZFxy1xBCvz5MEM++CNu0jWOGgjVmVehxAdfMok4LTaQW/WH5O603ZPkh7MrZ3yyv6xei6M+X93S/+5pA/EF27HjFY6a3GY+Gs5ebZZstcB5ag7LVks5y0QNbo2cZgmcMXB6z7vc3Wm402bRzY5HwDsGAC7yucV9tk14G7B1IghfxLNnzx49ejTuTpky5bfffoNreDgHfv311xcsWAA368jfunUr93LAAQMrAquABwA52DRm3cqjcmzAcmVcTGAc19TU4KUEbpn5kAsElZkn/KVis0AoD4lsqqx0hS+WfQRGQ86//1n+/mfCaEY+/d5Bj/9dFhJeWVkFnw52NaCVjWpTcS38u1mkIkFqdEB1pYYtBASqqqrg0h2+rPG+ZUfY5k+QY38eF7wK8yNHLwA97OI4Icf7CnxTf7O1cMHP2Yyf6NJo6aJ5Awcn92Ilr62tdU7upBUYSyCHM3B2JDgp2eotdDdLDq/IrRZoM5Mlt32jbZOEKwCZAR3GNlxq8xh4IAdu6Hoco+BNjmcH458fOWAPD2/fdIdrOyU8jADP8e2mlPAYjpgVtpVAcdtqc+gdVvuzZbDre/jwYTj6f+GFF1avXg37gQ1hGIZDhw7ZVoI0ND4eG/axxyDGCMZSEpuwK9nmTzwAeIpAzicqiFCoKy3QNTXKQkOFETE6vd7qkaetSyST1Z9Oz3v5DnNBiSSFCX7xs8BRl5qlfqomfP560WW1bwKmrE7boDHBLsD9fWSgRIdoYudLQXhIDvmxsOZEBZ8vbv8vAHSTnA3N5og7vkuAhJsPFj/3W0m50sQESL65te+UoTjPaqlvVII7JMcFGWynBfZSOvgNcug+ltxBkTayMWw47jyUIGpnySEJD3K2+Wg7ngge5Ny4Bfj8yMEaYx5/+Y0c8G0DX7rtMwh4zQAcPXrU9gwPJoy2cSQ2bdo0atQoLgdaALN4DCzo93nz5uEndHp6enpycrIdkohFZxuOLisrC2aDd0QXnEMFOYyNHZc2fxrqa3WhIbqIiOhBw3rFxrdZni1QeWBX4dNTTFWMX/9+4z75WRkQKhGJHMXlgFt/f4veP4SJCpHDvXNLFnjHghqKj3eVu10NOHeLhvMmx1FdaB9H5A1qw1e/nnpmU6n1jP+gsHfvHRcfcdEGCTQ4zDzvDR6wRkQz9J1do1z/yZJjmLlOYluSHTnsq6ptvotpTGJiY2Px6uliebtieEywpso7IAwsEGyPo4Fnx6vlT96tblkV5XQ2Au04lNKBomCEYQ8Ai/tsnfv27cM+MB42aI3Nmzcjc/jw4StXrmTv4mHGGSEc+4FSWLZsGTJREkP8119/HTJkCFvGyV8ekyCuNsyAeEyCjKomdXkxCIUhvYR+rj3DJmPO6uV7r7pCe5iJuv3RiSt2B6cNNBkMrXLHcc/8SmVNk3V+FxvWuvZHE6yit3/xx7btbkLniHzPydKb/7XlmXVZjFjww11Dv3thqp32Z4V3RM5J6CSBhoOcdw0gZGtwwsL5LSv03gOfld+5hE7uuknuTsOdSEW3OgMB77wBoCU33HDD+vXrcagfwwUz/cmTJyMTaaz/IoHZx9SpU1esWIGZCKaxOPDDTmemTZv27bffYmaKN4ZHHnkEb7udAYo7dSICu6okH4+QNDhEIvKDJmmzNkNTw9F//6vszUWMjklb8u9B9z6GKDGOqPRGc2GlEhu/2CxICPdHNElHJX0wv7pRiwBez2/MZ4zmh0ZHvTBnRHJ0K0dafVByEokQ6JYIOFQ0HmjtzJkzofFtlTjeebHCw7LGli8uLPvY7qSlNV94Qea9qtPZ7VIV5llMZrj6sUZ5zDn3iuOEaVNB7pHXn61buVaSxgz/dHPchCudFIbez69QGk1mqUSYHBkg61Ifx2Li/+LyYzuKlClh8kWzB8y8oo9EZN3LoIsQIAS8hYA3DQDabKv9W4XAVvtzBXxW+2vKio0aJU79K6yOOdvWbiW7/jr+zFR9HhN49aUj31wS2m8w18aWCZXWiJUfrG34yURJkQGSruPds16l/2xt+oI/8gDJw2OiMfFPigps2UDKIQQIAQ8j4GUD4OHWdio7rOQg0CN0nH9sUttRHs2mrJ+WZ9w5nwlgYh5+YsRzr0qDQ52IBx1aVK3GUSJ/uTg5KgAn/J0U9pFb+A4BkhzIKL/5vweLajSSQOnKuYMx8ad5v490EIlBCJAB6JgxAC//6pICzP1lYREI9eW8UuwSH3j12coPvxBKmNR3PsKiv/PXBXj3rMBHUgwT6i9NiDgX78U5C6/fxRn/epVh0Yqjf1ubxUiEtw+NWHTfuJiwdp+n8npDSABCoBsjQAagIzrXYlEVWzd+rUv/UbHOa1SXFu2cN1ib04gThmO37Y4ac6mT8phC47gn9k5RJjJYHh3q56SwT906mFUzYUk2ozUzEtGaB0fMujzVp8QjYQgBQgAIkAHogGGgKs4zGwxCkTgguRX/zLYMiv787dCV1wuimICxQy/5bL1/XJLt3ZbpvCp1QLDVp3NcuCKsiwRzr1Pq3/nh8Dt/FcGT8z3DI9+5f3x4YCufKbRsLOUQAoSAhxEgA+Au4JqKUoOyEV82+Cf2dubs02w6/c2SM3c/Cn6Jz/xj+BMLhHJn03l801tYozNaBHDukxQV0FWOe247Vjz9i8OqJj0TLPnf9fHzZ4xxF1+iJwQIgU5DgAyAW9Aamhrh7Rkr+PKoOJFjha6rrT74z+dqli3DZwsDf1yWdsudbXJV60w49Oknl/SJCZR1Ba+Y+DDtreWH399bxpgsf58Uf/+1aUYVkKGLECAEfBcBMgD8+8ak1ajLCvGpF5y1yUJb934lFIsbzmYdfP4u5ba9slRmzFeHw4eOdIVlg0qPTYUQhbRLaP9tR4v+78sjJfW63r38ls4fPmlkArDJa2z7IzhXoKAyhAAh0EkIkAHgC6zFoi4tRKzH5iiPrW/8CkWigl9/Lvngbv0ZJmzOTSNffj8gvo1Ff1Ya+DFQao1Y+g9S+HoH1TRqF6048s6uEvik+8fkxCdvHhYebF3aMplI+/MdWkRHCHgKAV/XL57Cod18oP1NOq1QIlFgI7c1jxQWg65q/cryt16Eg4fkN94act/jIoWrJzjVOqPBZJaIBXDy3G7JPEWACAQ7jhU9tOzYmWpN7zD51/eOvHxYvM+55vAUGsSHEOiKCHR/A9Dmx8ZOug20rZLrayqNygahSBgQlySUSFvWoKkqP/zGi7Wr/ifrHdjvxS9TZtzSsoyTnEa1EQFFAuRinKZ3Usz5LUfCO6fi7oIcO9vcT7sEVqje/v7Q2ztLcdTnH1OSnr1tZIj/RV6J3OfeKvJ2Yjj6yXJ3Ir8jQjYfhM6b75wcd91vPm/hwZ2Vv00hHRVwk9ydjnMkEuV3EgLdzQAg1hhCC7APDwYifJrD5ygPf86AG+RFRUVwPX0ROTy863XaorNYoBeHhEsrqhjLRZFehGKJsazw7D/ma07mWGKYwCcXC4ePy8vLbxnOxUmPnq3UYAVFrKvJ19QpVVa/yk4Kt3oLCLBhVXhHBQGMuOB41Y477ILWwCxam/vl8ZroAOHrV0VMGRnVUF1SV3mRkHDsWlpaCqcd/HxDwtsrQsLBpzE/cvQayOFaih0JrULkJBNMQQ43JLwDwoAc44e3Y2QMPAxdfu6g0WSQo+8QVcmu75w0mbsFcsTqYF3z8iPHM0gBYTg8fTzR3QwAlDVGMC7gjkGMC4Ek+UUEAy2ewIvIBQKzQa+uKESF0qAwWUw8vAZzHWydNYpF2T9+U3r/40wvJmLu7ZIb5/dK7RsYEGhbjCvfagKSK7UGuZ9AAU9vWqNMJuUXkwvaB3704EiVHznEgA6F8bAjR5ioepXuha+PrMtTzUgN/uDe4VFBEqxW2VkokEN1wqU+yFFPqy11kglySA7weZNDcvQRyJ1wcXILio/ten4hWVAzuCOgAuwfPx3KckcNPMjR9SAHLYYuP3JEYkD32XW9E7hsb4E7bL9tDqV9GYHuZgAw6HFxiGMcYzLC23kcZqBhYWF4mLkKVQU5spAQgVgSmDqAy+QSJz97r+SR5zET7vP6u0MefDa3oDA0KKi9cTk0NergYFlUqMLQZBGLJe0l54SBDsXFmxxPMqqyI9ebLM98tX1drmpSfMAXT14R5fjLZEyfERQFAHLytCsBDY6JJG9yTOFBywUUahdrtjDUH0YOJgE8aEHCcm/VlaErFbLk0OOuFG5ZBtN/yI+h2/KWKznQ4Hhw7LreFUK2DEad64WppHcR6G4GoCWaPCZBXCWgtSXXlBcb1EoBIwhI6sOVYRPa6oo9j81p/GMbfo78fU3StbOQMJv4BGWF3ze4/Az1l5Q3WKOasPXz+NssO39y6FA77tD+897+c3Vm7WXxgSv+NikqxNmHbC3J29UEN8nZtuMvXibaxZctzJHzoLWtoceS8244EXoYge5vADoKUH19Db7nwlK+IrE3Fvptq605nb5/zjB9NSNPYy5bfTYwsbft3Xalsf6DBRWFTARf+W4o/3bxdKkwtP+ct/5ck1V3WXzAT3+b6Fz7u1QjFSIECAFvI8D/kIm3Jfcof3zzBZcPVmef4dGSgAtLTBAie/Xy3ROG6bOYyDl3TdlS6472R211cKLAML7m+AHaf/67f63JrhsX4//j8xOjQ8mpp0eHHzEjBDoJAXoDaANYrCFYjEars0+zWewf4BcZwxGYdJrjn75X9OrLOOnf/8sv+s2Zj70B7i6/RKPGgFWb0IBWjpbyq9B9KryTPPHJzhUZtcMj/VY/PzE27MKOiPuVUw2EACHgRQTIALQBPpaDNRXFOPyD8/62zjubivIOv/Zc/fKfZcOYwW//kTDpmjYqcuG2UmNd//GTinwk1qMIJz6NlheW7PniSFVaL9mGFybFhZP2d6EjqQgh0EUQIAPgrKNwslNbVSbRaXDCE9qfi9VelX7k0CNTtXvqQ++4fuSr/wlK6eOsFpfv1ams6z/BCndfI1xm6KwgtH+j2rD0f/s+2leWHCbb/OLk+Aieh1KcsaF7hAAh4D0EyAA4w96kajRozBKFQh4ZI/I7t/CdterbjFvvZORM3CvPDHtyoSyE52E7O8bY8m1qXv8J8fcJ7/mQ5d/rc5dlqgf2kq9/YSJF8bXrL/pJCHQDBMgAOO5Eo8FYU8GEh0tDerHOPs163dH3Xytc8JYwgBn41Tdps+cyog4DEOd/jCaLTCLE/45l8tAdeKL+8OeMZWfUg0Olq5+fkBrL84sqD4lLbAgBQoAXAh2mv3hx910ibPmqi/IsRgM8uPlFx0FQZUnB/qfnN23aLu3LjPjqz9jLpnSs9Dj+jwrD/L2//aszml9ftv/tXaWDw2Srnr2ibzzPj7k6Fh+qjRAgBDocAW8aAPgqOXDggFQqHTNmzEX+dppbWVtbe/z4cfZ7VGSMGzcO/mFQHt8osijgfE6/fv2ioqI6HBRUaHX2qdcxQpHV2SfDVB49eGD6WKOZ8R+Xcvm3u23PAnUId3b9B1UFe/v8D9r48pf73t1dIlGIPr49rX9iaIc0kCohBAgBH0TAawYA3rK+//77hx9+GN+dL1269LbbbrNT5Xv27BkxYgSn7uGVBd/lDxw4kDUJcLO1ZMmS1NROCTWurak0NMEPmlgSnYCILpnffZlxx334ojTm7w+Nfe0/AmnHr9Fj9R/rPxKxUCbm6XugQ8YWTqC+sGT3B3tLwwKk6+4fnBDK5zPaDpGEKiEECAEPIOA1A/Dhhx8uWrQILYTXKpgBGIP58+fbNjgrK2v69Om2OTAAMTHnjuGnp6ffcMMNcXHWxZmOvYxqpRbffDGMIjZRUlB44Nl76pf+iFX5oRtWpUy/uWN5cbU1qK3rP728Gjwd2v+pT3d+tL+cUYj3LZwUE8SUw9cpXYQAIdB9EfDOfiNm/ZMnT+ZQZWf3dXV1XA4S8fHxp06d2rFjx5EjR2zzkQb59u3b+/fvb5fv/k+LyaguzscXv35RscqKsvwX7mpY86PfpcGXHt7XedofZ/+bNAgA4OUDoM/9d/dHB8oZP3HmwklpccF6Q7u9eLqPP9VACBACnkTAOwYAjh7tXPXC86JGo+FarlKp8E4Ab4hjx46FV8LFixfDPSR3d8WKFXhp4H46SWCfgNtFcFKMu6UuLsCqhywoqOr0iZ2z+uuOH46a/+Dkn7MjRo7jyrieAGtX3Emqrc7fGLlUYvf9F8hxuc7OriRYu0gOl9av/G//B3tKo4JkWS9P6ZdgXfd3ndyOL/vTu+TNyDkLaNOqzFwmO2xcRI+jsk2A1pWutyWxTbPy2+a0K+1d7u40vF3NpMLuI+C1JSDnosOPLgwAuzOckpIyceLEkydPDh8+HFTwMg9/s9zegF09+fn5FRUVeICRj8cAhgSLRW27ZQcBPODXVpma6hDwRH3sSO2nb1h0jGHGbGb63OzKanNpOT4JtuPl/Ce4FxYWws7BPbUjWquUAkFFo6FRY+rlL9JVSzjP+iBHXA68G1VWVsI1pnNeLe+CvLq6Gtvm8AzshByNNTPCFbuq3tpXzYgFH18XbmwoPlFbiAqBM3bp8bLlhLwlXzYHYMKcw8wjGAC/eACQPDc3F73MjxxUCAQEYSCJI/CdCI8mgxxzDgjAgxw1gxzQYaeKH3lBQQEOQcAPeXvJwRpdjw02TJ7gU5pH34EcHYeG8x54IJ8yZYojeCnfpxDwjgGIjIzMyMiwBQLD3dZrP9SWrTdzvB/s27ePNQD79+93MrxiY2NROVszBjGUIDaKbR362zK9kBYI4OlTz+gNUmHGfxcpN/+KmfeonzZVB0fEJyb6K/j4PsODBO2DVuBy8hgbTWZTuTLMbEmNDrB9AwAt1AeOSEVHRzshv9CEi1Mgx+YKLGViYqIjcqGA0RrN//7x2Ft7of2F6QuvGJAQhL1o1ARy9AiutLQ0HkoENSAYAKwXb3JIDhvAmxwGADWAHA25GBiXfqHJIO/Tpw+a4BJBi0IQHhMX2yHdoojDDMiMLsN2l5Opg0Pi5r7DHBwzD4RDcNT1zskx/ULDeQ883mEMnEhFtzoJAe8YAAyRw4cPX3/99Wyr8LTgicVSD8YrHjwoPhwAxR5A37592QI1NTVQZGwalsOJAYDGxMWBBS2Mh7DNuBxw9mlUN+pKizI+/pf66EHFkP4j3/u+1+ARjRkZcpmsTXKOnV0CDQEt/trl2/6EuwWRWOovEQYH2rvZAS3a4pzctiq7NMihSpyQ48Tnsl+Ovbq1BOv+R567fEjvCNTAubNjJbcF065+5z/RoWDNmxwqDLS8ycEdtPz0L9qFcQjhQY45hPNmOrrLkgNDRwWc54Ocxd95MUd33SQHaxgAVOKofuf5vGF3Xi3d7QwE+C8xuynNnDlz1q1bh3kilhqw2oNFHlSIWHQff/wxEviJhX6sgeA9GgFOd+3aNXjwYOTDVJw5c6ZjFxktZpO2vLh0/65jT89S7zvY64a5E1fshvYHOygCXEjwvtokx/dfKBPS2vdfVt5ucHdOi53nJWuPP7EmW6oQH3zm0hFpVu1vezknty3Zatq75KxI7sjgDi24g9ydGtyhdb/trPxsPTz+ui88D6ZEwg8BnhMcfsxsqTC7x0Tj4MGDmGRNmzaNXbfB3OHGG29EMaj4F1988ejRoyiAVZ177rmHnYthRv/MM8/Y1uN+WlmQc2b5VyUfvSsQM8kv/WvYo3/rQAcPzsXD+o9KZ4TbNQ8HAMCWwpJ16Y/9nMUEiHc9dcnofp3yMZ3zttNdQoAQ8DoCXjMAaDlWdbiFHRYIvHhy33YhjSNAdgDBMHAF7G7x+6mEV+eXn6jfvEmkYEYs2xg/8Wp+9fCjUmrhb8gM/89yqUe///pszXGr9vcT7X/ykjH9o/kJT1SEACHQ1RHwpgHwOnale7YdmTfZqGEUg6PGLt4S0s+6yuTJq6HZ/4+Hw798sub449D+QsGBZy8l7e/J7iZehICvIdCjDUDkiDGK0Zf49R485pUPEO3Lw32D4C9qncnD6z//+enY02uzGJHw8MIrRvallR8P9zmxIwR8C4EebQDEfv5XLNsokkiFneDep81+VmqMsAH+MrHt6c82qdwp8PGa40+vzybt7w6GREsIdCcEerQBQEdK/AO91Z2s/5/g1s7/dIZIS389+cRq68rP4QU09+8MgKlOQqDrIeC1Y6BdD6oOlbh5/Yc9/+MJG/ztpoz7f8jA912HFlw+ks78dGhXUmWEQNdFgAyAd/rOGv/d1Onx39nPYFf+eebOb0+inftevGwUnfnxTocTV0LAFxHwxPTTF9vtbZnqVQZ8YNap8d+h/fEl8NodOXP+l451/21PjRs38JwzbW+3nvgTAoSATyBABsAL3cCu/8BVUKDigteKDpcD30z8uDXniY0VjFS46eHRE4fHdzgLqpAQIAS6NAK0BOSF7sP6DxxA+8tFUnEn4v/X0Yonfi9nhMxvD4y8anSSF9pJLAkBQsC3EehEBeTbDfemdHXw/8MwQX6dOP1fvyvntm8z4ePzl/tGTBuX4s3W8uLNz4snL1ZERAj0XAS6uQGAHoH7IN7dC1p3HM+BHJcdd8R+0erNqDa0rQCQrZLb1dbqz7U7c25YcpSRCFbO7Xv9Jb1bLdNmJiRsKXybVFwBN8l5t50VgCXnbUXYYeNO80Hb4SOHw7bNRIeg1yYXRwXcabijOim/kxDobnsAmZmZCKbBPrr4C/eiCC4PH3M8PBSCHOFlSktLeZOXlJTAKT8ujjvW/Ru05mqlSSERNBYDfO6Off+COyLbwCESwgm47pHf30/y69GGhdvqsOv74jBhoD5/3z6l6+ScENCAgA4Xv6ggIEdAGNA6D0fDsbNLgBzBWAC+VquFY2e7u23+BDmoQA5fs0g7hNhBRSxJXl4e4uFAl/EgR8XgXlVVBZfU/MjhARcXAiK1lxysMXIwaBEDg184AZCj49BwhBPgMXJAjoAwM2fOdIAuZfsWAt3NACBQMC4O402bNl1yySW8HZQjqkZSUhIbmIyr0/UEwkLhKUJgJluSvAplrNYY30vRpgsgqABokKiodjhsWLPz7MKdhYz5/9k7D/Coiq6P7yab3hMCSei99yK9WxBQX8FesIC9F+yKWLC8CtgLfnYsYKOIIKB0EBCQ3jsJJRBSNnWT7xfG93rdTTa7dzdlw9mHJ8ydmTNz5j9zz5k5M3eOadbtnbo1Cc/NyalTt66+dtfDyG7cMBi+eo97vNHELVu2dL1GfU4UAIpTuQDSx7sYRnJB3r59exfzO2bDZUXr1q0NT2YRvviTMXylPlq/du3ahj2roL0YuhTi2C5XYpi4cPmuWwNPXyz6Q/8o4aqMgL2Boirzaow3A3MorSJoDUyCnJAXFPt/z2fiHxlatqspauenlVZm4Keluy99/09Tnu2HWzsO7dEQzjUHk2XSOmY4Q+1G7XYleEjubtvtalfkbqGnL8FDcoqiBBDQl+lWWDHgFok+c+WSe9JwfSskXAEIVH8FUAEgul5FujWPzOHBAdwB5zqVKzlnr9x7yXt/Inimj+l4SZ8mkCAFXCGsmnl8mvmqCalwJQg4IiAKwBGTcoxJyyxWAF7//uuXVfuGvbMWkf/VTe1H9G9ajg2QogUBQaAaISAKoOI6U/n/or4SHUAa5mPe6v1D3lmD9P/4ujZXDmpuuBwhFAQEgbMNAVEAFdfj+H/HKhMRYvHznv1n/pr95yP9TaYPLm95wwWtKq4xUpMgIAj4PgKiACquD/H/TmVenP4vWHPg3LeL5/7vXNpszPCKdmdWccBJTYKAIFA+CIgCKB9cHUrl/p+sXBvbstFhQQ6JRiIWrD0w+O3VpsLCNy9qevvF7YwUITSCgCBwdiMgCqCC+j+j2P5TFB5sUVc0e1hrsfR/8w9TQeGk4U3uutT4aXcP2RByQUAQ8GkERAFUUPcp+09suBem/7+uPTh48ipTXuGki5vcO7JjBTVAqhEEBIFqh0B1+xK4anZQfkGx/3fm/p5vAPy69sB5k1aZbEWTL212z8gOVbO9wpUgIAj4BAKVqQBmzZp16tSpvLy8Vq1acWGDHV4bN27kGh8+psdywvUsI0aM4OP+tWvXLl26lM/cycxtM7fccovhj/XtqivXx3Tuf/aG/WfhukNK+k+6tOndIv3Ltc+kcEHgLECg0hTADz/8gNxv3rz43DqCfsWKFXY6YNOmTaNHj+ZuKfVRKAFycr3MnXfeqV3RpSKrfjedPnP+J7as6z+dN2TxX4cHvb7CVFj0+n+a3D2io5e/JHZet6QKAoJAdUSg0vYAuK9KSX9Q7dmz544dO+yuEFFXEjL3R8orQc8FYcz6uaZKH1nFOwWzT76tKDvPxt0PESFl3/9TWnOWbjzS74z0f2V443tHdvLehwSlVSjxgoAgUP0RqJwVANdM2l30yI2bRMbHx2uQBwcHz5gxg+k/Qr9Pnz5cjsgFk1lZWSoyPT39ggsu0OfXCL0YQNPw86RA7n/m9jdbYVFkiNv3/2i1L990pM+kFZz3f2FIo4ev7OwiPx5y7tPkCiJPmuAJLbVD7kkJntB63nbFvyrHwF/PmTdQqZAYQ6DSFIDdFc3cmcxmgNYG7iLmcciQIVyIz8T/k08+GTlyZPErZTZrkW+++eZtt91mV05+fj75yUZRWIpYVeTm5rJoUHYkrXxXAiw7oOVKengwRk4TrNm5GTYTgZBIS15eruv3s6naWTes2pLca/JKLD/3d6v5wGXtXCwEBGCeegHEbmnlStsrlxwO4dww85DjDwBy+k6NBFearM8DYoqcrtfHux72pHZ4hpzuUxtgrleqcmrkFGJg3KquVwgYIwd2d3mW/JWFQOUoALxV4DVC32am+frtXPTB3XffrTIgvpnsr1+/vn///mPGjNEicTqxatUqIvXlHDp0iJLVa48MZSxu377dsGMN5ZPEMPnBgwdSjp1It4XAoZ81MMWdtQTMHz92dP3+7AdX5Jlspns7xtzYL27n9m0u3vAMAqyokP4Aa0wBsMbCIwqS1Bg5yzU6QulgfQe5GEaLHzhwAOQN1E4VUEFu+Dp+BJ9yK6Qfky5yrrIxcmiCYQlO7Xiz4TUxJoK50B9fAnikMUaOJyIaziTMMHldo14o3AJZMnuOQOUoAHxNLFy4UM89x4HwoaHFIO7ZIdBm9+gAJkRsC9epU0fzr0Kq41yj4ZmfVs6vv/6KrQlrkhbjVoAJIA5hNDbcoiVzRHh4vn9ovl9oWJB//fhQd8kX/BH01KrNpoKix3rVfn50D3e3a3j/AQ3E3K1X5Uf6o0IaNWpkjBzFQ0/pnfO4VQ6Kh05v27atW1RaZiQXIqxdO+MfSEOOQxjUsFamWwFGDg5h8OfjFpWWmSGXlJSEAtBi3ArwKgE+P7eotMw4FAP8mjVrajFuBRh4buWXzJWIgMHx7SHHDK/k5GSWqKocpoqMeKZ7zFiZOhHJ67ds2TKtlr/++osjQzxu2LBBi+TgUJkunxAExqaQqhZokURaje4GoOX8D3+jQ91WtEdPWS/6ZKs1z/ZIz9rP3+y29IdV6vWk7T5NTsP5GZjAqi5Ww4YS3O1xLT+0AKg9uhvwEHwPa4ecn7s8a/k9abhWiAQqBgG3BZO32HrggQdefPFFjn6iBnbu3HnPPfdQMnu8L7300nvvvcfUb+vWrZ9//jkTeg4IMQ9lUcmPHeDPPvsMP4WsBlAJiYmJ3uKnPMrh8H9eQWG4v19osNs4fzxnqzUzb0z7Gi+OMSL9y6M5UqYgIAhUMwTcFkzeaj+LgKeffhpLMYvloUOHqmJZtCL9Vfjyyy8ngDmyd+/eWqXK2TSrh169emmRVTaQmWNjHhUS6M8BILeY3H3k9GO/7Tf5mW88t6F7lG5VI5kFAUHg7Eag0hSAgh2zj3P8S/RMXcUn/lqLMnMK/YNNsRFuG4Lf+HEjF73d2CKsUUIZ+Gh1SUAQEAQEAXcRkPmlu4i5mj+voCgn3xZo8QsJck/Lbth1/I01R02Blqv71vYz+7BfX1eRknyCgCBQSQiIAigv4E+zgVtYxPkfi5uf7b40vXj6P+ncBs3rxhQWexCQnyAgCAgC5YKAKIBygZVCix1A4v4l1L0viVZsTv56c6opxHLl4OYoDhH/5dU9Uq4gIAjwfZKAUB4I5OTZcvMLLX6m4EB/t8q//4v15P/4kma1okNYQLhFK5kFAUFAEHALAVEAbsHlauYz/t+LQgL8uMvBVRqT6fd1B1ftz/CPChzRr5nrVJJTEBAEBAFjCIgCMIZbGVRp1jxm7xEhbkz/uTT0+s82mPxN31zWKsJNw1EZ3EiyICAICAIlIeDeAZWSSpA4ewTO2H9sgQEWS7H9x1Uzzoyluw6esEbHhAzr1di+RHkWBAQBQaAcEJAVgPdB5fwPt37i/z3Agjcbl8pPt+aN/HozyuKra9oGsXUgP0FAEBAEyh8BkTXex1j5f3fL/e83C7abrPlNEsIvOKeB9xmSEgUBQUAQKAmBaq4AuNVLXQ1dUtvLjoPWXfLcgkI+AQuw+PMFgPILUmY1R1KzbvlhJ3c+Tx3TRZ/ZQO3eJdeX5m7Y8D2aqiIPyVWvudt3WhsV8obJKUeVoBXobsBDcsWAu5Xq83vYdn1REq7KCFS3PYCMjAzNDQhChIuGTp48aexeeN4BrlPmnmpuNna9CzHm5GbmWYItJ08W38XPpcpcrOjkWspily/rDjzSOiA8IrJpTcuJEydUXdTO1XgwwKVJTshLYwxyoOBKeq50NkBOsdxHD5KGyaHlmj/D5HDOpZL0nbFrKaGlBO4lNibIQIyLaekLbqUtDWHn8bQd5g1fBw16DDz+Oq+lxFSazMDjL63gV2IeJ5EQMvBouOGBx5g3fJW0E8YkqTwQqG4KALHFm6Nee/7iCQC/Fsamk5DzIlEaWsT1FwlfZBEWHAGbjh3L5Ko7xBCvsXPyNvVC2zVsgcfH5JSjWh9TO22Bc0pwTq6R6AOQU7uSnsbIYRv1Y9ipiPJFY4ychsA5teMX2gDzkEMFuaZN9ci4GFbkxkYOVYAeCsCY/tAGHoLYQPMhhxD8aYIxcg8HHjMPF0GWbJWOQHVTANwTx0+Ddfny5TiWYS6jxbgV+PPPP7mPGudKblFpmbds2cJldnFxcVqMW4E9e/awdsEZsltUWmbcQiECDHt0YfLOVayGPbogg/DFZpgctffHH38YJkd/4C2uZcuWGhruBiCndmMSnLrWrl2LQxjDnohwgNGgQQO9iyS3+N+2bVuNMz+3qLTM+ORg0mN44OFPBsWDHtIKlECVRaCa7wGAuyfuKRjH/Ax3HrTGLBiqxjOVe1S7J8zDeeWSG4YdQsW8Yf5Vr3k4cjzset8l94RzTzpdaA0gUP0VgAFQhEQQEAQEgbMBgWquADyZxNH9zGUMzyIhh9YTcmrnZ3gUUrVPk3ved4ahg9Dz2j3set8l94RzT7pMaA0g4P/EE0+wa4/BsFra7HCrjQXfcNPwVEwJhg3BWPAhN7wDwRkSmMcaa6BfIYGQ2o2dgIKcVlO7YSs226fUDoDGmIec2sv0F1Ra4fQ4tfMrLUOZ8Yrcw5FjeA8Z2GGgEgceyBs+wsTA42cYujK7RjJ4EYHiAwP79+9np9TwYPUiN1KUICAICAKCQIUhUM1NQBWGo1QkCAgCgoDPISAKwOe6TBgWBAQBQcA7CIgC8A6OUoogIAgIAj6HgCgAn+syYVgQEAQEAe8gcBYpAO514PvSEmHju3l2wg8ePKg/wcZ1LnwSeejQIX2kItd/7M7nvo53tkC7ceNGT05h2vGprpSwi9Qe+eh37969VKrFEEhOTuZbYi6E0EcSTk9P11oEFTcW2GXgcdOmTY6NcszmegwnzUrLDAO7du3ixg59BiJ3795tF8nFEjt37qSxKic3FgCynkqF6Ue+YXaMNxxDRaV1Jd9agyHfvuoLBzqQt4tkzNBMRpQGPh8b66lUmE5ct26dY7zhGFdGjl1fM+bthhOHYomBf6BQnNCQEkHmE+jS3jLDTRDC8kPgbFEAXGzwyCOPlDhkSZo6dSr3xvDGvvvuu2r4cgXQZ599hthCMXzxxRdoCK0PfvzxxwULFmiPH3744YoVK7RHFUAwjR071k4i2+Vx/REpM3HixM2bNzuSwNhXX32FrETQv/fee0rOImKI3Lp1K5HTp0/n1dUId+zYoWds3rx5H3/8sZaqAikpKVdccYUTkW2Xv8zH33///ZNPPikxG0m//fYbSXCiwbho0aLFixfTEURymYciXLly5dy5czlTiw749ttviaQ327Vr5yia33jjjRIVQ4kMlBmJHr3hhhvoAsecKCTaBciw9PXXX6tPB8j/6aefIlK5j4FIJe4ZRV9++SUtYly9/fbbajh17959/fpiF9D635IlSx599FF9jCdhanzppZeYozgWAm50PYqKCQF8KrVKflqEcEddffTRR+oyJQ4KkoHxwEHBb775hsFGaQsXLqRFdsWSs0uXLo5zDrts8lh1EKjm3wEooHmpNmzYwB1B9erVi4qKskP/u+++u/TSS7l6pU6dOtycg3xHrLzzzjujR48mBpLo6GhEEhfL8D6gDFq3bs3LoF1Tw1yViWqbNm3052h/+eWXiIiI3r17G/4IQGMSOT5nzpxmzZpxthp+tHgVmD9/ftu2bTt27Mj9i127dkVP9OnTB0nKFUbdunUjkrag3giTn9dYvZydO3emNGJ41VVb9Af2Ecp8F0IeTuLbVefuI9V9/vnnCQkJyErYsyMnkqXGZZddFhsbC6rI+rp166KPkaFDhw6FByJpO2eU4Xb27Nk33XRTTEwMPcUlcRwz50cGRBWN1UpG1KJ3uUVHH6mluhv4+eefUZ+MHEB2PBf/+uuv33///YAMS5zZ5wIfKkWA3n777UQSZkhwIVL9+vVRw6NGjQIHikpKSqLVRMIMg0cPC/KX9pKHTnSXVcf8TNh/+OEHBiqfFFCpXQaUK6OC4QFXwAiHnTp1IrJHjx4MZu6w6tChA9MIhgHIw0/Tpk0BnzyTJ08eMGAASgVA6C/9txpo6IEDB/KmGP58xI5JeSxvBM6KFcA555zDlJYxra2+NViRQYxgTdJxcRtSHmOR/iYv3lWmNtCiCW655RZeDF5UrQRkEzH6SdaBAwdatWpFZi2PJwE4ufHGG1EAjlNd2EBWNm78twtJxA3vM3NPJBFUqlKkJJGIIR5HjBgxZMgQvaynBEQtClLjkLYjQKnOESstj+sBBAHMI1P0iGnkTOoRJdoj2bh9jx9SRou8+OKLMZUw2UcEa5Hx8fEwCSDt27cnvxZPgDKRax5+x6sVeO655wIa+skRDcYJFQGvyozUYyyxeIJ5LZJhwByf+T6aW/sij9IUGpQJFT2oVcfkGn1QIlZaHtcDjNsxY8ZQhePIoRAWSUxuVGmMYcLUzqpFuwOOSMJMbpD76CStXrWuhcm+ffuCthZPDG1B7TlipeWRQFVD4KxQAGruVuJrwPjWXgPVN8g+VvT6EU88SoLJWonTecY9JJqlgsxMo1A56j3xvL95DymkROaZ/NpNS5nNYQFnuqeoVO2wp9byKlL/flIswhSxpfGJDRfThD6PlmQggE5S0rDEAoGImbJWLJijaNFALJ60SNQ2zNOuXr16qUiKWr16NdIN5vn1798fu5ZKoo8AhDKJ10rwJOAEfLUy0BdOZsx0+haRCgIwfMEFF2g5MZ4wm+YRsxI6j8WllsQFqEwmvDtyStOFdtfEMkjQsnYLBfIwxkAYDrH+s4x+66237rzzTh4plhcHi5Zm8CEVda43lmrtkkCVReCsUABO0EdS2F0YwBvLVE4/TYYc0V/ayKYE5rmshdWmH4+88Eg9b8kgJ8zzEuoFPTl55I3VZqCKFu1VmhQgA5l5w7UNSSbUSAEn+Z3w426SXS2qLdhS7BSttvFI+QjHDz74AEHDhBqc6RRsFBhqVNVor9JWG+7yVmZ+xLedoYORg/3K7v4G+NSPhBkzZqDelCmP5lMCKzbUHtWh57BPEqPPXyYbhjPYDRLeAqYydu8ChSveCNBewGeJzABTlZKfiQ7mLPXIHAjFRqR6lL8+gcDZrgAQOoh7fVcxA2Xiw9RGH8nod3w39BlYDmM6Jwaj/LBhw/RJ5Rdmtqu9n6oWmGfKbHeoA4u53UJBzxJiCCP7zJkziWTpg1GFALJVn6ecwkhMfclIFmQfUsaOf2wmKhsz/RdffPGqq67C5q4nZOmglATGN9YKFSNA2Waw2ydHG2GOs5sosMRUKoER9fTTT2PI6tevn555HtXuN/veyNOKYR4GGCp6NlgFoqv0a0FSkeZYR9W7gOjHPAX47AMrQlhl3aD22yFUu2IVM3L0nEvYEwT+9QZ6UpCP0mLl5NCbnnmmwNgxEYVaJCKSkxJ6u4SWpAV4T5TYwp5uZz7S8ng9gAxisqx/5bA40yIEk7Ywp1LmaLyozmtnl4/ZK6ebNEuL8/xeSYVVtri1ohDfbF3YaV+6Q4l7LCfsZIwbN87RTQqWeo4SYb/Wb6hqxZZTgJ1S/cYPUwTGCQZ3u4M9WNiYZLDZzlb8M888w9Cy44cYjuIgZNF/JW5T2eX31iMLVr2yAT12LOysT5ikGGAci9BXSsPVoxp4DC2sjgx71l76bBL2CQTOIgXAcNfLSo5vMllj1sPUBpuy6i1O71x55ZWEWYxrskmdl9B3p/7N4bVXSYihJ598cvjw4eqRwvXV6ckNhClKXymiR6koDCCc9FAFIh+RIITZ1/3+++9VJKdiWAFok2giNYZVWDEJ85hWOHijzC/k8SLzVKS3DGBrBlIimfDCp0pCfWIeYS3CHHnatGlKEqHGZs2axRknmoB9X8MWWn4aJrSO7UdmppDr48/k8sIfPWJwq44/KuM+kl1VwBKK5rBMRNwj0FUkkA4aNIiO43gMpnM76xA6Q2WDkJOjnBnjkUbZrSFUHsN/NZRUCUhqFC3hkSNHcv5NRTLUycYUZ/DgwdrhTmz6GDZZUaFxOVagcmKnUjsuWofSOzQTiNRJCru3TFHJ3yqLwFlxDFShj0xBUmin1nh1WbTyGjPrJIwFkx089rWUDYTZEGtbLOOIWuakHKTRdyFFadN8wsxkSUVnsJi48MILVU7eEHU6UE9oOExpSBZNjmPxQJpw0IgJJq8c1ie4JQ+inyp4k5FETOd5h5GhnLPUG3xhmBdbCSPCvOFoQTIwA2X6r/BR8XY2bsPMQ4hQUwcfCVM4axSgplL2PLHgM81nT/Waa66BE5pJv6CJ+R4KPInEfoWZDqnKuUa6A2lFYxE3CCaS0N+UCQ7svqq1AsKIku02Yz1hXqGhVCNoo8DUISuY5xgMswemz+gttcziL5wzcohkUswEn9k9MpczNop5hpk6ekBRioR+JAZdrjqFeLVL7AnPGi2jApQ4yaNi0KMMD6CjUgY/I0cJd2W3BEx4UJF0BGc6QZLuoC3YqchJ22+++WYiwYRhDyYUTjkc/1cKgHj6gkiNAQlUZQTkOuiq3Due8sa0jnfV01LKnx6RhxCxq8dXmLdjWz36NPMltojIEruptMwS7xMI2L91PsG0MOkiAj4h/WmLo/Qn0leYL7EvfJr5EltUWjeVllnifQIBUQA+0U3CpCAgCAgC3kdAFID3MZUSBQFBQBDwCQREAfhENwmTgoAgIAh4HwFRAN7HVEoUBAQBQcAnEBAF4BPdJEwKAoKAIOB9BEQBeB9TKVEQEAQEAZ9AQBSAT3STMCkICAKCgPcREAXgfUylREFAEBAEfAIBUQA+0U3CpCAgCAgC3kfA4v0iq12Jh08UFOFfxP0rFbiFgYtS4iNFy1a7MSENEgSqBQI+rwBST9sK3b+7nntBa8WU3fasvKK3fs54dGOeCZ/w7isAk7+pTpB58Q2RDWsVe/WSnyAgCAgCVQqBsoWgh+zmFxRZc4s04cmkOCL0nxlxetY/rvsQ42HBZou/ltelmkd9cnr2yULTP0W6RGVKKyya+I8zwtJojp2yPbop7/xwc9s4f+5Hdsxm8TNb/Ew5xU6Q7FPNJnNmftG8o7bPllifGWnvid6xKIkRBAQBQaCCESh3BbBhd27XjzJMoeZiCWkrGpHgP/2uvx080dSol1K59Kt4cs2/rKJd90U1ru3eRbLxweauYeYgNxXAUptLBLaCQlOeqW8Ly+OlSPCcgqI9Rwta1S55gn/kRMEHb6bl5NrrhgruY6lOEBAEBIESESh3BcCM3hJh7h5SLKOzbebEsH9N8HtF+VnMJv7lFJqW+XEr5L9SS+TYIbJ4Zl6OItZssv2zSrGv/IO5mff+mXv07pia0f72aVx8X1DMl4EmORYlMYKAICAIeB2BclcAcIyb6FFtA0efH+HI/dKHir158Hv2m9PLtmBod/t3Mte0xlrktgkoywsqY8ba7HuX5pgCzNd8cfrHMTFhQSLq3e4+IRAEBIFKRKAiFADNK7CVIXALDOzknoHtp3v+MShVMI4XdQ4p6hxSwZVKdYKAICAIeAsBl0zhnlRWLPrTCrPyyigjLbs4m5EDPWUULMmCgCAgCAgCJSNQ7iuAzs2CCyYGl+mXcPKNMZOKTP7u66Pbp5zakV4Y4Kb1ZV5mUeFTNUqGpPTYaSutl3UPLT3975R1e/MSo/0TYtw8z1RmuZJBEBAEBAGvIlDuCgDR74ogLN79dVOIKxysBUUn803BbmqOorwyTFJ6kDnomWYtfHVGxtd78l1RANlWW+Jn6UuuCK8dVcLOsL5kCQsCgoAgUIkIlLsCKO+2peaa1hvYBM50QwGwLjmQapu4O7+tawsNP5SetWjxjrwRnYKNabXyBk3KFwQEAUEABHxeAXw8KqqopE+0nPeuW1vOuQWmdnUD9t8V89SPGc6LVanseswaGTa0a+iBo/nleUDVFV4kjyAgCAgCpSJQ7gpg95H8aSusneoHnNfFmfV82pLM3cdsNwwIS4h1j6X4kg7gl9pcDxLio/wnXe3SB72dmgaFBBbbs9xSMx6wJqSCgCAgCBhBwD1pa6AGLnt4bG3eW0Hm85wSr91f8PKW/Mt7OFMSJRbwzaLMLK6acHP/wJpXdOfQyBILdBIZ7JoJSEl/J+VIkiAgCAgCVQGBclcAxaI5wBzIx75Of0EwEmB2V45T5P/9mTuv+C6gMsq3rzyt0IACsCtk+fbcfUfy2SIu8jNf2DUkwt2daLvi5FEQEAQEgYpFoNwVQHFz/E2bUwp+WW3lm4DQIPPAjv9M82euzELoB/ib96W5f6HbGaReHh7+XME/l825iF6+zcWMzrJ1ahg4fk7W3BTbwqvDRfo7Q0rSBAFBoEoiUO4KoBBDeF7RB4dsk/da1WVwegVw0WyrugwuhuvU8gzs5po6NAkqb2BLW10EB5rfvipy1vqcAe1K/h7YwGcN5d0WKV8QEAQEAQ2BclcAiXGWWcND1SXPnNZhBaDVTWDORWcuhzsTx8XRMRFunuc3mVZuyeHOtX8Vqq+glHC+rUivh0rJZSo8cw3c4fTCNdtzSlw0YP/p3dACD46nSlEbR07aAi2moDMbwqVVIfGCgCAgCFQWAhWhAIbGlVrLBV3/MQcZg+Bomo1N4NIm6aWVmZ3vKLFLyFunVsBFNfzePWx7d2pmCcllRqHOCorqxZfa/DILkAyCgCAgCJQfAuasrKz9+/c3b97cz8/t2Xf5sVV1St6RnL9hb34ON1m7v0PN1UatGgR2LX8jVdWBSzgRBAQBH0JAJqdldFazxAD+lZFJkgUBQUAQ8EEEZNbvg50mLAsCgoAg4A0ERAF4A0UpQxAQBAQBH0RAFIAPdpqwLAgIAoKANxAQBeANFKUMQUAQEAR8EAFRAD7YacKyICAICALeQEAUgDdQlDIEAUFAEPBBBEQB+GCnCcuCgCAgCHgDAVEA3kBRyhAEBAFBwAcREAXgg50mLAsCgoAg4A0ERAF4A0UpQxAQBAQBH0RAFIAPdpqwLAgIAoKANxAQBeANFKUMQUAQEAR8EAFRAD7YacKyICAICALeQEAUgDdQlDIEAUFAEPBBBEQB+GCnCcuCgCAgCHgDAVEA3kBRyhAEBAFBwAcREAXgg50mLAsCgoAg4A0ERAF4A0UpQxAQBAQBH0RAFIAPdpqwLAgIAoKANxAQBeANFKUMQUAQEAR8EAFRAD7YacKyICAICBM4SrgAAEAASURBVALeQEAUgDdQlDIEAUFAEPBBBEQB+GCnCcuCgCAgCHgDAVEA3kBRyhAEBAFBwAcRsOh5Lioq0j9KWBAQBAQBQaAaI+CH0Nf/qnFTpWmCgCAgCAgCegQsSH+z2ZybmxsSEkKAR32yhAUBQUAQEASqKwLmtLQ0q9WamppaUFBAI0UBVNeelnYJAoKAIGCHgBnRb7PZCgsLNUMQOUQN2MEkj4KAICAIVD8ELEz84+PjVcNE7le/DpYWCQKCgCBQGgIWpv+Y/lWyFigtt8QLAoKAICAIVBsE/FAA1aYx0hBBQBAQBAQB1xEQBeA6VpJTEBAEBIFqhYAf27/VqkHSGEFAEBAEBAHXECj+EMy1nJJLEBAEBAFBoFohIHcBVavulMYIAoKAIOA6AqIAXMdKcgoCgoAgUK0QEBNQtepOaYwgIAgIAq4jICsA17GSnIKAICAIVCsERAFUq+6UxggCgoAg4DoCogBcx0pyCgKCgCBQrRAQBVCtulMaIwgIAoKA6wiIAnAdK8kpCAgCgkC1QkAUQLXqTmmMICAICAKuIyAKwHWsJKcgIAgIAtUKAVEA1ao7pTGCgCAgCLiOwFmqAHCDs3fvXucwnT59+tJLL127dq3zbJIqCAgCgoCPInCWKoCAgIA///zTeZ/l5+f/8MMPWVlZzrNJqiAgCAgCPorAWaoA6K2QkBBX+ky8pLmCkuQRBAQBX0TAoALAj9iPP/7YvXv3Nm3avPbaaydPnqTxR44ceeqpp/74449zzz33+uuvz8vLI3LBggUjRozo0KHDE088obe6bNq0aeTIkc2aNRs1atSyZcs07FauXHnRRRcRP2bMmPXr12vxTgI7d+58+OGH27Zte+ONN1K7lvOOO+7YsmXLN998M3jwYMqcP3++Srr//vsJTJw4cdy4cYpJjcRJ4Pjx47S0Y8eO559//rRp0zRPahkZGZMmTWrfvv2gQYPeeOMNq9WqCjlx4sSzzz7bunXroUOHfvrpp7m5uU4KlyRBQBAQBCoBgV27duESwN0fog1eZ8+evXDhwssvv7x///5I0n379hFZt25dBCLinjI//PBDYhDBS5cuHT9+PGEkMvGIbMJTp05FXr/zzjuEf/31V+KR/oS/++474l999VXC6ADnvP31119ke/3111esWPHFF18QnjdvniIh3KpVqwkTJhBDBh4pnyR4Jvzkk08uWrSIzYDSykfik23x4sVkSE1NTUxMvPnmm3n8+eefiX/wwQeJh/zuu+9Ghy1fvvyXX34577zzzjnnHOJzcnLAZOzYsdT4008/NW3a9N577y2tIokXBAQBQaBSEDAZUAA7duxAAq5Zs0ZxjNy/4IILdu/efeDAAeLRCir+4MGDPCJktYY9+uijrAx4nDlzJklaPBNqle3dd9/t27evFv/++++XqQAuueQSJuYayaxZsygZbUQMgccee0wlMWG/9dZbldRWSeTUqEoM6BXAe++916VLF8S6yqkQQOijGKiFFYyKB8zJkyfjZA00iN++fbuKZ7/hk08+Ib7EiiRSEBAEBIFKQcCICUhZZjp16oSM41e/fn3MQY0aNVLeJZl0q3hl8OnTp4965C+mmM8+++zUqVOYa3i8+uqr/+///o9sF198MXKfmG7dujHFxvjz+eefHz58ePTo0ZhWNHLHQHJyMlX7+/svWbLk999/h1bt2e7Zs0dlxmKjAn5+fi1btkxLS9MK0Ww4WkyJAfYA6JjVq1dfc801QUFBKg8zepq/bdu22NhYrF69evV66aWXUAPx8fH33HMPJCyDrrrqKvQiKw/OEZGfVYJsJ5SIsEQKAoJAZSFgRAFgzo6JidGLMyUZEZQ0A3GsGqOs3vpsNWrUICk9PR2dwRyZv8zx0RxDhgzBNEQSUnXjxo1RUVHYberUqXPFFVeoZYQq0PGvEveYWTCyo1qYZbPlwEwfy4zKHBgYqFHpOdEiXQnQEKbwSUlJ+sy1a9fOzs4mhkUGpieUYu/eveH8gw8+IJJTRiwaMAGhmVg6RERETJ8+XU8uYUFAEBAEKh8BAyYgZcDhlKS2ZsEszkRemYD4q+KRfTRPb2RX9pljx45Bi1QlW2ZmJvsBLAXYpEWeEqmK5Qw+82smznfddRdTda0iu0BKSgpVrFq1Sh+v5SeJ9YGWhHEGI756JAnTvJZUYkCZgGgFqSgVdJI+W5MmTT766CNiYFvFHzp0iN0OSmbKr8XDDIBg2iKelupLkLAgIAgIApWLgJEVgDL+IHYRavwQcBdeeCGGHbspNgdgSFVCnwB2eWbBiGDsJEzV1aIhLCwMScr3VohyLEgvvvhijx49yBwZGdm8efPbb78dKawsS0Q6/mrVqoXywF6kTfnZQGYJgv5wzGwX46RYfU66h0eMPMz0NQsSCwIUJ0ebOPjEcVJMT+RhTcC5JgIww4KG+K1bt2J6whx02WWXEe/6iSMyy08QEAQEgXJHwMAKAJmIyQXOpkyZwhSbAKcwiVSW9/3792s6DXFM6ssvvzxjxgwM4oRVqpq5I7vZMWYbgHgyQKUMQZjR58yZo04H2c3utZK1gNppYBeBGT2bzBT15ZdfqlTCqBwtJ8eKsCmpR45ykvr4449r+7paNi3ASoU8v/32GzFkYwODx++///7jjz8mwKEm4tEinCbiESsQCyPysOvArB8dwOEo4r/99ltIUHKsIbSSJSAICAKCQFVAwIwCaNy4MaLK3R/CmpOXbMNy3hHzN+TMu4np2bMn83etNMpnB3Xz5s1du3YdOHAg1nCVxKcDHMdkKs2GMEsK5vsqnr1frC7r1q1jQxjrOfsEWlGlBTiJzzFTauGkJlN1bReao6VsMGitY3cBi5NaYcAq544gZG6u3yfQV4Fth33dFi1asBuh4iFBIbHCGDBggFoGEU8vchyIeIQ+DPfr10/tgmDLoiGcZ2WV07lzZ2DRFy5hQUAQEAQqHQHjCqDSWRcGBAFBQBAQBDxBwOIJccXQstTgtI/dBgNVY37hZI42PTfADF/tYqZnCu9Iy3SeJYuayzumSowgIAgIAtUAAR9QAMho9lrZTbWDGxsLpidPFACWfUw3KBJH7YJuwG4jCsAOc3kUBASB6oSAmICqU29KWwQBQUAQcAMB+2m1G6SSVRAQBAQBQcCXERAF4Mu9J7wLAoKAIOABAqIAPABPSAUBQUAQ8GUERAH4cu8J74KAICAIeICAKAAPwBNSQUAQEAR8GQFRAL7ce8K7ICAICAIeICAKwAPwhFQQEAQEAV9GwM/xGyhfbo7wLggIAoKAIOAqArICcBUpyScICAKCQDVDQBRANetQaY4gIAgIAq4iICYgV5GSfIKAICAIVDMERAFUsw6V5ggCgoAg4CoCXLIpViBXwZJ8goAgIAhUJwTM3LTMQSDlnKw6NUzaIggIAoKAIOAcAQu/EydO4ExRzoM6R0pSBQFBQBCoZghYlM+TqKgoUQDVrGulOYKAICAIOEfATymAEt0iOqeUVEFAEBAEBAGfRkA2gX26+4R5QUAQEASMIyDHQI1jJ5SCgCAgCPg0AqIAfLr7hHlBQBAQBIwjIArAOHZCKQgIAoKATyMgX4H5dPcJ84KAICAIGEdAFIBx7IRSEBAEBAGfRkAUgE93nzAvCAgCgoBxBEQBGMdOKAUBQUAQ8GkERAH4dPcJ84KAICAIGEdAFIBx7IRSEBAEBAGfRkAUgE93nzAvCAgCgoBxBEQBGMdOKAUBQUAQ8GkERAH4dPcJ84KAICAIGEdAFIBx7IRSEBAEBAGfRkAUgE93nzAvCAgCgoBxBEQBGMdOKAUBQUAQ8GkERAH4dPcJ84KAICAIGEdAFIBx7IRSEBAEBAGfRkAUgE93nzAvCAgCgoBxBEQBGMdOKAUBQUAQ8GkERAH4dPcJ84KAICAIGEdAFECp2E2aNOnRRx8tKioqNYckCAKCgCDgywicvQrg888/X7VqlZO+Q/QXFhY6ySBJgoAgIAj4NAJnrwJ4/fXXd+/e7aTzzGazn9/Zi48TZCRJEBAEqgcCFi82Y+XKlV9//XV2dnaHDh1uuOGGkJCQkydPvvLKKyNGjPjhhx9at259zTXXWK3WL774YvXq1ZGRkcT37NlTMQDVtGnTli9fHhQUNHz48MGDB6t4ZPRnn32WkpJSu3bt66+/vkGDBk4Ynjlz5qxZsyjhP//5z4ABA8h54MABZvqXXHLJ999/TyFt27YdM2bMsWPHXnrpJXj78ssvt23bNnbs2PDwcCfFqqTZZ35oBUqmQIvlb+jmzZtHvTabrVu3bjQwICCA/MnJyR9//PH+/ftr1Khx3XXXtWjRoszyJYMgIAgIAhWNQFZW1pYtW5BfWDw8+a1YsQLWEfTr169HUl988cWUdvjwYSLRBy+//PLSpUupBav6yJEj//zzz59++omkOXPmqErHjRs3cODAJUuWoEKInz59OvGIUcKvvfbaH3/88cQTTxAmpjQmIezUqRNsLFiwgJwLFy4k59atWwkjsmFMlYxczs3NpUAyU+natWvz8vJKLHPixImPPPIIViBSv/rqKzTW/PnzUXIooaefflqRoLQoHwVAUvv27W+77Tbi09PTUQMwjJ5D05Bh165dJVYhkYKAICAIVCICJq8ogIKCgj59+iDTVUsyMzORejt27GDSTWDZsmUq/ueff+bx9OnT6hGhySPimBgm+IhRFY/sfuuttwgrGz2LA8JUgRbZvn27ymP399ChQxS1b98+Ff/rr7/ySJj8BFhGqHhWA+3atVPhfv36TZ06VYVL/KsUAEmq8I0bN6psR44cocxNmzbxeOutt6JFVPzOnTsfeughwn/99RcZjh8/ruLfe+89DQEVI38FAUFAEKgKCHjHBISgZ/KOmQXBxy8sLOzUqVMREREYW3hEuJ+JNrE4uOeee5hKq8cuXboQ2Lt3b/PmzVkZYPn55JNPunfvzoSdH0lNmzZl9XDHHXdceumlvXr1wlajCB3/UjKRv//+e3BwMFYaytTniYmJUY+hoaFKOvPIcsTFPV7WK+ecc07jxo1VIYmJiQRQLRi1Lr/88kGDBtFS2Ovateurr75KEs0ZMmTIjWd+xKMkFKH8FQQEAUGgSiHgnU1OZvF2rYqOjvb390fFEa/JWR6RlVpOhDXhEydO8Peuu+5iTo2FB3M58djuiURwsytw7bXXsg0QGxs7YcIEx4pUaaw5atasib5hEYD0p+rnnnuORYNKVWyosIG/2Ihq1aqluFXknTt3VpxgtsLQD2/oLSpFA5EhMDBwxowZqDQMXAkJCffddx/LLAP1CokgIAgIAuWKgHdWAEz54TItLY09T8UuC4I2bdrYnaJBK2zYsEFrjzKtMLNGPrIZi1GeOTVy86mnnho2bBhTdYwq7LUiZPlheGEl0bt3b2xNWglaoFmzZkh/1gqKEy3eTvTrHwmzVtByOgmgkxDobBrHxcWpbOwcREVFodhYHMAzO9782A9g4YLWYT3ENgBzf36sCVjEsNZBjTmpQpIEAUFAEKh4BLyzAmCCjCn8scceY7JMGzB59+3blymzXuASjyXn22+/RZgSRnqyQXrnnXcyR2b+johUOwQkoSewzxDAdt+oUSMm9YTVrzSRjYxmDxZru6qRrVqY+R9Ryf9jiVJbFCUn62KxRHEq6dNPP1Vxb7/9NgFkPertwQcfpBVqNaCOErEOYOu7VatWqDSywTBHj+xw0JUtQUFAEBAEKg8Br2wCI+AQ4vfff7/WDo7fEHnw4EFi9GdgmDtrea688sr8/Hyy8eNYjhbPTJ/FAZEoicmTJ2vxmIC0/GeI/vUHPYE5XsvM6oHkzZs3E6Ptx3Luk0dFpg4L8cjC5V8F/e8BI9Itt9wCD0TAD2Z9rXDtMBIBfaWcdFLUbGZomYEFkP9XqvwvCAgCgkBVQcCMbMKKzb6lnblGk19uBVAD/DDHq9JoJSYRzDj6mTuze8w1HJTU7EWqCkRtamoqf+Pj4/XMsKpgqo6dXU2xn3/+eWxEeq44V8r0nA1eqkOaUz6bB8zEyaMYUGfzeVQbv9qj2n5gOxfLkr5AwhxhatKkCfm18/4UxWmljIwM9JOePeKJxObDUkbLTAk0HPWApUjb9LarQh4FAUFAEKhcBLysACqgMYh41gH6ipD1SHy9jtGnlhnOyclBfOvJkekUqOmJMkuQDIKAICAI+CIC3tkErsiWs0Pg3erYq+Dn3TKlNEFAEBAEqj4C3tkErvrtFA4FAUFAEBAE7BAQBWAHiDwKAoKAIHC2ICAK4GzpaWmnICAICAJ2CIgCsANEHgUBQUAQOFsQEAVwtvS0tFMQEAQEATsERAHYASKPgoAgIAicLQiIAjhbelraKQgIAoKAHQKiAOwAkUdBQBAQBM4WBEQBnC09Le0UBAQBQcAOAVEAdoDIoyAgCAgCZwsCogDOlp6WdgoCgoAgYIdANVcA69at4643uza7/ohjMjwSu57fVli0JyXjwPG//X/hxIaLQp2T//j7jpc/Xv7N/GLnAXY/7kzFs6ZdpOuP0FKC6/ntcsK58n5sF+/iI7gpt24u5rfLxl2qeIOwi3T9kTtl9+zZw6V+rpPoc0KIXznugtVHuhXmht3SvNe5Ug73qFutVldylpiHa2i537DEJFciuT7dw4EH/q5UJHkqHYFqrgCOHj3qyWuMEHSL3N/PnFdQeDIzNyevWHZw1bbdxaWO/d2wTsyjcw9d+fWWg8cz7VKRoZ4IEWjd0l52tcM5/NtFuv4IbmUqPyelIUG4fNtJBudJSHDPyT2RYshfT8i5Ux0V6LyNTlJRHmUOPCfkzJk8mTZ5MmidcCVJ5YFANVcAXNyvv+fZXQQN0EaHBVLlaWuxZzRXyNs2jn+of21Tnu2LX+0XAa6QO2+RhyWcteQ0XP2cw+skFXInqWUm+XTtZbZOMlQdBKq5Aqh4oKPCAqk0I9vV6Zuf2XT9+S1MhabHf92XfMr4qr/iWyo1CgKCgK8jIArAyz0YEuiPIciaW2Bz2f7cpmHcvT0STIVFH83e4mVupDhBQBAQBEpHQBRA6dgYTcEKBGm6NZ/ZvSs/co0e2tqUV/TUrN1HT7ux5+xK4ZJHEBAEBIHSEBAFUBoyxuOjQgOQ6SiAQtcXAQ1i7+yZaAry+2zOVuMVC6UgIAgIAu4gIArAHbRcyxsWHMDeM1YgTgS5vhd467BWprzCsbP3HE2TRYBrQEsuQUAQ8AwBUQCe4VcSNUI/IsTCNwHWXA6DumYGMpnaNoy7qWstk7moxG8CSqpH4gQBQUAQ8AgBUQAewVcacWTImbNAOe59SXTvJW1YBNw7Z2+KHAcqDVmJFwQEAe8hIArAe1jqSgoPsXAWKLegiHWALrqMYLtGNUa2r2HKt33/+84yskqyICAICAIeIyAKwGMISyoA8c95UI6C5uS79038YyPaUt6dc/YcO52DCimpbIkTBAQBQcA7CIgC8A6OjqVwFojIjOx8xyQnMZ2a1RzZItaUU/DT4l18SewkpyQJAoKAIOAhAqIAPASwVPLwkACLvzkzxy0jUHFpYy9tbfI3P/vr3oPHMjhNVGoFkiAICAKCgGcIiALwDL/SqQP8zUEBfrn5NnUxXOkZ7VO6tki4rGn04YzcGSuPiAKwR0eeBQFBwHsIiALwHpYOJUUEW9gCTj9zMZxDorOIsZe24fzohHVpqemu3inkrDhJEwQEAUGgJAREAZSEipfiwoKKZ/BcDOfGSaAzVXdukXBjqxhTbuGcNUe8xIsUIwgIAoKAPQKiAOwR8eIzViALHgJshTl57k3kMfzfNrQli4Dxvx36c/tRL7IkRQkCgoAgoCEgCkCDwvsBpv/hwf6FhUUZVvcUAKx0bpH4bP+kU9kFgyavWLfTuGMv77dKShQEBIHqgkD1VwD4hDHcWdB6Qm42+0WFBXE9dGau2wrA3880anD9h7rFpeUW9p+4fMOu4+62wkPmIffEq4mHtfuDmscdZ5h/CD3kv3LJFf/uDhgtv+fMa0VJoIojYKni/LnLHm5sT548qd58xjGe7Xbu3BkaGmrAPSzkeHbNy8sLCwszSr4vPCLqdEGwzVaYmxaMRcj15lB7SvKRIa0Ds6xh727K6Pn68lmjmjWuFcQFc64UAgJ4dsUvIPwb8E0IOS4V09LSCBgjx50knmkDAwMNkNNAOMcvbkREhDFyHFLikTgkJESNBFcQ0+ehu6k9ICDAYjH4glA7hQQHBxsYOfDMwMOdZ3h4uDFymOctiI6ONkDOwKPjUMB49DRADvP4Ya1Ro4YeTwlXWQQMju8q257IyEiEjnrtGcS49o6NjTUmwSFHhkJu7D3kRUKAxkFuCjmZkRsYGlQjIqjQZTflkCND44ODxt/cJva7zS8sOTLw/c1L7u7QsWnNvPyyrxiCHKeySP+4uDgDMhQAKYFeNkyO7EOEIQjccqqsxhW1wzngGyan7ceOHYOcotyVYpCAGFKMtqMD3CVXTVAjBw1kgBwGEN8MPPSfAXI6DtnNi2Cs7yCn49B8oGdg5EDuiTviKitYqitj1U0BIKz5ab3FOOZF4j3UYtwK8BbFxMSgP9yi0jJH8YuOrhESUWDO8A8KiIp2rxxmcCizGjFRT47qnlOw/LWVKX3e2bTr2QGNk2K0KpwEWP0gRmHBSR4nSYge3n/D5IhOHKMDoJMqnCQhwaE1TA7n0NJ3TqpwngQ5I4dJgPNspaWq2tGCpWVwHq/I9SPZeX671OKBd+ZnF+/iIy7peXEowMX8dtk8cShvV5Q8ljcCxu3j5c2Zt8o3MIfSqobWY/JCLgXiLFBWTn6BOxfDwYNWO/4FXhjd89YONU35BU2eWrDrcJrGoZOARu4kj5OkyiVHgnuMfHELnDTQSZJqu2FySlYlOKnCeZKvkztvnaRWHQSqvwKoXKwRQZwFCgny50aILDfvBdJzHmTxe+uevpc3jzEVFjZ95rfdR07rUyUsCAgCgoABBEQBGADNbZLo0GL3AKfd/yRYX5PFz/Tlo4OuaBZtKrA1eWbhzkMurQP0JUhYEBAEBAE9AqIA9GiUV5iL4ThTme7+1wB2DFnMpk8fHjSqVSy2oM4TFm0/eMougzwKAoKAIOA6AqIAXMfKeE6uBQ3li7Cioswc926HdqwyyGJ++96+o9vWyMjMu+S1pdsPiA5wBEliBAFBwCUERAG4BJPnmZSTyNNZnioAOAkLsrx2R+9b28dvO5lz1aRlsg7wvHekBEHg7ERAFEAF9XvkGf8wHm4DaLxGhgS8dGvPWzvGrzuRfc2kZbIfoCEjAUFAEHAdAVEArmPlUc5AS7GTyPyCInfdA5RWa3RY4IQxPUa3j197PPuy15fsSZZzQaVBJfGCgCBQMgKiAErGpTxiWQTg5NFbiwA4jAkPeuXWHqPaxG1IzTlvwqIDxzLKg20pUxAQBKorAqIAKq5nI0MD+Swg3eqFbQCNaXTApDt6X9YsZnd6XrfxCw+nZmlJEhAEBAFBwDkCogCc4+PNVExAGIKseTYXL3Rzse7o8MAP7+97fv2Io9b83s8uOCI6wEXgJJsgcNYjIAqgQodAsRXIZMr04JPgEtmNCg2c/tig9nGh+zLz+o9fkHzSWmI2iRQEBAFBQI+AKAA9GuUeZueWOtKy8rxeU3iwZfkL5zeJDt6ZkTdw/IKjaTkBfDosP0FAEBAESkegMm8D/eabb6xWK3cHtmjRYsCAAXZMch/9m2++2axZs5SUlL59+3bo0OHEiRNfffVVUFAQObmmkUvPhw8f3rlzZzvCqvwYFsy9cKasXBsXwxWHvPoLDfT/88ULLnj21+UpWZe/vPC1a5onxRRjJT9BQBAQBEpEoNIUwNSpU3v16lW/fn3YWrNmzeLFi5HyehZfffXVcePGKY8c06ZNSzjzu/vuu7U87777bqdOnbRHXwlEhAbwORhWILUa8C7bEcGWWU8MvPKV3+cdzLjt/zZOGdM+qbZ3a5DSBAFBoPogUDlWAm67ZTqvpD9YdunSBc8tes8h+JRgyq+kPxm6deuGcxU96lOmTLn55ptxnaGP9Imwkvuny8EKpJrPuaAvH+o3rEHk2pMFT321/YjsB/jEsBAmBYHKQKByFEBqamrbtm317cX3BZFaDH6Fhg4dqj1u2bIF7xza49atW9u1a4ezFC2mnAIoGA91jCN5BBfD+Zk5C8SN987ZPlO5EQ1XIzL4/+7r859GUbMOZt751tKUU0b2hB05d86tXWrlkitmPOHBE1pqh9yTEjyh9bztin9VjoG/njNvoFIhMYZA5ZiA8Hhn5ywJF0h6T3KY+DU3XqtWrWIpULNmTa2F33777TPPPKM96gMUgjMpFQMVfkXYYzDm1wkllJubC7m2ENFXVGYYchxyUQJ/9a5FsPz7m2xZ1oKT6f7c6FCaEuAtghZCO/Iy61UZ4sIDnrms8aEPN/y4N8PvrWVv3dEjPjIQnwQukqvaqRo8wdBFKn021XDD5FTtSe3KHSZ9Z0wY0WRqhxy/ZvpGuR5WI8dY7Qp8SmC7Sz9yXKwdcsU8f42RU7VCwBg5uLnIqmSrdAQqRwHg7JStXX3j2Q12FNOMYGR9gwYNzj33XC3zrl27evTooT3aBXAKjzNYhC/xSoLv2bPHsFP4Q4cOIUoMk+MZnK1stJ3+RcJqlWa1HUvPz0j1rxWFy1m7Fvz9yGuMb24EEP759OQl53aIhTzr1MmxA8I/+qPg+z2nj7+8cMJl9WpEBBbYSqnv3yVADuf8CBhQAFDRoXQxIswAObwA++HDh5kWGCPHnAi5YV+eAE7Xs8R0HJP/xqnUJ8hJUwcWSs1USgLoQY4YZVlsrOshhxarqTFy/CHTcMMDD/K6deuW0jiJrloIVI4CYDo/d+5c/a4vgxUXuHpsTp069dFHH11zzTWJiYn6+HXr1vXr108fow/Xq1dPG3woAHxzc8TImCBQso8CjblmhZylA5YrfHPbvYe5+badyRkB/uYmiZH+pZwFghy2kUE0345c397SwpCjP+rVz+/bu8bIVxYtScl6dc7xTx/sFxXG18hl6wDIscihupo2bepKfkc2EB+s4Vq2bGmMHN2PEDdMjv5Q5DTEkbcyY9A6lMDIMbYCoMmU0LhxY20VW2aNdhlgOykpCf1nF+/KI7QoHjz64tXdAPiQMz9j6BoeeJ64YnalgZLHiwhUzh4A84uMjIysrL/vLeBAJ2MdoxCv/Y4dO1TzPvzww4ceeshO+jOgN23apDcH2WHB8EXuq59KIkykgR/kUHlITgmqHD0DwYGWkEALF8PxSbA+Xh/WqLSAPrXMsKLCeFMzJvS7R/o3jw3+aU/a6MlLMnMKyqQlA+Sq4cZqp4SqQG647xShh+Su4Owkj+HatS7TAk5qcUzSqLSAYx4nMVDBOX/l5xMIVM4KAGjuuOOOiRMntmnTBqGPrWDMmDFEYjdA7nMAFCnPMaEZM2aojQFmc63P/MhMnopE1sAcyhX2IkIs1twC7gUKDSr3LoiPDlk+bnDc2LnTd5y0TFo85f6+eBQok0kPG1655Kp18ICoKrOlksG7CHjY9d5lRkpzjkDZgsA5veFUFgFM8DF0stjkp8rBCoT0J4xieOWVVxwLZ22rMjgm+VYMlzccP52bkZ1fKzqkAmRUbETwqVcviLlr9tfbUi2Tl3xwX18uJvItxIRbQUAQ8DoClbxYw+yjSX+vt82VAq3Jh/JOn3Qlp3fzBBdfDGfOzS9kP8C7JZdWWnR40PE3LjSFBH6x+cStkxez/igtp8QLAoLAWYJAJSuAykU591TqstHDFl83+PSevzceKpIfPgjAS3C6ty+Gc9IEvg84/PzgetHBn2868cC7y0QHOMFKkgSBswGBs1oBZJ88UZiVnLV+3bKrmx/6fW4F9zfuATD+ZHjVPUCZTUiKC1v8eL/WcSHv/3ns0Q9XZOdV0PqjTMYkgyAgCFQ8Ame1Aohu3LzX56tih1+Vv9+07sYLNr77WpGt4gwjWIEC/P1y8m15FWUFUsOrfq3Inx7s3bFmyJt/pDwxZUVugeiAin/vpEZBoEogcFYrAHogvG6DLk+90vCxF0wBpj3PPrTi0TusKYcqpmf4AoA7nPk6l6OZFVOjVkvj2tFf3durfXzIxJXJT05Z6eLXYRq5BAQBQaB6IHC2KwB6MaRW7SYjrm75/MeBzeoc//bDRSOanNzyV8X0Lv5hqMi7TiJd5Lx53Ziv7unZMi74v8uPPDZlednfhrlYrmQTBAQB30FAFEDx517BtWonduvV6eXPInsNzjucu/Sc9ntnfEsn8jlTuXZlaFAAbluseQX5NiP37XjIW8v6sdPu7ZkUFfTfZUfGvr/cw9KEXBAQBHwOgfIVcL4Ch58lAB0QEl+rw2MTEkfdVxRr2nDxFRsmjrdZs8z+5Xhe3uJvxosLFhhrhVuBVNe0bhA394FeptCA/y499MRHK32lv4RPQUAQ8AoCogD+htE/JDS4RgKaoPWNt7d+eYo5ybTvtWcOPH599oG9XgG6tEKUFeh0xZ4F0jPTpmHcX2N7myz+L/5+4OlPVumTJCwICALVGwFRAP/0b2B0bGBsfH62NaF7377zNwQ1CCzYtWXl4NbJKxb9k8nbofBgjgKZM3Py+SbA22W7Wl7bRjU2PNXPlGd6bsH+cZ/+4SqZ5BMEBAEfR0AUwL86MKRWkiU0vCDbGhAcfN78U5EXXm87aVrVs/9fb7xoKiyX45LsAXArA2eBKvezrHaNaqx9YQAXeT37676nPynWAVyh9y9o5EEQEASqHQKiAOy7NKxuI+z+BTnZeadSE+96vOH7U/w4Ifr8E0vuuCo39bh9bm8880UYs38cBXujMONldGoS/8eTfUyB/qwDXvhija34JjXjpQmlICAIVH0ERAHY9xEnfyLqN+EWyfz0tJxjKU0vv77H6tVBjYNO/jjtt0uaHFvj/dMy3AyKqK3IOyHs2/y/567Nay1/sKcp2P/JBfsmfLWBS53/lyL/CwKCQDVEQN7wEjrVLyg4JKE2Jnnb6dS8U8fj23fpP/NAjatuyF6avnJorz0/fm0q8uapzUCLP1aggoLC7LyK/iLMsfE9Wicuvae7KcDvtTWpn/522DGDxAgCgkC1QaD6KwBjN8IHRseF1KiJa4uco0eK8nIJ93r1g6ZT3jTnmDZef9Wa5x8tsP7tzcbJUKBqV5xjMP2PCgvic4T07H8pAGhdIS+NAcPkvdomLb27e0JowPhlx179am1p5TuPp3ZjyKtivUJumAHVcYbJaYJh8L3SfMW/8w5ykuohOW13UrgkVSkEKs0fQDmhgENg5cmW8nE5gHtrHPMaceqL7MYtYqbVfPhQemZWUFJ9c0BA9PkXJ7wXceThGw69/mrWgd11b3k4sHa9QpvNVNIBHl4DHBTjvgbfZ85dZPC+ZefaUlKzT53wy68VWnjGdTvkyi0tPnAM+MWFHCigBQd3yeGnYaz5hcE1bp6VPHbeXnzy3DC4ro196pKaWWI/UgLu3vBJiXNBd2unQMipFDdB+PXFF1CJVTiJhBwqyI35JoQcnmEel5y4hHS91XqWIIeW284NkMMAI4cmGPMJrAYeo86YU19FziXthgceyOONUo+GhKssAtVNAfDWaY5YUQCMZtzq8jPQAdAG1aodUJjrX2QrTE0JqdOQQhqcPzyh5ertz92fNuP708u+r/fs50n9zysWjA7CUasaJzZlSoGAQFOYtTC3oLDIbAkKKp5AIQVgm+bwt0xyx9ZBDi3xhsn7ta35dk7eA7+nPfHboVOZ1rGXtSly56QqEhwGaLsBBaCar9pujBzpqcjBwRGcMmOoVDFv2FmFqt3YwIM9T8hV11O1KwPPEQpFTsMNjxzDrXZkRmLKG4HqpgCY9OnnfbxICQkJmkpwF83UkwlJcbF+accLbQVB/qagGrWKS4iNTfpi7poJT6S8Pengg9cFv/B+8ytGmQODHAtPT0/HI7yds3vHbCqmMCA7NSM3LCo0NvxvdcXyhXfJiQPk0orS4inBMDlVD+nu37JF04ETV/x35cn6iSl3/aedVnKZAZY+1B4fH19mzhIz4JMdn6CGyZHgx48fp+tLLNyVSEXOHMKVzI55YJ7aEcGOSa7EsIqtVasWKwBXMjvmYfnFqNO/CI55nMSgvFEAhkeOk5IlqaohUP2tdQamz1onYd7xDw0Pik9k8pt9PCXv9CmVxGfD54yf2OLtT0yFpm033brs4dFakkZLgKpdn8BGhlqYYadl5molQM5Pe3Q3QNUekmP2GdCxzty7u5kKCu+etu3tH924I8/z2j1hvhi4Mz93QVP5/0dtHHxKcL3rHZn0ED0Pa1fNd+TKxRhPGu5iFZLNWwhUfwXgIVJFhbaAqJjAGGyaRdnJBwrz/hHQza8c1WP2moCWphPffPHbhbEnNqz2pK6wvz8JLlB7AJ4U5V3a87rW//nebqb8wru+3fbuTxu9W7iUJggIApWIgCgAl8DnVGhAeBTT8Yx9O/Xm/hrtOw/67WjcJZdl/2Fa0qHbru+/dKm4kjJhq+aDAFKqwgcBdgwOOafBjAfOMdmK7vh6i+gAO3DkURDwXQREAbjad6FJ9fwsliKbLWP/Tj1NcFzNPu992+TtidgLNl117cqxYwpzrPoMrodjwoqt/6ezio/uVLXf8B4Nf2Id4G++49ut7/3khi2oqjVE+BEEBAENAVEAGhRlBLgfIqxeY85I2LKtOceS7XK3ueW+nquW+oebjn45ZeGIblmHD9hlcOUxIqRYAeAgzJ3jNq4U7J08F/Vs+NPtXUwWv9unb/9wptiCvIOqlCIIVCICogDcAN8/KDi0TgNTkSkn9ajjrm9Ct14D1u8N7dIlc/nmRUPrH1k0D53h1m0KHFnESSTiP6uS3AOUicVFvRr9OKYj64Bbpm/7aNamMvNLBkFAEKjKCIgCcK93AsIjg2KLjzZmJx+05WTbEeNheODXi+o+8njeBtPaUecfmjqlKC/HLo/zx6iwQDYDKtE9gHP2SL24d+Mfbu7AZaGjp297/6e/cGpfJolkEAQEgaqJgCgAt/uleEM4LJKjctbD+4sK/nV5A2VxQrTzoy+0mva52d90+u0Xt7z8RLaDvchJlREhASiArOxKvhnUCYckXdKnyfc3djAVFt323fZrX164ae8J5/klVRAQBKomAqIAjPQLhiC/gABbXk7WkQP6Q0FaWc1GXttj5oagDu1TP/2/Zbf+58TGP7Uk54FAi19QoD+fBOMhxtBHrM6L91rqf/o2WfFI70uaRH23/VTb8Yte//bPyvVn4LWGSUGCwNmEgCgAI73NHW+hteub/fwLstL5QKzEIuJatWs4/v2IESOzVqxaMaTzru+nlpjNMTIqtPgKh3Rrvtlk5BoDxwLLKaZ768TPHx7w3uUtTX6mB2fu6vPk3JVbko1/OlVOXEqxgoAgUDoCogBKx8ZpiiUkLDSxLnu8uanHHDeEFal/bM32L7zddMJbtqOmTSOuWfvSEwVZmU5LLU7EPwz30GVk5/NJWJmZKzdDeEjArcPbbhs3aFiT6D+PWnu8sHT8J6uOpRVvjRi7hKdymyO1CwJnGwKiAIz3eEBkdGBM3JkvhA9yNtSxIL4i9rMEtrrxzq7z51nqmA5OenHhsIbZx4865tTH4BvA4m9mczW3aluBNJ6b143+8enzprEzbDGPW3ig1qNzf11zgCtN8XWs5ZGAICAIVEEERAF41CkhXBcaFsGGcNbhfVz+4lgWV8IQWbvfuQPXHA5uEmDdduLXmglHlixwzKmPiQzlFuLiDwKquBVI49nfbBrZr2nypAtv7lDTlJ533usrb5q09FiGHBDSEJKAIFAVEahMBZCTk7NixYq1a9dy+2CJ2OzatWvhwoVk0FK5JXH16tXE8Fu1alVVuHYKH8L+XHqcn5+5f5fGp2MgtFbS+Qsz6t3zKEJxZd/Bm95/3YlbsajQQHaAs3KL7+B3LKrKxiTEhE55oN9vD/cMjwuevjO97zs7Pv1lS0EJarHKtkAYEwTOLgQqTQFw3e7EiRMbNWrEpbUffPDByZMn7YCfOXPm7t27e/ToERkZ+eqrryol8fvvv+NkI/bMj5uWq4Sh2WwOxV0MXwjnZFuTD9q1Qv/IldGdHpvQ7rsvuWJ419gHl959XfbRI/oMWjgkyN/i54eDSJsPSs/+HescfmXIi+fyxVzhDVM3D3rql/W7jmtNk4AgIAhUHQQqTQFMnjz5scce49JzdMCoUaN++eUXO1CQ/ueffz5X+Tdt2vTCCy/cv38/Gbhivm3btg3P/Jo0KXbdbkdVKY/+wSEcCqLqvLSTeaftNZkdS40vvbrv1o2hPVqc+Gzq4mv7pW7ZYJeBRzaBw4KLb4dmEeCYWvVjIkMC7hvRetpVdS9uGrN4X3rH5xf995s/sWhVfc6FQ0HgrEKgchQAxp+uXbtqQOP4AhdOmHe0GNyJ9O/fX3vEuxYeKqDC2o6bwI0bN2Id0lLLNeCijgmIiAqpmWAuvjKaL4T/2RAukTymRZu+Xy6qedutOX/sWjGyw87pXzg2gcOg0Np5CXbM5iSmxKqd5LdL8pCcS63bNoia9sTg/7u2NdcHPTxz16XPz1+84ZBdLaU9eli7KtaTQjyhpXbIPSnBE1rP215ap7gY7znzLlYk2TxHoPj+4Yr/HT16FAOOvl4eEfpRUVEqEi++HTp0UGGMP8uXL7/22mvJsGzZspYtWyYmJuLpd+rUqVdffbW+EMJ4ktI2BtAZKAzIiSRgl7PMR7QOhK6T+0XGFp1O46Dn6b27whs19w8IcEJuianR+Zn/bm/Rdv/4u7bcfF3G3p2tbrnfEhHJbaOKsQA/jtcXpmXm7DicVjs2JCjAz60W8BIqztGsGiBlNlnL4BXy3Nw8i6lw1PktujaPf2XaX59vSft14srxA+uNvrBlrZiQYnc1Wn0OAcU8nMO/Q2LZEVABPl5tjQkj6oUBfgaGjWIOWmpXI7Bsdv+dQ4FPCcbGrYfkDHuqhiPDtStP1P9ukzxVUQQqRwEwwhhnekjwvVfiy4aq+OGHHy699FK865Hn6aefVv6mcRaI2+tt27a1aNFCXw7etNldUK89VbBoYK1gxCk8dhg/v4MHD8KqG+RFRXnJyUX5eX7JyUGJ9fYfPJiWlpaamlpi0/iGILDX4JrPf5T69m0HXxx/fNncmnc+E9qomS232OeMn5+5ID3lZI7Zas3aaTbVigoMD/J3KjP1MBTPQPFKCPMYzUqs/V+5HR4gZ0Gm1mTG9AfaGufg+JWEPNDf7+FhtbvWD5649MTT8/Z9tTb5vj41eraM4bPnglJ2OZA+hw4dMuZTntagACCn4xxa5lIEiEEO84ZdQjJyKMSYS0jAhzw7O9uYU3jIYZ5NNX7Gup6Xjobzfhkmr127tktAS6bKRqByFADuRjHx69vOYHV8WzgjtHXrVnYIlFNfZJl+3cBSYM2aNXYKICkpSfMEyyBmKLNVEBYWZmAoowCYy9SrV8+N99BsLmzYIGP3Nj4OCI4MM9WrHxsbw461k9rNTZta+w3688k7T8+afWTnhW3e/bV2r/7MjdkGCMLPsL+lKDDqtDWP86DhkYEJMSHOps06QJECSE/4r1+/vjEJfkaAnGQDxhh5ZmYmDpk1cvZq2rZqNrx/xvszt7607PCts1NuPZA79ooOTRKjKN+xUXBOdzdr1szwCgByw7tEsKTImcLrQHUjCP9sbrkxddCVTd/RakYyxx+cjBwdxb+CkDPyWUzzshggZ9hzzkI50zZATu2GXRn/qxnyUCEIGBzfHvLGyOYQJ3u8qhzGGXNVJdwJM4aInzt3LgPxpptu0upatGgREoG3WsUg3B2dhtu9sRTFULaL1AosMwCh2+QWi1/9RtYjB4qy0k3WnICEWmXWHlm3fu93vtrQ4qVDz724vt+5mVPeaH3TXczh/S0BQUGBtWpFHU3LPnY692SWrcicV7dGWJlsqwxwjiBDFvBzkUSfTTXcu+QNEqInjOlxbucDY6f+9f761Pe3LfrispbXnPuvNZzGA7ghjPhpMa4HoIKcKbzrJPqcDEIjXa8rAvSovcyu11H8Kwg5P8PknjNPCfz+xZPLD4Zhd7kGyeg1BIy8XV6pfPjw4fPnz6coXravv/76nHPOIZyenv7OO+8QOHXqFMd++vXrx1RI/cjWp0+fGTNmMLUkA9aJ2bNnd+vWjXC5/qjX3fIDImOC4mpBmJ+aUljSF8KOBVrCIjo/9kLbb77h2OeO0fesGHtLflqqn+Vvs1it6JB6NUL5sPZUZt72w6fz8l06HGqAcz1j5Uc+sFO9X8ed9/KQRqas/Gs/+6vX43P2JP+z/6948LB2zwvxkAHIPSnBE1rP264fBgbCnjNvoFIhMYaAQSVvrDI9VceOHTHvTJ8+nYlGr169sLSQihWIMAEW4KygWQQoEmayjRs3Zvp/2223sQ5ITk7GyPjII4/oC6xS4eD4hGLHkEVF2Uf2m2olYNF3hb1Gwy+P3th05cWdjn46ZeGKqUlPf5vQtacixE8AB0N3HslA+u84kl6/ZhgXR7tSZtXMExMeOPaqzhd2bzD0jRXLD2Y0vueXyVc2v+2S9oF8Uiw/QUAQqCgEKk0B0ECM+Pz0LUUBqMM/2PE1U74+A0bVIUOG6GOqbDiUL4R37y4syM86tBdfki7yGdum4/mbspbfd33qlO/2XDys6L1340fdyqlCyC3+fi3rRu07lslFoXuPZtaIDEqKNbjJ6SIz5Z2tTcO4Pa8P+3zulhs/3nTvjzvvnbd3zQM9OzevVd71SvmCgCCgEHBpZipgGUMgMLGeyexXYM1iS8D1EnAp0+fdr1t8+oHZz7T3kdvnX9QpZdVSjbxBzfCk2BAUwon0XNQAl0VoSb4YYMZ/wwWt9r954eWtYk1ZBV2eXTTu45UnM3IDAypzauKLSArPgoABBEQBGADNZRKLJahmEhvR3Bedl5bqMhmHQC0trh3TfMac4F69MmeuXzWkz7L7bji9e7sqoUZkcMOE8ACLH1dGYw7KyvX5L2zr1Qz/5rFBP93WKTQm6NkFB7o+MXf2sl3Gtn/dAFmyCgJnPQKiAMpzCBQVBUbHsh9QfGX00SMF1rKdAei5CW3Rrs1rn3eaOyO4bdyxyZ8uPq/FX2+9nHuqWJGEBwc0SYxgVwCnAfuOZjJl1hP6aPii3o33vXDekwPq7UnLHfbBhtd/Tt6Xku6jbRG2BQGfQEAUQPl2E3dEB8bGcy4I3wBZB/c6+hB2Un2RrcASElrvvOEDZu5q+sFkU55p992P/nZJ+72zv+fiuQB/v4a1ImpEBHGK/vBJ68ETWU6K8pWk+OiQ8Td3X/xgj/PqR3y2LbPZuAVfz99W4ONmLl8BX/g8CxEQBVARnR6aWMc/KARlwIawW/WpE3WBkdFtxtzTd/GOxCfuzl16eMOwEUvvHZWyerlfUWFSXGiduDCsTJwQ3Z2SkV/Kh7VuVVq5mdne6NO+zpcP9nmgU3i+zXTVRxtu+e9vW/eXccVe5fIstQsCPoqAKICK6Di8B4fVbYgn4YLsLOdXRjvhJqph0+7PTe7+x8qoa4ec+mDaqu69Vo17MOPAXo5UNk6M5LKgrJyCnWwJ5JTsXMFJyVUwKTo8cNSAhA2P9R3ZJu7jDSdaPb3w49mbrHKfaBXsKmHJlxEQBVBBvecXEBhWuwGV5Z06wa3RBms1m2t1PqfflO/afPdlYO+k5AmTfu/WaPPH71isac1rR+Gh12Yr2ns0K/XMlkDVuCrbYEO5HoK2tGsU9+UjAz+6urUpwO+mL7dc+sL8P7eX4VDTYH1CJgiclQiIAqi4breERwTXSOCOh+yUQyX6EHaRFb+gkCaXXj3wxw2NJr1s8jftuOnO+QPi9s7+rlGt8PjYMI6FHk61HkrNxkNlNfiqigvjbhraeuf4Qde0rzF3X0bn8YsmfLE6s1qsclzsbskmCJQfAqIAyg/bEkrmRFBAWPENX5kH9/CNWAk5XI4KiqnR7s6x/ZbsiL/jipx1pvXDRs4/v17woS0Nk6K5wCctK2/vcatrd0a4XGXlZWySFPX5IwNn3dnFFOT/+C97I+6fvfBPZ87XKo9TqVkQ8CUERAFUdG+F1WuEOQgDh/XQPs/rjmzUtPfbX/detyqstylj3sGFrTtufnJMUtGpsIjQPJtpzzFrRrbPfyWgUGI1M7R7w+OThj3Sr67pVO6gV5df/+pvJ9JzPMdQShAEzloERAFUQteHN2jy94bwEe9MY+M7dDtvsa3dD18F1jQlT5qytE2DzG/eDMlLN1sCDpywppzKroRGlk+VXIDx0pgeK5/p1ywx7PMNx+Nv/enb33aUT1VSqiBQ/REQBVAJfexnCQir05DNgLzTqbmpXnKYbvZrfMmV5+042fiNiUUZpj1PPnjigR5+a38zF9lOZBXsScmsTnc0ntMyYd2EC974TzMuSLrio/UXPTtvr8N9opXQr1KlIOBrCIgCqJwes4SFB9dI5Ja3nBMpeJH0FhOBUTFtb72v3+7tNUffbtuakzVhzN6nRqdtXmfNs+1IycrGKlRdfqFBlrtHdNg+YdCwptEzt6c1emL+Z3M2W33/Vozq0j/SDt9AoPorAOVexlhvQOvJjTTOyYNr1AyOiTNzZXTyQbxIOnIIOT/H+DJjoho16/HSO62mzwnsOrBw3vcn7uu3463n0g7u33cyL83q6pYADTdWu2KvYsib1Yn58enzvrixfe2wgFFfbL5qwsK/dh2DAf8zzBvmH0IP+VcllNlTpWWo3Nphnl9pvJUZD/Nl5pEMVQSB6nbnIo4Ms7Ky1PDlMAyOxnAeYNglJG51uaHaMDneifFti2+Dks0vZnNRoV92ekZRXk5qWlowRiHdj7cIctxCEVcyuS6zYxDy3KQGUXc+5Tfo4qPv32v7+LVtC/8v8rrnj57Tr2bNmNgwbuF0VioA4s0YtzzGvPJCTi+A3rFjx/Dl4Mie8xjIcamIL2UQwB1QmZnP7xhbP675u3P2TN1+esZzi/97Qf3h3RPT004eTUkxdlcq0MA8I8ewVyyYxzMiLu2coVxKw2g+DacHAd8wOd6wwdAAOfXSa6rhBshhHl/QynF3Ke2T6CqEQHVTAErgMgrBmKHMCEYHGH4TFC1S2MCbQO2K3Fnt+H2skZhzeG9+xun8A3uCEusiltXoUOSEnZGXPpBAINdqNQWH1r7o8pp9Bh3/+Xv/t54+PenOlKY9k4ffULtb97qJMQEooP9VZ1cS5NSr+DcswQ2Twwy0/NCdLtbesFbohOvbXrTt1OR5Bx6aufeD5YduauufWDc7PCQwv8BWSivtGv3PozZsXKz9H8r/hWAeANUI/F+cq/8D/pnWGxy3HpKrgQcChgcehK42VfJVNgLVTQHgbp6fhurOnTvxNaZ8ymuRrgdwPwk5XmhcJ9HnRH7h6Dg6Olof6RguTEzIPLCL+GDEVUwNfQbcq+qbo08qM8zahVcxKSHBlJBQv2XrtOEjNr/x/JEZX504ssa2fXTolTe2OadzcECpK32mnzS8Tp06ZVZUYgar1YoQMUyOIs/IyHCXvF7dOv06N35/5sZxvx96dFXRvqwD1w5u2qNNPb9SW1ki78WReCetX7++MXfKkKuRQxeUWoHThOzsbJzCG/auDnqMupiYGKeVlJqICmHS48nAK7VoSahiCFR/ax1iyDDmTAANzwGplKpdIfcLCQ2KTyKn9d9fCBfX7b79RGusHXl0s1a93vqy99e/NWrc1jr3nRWju/707NNHDhzW8tsFIPcQukohT4gNfWbUOUse6N4l2v+9dcd7P7+k/xO/zFmxp8Cda/JUx3nIv4d950ntin+7DnX90UNyTxruOpOS0ysIVH8F4BWYyruQoNgagZHM18yZB3a7dWW0m4yZE3v2P2/asv7jP4ryNx2vemc7AABAAElEQVT94e3pfeos++IzLpd2s5yqnr1326QPrmuw4tFeQ9vELdmXfuEbqwPumDFtwbZMuU6uqned8FehCIgCqFC4nVQWWru+f2AQS4bM/cXmoPL7mQODml110xWLj3W56b6iE6Y1z93yUefQffNnl1+NFV+yzVboZyrq3ipx1lODN784eHT3RFNa7uVTNkTc/MOb361PPmmteJakRkGgCiIgCqAKdUp4w2aYX215uVYvfSHspG1BcfE9H3z6mr921Bp4rTXN9NOYYbMubJu2Y4sTEt9KwvCnrCitGsR+eG/fvW9c+MLQRiZ/8z3TtyfdOevB95bvOHTKt1ok3AoCXkdAFIDXITVeIIfPwxs0NRUV8oUwt0bzqbDxslyjrNG46eXvTOn/5eLgZn32HDj8U6/Wa8Y/nH0sGWqYca0M38jVoFbE49d2TXlj6JRrWptig15fdLD54/OvmLBgw87i7wbkJwicnQhUq5e8GnShf3BISGI9xG/u8ZSCzHQ+SCrvRnFCpkP/PhdPnd3o7v9aW7Ra9/F/5w9J2vHVx/npp7lKqLxrr+Dya0WH3DysTfqrQ76/vXOXxLBvN5zoMH7RiPHzFq07mFNtrk6tYEylOl9GoLodA/Xlvvib96CYOFuOtSD9VP7xY4URURXTooS4iAE33rije589P/+YPO3l3LtuCu/WNuHmR00N6lcMAxVZS0Ro4H/6NhnSveHv6w+9O2f791tPfb99xbDGkbee37Rvh7qRYYEY4iqSH6lLEKgsBEQBVBbyzuoNTayblZ/HcaCc5AOFNeP9Ag0eJ3dWh0Macq91m6aRtW5P6dn/wPefpc19K//Oa7JWLm45+u7YFq0dsvt8RHCg/wXd6vfvWGf1luSpC3e9t/7ErDfX9K2/7eYBDc7rVt9iKb5NwucbKQ0QBJwiUO4WBqe1S2KpCIRxKCgouDAnm0NBOcePal8Il0rgjYRAf1PDWuEN2rVrc+8zUeO+z+g+MPWT95e2a/Pnf8fleOvWUm/w6cUyggP88UH/xj39NjzV7/FB9RafzB31yaZOTy/47LeU/UfTvViRFCUIVEEERAFUwU45w5K/JahOA/+IaFNhYc6J5Iy9O225FXFany2B2jFBifGRsR3Oibz7hbCXPzF3jjr41LPzO9bc9eO3ttzq6YAlwN/crkn8C2N67Bw38LmhjfJtRa+sSW/0+MIXPl21+0ia8S8Jq+rgEr4EAYWAKIAqPBLMfiFJdcPq4kEsyJZrzdy7I+fM+ZwK4LhGRGDdmMCA0PCkCy+r++4fcY88gvPhjf+5YuFlvVP+WFYBDFRWFU1qRz95Xde1zw56pFtUw3DLk/P2N7l/7oNvLVm745jbNwpVVhukXkHAZQREAbgMVWVk5Lsw/9CwyCYt8ADMzW05qUfTd29ji7gCeAkN9K8bGxjsX+QfW6vGHeOaf/VH3HXnZs1du/yc3sseHHN6b/l+rVYBDXRSRb2aEaP611rx3ODPR7WLiw+euCK5y5MLL3t+AY6ICwplPeAEOUnyMQREAfhEh5lDEupENGpu9g8ozMvBHJSdwh0+5SuJuAko0OLXsGZYXHhArjU7t06LppN+7Dbvl9B2pmOvT/mtUdN1E5/LO+kld2ZVrBP4giwvvxD3k9ee1+LwpItm39U1ISn8h82pOCKud/+sH5fsyi0oX/CrGB7CTrVFoDIVAFceLlu2bPXq1aXdH8tdnvPnz1+zZo0e/u3bt//6669Q6SPPhjCfCEQ1ax0UV5MN4dxTx9N3bsnPyqiAhifFhtaPDyuyFZzIyMv+f/bOA76qIvvjr+alV9IILYTQq/SOqIAoNsResWFva1t1dd21u/ZV178du2IBV8ECiAgoIiBKhwCBQHovL3lJ/t9k3OvzvZSXm/peziWfx9y5c2bO+c3cc2bOzJ0ZOOWY9fbB77xMu9l/098+i4r58Z4b8nZvbwM22r4ItYugzWKcPT7x8JMnfnfnpJMHRR1OLzn1+Z/9L/zwhU82Z+T75oxI20MtJbYXAu1mADh04sknn0xOTo6Ojn7xxRc5QMMFgk8//TQlJWXy5Mnsavvwww+zQzoJ3nrrLXYJnj59emhoKJEuJJ3hNiCma2hSf7MtoKqiovjA3pK0A9WNHZnSfFjCg/z6dg1jprS4uGx3Rlm3efNnZhzu9eT9FpPh0H1PrUwesPL8mRlbNjW/oI6cw6QhCZ/cdeym+4+5aXICfF75wY7Ya5bct3D93sOyWKgj15vw1hAC7WYAnnrqqTvuuIM9x3v16nXxxRcvW7bMhc19+/bNmDGDHdWTkpLmzJlz4MABTvtie/1Ro0axS3u/fv2mTZu2fbtv9j1doHC5Ndn8cQf5x3ble6Xy/JzCvds5T8YlTYvf2qym5K6hoYFWh6Ny98G8Av/I4df/dUb6kQFvvew/xpD75pdrho74Zt60A8uXqiOOfWwnCQ3PYUld/nXlxL2PzXzkRHYWMt2zNCXprq9ufP7731KytDQSEAS8BYH2MQCcVzdmzBgNI85c5OS//Pw/tBjHiUydOlVLQIAz6vDMjhgxQotMS0tjHKDdtlKgmR+FNpO8AaH8o2IwA5agEIYCJYf2Fx/ch5emgfRNfeTOudlk7BUTHBsRwPdRR3JKUrNKrFGx/c+Zf9y3xUd98UnomdMKP/z252OOX37W9F0fvlWamW6qPc+yqeWq9O6l68inRTKps9zE+NBbzh6V9visf5/eLzrQ+uTqQ4P/8e1l/1qx+pdDaksJim690utkqeNEdlrBO04VeM5J+3wJ7H5qKIeIovTDwn7f+YCzqIYPH67EwPmzdu3ac889lxNWVQwzBz/88EP//v05NclFVOdjTBgoYDMwLfxyuaRs9Jaz8citfcmVOPUyb/ULTOhlz80qyzxiz88tLyoIjOtmDa05gIyXEFouwuq3UXmdE0CO4HWSR4cyN1x9JLcsp7A0v9geEmCJCvXvOevkhKnHHrlqTcrH7+YveeW3eef5jbWFnHBj14uuCE7oweGcMOGcf6Ph+kpvlFAlUOT86lNGCA7mkNMGGigxPjLwqlOHzZ2a9MW6fW9+n/rSluyXNq89Izns/GOTov0cRkMVszX1HbrZQLaq7ii9ptU2vd1qVd9e5HDegHTyqEMh0D4GwP3NrO/0VA7mZjJg7ty5mvYHPpxCnNS4ePFiRhLO8Tw6ePAgx3mr1548ScBMMuZEx4sE+f79+znWUfeh8JBzrjr86CgdEZCdk/k4m7ARcgxVhcGRnVlVUmjYu8ccFGqJiuFogcyMDGwnCDRCXld7pHQGZFzQutsPtkhgkUxmYUVZBU+rDUYDa0bDAi0h8Ykxl97gf/Tswu++Lv7shdx/P7Ty3w+Fz78udOpM/+49OYegyuHw8JNmOMfpx5mI7qXXxa9rHA0MctqGagmujxu7R2rI6UB4cig8W0ZM6uvXPzbh1F15izfnv78z7/0dG4ZHWy8YXTB5YGR4kBURKpuyeBSeaTn0hxBfX92lpqZyoifzavrI6Z8heOMNry4YYR7yOA4ilcsbEGgfA8DE7+7df1pIjn/f/QDVNWvWoL4vvPBCdajv3r176fLzVqvmdeKJJ65cuXLWrFnOOPOIwYSKoRGjQ5lj0K3BWafUvXt3fe8h9oOjWSNrLx3vIeQoIM4ERqLGyWt2renvyM8pSU8z0O2tLvcP6+LPmcAVFT179NChQ3mNqRGuxMTEOslryuPAekdVblF5XnE5n86Cud1oDAqPTD4mOWDWnPzLrvtt4X/s3y/JefDpvAefDrzszMTTz4sbM9EaHELXulGJMACA37t3b33dSahQoDCPIM7Nw8Mw7BUXF0PuiQFQeXbvbhw11HjeiY6Nu7Jf/XLnG7vyNy1L7/1DzoKxcadOTuoZE4KdYEzlCQPwzLq4+Ph4PJyNAuWeIeRQMZhmwkwHOQ2PbgeCe9Tw3IqndJc+mVsSiehACLSPAaBl48OZOXOmQoJmiq6kvXKLuqEJEli6dCmKG+2voaWmfFEKKgbl6N6+ieTSSMiK5uhuWrQEDQcgbEdyikaWJjAfEx8U2aXkcKqjqLA6L6u6qMQWGe2MRsPCujxVgjdMjksuNDiwe1V1QWlFVoG9rKKSIxdTs8qCA6x+EV3jzr8q+aa7Ur7+POW1J4pfeu+X/3tv/xlTEy+8vvvRMywBQS7Fudwq44cm4nJ55MktBgDOVb/Bk/QuaWhXquo9NwAqhzCr37QRgZOGdZvz1fovthS/si331qUHbl2R9vC07qdNS+6T4Oneru3b8Cgd2JvQ8P4Mn27Y/5yN3LUFAg25OFu1/JNOOonl/LxpvKvvvPPO2LFjKY5R53PPPUeAviejYNaAYhjoDHKRctKkSbiD6JoRxjPz6quvjh8/vlWZpCCu5hTRHPLawptWOjv4s3VEQHx33O6VpUWlB/ZU5Lmur/VQHM85p2/LOtE+8SFJcSERQX7sJlRUWnEwq3jvkeIcc2jveRcc99GakcuXhZ42tuDjbzedcNqXI4K3vvky8xYNcOJ56a2USXMYsJiMg3oEvXD9xO33Hn3zxAQ6Nbct3Zt8+5c3Pbd68x6PPp1rTukKkGbm0Bzy5tA2UJvyqDUQaJ8RAJIwx0uP/uOPP6avgaLH00IknY4pU6YQQOPziQAWQsnMsICOPzGXXnrpd999h2MHJ9LVV1+tTRqrZPKrEPALj/QLDcsrKa3IzuSb4bLcbPYWbYM9pQP8zN26BMZXBeSXVKRWlDkqqzLySg/nFIcG+kVPnHHMtBmZG3/Y8eKjWS8s2nH+pTsNhm733558xoVhfXAm+dqFl6vSUdmve8RjV0y4/rSit77eccfSlCfWpj2xOm12/4hbTxowZXg3Pc4pX8NJ5GlnBNrNACA3y3i4nAHAAAwdOpSY2NrL+ZEKM7Xl4vR3TyMxNQiYzLb4Hg6rv9FiYu+ggj3bAqLjbV1i2wAcVotGBvv5dw2pLPaLDLZlF5YVlpTzZzaZovscNfH5D0v/unfnBwv333xv6p0PHbjzoaizpwy47V8xw0a1AW9tWYQ2duseHXz72SMvnzPk01U753+6+/MduZ8/ssYQZnvphKRJw7v161azaksuQaBdEGg3F1C7SNupCmWBjiUkLKTPQL/wKNbelGYeLtyzrc32c2Z1UIDV3DXSf1CPsITIQLbdZ1XpkeyiLftyM/zjkq+7Z1ZWxsC3/i9ggCHnnVVrho9ednTsvi+XVJYUqTrSN3nbkesXo3jx7MElz8z57OpRAxOCDblll76ztf8dXw2/9fOFS7fuSfvjI5iOLIXw5mMItOcIwMeg7Iji1E5gcL6YX2gEH4tVltsL9+7w7xLLXxucOK+6wJyrFRVq46+4zJFdaC8srSgotheW2K1W/y5zLux+4pk561ftfPr2/MW/blp50rbJ0YkLHuw140QrHDa4Br8jou0BT3jJTpjQe/aE3j/tyFi7+eDdqw9tPlh0wcJfDbZtJ/UMOWNC90lDE3rGhlprziPzIDtJIgg0DwExAM3Dz0uoLUHBIX0GlGWk2XOyyrLSKwrymCi2BDayFKdlhQvyt/DHxEBO7cpRe7kjLbvIYjGHHHXs6LfWF/2yLmXRwszHX9n+3aV7x/pHn/IX47AxhmHDWpaHDpIbun10vxj+Lj9l6MZdmas5mnj9kcU78hbvzDMEb78oKWxwV+uJk8L6BYd0EIaFDV9FQAyAr9asq1xszsOe0taQ8JIjB3EEFR/Y4xcR5d8lzmg2uyZtzXuL2RQT5s9OywwIcgrtRWWO3PySfJMxIHls8p2jEs+/+vBXnx559b5Dd/7TMsiwbvW1SWdd3GXAYFY3tSZT7Za3v59l/KB4/q44pZwFQt/+nPrapszXNmYatloeWJtzdnLYzDE9RvePjYtsU1PdbnBIwW2OgBiANoe8XQusGQokJjMIKM/Jsudk8sVAQGyCpc17mviFQgKs/NV+SmbPKyovKS4rNhptsf1iLrql62kXpX/58aG37jnywDPpDzwTftW5SfMuiBs9no2P2hW8Viw8NMhv8tAE/q4qKNu0K2PRyh2L95X8+8eMf/+QHhXlf+ngqOlHdR/ZP5ZdN1qRCcm68yEgBqDT1bnRZGZPaWtwKCtEOWe4+OBev7BI/5h4o7kdGgP7CsWGB0SH+heWVeQUlpfYK3LKDWZbVMApl9l6DOpdlH5o0ct5z7+1/rm3QuZO6nXuVd2PnmkLj/ThOosM9Z8+skf30LK/BERsP1j81U+pH2zPefi7tIdXpVFLN4+MnjCs27iBcUH+vjkk8uGa7ZiitcM73zGB6GxcWQKDgxOT7ZnpjAPsedkVRQXMCmAV2gUHZjzDAv34KyuvzC225xdXFBXbc/0iek6fljR2hv3in9Le+0/R8s+2LFq9a5Shx1XP9TxudnC3nu3CatsUyuZB0eH+id2ijx/X65ac4vVbj3z2Y+r7u3L/9nWq4atUQ4Tt4QnxE4Z2G90v1ubXph68thFfSmkzBMQAtBnUHa4go9FEx5/dQ9UCIY6X4QuyAI4ZaI+hgELH388c7xcYG27Izis+dKC63G4vNQWYhk7rOWyyfcemrI9eyl/25u4rrtpdYYi/79Y+c8+JHOibs8QsoNK+QI+PDDppUhJ/d2cWrd96+O3V+xftzb/t832Gz1IMUQGPT0qYNKLb8D7RVrMs6e5wr1jHZ0gaTcevo9blkJMmQ5L6+0ezfWN1eX4uh86r42XacSU+KyAjgv16RPolx4dGBFn5pra4oroy+aiYO5/vvXBD9F23WRINR5545NtBw1ddOOvI+u9bF6AOkzsflJ02NfnDO49NeWDGWxcNHZsYZsi337Rkz5h7V/pd99nTH/y8ac/vh9LIGtIOU2kdnREZAXT0Gmob/vgywC80vCg1pYoFQqkpfmERVdaANvhWoD7p2EafHZStZkO3qMCEqMDcQntWob2szGHq2ifi8nvC516e9+3nuU9em7dk2do3lgUcZUi+94Nex8wyBwbXl6EvxfeKDel1XP9zjuu/53DBdxtT/7Vy/6/7C6//ZLdh0S4+MH7++N49ogyTh/vyTIkv1Wb7yiIGoH3x70Cls1kQpw3b2T4oI41BQHHBgUq/wI7AH6vmI0Ns/JWWV2YXlOWX2KtCosJPuTh05rz8H77Nf+Pu8u93/jpv3tYQQ+8Hn0488bTAuIQ2XtvaXiglxYcmxQ+6aPagnYfyv9t44PnVBzekFFz5wXYDhzWH7Hv1+N7jhyawH1F7sSfldnwExAB0/DpqUw5tUdHWEBYIHTTk5ZVnpJWkRvjHJrTBRnKeCFm72VxQfFV1fnF5dpG9zGILn3J8+KQZBVvW53z6muOTt/fccl3KQ9fFXXBzz9POM7Enka7DADzhpKOl6ZsQ1jdhyPwTh2zdn7N286HnVh/YmFZy8VtbDR9snxIfeOm0XhOHJPTu6ulm1B1NOuGn9RAQA9B62Hprzqj7oB5JxVWG/MJCR0lxYcpOvhfDMHQQeWo2m6sdEJTYaz4lyy+tChk2LmTwqKKzr835erH94wfTnvjXkdf/5TdxdsT8G6IHDamd3uggvLcuGwyVBvWM5G/aoKD9OZW7DpW8vPbQqgNFq17/1RC4fV7P4NPG95wwpGuP2EY+pyCfTmM6W7dGOn7uvm8AmtMNhLbTkvtFdPHvlugXGmzPzyvLPOyoWSfKUMDTD5E4iqc50HlCHmiz8MeAgCPJ8oqNpr6DQ5IGlJ56fs6a5cVLXqz68fNN731uHRsTdewF3WaeEtm3vy2i5sQhT67aavfiqsdGju4bfczo8AtnD/41JXvNL4cWbUz/YHfBBzt/MQRuvSQpbM74nmMGxMZ3qXvKpKi80q9a//KQ5tS7J7UjaVoQAV8zAIWFhZyCq5ogp0pxngxny3AmsA7I0EEcisuhvpzPp5tcKTIdR2QgQl5enjoRTDc5nLOBtj5yBC8sLbPHdy0rtduzMwz5ecaMI2wobQ2LrNmmTVulWA80nNtTe6hwfp0nStZD9Ec0B0JwOhA5eEJuMRrCrdVFlY48viMICgk57hTLiPH7li0OOrjLsuntsn8/dviex/yP6xs189yY8VODe/c11m6CxG6p9UlBoZROy2nqiWCaAJADIKeqaTGeB6h6BOdcLVqvvrqj5SBC7TnE1f3jbcmxiXPGxv+yN2vF5vSlewpe/jXn5U1ZhjC/mwZFHj2i69DE8NAgi9loYO3p4Rz7+ytS1u3JsZoM4/oePPvoxPBgzjRW2/p5JAHMIzjHdXiUWhK1NwI+aAB4b5UBQPnyCnE0NqfL6sBZGQCMBxZFB7nS4LzANW9iY+rSPX9Fzmk56CB95LyHqFH410eOEkENZefkVFWbqkOjHNnpnDlsyMk22gKskTHmoOD6tKeShSN5a8izsxHfXbpGY+C8qeQ4LoLNBlt1dZG9qtjqVz5kbPysuVVFV1bs2Fz+6/ry9a8X/+Oeg/kG84SggMkXB48cF9Knv394JEfaV1dVusgCz6p03QYActqh7lMVIcf26+t50HIg56A9QNaqHpfO8J6Bw3omzi+t3H24ZP2ugsXb8x5fl/74miOGIPPFvfwnDQwPD7TMfT/Vz2TqZuVUYcOnaakPfH/wk/OTEuOCPT/UntKp9EbrVxJ0EAR8zQBwajyXBm5qamrfvn31dcTIhDewR48e+gYQkPMycCZ8eLjOEz9QH2iBmJgYTZwmBej7w3+3bt2aRKUlxgCgwrQTmGsOnS/ILc04zBme7Ctniwj1i2yol4cBgHknci1jjwLoL5jXTc6eo0Hmip59+hSXVTqGjXDMObM4/abcTesK1nxt+vUDx1vPlrz+bOnECdEnnJcwYXJ4Un+DxfVF4MxRWg6dAI/YdUsE80lJSboNABYoISGBM7HdMvYoArZpdRERda//GTbQMPcYw015pRt2HPlyw6GPt2W/urP01V1lBquxn3/NUEBdA83mVHvV0l+KH5owrEko0PA84lISdQAEXNt9B2CphVnQ1wNVTEAr5Fp9WPgqKyCYBULsG1GSfsienxMQk8DucloC50AzocMANAd5/BmBVmPXiACcF6V2R36JyRbQO7RnUuXM0wrTbs/esLbg2yXF25blfLVmZ4Eh9MTB8Wdc13XcpLCkvmyUhBR0nBX/ug1AM8VvJrni37k63MMx4QHHj03k79ackp+2H160JnXF7hxN+6v0Xf2Mj27O/EtuSUxEE5yozak4dz4lplUR8H0D0KrwdbbMTVYOnU8sL+Co39TKsrKiA7v5kNgWEc2HYx1t4UjtbgrVDMJ+nys2GJggKCg1BwT2C+vdr+qkswpSUzL5jOCLhaXbNuddc3mNJZgzJOHcv3QdM54NU42MCfR2/72rVcRHBs6ZkDS6f3z8VUsGdrE5M28yGA2lDjYmco6UsC8hIAbAl2qzjWThm2FrUHBZxpHyvOzKstKStAOl6YdsEV1skV068sb9bDTEX0yYgT2oC0stQUEDwpMGVM2bn5+yM33NNzmfPF7w05acby7cXmIIOb6PddZlxQEWTEUbYdrexbA1tyHYr7K62uy0ArS8unp0TIC/v197cyfltxYCYgBaC1nfzpcN4wLiuwXEdbXnZnPKWFW5vSw7nT/2E7VFxbDVKPMEHRYB9qCOCrHxx9xmYak1NHhoVP/BjvOuzNu1NePH1ZlLnsvdtjtk520rrr8tbEaXnte8EHfUqMCEnh1WnBZhLCjA+tKpyZe+tXVg5O/qniHUnuzyh+cNCg8WA9AiGHfETMQAdMRa8RqejCZbZDR/FUWF9pwMR3Eh0wOELf4BFXwx0PS1T20sOOvlw4P8woNqOC0s84sIGRk9cJj9zPm527ekLP/C8cPi3N07M8473T/AEDlzVrczr4gdMSowXuekehuLpqO4S04YXFjquPG9HYZQS82a0NKqF88beOq0vjqyEhJvQUAMgLfUVIfm0xocwh8nTeIUYktR/EJludnlmZn2mChrWEQH2UmiAQRxe4QGWPnDEpSU++dFTigKCou/9PqCnb9lbFiTsfzdrFVLDy1ZGtjd0GXqeQlzzooZOoJ9sxvI0Esf3XD68LOm9/1y7XY/q2XaqD5xkU2Y+/VSkTs522IAOnkDaEnxzTZ/Dpj0j4rl0Pmyw6nVDgfDgrKcTI5yZIagvvVCLclBs/PCEgTZLEF+5swIS++e0WXdpncfM77grEtytv6S+fPa9O8XZyx+c9/CN0MGRkRPObv77NOiBg7xj9K5TrfZzLZKBij9446KtVqsXUT7twrAHStTMQAdqz58gBvWz/hFdgmy2iylFdbgsPLC/IrCvIqifIt/INtLMEnQ8bfqxP/BavhAP3OoxRIdbC2PDi7s0TV/wuT8sy/N+nVT5sZ1h9d9dvjd53Y/8VzEhF6xx53bbfqsyH6D/MLrXnfvdXVaWcnqKVn543X1podhMQB6UBOaxhEwmcxBIQEJPf3KSvELsb+0o7SYP5PV5hcW7hceZbJ6x9QiYwKbxWQLNkUFh1XEhBYl9cg7+pj8jCvSN/2YuWHtwZ+WHnrx/m333R81dUD87AsSJkyNHDCEsQ6T5KxAbRwlSSEItCsCYgDaFX4fLhxveu0kMB8KBMR1YxMh5gYq8nIqWS9UcxBxFoMDv4goS+22PF4BA+rcz2KMtJgjA0MrokN6J/fKm3VidtrVRzasy/xx9f4t36U+fof/Xw3RUwYnnHxJeZfYyvhYg669gLwCDWHSNxAQA+Ab9djRpTBZrPjK/WvWCxVwDH1laUl5QQ5/loAg/EJ8WNDRviNrGFCr2RgWwF9wQnTf5IHJeSfNTU/Zx5ggfdWylF0bDjx8Y6DRkNG3f88zrksYNzE8eUBH/jyiYUnlqW8jIAbAt+u3g0lnNFpDwvirLCux5+AXyqvxC5UUl2Wk4RRiorjm+1uvutiINMTPGBIdlBA9qHTEoLy5Z6Tt2n34p3VpX7yXf2TP/n9eZdtriDthSOJZ13UbNzG0zwCvEk6Y9X0EvOx98/0K6RwSmv0DA7sGVld2ZXqAvaarKirKso7Ys9KtoRGcPIPXqCN/R1ZnFfHZW5DVEBQVlBA1rGz0sB/HTjAXF6evX5f+yYu7t21J+dtlthRD/OxByRfeljBmXHCv5DozkUhBoI0RaE8DwC7NGzZsYMfjESNGsHOku+Q7d+7ct28f+xqOGTPG+SmbrbNLV3Bw3duQOaeUcEdGgOVAfDbMX41fKCvDUVJU4xfKz7YEBBpD2ELVWydR/c2GLoGmPsOPMk4bl7/gqgO/bklZtTJjyau7f/1t720X2PYZ4qbH97/uqYTBQ4OTOstWEx25HXZm3trte/309PQnnniif//+cXFx//nPf9z3EP/444/3798/bdq0Ll26PPjgg+yvq+opIyNj4sSJGIbOXG0+JjtrQ4N79QlJ6o8jiPUzbDNXmpZacXB3eXY6gwNvFJYjICodlVaDoUuI31HjR558681nfP7TtHd+7Hblv6onDUnZe3jptWcsnNn/7eOGrXvhibRfN1cUF3mjmMKztyPQbiOAp59++v777wc+9Psll1yCuj/33HOd0WQr/+uuu44YNoU/5ZRTMAbJyck//PBDZmbmiy++qM67cE4vYW9HgO/IAuO7V8d0Lc/PqcxMr/mOLCujNDOdOYNav5AXf5XKVEFUqH/UhNFDxo8uWHDlgc2b9v+0Pu2zdzK2rsvYedN6f4t/7/HxYyb1njknJjExPD4OK6h7J2pvbwbCf1si0D4GgDO2xo4dq8nJiSvsIc4xRmFhYSqS40SmTp2qJWB/c3U2E74gllfjONIeScDHEKjxC0VGG4NCLVm5lpBQtp4uL8jlMwImBmq+IwsJ87rpAecKYsP9iNCAiMnjh04eX7bgivQd2w5t2ZT6zedZyz/YvfO7PQsf9Os1NmLEeFP3pNLJU7sl9wkOCrCYDNgPuQSB1kCgfQzAkSNHoqL+dEI34wCUvmYAMAnDhg1TAuP8WbNmzXnnncet+rhGO+iuNRBxzpPiVInOkZ6HvZ3cc0ndUzZXdr4jCw4N6NrTFhVrz83iSHr2F+IQAnt2ul9oRM3+Qg1+R6ZKb2bduQvleUyj4qPSAwJsvYYP56/yjLOKMx87uH7NwQ3rM75bnLf0pSr/8OyPXg4YMLrLsFFxoyeFdIkKjYwI9DP5W00Ws8nl2JY6uWqO7GTYHPLm0NYpi0S2HgLtYwBw4Li0Ega8dar1tLS0xYsXn3766R4e64jjCB+RGj7zyzzz9u3b9Z2LC3lKSgo56CbHbZWVlYVVq1O0hiuV0pGdGXJ1wHfDid2fQg4OnKyLWdVxQhO1w4CMS9/JXJBT7uHDh2FDX+lwvi8lJTAgoIbcaKy0V1QWFVQVF1RXlPN9GefSmINCzSHhRj8bT2vgrf3oTOFA6bDNLFGdKwvcsXKPIUPIYV7fgcxkSMtBBBqtJ1XPmAaercPGdO8zIGr68YUcULP++7LfFpe8t+ngN/+XGmCzDT81oP9wv159bTFx/qGhfhazzVTlZzH4mY1Q1my8DQA1YtSUBtu8BSyRYPWEJ6W7iA85FYfguhse3Ts5FN4F1Q572z4GgHNud+3a5QwKx8+6q/jVq1fzIl100UXuj5xpncO0PO0MXrPZfOjQoe7du3O2qr43oaioiDN1eZf0kTN24VxWjgXWR45cGID4+HgdOpTXGPUHA5xprIMcfUSNYPl0kwMdUusmh3NygPz3yR4Yqj29vZKVQrnZrBfi0EZjZZmxsrr2O7IwnOY1INeaAdI6HI7CwkKqvoau6RdZKXLw11F3FFhcXEzpAQEBTSM3xhn79DVNmrJj8nGxQQ8bsjMOLV+asfbros/fta9+tyDQUNVjsnXQuIhBw6OGj62w+BtsNpOhymYxsm0RB59xzgEbXJvNpkqHIzQslEF2VVWtXWgKArQcQMMA6G54gNaUAiVteyLQPgaAY6N//PHHWbNmKdHRULyxKEpueeFR3AS++OKL0NDQ888/v0nwYCqcrQVZof1RZE3KREsMYXPIef+bQ07pKHEy0fhpUgByXmPd5JyKzlSNbnIUHwzoJkeJQOtclb/LztrfmDjcQeW5WUwMVFdVGvKzq8uKreGR+IUwAyoZLQpy3QuFYR5y6g4Am4S5lliR18G/lqLBgL/NFhzbNTgpOW7MxIr8G/MPHkz7Zknm6m9Kflhh3/BdaaghJSLQf/rVYUNHRwweYQqLLKz0KyyutpgwBqYgf5MD628DfP3NXuHfII/1PgS3ep/Jgw6GgM723XwpTj755C+//PLYY4/lXX3//ffHjRtHnizwf/3116+99lpWheI/WbBgAT1BVRYdE+1thISr+Tw0mgOKgKvRZA0kaA55beH6S28+eQNyNfqo+aU3AF3N/kLx3W3RcewvVJ6T5bCXOY4cMmYesYaG8zkxT0FNMaBvBIB0DZTeqOyKvDk51DL/ewu3hkV24W/Q0Oqrbsnftzf9p3WHFr1c8tPqqkceLYo35LFX3fS5QVNPDB0wNKBHkqOqurii4lBuOefy5Dv8au2BJaD2LEw69p5wrtI0k3nPC5KU7YtAuxkA5nh37NixZMkS1PqUKVPwtAAEPSYW/hNgHMAnAsuXL1focJuYmNinTx91Sxh7oMLy22kR+H1/oagYVgpx8EDN/kIcR5OXbQkMsdQeUq9b+3dMSI0Wv/A+/fnrd9ZFhft2Z+/euf/VR/Le/bbqzUVF/11UbDdYR/sFn/oQDqLg8BiL1VJRbSwrdeQW13ShLCajzWrGEgQHWPytZsIdU0bhqo0RaDcDgJz9ai9ngfF4DBkyhBgmCbicHzmHZYrJGQ0Js5ccf5xHxq4SjAkcJYX2wrzy1H0lMZF+IeHmwCCvXjlaZ/2G8N1crz69jp1d8e+cjG2/Hvrio8P/eKrs23L7mptyQg10jmIWXBUw4/Tg5IGmiNj8otLi0ooSu4O/rEI7GbKZHcaAaYMgPkGwmllZVGcpEunzCLSnAfB5cEXAtkSg5juyrj04kozNJEqzMgx8R5abxYAAHjiLxhIcysFkbDLBqqG25Kq1y2LyI2H8FP4M9z2ZsWFd+uafDz5ydfl+w+Hnnzv86HNGmyHilOndz7uxd/+Bfj17F9urS+3lBcWYSr5TdnACsGKPZUWBNjPGAJPg78dEcmtzLfl3FATEAHSUmhA+WgSB2u/IYixhUZacAn825S8pZudRR2nNnyEr3WgyYwMswSGWwGCmClqkxI6TSczIcfwNmX/V9lXflKTsLfzyvZy3v8letDx74XJzkiFg6Lhup1wYO35afO8+BrOlpKK6qLS8zO4otjvsjspyR2Vecbla/5OdZe/T1dZx5BJOWg8BMQCth63k3J4ImAKCAmJqzm1nUx5HcRErR9l3uqq8jI3n+KtZem+2cByNJSTM4h/Q8Y+tbxKUfj2SYoeNijh/ftnTuQe/X5G5fk3Wg08W7Vm3/eN1O6MMtslTYiZM6Tb79Ihu3a3RtUvvqg0Yg+KySke18ciuXUd+2RgRFxg0YlRI915NKlcSex0CYgC8rsqEYc8QqF1JQ5cWRc/qIP74dKC6ogJLwM5rVWUllawwYxFRfi4JWDZZYwyCQhgWMLdc4ybyak8Ri+QcDoPJ7B/Vpc9J8/qceJr9L/dkbvzxyMYNWR/8teSTVfu/XrX/1n8GzB4RPurohONOiOw7MCwmLszm2PTcvw4/dbvZZtiTZ9hrMiTd/8LACy7nu2DPEJdU3oeAGADvqzPhWB8CfDPLl8OskGfPUT4gqCovdxQX1hgDexluIhYRcVQZ2t/k58+8cbW9tKrSYdD7HYA+DluLymS2hYV3mzaDP/v8BUVph/YtW5K3fmXh21+WLt+Ydt/jtqMMEXNuNVWWHXn5aUtiV4PRZIgwGqocOy9c4BcW0efkM1qLMcm3vREQA9DeNSDltwcCTAbQ2eeP0wiqHBVVdntFcSH2oKqinEVENW6iwweKQwMrmDeu+QsyWfx8YymRLSyCv6gBgx1FVxf+MyNtxbLMDT8WLnkj/b5HzMMNlt7d/6gNPrwZE5Hyf/f3OHpGzZmdcvkiAmIAfLFWRaamIFDT62fZfBDnC8VjAGo+My7MN1oPsB81h1ZWFOThDmKJkTkg0IoxCAxmntm7HUS14LAsKoK/xD6Gi68svOMfaWu+3XHHBYY/fztstASWrfylorTML7QpgEpa70FADID31JVw2voIsMkof8wMW7Pygnr1NjrKKwoL2ICoqtxeu/9ENuMAsy3AHBRsDQo2B/jEFwZGU0i3Hr1mzdl6psEczzfUTh7/6ipTjUNIvhJo/ZbXTiWIAWgn4KXYjoxA7b5y9PoZFviFsU6mGu3PbIGjqLCSU+zLijnLnhOMa4xBYFDNsKB29rgjC9Qob9aAwMiL5xRsXGIK+sMLVJV/uMsZNwR26dIouSTwUgTEAHhpxQnbbYmAkVPs+TNExbBJENq/ZvaYIwr4vKCokD8shNGMEynIGhyGj4g1RdgG79qIgnHPUX9/ekWPJZX9U03hcUhUlZlushsGX35dzZywXD6KgBgAH61YEauVEGCr5MBg/vyj46tZbMmi0lofERtRVBTk80exzChUpKdVREfZrBZtg9JWYqcFs2XV/4zMI7+9+lzG+/cZLYb4mx7tf+aFHMbZgkVIVh0NATEAHa1GhB+vQYBuPsfZ8wfHHF5f860ZS4lqPjezVxXllxzaV5WdzidmOIiswXxhEFgze9yxL76dHnnL3w+ec5nVYo2Nje3YzAp3LYCAGIAWAFGyEAQ4pIwl8/zhI6ost5vziyxBoRxWxuQBf3xhULvwlEmFEOYMTDb/Dj2zWnOyjv59yKUxeBECPm4Amt+Om+nJbQ45tM0kb05DbE7RlNsRyJvDg37a2jWjlrDIoO6J/v62mnmCkpo5A3xENRsTlRSXZ2cwFKhdVBrKHHKdu1DoL/1/Vd7MHJpD3hza/7Ev/7cRAr5mADjEirNYVRPkLeOwAQ7n03d6DEcOcA4B5PqsCDyoI3nZ41pHDpBzJhec62MAcgWFbnIOQ4Z/feQ0Xs4E5pQ3fScSQ07RyK6bnIqDnGMdVUto6ssELTlwJqV2BlFTc3DUQueorKwh9A+u9gtgd9Kq0uKaqWN7mcHoMJSWGbKzsQQ1q05r1xHx3ZVWimo5+piHik0uqD59dQc5DU93y4ecojVBJNDBETDSSjh7i535feOIlQMHDmRkZGiyoARtNpu+F4magxz1reXW1LrkRVJGqKmEKj2vMZzrPmEVJYLhgX99paO+0QKgp48cHQr/us9EhHPA52BFfaU3k5xC0WIw38Itp3aLIadl9jXC1fpa/nSoPZHITr3TeGpSNP1q34YHdBMmTNANXdPFFQr9CPzR6dCfR0ei5BhxLo2jNWvWcPSY7n7czz//jGnUfcbptm3bONaGs7k1fpoU2Lt3L/o3ISGhSVRa4rS0NPRI7969tZgmBTiVMz09feDAgU2i0hLTseDEt6OOOkqLaVIA88Op0brJMQDr1q3TTQ6rP/zww/Dhw3Wr4A0bNgwaNEi3/fvll1969erFmdhNAk1LvH379i61lxbTpMC+ffswP81peOAvBqBJmLdXYt9f4Us3Vje4tGN6srrJoW0OOaVzNad0rybXLTiEVHoteDrRU7XWji0H5pvZctqRvDlFN6fShVYHAr5vAHSAIiSCgCAgCHQGBHzcADSzM9KcXiStp5nkME8Oulth83uR7Vt6M+uu3cmbg14zmafo5pTefHLdjVYI2xgB85133pmfn4/D0Cd9dszfhoWF6Z7FBZOQkBDdUwiQM3+gexoW8sDAQN3TsJDjg9Y9j0pDpGjd8x+Q40cGPX0NGuaBXbcTnEJx31P1+kqHijZD6bpbDoTIrnsKAfLg4OB2bHi0HC596FF3tBx+9ZELVVsi4GurgNoSOylLEBAEBAGvRsDHXUBeXTfCvCAgCAgCrYqAGIBWhVcyFwQEAUGg4yIgBqDj1o1wJggIAoJAqyIgBqBV4ZXMBQFBQBDouAh0IgPAdz31ra5j3VteXl5BQYFzRalI9pNxjlRhNjnQIvnglk0XtFsVoKCDBw82ZymeS4ZkxcexLpHaLUzm5ua6FMdWNkS6f83kzHxmZib77Wj5aIFDhw65C6U91RFgc4L6qPhcOScnxyVBnZHsMcD3yVqNQAKf7tkiNQvb3ON1x6hNNeokp6IpTmNJpakzkkwQk2am5cOX3lpYC1BfbGei3TY/QKtwbwNatnAOSy7vBS8CkS7NiRj411pFVlaWi9QqT/aVcclNK0sCHRABH18GqiGOQnnkkUeSk5PdFyaiAd9//31eY76AZ/8Ato6AivRvvPEGGof9DNjRYcCAAVpW33///ZYtW9giQsXcdttt7PfQrVs3LQGB33777bLLLjvrrLN0r+Rzzo0wzLCuLj4+3iWet/S///3vzp07eWkJaLsXfP7551u3buUVXb58OXtjsJxUEbJR0v333z9lyhTF2GuvvYYso0aNcs4WJTt69Ojzzz+flYjO8brDoPHRRx+NGTPGPQdKh1WY+fLLL1lyynJk0mzatOnrr79Gj/BLfam9NKgIqgmW2GUB45qUlISuYb+Ee++91yXbf/zjHyy9JYFLvL5b7Ogtt9wyffp0902ZaDkvvfQSfP766680HloXRVARr776KvWCFKmpqX369CESdfnKK6+QA8m+++47qonIyMjIc845x2WnEFogyC9YsEAft+5UlMtq1Lg4DvlyvRYtWpSSkoJa//bbbxMT2bu0Zt0nNUUkr8NXX33FRiDwTI+BSGLY3uPTTz8lJTX19ttvs1GKS51iHhDn5ptv1r122ZVFuW9lBHxtL6A64eKlorHSNOvsCn388cezZ8/mbYQWzcLtqaee+vzzz1966aXKWqBJv/jii+OPP5534N1330WzOHeOxo8fz9Yr48aNcy6anWTYD8s5Rnd4165dS5cuxcZonS/nrHh1u3fvrowWPDzxxBMYJIQlvVLrkyZNevbZZ2+44QaoeI1RT0iq8c+hH5g3Xmzn9f6QX3DBBc6l6A5jQd95553w8PA6F9TDzNq1ay+//HLyh3mUKbYKhUvX+KKLLiJy4sSJDz/88E033YQawk7ceOONRGKcPvvsM2oKvfboo49Suc7gHz58GKvQUnYXCwT+7KdUZ6/2scceu+OOO5RhWL9+/erVq0Ebm4r6Vh9/IN1PP/1ERdCoaE4KZKw18TSbu+66C/Pct29fhFIXpZDPvHnz/hfRrP8xNtQ4LaFO5rFD6HfVs6GaFi5cSEXQnNjCSHVuxo4di5aH7W+++QbMQRVu0Ph///vf77nnHroUdJIYaTl/bLFy5crHH39ca13N4l6I2wSBTuECQj9eddVVdNDc3wS6PygLpf0BnI482hw7gSbSxgq8J7hKeMoLfOWVV86YMcPZG0O/ibcINaHVF71sVAYvnhbTnAAsXXvttegLd+uFOPSClfanCDhBRnhDvwwdOlQVinRDhgzZs2cPt9g5FCvqWHtFyfPkk09mTKNxCDmYDB48WEujPdIRQDleeOGFp512mjNiWj5oTFjSbull07vnQo1qkWeffTY7+uHqQTFpkYiJnUB8hjLoUC2eALJAXmdxzsk8DGOBqHFMrDsaWC82m1Pan9xQkbhu6OmDtvbpH7XGlnaAjJbUTCyaFIsLCQaMZgbaGjOMGKhN94rWEjQpgHXHdk6ePLnODAG/f//+KkO+FuzatSv8A742tIXhiIgIxGQEoLS/SqzcayBMfWF9NZaAiK4S/Lu/ZVoaCXQ0BDqFAVBfw9b5GvD6OTduqgfVibfBZStEPA90/7UX27kWeRN45+klaZF4M1BMLk5t7WlTAw0wz8vp8q0pqorXmF9nVjFg7OtJuWqM7/x+gglvPjZP42rjxo10/VrqM07Mj+LQXYFSIh1PZ9cZmONoxm3l/AEwKhLryyPsrsYkJoF4BOGaOnWq5jQHEKqDxMRriZsTUODXmRsmH5ydM8cY4AtyloinIMx1xhlnaCnRvHhRuKUH7WLAeIQhcZ6k0ah0BBTz9dlCFL1zLdOJwcq6+BiJpDroIlA67RmcGc9dccUV3JItZhgSrZ3j6DvmmGNainkd8gqJDgQ6hQFoABdeTmd1Q0p0Fs1a668pWiJ5XevMB+3AcJg3R40SSIMWQ+vVqTXqzEF3JC+h0ulaDvheMVTOLzaPGMrUpwV4inNm2rRpOKxVJuhWLGID6bWymh9wKQXmYQbonK0XpSCRc1k4K/D5gDkIkwOddCY/VAKkaMHuv3Oh7mEmALSZFfUUznFAKYOnpVd8arc4SSBU8xO0PToWdEEwhCRATBCg4RGvpW+9gEsjYYyCj9HZn0PRpMGmKh7gExXP/JD2agA+FaH1HvApMXB0qdPW419ybhEEOrsBQLMzbHeGEkVPB81lGQn9Gq3dOyfWwnTl6L5xy2twwgknaPGtGqDLqZwJWimokujoaJdeGAJqngotpRZA3TDowatODA4lhgsE6uywayQtFXDRlciCTkcNudhabZqUBVfM5NPNVJ5rjU+qRulQeuV0wNvA9FI0ZlVTjgoQdB8tx0UDkkbNfyDdM888g//txBNPdAYQ64vjhRg1MdA2yFOc1nNXzNB9wfxonRgVSdtATLxt3NLFYRrsmmuuefHFF9VTcGYCgxkjbmljajzUZvwrHuS3mQh0dgPA6gimE51B5CwOVm4wgaZF0tBd/BLaIy2AtxTXP7crVqxQA3ztUesF6K+5TK7SHeM9pLPmrEOZ5m2UJfpuvOfMATr731uPc5UztkpNTqhbKgL1jQPH2frCPDEkoHe/atWq66+/Hp+VC2PHHXccrn+yQgqXR613ixHavXu3lj96n4tIrUesHjG+ZECDC46By3nnnafW/2hUBNChkGAelKerbawX5YK2s7LG7UmzV7peY2/z5s0MApYsWaLFENCmlyCn/0R7O3LkCAPHllr14FyWhFsbgU5kAJybO7DS5+WNpfOIq5SFGQpopuyOPfZYwvR9NMPATJfqF2uV4ZyV9sYeffTRdI5wSatk5OCcTKPVFyArrSByYJ2McnzDGCZH5UkHGXvAG8vsnDYnQdeMVX3OqwCd8yGsmJw5cybuXbrkmte7BZmHPedCsU+skiISBwLLChXzwMVSK2Z60S+4dFTpjAY++eQT5iQwCfgfWFZLL1WlV78qGZ1TtP/ixYuVbiJSxTunbE7YmXn4ZI6H3OAEbLlUzvh2mArCJNNXUDMuxCOdMqg0NpbT0EtwZoPmp26xWywT0hbjUoRzsmaGXdDAjir2Zs2apc2f09Rhm5EW/jSt5dBseDuU9dX6QxgqKgKWNExgm5bDQEetpHAprpnMC3lrI9AploEqEGnfzuoDHarUBHNcaBwWpJOMEToeBgKsg+SdxB6QhsXpLr0b5xU+WoeUrhzL5ubPn6+KY4rMpXuu4vX94ut39s/SwVeZjxw5EqfThx9+iO5GXZ577rnkj4sZjfnBBx8QiTa55JJLnAvt2bOnxhiqU3mH0PtYDs16IZSLI945Bx1h5yEIfCpXA9WBTkd9EMAzzqJ4CuVivpdIRCalYp4+JhJRI0rvoDoxHqgn6ksxgwuOClXWC4k0r5EOVt1JQEnzVsGAGuqRjAVC7733HswTydgFA0Dk3LlzaU4gDPJEMteKsEhHpNL4JAZeVK22fAsDQF9bTQyQQ8uOY8BEW89G5mhq1ZAoBaVPy4FVLiqCpwxQGGbxfQCNAZGVM5NV0dhmVgfwLmAA/vKXv5AS0RQmVAFh5q6J5GJUx7BAheW34yMg20H/Xke8ljR6Luc64x3W3nzn+A4YRrm4v3go0Aa8/x1HCvzRqFEXfpjJwBi4RHbAW69GHjzr5N89khgul0UHHbA6hKUmISC2+ne46AS5A+ct2h/O3bU/kV6h/eHTXfsT6RXa39uRr49/9+ZEjHsk5HJ5NQJ1aD2vlkeYFwQEAUFAEPAQATEAHgIlyQQBQUAQ8DUExAD4Wo2KPIKAICAIeIiAGAAPgZJkgoAgIAj4GgJiAHytRkUeQUAQEAQ8REAMgIdASTJBQBAQBHwNATEAvlajIo8gIAgIAh4iIAbAQ6AkmSAgCAgCvoaAGABfq1GRRxAQBAQBDxGQL4EbB6q4rMpQ3XiyOlIYOfXJ6G/90/YSdSSTKEFAEBAE2gMBrzcAFQ49uhkaP0vjerncUf3e6uILviszVBgMjSd3q0CLYXSIaclFYbHhZrdnEiEICAKCQDsj4PUGYM4zOctyqgymJqrnvKrqpxo/szctp/KC1WVjbMYqm6GqLkPjZzL4mwwFv2/r+6e6ZFs5tH52SdUrK4vvOCX0T8/kRhAQBASBDoBAqxuAn3aUjf6/QkNgrYKuqj411vzRtZGa4Ma7Mmt61moPzuLqXTeF9Ulo2gaQ8YGm0RUGWxPnMlZXe0TgqKgylBtO6W+54/QwjWeXwOGCqvjQunM7nO3o+nReQVHLnE/rUq7cCgKCgCDQTATq1lzNzNSZ3GI2BoQbBykD4Pzgz+ERQUZDmNHc1I78nzNplTujwVG/An/lm6K+z+QUlNSdwl5RM2r48w7TrcKjZCoICAKCgA4EWn0EAE+lFYbrRvpdfnyIO3/V/4xWkfe8m7dxW12eFHeaP8eklVStL2y6Cyi/bpX957wbuftmS9klX5bi6Lnszfw35ofbPJhUaCRHeSwICAKCQBsi0BYGAHE4eLBhoer0sDdMop7+97o/HEqepFdpGmPHo5ymDLCV32+jgw/zDHQ8opFEgoAgIAh0GARa3QA4KqsNhVUMAhq+CspqklU1vV/ejprXKl3+hitVngoCgkDHRqDVDcCQ3rase6L8GlsLf98ZYX87rTo0qMlzEo99nJ9dUt3UuYPc0urnLvnTCd2eVNOqrWVTBvo3mnLHoYq4CHNYLBhIqAAAQABJREFUYFOZajRjSSAICAKCQEsi0OoGwGY12sIaXwUf1nTVr2DYdKRyQ151QBMNx8aCquc8hhHvjr2y+o3lRX/90Z7pgQHIynVc/l7Bf84M9W9cbo+ZkISCgCAgCLQ0Aq1uAFqaYdf8iioM28urDU00AIYy13wauLeaDT+nlF/+nX1CiEeOftYyrcqsenZF8dVTA/V8PtYAK/JIEBAEBIGWQ8DrDcCjpwU/1PSZg0pmJjy+yhyG8X1su640XfNhoSdETHu8cHTAZccFH8yo0LmHhCfFSBpBQBAQBJqHQKsbgIOZju9+K+sdaxk7oCHv+YpNpUfyKmeMCIjywF/kLHJyt6Z9OOZM26Rwn3jr+5eGe0IypI9t0pCaIYnupU2elCJpBAFBQBBoJgKtbgAy8hznLC19eqKtYQPwxRb7o79V7Olra6oBwHKUVVR75JpxgopNfk4aH+QU4VEw1LOphjDPknlUpCQSBAQBQaDVEGh1A2BinbzNyFRwwyIEWA0k0/HR7EMrSr7UtxdQ0w2Aiwgb95bnFVSy1gd/0uj+/iH+jcjoQi63goAgIAi0LwKtbgBqxDMZ9udUbtxlZ+dOLMGwPjZN5h+3lTFNajUb04uq9c2X3jI14NqmjwDsej461rj+PZAYa5n4QeHWnKqPTg0S7e+KjtwLAoJAh0eg1Q1AjR+80vByiuOBHYV0lefGmT+85g8DMPbtwpq9coyG7sRVVuv4QPfYowJbG+T6xiXhQaZ3zgr5+OeyU8fVzYO5qWuTWlsSyV8QEAQEAScEWt0ARIaY/jPe5ldbDsaAz6OcSje8PNm/Vv/XxNErDw5oshdl18FyR2WTN1xjFdCgxD/skDNLzmFlkHJLqvcdrqioa+FQgNkw7yj/3YfK3U0Xvq/ULAc7BckHw86QSlgQEAQ6DgKtbgB6xVkvPx4Hf93X/Bl17BBXd9J6Yr/+pQwF3dSPbgvt1fd7YACiIy3HRJjeTHU8/kJ+PeXXHhRT35pSPgSrMsRFyfdg9YInDwQBQaAdETAWFxfv37+/X79+JtOf+ubtyFOHKvrHXfbVO8tLS6qaPDYx1OwjPbKP9bhhAZ6cPtahpBZmBAFBoDMg0OojAG8HcUyyjT9vl0L4FwQEAUHAHQHp9btjIjGCgCAgCHQKBMQAdIpqFiEFAUFAEHBHQAyAOyYSIwgIAoJAp0BADECnqGYRUhAQBAQBdwTEALhjIjGCgCAgCHQKBMQAdIpqFiEFAUFAEHBHQAyAOyYSIwgIAoJAp0BADECnqGYRUhAQBAQBdwTEALhjIjGCgCAgCHQKBMQAdIpqFiEFAUFAEHBHQAyAOyYSIwgIAoJAp0BADECnqGYRUhAQBAQBdwTEALhjIjGCgCAgCHQKBMQAdIpqFiEFAUFAEHBHQAyAOyYSIwgIAoJAp0BADECnqGYRUhAQBAQBdwTEALhjIjGCgCAgCHQKBMQAdIpqFiEFAUFAEHBHQAyAOyYSIwgIAoJAp0BADECnqGYRUhAQBAQBdwTEALhjIjGCgCAgCHQKBMQAdIpqFiEFAUFAEHBHQAyAOyYSIwgIAoJAp0BADECnqGYRUhAQBAQBdwQsLlHV1dUuMXIrCAgCgoAg4JMIWND42uWTEopQgoAgIAgIAnUiYEL786C8vFw9Vrd1JpVIQUAQEAQEAV9CwJifn19QUFBYWKikEgPgS7UrsggCgoAg0AAClsrKSn9/f7PZXFVVpXxBDaSWR4KAICAICAI+g4DF4XBEREQgj/T9faZSRRBBQBAQBDxBoGYEYLG4rgXyhFLSCAKCgCAgCHg1AiYMgFcLIMwLAoKAICAI6ENADIA+3IRKEBAEBAGvR8DE3K/XCyECCAKCgCAgCDQdATEATcdMKAQBQUAQ8AkEZC8gn6hGEUIQEAQEgaYjIAag6ZgJhSAgCAgCPoHA71tB+IQsIoQgIAgIAoJAExCQEUATwJKkgoAgIAj4EgJiAHypNkUWQUAQEASagIAYgCaAJUkFAUFAEPAlBMQA+FJtiiyCgCAgCDQBATEATQBLkgoCgoAg4EsIiAHwpdoUWQQBQUAQaAICYgCaAJYkFQQEAUHAlxAQA+BLtSmyCAKCgCDQBAQ6qQFgC7wjR440jBPHZN58881btmxpOJm3PwWKu+66a926dQiyY8eOK664oqysrGGhIElLS2s4TQs+PXTokCe5eZLs+eefX7RokSe5SRpBoDMg0EkNwPHHH//99983XMF2u/3xxx/PyclpOJm3P0Wb33///UrMoqKiF198sdEjIu688842U6M33njjihUrGgX5qaeeev311xtNtnXr1p07dzaaTBIIAp0EgU5qAL788ktOQvakjk0mH4fIaDRyJJwSkzCYqN8GwGHw5CF6DWTi4aOvv/46MDCw0cTbt28PCAhoNBmScvx1o8kkgSDQSRDQeRgkBwj/8MMPn3zyCUcKjxs3bvbs2bylmZmZixcvnjx58ptvvhkaGnrDDTfwvv3yyy9LlizJy8sbOHDgiSeeGB0drZA9ePAgXbbs7OyuXbuecMIJAwYMUPF00N59913S9+jR45RTTunVq1ejNYE+ogjcF/Hx8SeffHKfPn0UydNPP83t/v37v/rqK6vVetJJJw0fPpxHdHL5/eijjyj97LPP5lGjRZCA3vHSpUt//PFHFM1xxx03ceJEpSgZKHz++edr1qxBJ44cORJZVIbFxcUUsXnz5pCQEBJPnz69YVsCMzB29dVXv/XWW8gyePDguXPnquOaKb2iogIpvvvuOyA95phjpk6dqkqndwwDFPHhhx/CFRXRgCx79+6FVQBBhLFjx5Lez8+vgfR1Plq4cOGmTZvwj0ELevxSZf/973+pUCp90qRJ06ZNQ8nCFSnhEw537dp13XXXde/efffu3e+//35ubi6NZMiQITBz+eWXK7hSUlJoPKmpqb1796beaRUMTYAiKyuLeKp4wYIF9QFIO/z555/xSlHuxRdfDBoMaD799NNff/0VAKmRESNGuMsCV++99968efP69evHUxiABCloPzQb2hKRVPpzzz135ZVXfvPNNzR4cqNSkpKSVG4bNmz4+OOPkZSYU089NTY21r0UiREEOjICOru3L7/88vjx49HOw4YNu/feexmn87qi8i699FLeKHQZbwuqCr1MAnTEmDFjvv3220GDBvGGAwfuWtQBr+vRRx+N2cA2oFiJ543lbUSPoERQgomJiaiMhuHbt28f7ypmBiXLq5icnMxrqUiur73efvttdAp6BC1A/jxSvVe4stlsjfZ2VVYognPPPfeJJ5446qij0E3or2eeeYZHGMJHH330sccewwqCxgUXXHDttdcSjxcFVf7ZZ5+hEKOiolC1//rXv1RW9f3m5+f/9a9/Pf3000tKStBZjFFQQxhC0mNl//KXv+CdB0zUE6A9++yzKh8ER+WhXpFdJa4vf1Kip8h8ypQpMTExGGMc4vUlbiAe3KhZhR4BPGlUmapKNO+xxx6Lioe8vLz8tttuoz188cUXWAiaB+4XagcksT0YjLPOOgu2MWwkxqJQRxgG4KJCExISsFWoeyqIXywEVdZATSmWYAbDBifUNbktW7ZswoQJQUFBVBkWwkUi2lXfvn3DwsL45dH69eshAUAY4BFVjD0gniaNFNTp6tWrYX7jxo3gryYbaOGjRo0CUqoe8x8XF0e5LqXIrSDQ0RGguaPFmnQdOHAAqZYvX66otm3bxi39KRVPr414dBaanXjePZWMV/2yyy7jXeIWw8AjFc8vClQlQyXRvdXib7rpJvq82m2dAdQfc5jaIwYf5Iy6IQZdgNKEE8KUfsYZZ6BhVUrSoJ01qjoDiv9Vq1bxlM4sWgYFoVKiL8iB3j2mjgDWS8WjIJCR8J49e4hHi6l4+o/33HOPCtf3S2JIQEAloKz+/fsjDre4QXhEWeoRVcYtpXP773//m/Bvv/1GGBlVgjp///nPfzKloT169dVXsQQoOGwVqhM1zSNlO4nUktUZuOSSSxis8Aja22+//ZVXXtGSMZ1w3nnnof1R9IzqMDPUBclIwPAIHlRKWMVAYuAxSMQggspQPX3ggQeoVhWm68B8gwo38HvNNddo0gE1oz1VKCRUNPmjtQkzKsWK080n5qWXXlIZlpaW0oegT6Pl/+CDDzK44TY9PZ2UjzzyiHqkqpuBHbe0OkYGKp7xEBaa7oW6lV9BwFsQ0DMCUOqPXjDvBhd6it4rPSNedW7pdvGL8qULT2DOnDn8cqFlTjvtNDrOBQUFahDNXCL9XGhZbENvlzSoDHQl7z/WBdVAr5keWS113T8ZGRkoMnKjj0kHE0ewGoYr/YsKwPECJ6p0hizqfVZ58bTuTP8cqzqeLJK577776DCqh3T9YBV1HxkZiRPjoosuwm0FD7iYlH8JhULHEGuH8sLlAhuMk/6ccd13M2bMUA8oizGHGrKgl4nHJCAjFzijFtWYiTCZM4SCCoTrzrQ2FrQZEWEOcY9Qg+RDG+W2AZL6HkGlCOmeoysvvPBCbun/rl27lv472l+1BNCmWgGQZIcPH0YKRkIqT1idNWsWvWweQUIkiNFg4AoY6YxTraqyaotqnElqU7FE68JaO/uLGJSQP4MMfhlM0HPHjUOVYcaI4YIN2GMAR/tRDMDMO++8w6BK1T5DFpWS8QQDF1odt9Qv1hoDTPefbBkIMsBVyeRXEPAWBPQYAHqIuDWc1Q1OGwRGofCrFC4BOlb88obzqy5G1gQY6aM9f/rpJ3T9zJkzw8PDeRtVp0y5jBlDMA7gZcOzROL/UdfxP8aDWDrdeCF4/chWqRgVzyN8CBqZxpgW42EAQRgHoJWc0+M6oJNLzMMPP4xSoMMID+gLNbjBF4F3mDS4dNAs+AfwgDmT1xfGOaM9AmSMB2ijE7GU2Dlk5CJbtL+Cl8Qk00gaCDCgueOOO1BVpMdoAbhzDTZA2PAjfHrYFbJldufWW2/Fumg1Thi3iSJHLxPAXmq5KbZBTI206CjQk0A6YARPMqEHoCX2PIDWZizF2EIjUW0A80kM2WKqR48e/dprrzFmVWkUkjQ52o9igAbJIxhTBgDXlkrJr9aKcNMx78LIA98jHioGN/oY1nKWgCDQ9gj8oZ09Lxt1r8bCGgmdO7pF6s1X7wyP8JloCVRA9enUC4lDgG41+dBtpJ+L7sAtQDK8/zg0SInz58knn2T07ZKJ8y1GgltKdxlwkblzMhVW9sk9vtEYNburTJSWmG6jEhADdvfdd9PnpfNINxDnA952kqH0cUrQHeZLAhwayKW6uloOdQY008VTwEFjImO3bt3wXbjIqCYbiNQArzNDLZIpEKZS6WXDEmMXHGING1eNsIEAWTFXT70z5KIPTpUxUkGfalCrJkEOXbp04Vf1nVWGymNOSjWuog/uLCC5KSWuZdUAGzwimcIBU8StctMrEjUyUK0Ohv/2t78xDMW7yDBOZa7qEXPuzABhYFdDGeeiFYmKYSTBLZ5PvEy0AeWOc04sYUGggyOgxwAwFYlUakxNgDeZdwkVoN527Q1R02s4czQI6B3jBUIzsrJCva4oOIbSuEcwAygOfPR0xEhPRxjnD0NslLsyDFomzgGGFDhAWMuhRTL1R86eTMdpXTmNts4A4pAhLhd6+hondMnR8vQWUTQ8hUl+6T/iCCYTersofWLov6OPWM+Dv1jF11mEc6T6IIsYOv6IT1+V8NChQ7GFmm1AcZO5mmt1pm0gTL8YVnF6UCmwhFBMw2LY3BVcA5loj4AOBrhVY6D58+czPKL2UbVwpSl9LT0B+vu4g+gyq0j0O2HGNDCg2glLfbT0//nPf8hfZa4UuvaovgAsqYaHPcaj+MILL2gp1Qcf1CAxpAkODiZAS6Oa1Ex+z549iXHGU2ufSkwtKxVQkfwyoiKG5QyMBpgrZsTjklJuBYGOjoCOSWDeIr67QTCUCAqL1xgvMG8ynWIi0XokUBfzbMTgk8XxiluWMG5WHqmUzB/iCGL6kXg1dcwtYVQtq/rUyg1tqvl/Wbr+T78bEtaTUIR6n1n9qRIR7zx/yIzCOeecox6xMoSuK2tp0IyuOf7vXrkmVq5cSQSzfPRhsXw4fBXDajZbTWZSEPO0uGWYcyZMnnxMizsL9Y3vCBLWNXKR+H951/G/mgSGnNWxmEPsIiZQzceiWBlA8Ijc1JQmt6qjzSCDpZ91ZOcWxZQvObAok346ZlgZWvrd6GLi1ZS4mt1x7wu7ZIbfAxI6vJgikMRpg5IFKHzljABwf2GrlPpW6Cly9fEwMzFMw9JmmC9hJaiaBIYlMmTuh0pkCpcwKzIVFf4lbmkS8OnChvOtapD0JKgpNVZjRRY1gisfchqGSsy0rTYRTV+BRyxh4JEKM9ykPSsGFCCMmUhDs1TkVCtmjGbDLQ5MHlFZmH+mwQlDq5LJryDgLQgYMQC8jTTfpl44tdFHvCGsSuStoKdGv5v3lgXRzq5eXmmW0PGaMUrAz8sKP1UQ+o53jKUsdJxRCjhSVTzuIPzd9K/RLGhAfhtljL4tK/R5DxlboBmVpoOKlxO/B91nlQNsoLDUpDQ6gtLpA6JfnOcJnMtChbHegzlt5reJR1UhHTqFzibakzlM1RNEEWMSUNloKLgFCuUyRglSBCqVLyRwSaGm0YzO+buEMYr0o8mH0QOlIAVIqu4qKcmc0tFEmATgQgrl/kIoDBUpXXJzv4Uf3N/kQI+bKsM7/8EHH7BKh2lzrC+cM5OMNwNViOqsDxOVLQ46OuxkyOwLThXyIVtwZrqV3LDcLAVGakZmsKrQU4Q44rFA6GhMKWUBvmb2GFShUjH8MMZEjrZyH68RGWKo6LM3MBqAGVgCQ4wTzY+mCPi0PWYmsEnaUgJqilpQ1hSW3njjDW7hljBNFN4Y14IPgqhPRmghSMFMFd0F0mCJyRYZkYtbKgsSRKB+yZOWTKRcgoAXIaDfACgh6fjXOeR3gaC+ZOi1Ol0x9cW7ZOt8W18RzmmaH8awk4m7Z6CBePfEdbKhDADzBHgkGpCFgjzMsM5SdABbZz4ukZ5ki/OdiQccYoqWTj3aE8vtnFUDgjsn8zCsI7e2IfGQf0kmCLQ2Ag0tHPSkbE+0P/nUl6xO7U9653hcB3Qz3bUeqpD5Q+1b2fqKaEAKsqVrWWcCNBqTkO6Fuscock/i6bxrU+XOhSIIfVuVA+XyqAFZ6isIKoYCFOGegPwZLqhvsJ2BdebBOYwLC1jc8yENWalZX+f0hD3JFlp8PowwGGrgMsIBiAFwyac+wYHFZRJeI0RlM6x0X3FAgvpy02jdA21D4l6uxAgC7YJAcw1AGzCNH4Z1Qe5rFvG9sFIeb4ZuHvBI4I9mXO+u7FiBw0rBOtWK7uJweuBl1rw6Wj6oMLQzH0YxScvErBbf1AAOd6Yi3HnGzuEGYcmKhxkiO6v73fOBHNvAFIs+Jlknw+CGBf7MEuECwu+veWYaZQzne301hfOHxVfq+49G85EEgoAg4IxAc11AznlJWBBoFAGsHdqcGYJGU0oCQUAQaG0E9CwDbW2eJH8fRgAfi2h/H65fEc27EBAD4F31JdwKAoKAINBiCIgBaDEoJSNBQBAQBLwLATEA3lVfwq0gIAgIAi2GgBiAFoNSMhIEBAFBwLsQEAPgXfUl3AoCgoAg0GIIiAFoMSglI0FAEBAEvAsBk/s3UN4lgHArCAgCgoAgoA8BGQHow02oBAFBQBDwegTq2AqCPVu8XiwRQBAQBAQBQaAxBH43AKL0GwNKngsCgoAg4GsIWJgDQPury9eEE3kEAUFAEBAE6kfAgupnfy6zmckAmQ+oHyd5IggIAoKAzyFg5HgmNig+lGnfsTu7ssLBYSc+J6MIJAgIAoKAIFAHAjX+n3e+OHDOvTsNTAeU15FCogQBQUAQEAR8EgHj6o2Zk67dbAgy/W1O9NDEAFkB5JPVLEIJAoKAIOCOgOXJD/Yb/Awf35p8yvRu7o8lRhAQBAQBQcBXETCt3VNkKK6eMSHeVyUUuQQBQUAQEATqRMBkJrrS4GeRud868ZFIQUAQEAR8FoHfl37Kt78+W8MimCAgCAgC9SAga//rAUaiBQFBQBDwdQTEAPh6DYt8goAgIAjUg4AYgHqAkWhBQBAQBHwdATEAvl7DIp8gIAgIAvUgIAagHmAkWhAQBAQBX0dADICv17DIJwgIAoJAPQiIAagHGIkWBAQBQcDXERAD4Os1LPIJAoKAIFAPAmIA6gFGogUBQUAQ8HUExAD4eg2LfIKAICAI1IOAGIB6gJFoQUAQEAR8HYH2NwDff//9jh07fB1nkU8QEAQEgQ6HQPsbgAULFnz99dcdDhhhSBAQBAQBX0eg/Q1AfHy8n5+fr+Ms8gkCgoAg0OEQ4CBgnVdJSckvv/xit9t79OiRmJhILoR//fXX/v37BwUFqUzLy8u3bNmSnJwcGhpKzO7duw8fPkwgOjqaZCqN/AoCgoAgIAi0CwI6RwC//fYbWv7ss89+9dVXe/fu/cQTT8A958vfcMMNL730kibJ+vXrR40aZbFYsATXX389lmDhwoX333//gAED7r77bi2ZBAQBQUAQEATaAYEeZ640jPqqvKIS9e3hlZ+fD6MoepX+4MGD3K5YsYLbt99+e9iwYah79eiUU0557LHHCH/00UekycrKUvHLli3jNi8vj9vjjjvuxRdfVPHyKwgIAoKAINBmCOgZAaxbtw71ffzxxxcXFxcWFuLPueWWW9RELtp88+bN27dvJ0FOTs4nn3wyc+ZMwrNmzcrOzo6KiiIMiclUU64yJATkEgQEAUFAEGh7BPTMAWRkZMDo7NmzVRfeZrPt2rVr6tSpzAp06dLl1ltv/fDDD4cMGbJmzRqSDR48mN+AgIBt27Y98sgje/fuXbt2bdeuXdteVClREBAEBAFBwBkBPQZA9d+XL19eUVGh8rJarUaj0d/fn9uTTjpp0qRJd911F17+9957TyXABTR37twnn3zy9NNPZ8Y4JSVl9OjRDHOcWZGwICAICAKCQFsioMcA9OrVCxbx/3Tv3l3x+u6776L98fhzO3LkSH4XLVq0adMmhgUqAXO/DzzwAPPA6pbJYQJVVVXqVn4FAUFAEBAE2h4BPXMAY8eOpTt/9dVXp6amFhUV4fBhOZDq/iMAASZ1iaGzHxsbq0TC58Ms8YEDB8rKyvj0l/kD4gnzW1lZM/+sksmvICAICAKCQJshoMcAmM1mlgANGjSILwBCQkLmzZu3ZMkSpnk1pqdPn05Y6+8TvuOOOwIDA3v27MlkwNNPP71//34iDx06xG9ERAQeJAJyCQKCgCAgCLQlAkaWgR7YU1G+drrV0mRjgA+HtT2sAvKQ49zcXAyANlbwkEqSCQKCgCAgCLQGAnrmADQ+mA32XPtDRWdfo5WAICAICAKCQPsi0ORef/uyK6ULAoKAICAItBQCYgBaCknJRxAQBAQBL0NADICXVZiwKwgIAoJASyEgBqClkJR8BAFBQBDwMgTEAHhZhQm7goAgIAi0FAJiAFoKSclHEBAEBAEvQ0AMgJdVmLArCAgCgkBLISAGoKWQlHwEAUFAEPAyBMQAeFmFCbuCgCAgCLQUAmIAWgpJyUcQEAQEAS9DQAyAl1WYsCsICAKCQEshIAagpZCUfAQBQUAQ8DIETA624jfyTy5BQBAQBASBzoWAqXuUnyHQuGV3fueSW6QVBAQBQaDTI2B887N95z22e3L/gL9fnNS3e1C1oeafXIKAICAICAI+j4Cxqrr6ygc2/WdZtqGq2lBzRKNcgoAgIAgIAp0CAWNlpaPaaP7km4PLf8goL3UYZTagU9S7CCkICAKCgMGYlZVlNpvCw+WsLmkNgoAgIAh0LgQ41dF06FAap/t2LrlFWkFAEBAEOj0CJjP9f5N8DdDpG4IAIAgIAp0PgZoRQOeTWiQWBAQBQUAQMJiMMu0rzUAQEAQEgU6JgBiATlntIrQgIAgIAgaD+H+kFQgCgoAg0EkREAPQSStexBYEBAFBQAyAtAFBQBAQBDopAmIAOmnFi9iCgCAgCIgBkDYgCAgCgkAnRUAMQCeteBFbEBAEBAExANIGBAFBQBDopAiIAeikFS9iCwKCgCAgBkDagCAgCAgCnRQBMQCdtOJFbEFAEBAExABIGxAEBAFBoJMiIAagk1a8iC0ICAKCgBgAaQOCgCAgCHRSBMQAdNKKF7EFAUFAEBADIG1AEBAEBIFOioAYgE5a8SK2ICAICAJiAKQNCAKCgCDQSRFoMQOwcuXK7du3u6O4d+/e5cuXOxwO90cSIwgIAoKAINCOCLSYAfjHP/6xYcMGd0m2bt163XXXVVZWuj+SGEFAEBAEBIF2RKDFDEB4eLjVanWXxGw2x8bGusdLjCAgCAgCgkD7ImBpqeKNRiO6PjMz89ChQwEBAX379iXGPfPCwsKUlJTq6upu3bpFRUWRoKioKDU1NTk52WKpYSY9PT0/Px9yd1qJEQQEAUFAEGhBBFpsBIDSf+ONN84666znn3++f//+9913nzuXhw8fnjZt2kMPPfTiiy926dJl586dpNm3b9/AgQMxAyr9V1991a9fP3daiREEBAFBQBBoWQRabARQVVUVHx//4IMP4gu69dZb+/Tpc/LJJw8fPlyxy2jAbrfffPPNV1999fz584mcOHHimWeeuXHjRpOpxghpwwWGEYMHD25ZISU3QUAQEAQEAXcEWmwEUFZWNn369IiICFR5UlLSZZdd9sMPP2jloeXx7bzzzjvjxo3Ly8vDyTNo0KBNmzZBpal+LbF7jPZIAoKAICAICAIthUCLjQBgCM++xtawYcN+/vln7RadfuTIEW5PP/303Nxchgu4gLjNyMgQda+hJAFBQBAQBNoSgZY0AMqZo7hnXrd3796aJNgGpfE/+eSTmJgYDACPSI+/iHWiWjIV6XwrYUFAEBAEBIFWQqDFXECsAeWbL8Uljp2HH354zJgxGtMYALxD3B48eBClH1l7HThwgBic/vxqX4rt37/f399fI5SAICAICAKCQCsh0GIGoLy8/Pbbb1+wYAF9/PHjx7MoaOrUqTDNJ2A4f+jyYwCWLl16zDHHPPvss59//vnkyZNZDkSCxMTEkSNHXnrppUuWLLnqqqtuu+02loq2krSSrSAgCAgCgoCGgLG4uJhONysvnR042mPPA99++y2qHF3/448/9urV69hjj1UdeYYFLPTEGKiePh8BMDnMPDBLRVkSqvJnTnjZsmXQMmhgbMDIAHLPi5aUgoAgIAgIAjoQaDEDoKNsIREEBAFBQBBoRwRazAXUjjJI0YKAICAICAI6EBADoAM0IREEBAFBwBcQEAPgC7UoMggCgoAgoAMBMQA6QBMSQUAQEAR8AQHfMQBsJ/fYY4+VlJRo1cLuQ3x1zDKnp59+OisrS4tXARahsnGpS6TcCgKCgCDQeRBosVVAq1evttlso0ePbi/s0PUsIWVFaWhoqOJh4cKFrG1lT7qQkBC+TGYDamfe2IKCI8xk51FnTCQsCAgCnQqBFtsK4uuvv2Z/f2cDwMdfKFkuZ0CJdP7ggC+EXRKoxM60LiTOt85h51JU+IILLsAekAYtr75CcCnOuWiS1fD6P26dUzqH3UuRGEFAEBAEvBSBlnEBzZkz5+9//ztHP15++eUAwb7/bPWMzkXXs+GzgiYtLW3WrFlEXnnllddff/33339PPAl++eUXF+w4V0DRfvDBBzwi57ffflulwcywnRzh3bt3s+coyThLoM6ziDdv3syXyYwGUN9xcXHvv/8+yp3i3nzzTZfi2IUCd5Aq8ZtvvuEpX66RUh1jyRE3bFzKVqYuVHIrCAgCgoDXI4CLnO3YUHYoyuZcbP7z0ksvkUNFRcVRRx21aNEiwqtWrQIgdokgf4zEW2+9ReSOHTuIxGVEmM+G8doT0C52iZgyZQq3ancgzAP6fe7cuWRCJCcNkCcBcuAIYgKfffbZzJkzlbufSLr8RHJhjdauXUuAr47ZaoJpAPr4GCHSsA018QTghMATTzxx9913E2BrUiJJw+fKBGCASLWJKZGE5RIEBAFBwJcQaJkRgFKX6GgCnOyIambPH/Ssn58fMcSjQNnq5/jjj+eW4x4vueQSQCSMy4hdgwioC5373nvvXXPNNXTw2RDihhtuYADBaZFsIUdPHFPxyCOPDB06lMSQd+3alSII09kvLS3VvDfEcIvG5+wBlZLNhebNm0cCjqzBCGEYiOcihjyZKuAkg127duXk5DBhwN4VdP9VApVGNqfT0JCAICAI+BICLTYH4AwKShz9y+FfSsvziBEAul47NV55V5xJVBhTQY8ebwwOIuWUx4eDOsabv3Llyp49e+I7CgsLIzFHj3HewGmnnYZhCAwMdMmKxK+88ooWCRtaicHBwT/99JN6hAFgxEA+OIgUb927d6d0jVBL5hIjt4KAICAI+AACLWYA0LBKddL75mRgteoGJX7eeeehZxkKrF+/Xg0RQE25d9zhY/QwduxYthR1nkwm2QknnMAhw5wocNFFF3GLj/6vf/2rsi7MN9x7771EOl+zZ892XuKp2SHSMJgYMWKElhjjERsby2xBjx49tEh2xyMM2/wiFBJpjyQgCAgCgoDPIPCHr6OZIuHY+e2337Kzs5UZoJeNasa9TrYs0MT3gvv+hRdeIBLvPzOxSr2ydzSOF61o7AS+e6Z58cPgfGduVs0hqxWcK1asoBQSqy7/unXrMDOPPvqoOmJMZcJwAfWNmwgHkZYt61OfeeYZTAKdfcYlkyZNUo8wWown/va3v/3zn//E48R1xx13kCfjABIwGwG3HGOp5SMBQUAQEAR8CQHznXfeiRuEzrXSyLplox/NEh10KBv9M2eL3mT3f5z+bA3NNtE8ZatnnPXvvvsu/XEW7NPHp9ONXqY/zq1WLqfJs7AHBw6TvehxtozGoqDWUfrcMr1MyqCgIPaXZs4ZG8DKIkhw9zN6ICXpKeLcc8/VziNjUpqByKhRo15//XW8/0899RTHVZIJ/foJEyZgAGCAMDMB7FPN1AUpgYIxB0uP4EFlTqS7o0njWQKCgCAgCHgjAi32IZgSHuVenyHBb5OQkICrHf8Pv9u2beNIgOZAVl9Zr732GlO+GInmZC60goAgIAj4PAItNgegkKpP+/OUiVlW2eN+wZOD16WZ2p8M6ytLzRP4fM2JgIKAICAINBOBFh4BNMwN3nk8/qz7bL72b7ggeSoICAKCgCDQKAItPAJouDzWcXI1nEaeCgKCgCAgCLQNAi22Cqht2JVSBAFBQBAQBFoKATEALYWk5CMICAKCgJchIAbAyypM2BUEBAFBoKUQEAPQUkhKPoKAICAIeBkCYgC8rMKEXUFAEBAEWgoBMQAthaTkIwgIAoKAlyEgBsDLKkzYFQQEAUGgpRAQA9BSSEo+goAgIAh4GQJiALyswoRdQUAQEARaCoE2/RK4pZhuIJ+srKyCggK1TRC/HHjJrnBsFNoASX2PIGejaQ6Q0U0OJ2xwzYFizgcS1FecSzylFxUVqW1Q9ZFz2BnboMK/PvKysjLOb1CHKrvw5sktW/7Bf0REhI7SyR/OOcszKirqT+Sc6mwvMToqDEZDlX9wtbHe7gtU1J3zLrOe8OycBvLw8HDVkJzjPQzDfEhIiNoa3UMSLRmFskFvQFCwzVAZsvc3Dqc2OCpKY7uVxfQwVFdpyeoLQN6+DY96Z9t23dDVJ5fEtwYCejRja/DRUnmyrzWXlhtbUrNxtPOpk9ojTwK8h9HR0bo3FmWXaXQQStCTstzTcDIB9oOdtN0feRLDQfYclcwOrJ4kdk+DBuR0B93kmB8u3eTYjzpKt5dWZhQb2PY7Ks4QVHMwXH0X9gNydYxEfWkajoecIyX0aXByxgDExcVxEEXDpdT3tLC0LNJqMPzfXfaUFYayw0ELPukyouZ8Uw8v6h3rpdv+cWgS+/U2p+FhgMUAeFhZ7ZvM1wyAC5r0oP/UhXR57MFtO5JTdHNKbw4twDSTHBXsAbr1JqmDvLLSkZHK0MAYEt6w9teYRwR9akjJ3o4IBO7/zfHuNdXGCGNoz7AHvrd0TawXqXoeNJP55pA3h7YeaSS6tRDwcQPQWrBJvm2OQGVmrfb3DzRH6BwStTnLOgss+fL9oHduqjYFWIceGzr/LmOIzhGkzuKFrDMhIAagM9W218pamX3k/9s7DzgpiuyPd0/Y2Z2dndll2SUjIIhIEEXU/wliOAOmM+KJpx56ZjAgenrHqXeGS3pmwHDqGU9BBbNn5DwjJgQVRdISd4ENsxN2Qnf/f7MFbTPTk6pndnpm39x9sLe6XtWrb1W/V13dXU8JdSBAs6W2DwJBFG070igu+9t8T98d/vhxtNB+zNWeUy5II0CniYAxAuQAjPEj6fwTUPxexd8Ku2+t7SvasDRemr/olob2B/4gNXxicda2HfL7nkedVprtpFaZiQA5ADP1BumSSCAalrZvFhTBUt1TrCjZMJ8dn/838NTNcusaa7/93Bf/pckfFWQpEQalEIHcEiAHkFueVFqOCUhbN+J5NEy/xVOb46JNUpwi+197OrjwZlj8sv87x3XaxbGWrlhhEu1IjdImQA6gtPu3aFvHFvpbm5RwSLDZrXWcL7OavP2yt9n39F3hT58UrI6Kk2dXTj5DSP5xg8nbQuoVIwFyAMXYa91BZ9ES8Mq+SnzzZe3ZtyTNYnTLeu+dl8jbVoqega6z/uTYZ0J36Fdqo6kIkAMwVXeQMjsIiOGgFQ9+hb6Wml6io6L0uHR89p5v7mmCpZe1z2j3jDus9f1Lr43UIvMTIAdg/j7qfhoqMpb+8SqkpcIllN5b8FLUt+ifHc9fJNgHlh1wctXZvxXLS/bhdvcbu0XWYnIARdZh3UFdedsmRYoK9jIBb/2X1i+6eZ33nhly0w9KRKj89W3On9O7nqXVwcXWGnIAxdZjpa6v3NokB31Y9I+460qsrYE35wfmXyfIZZbq/lW/e94+ZK8SayA1p+gIkAMoui4rZYWVoE/2tuCbL0tNveLdUDJNxZfMvn/fHfligaBEyg65sOrUi0RnVcm0jhpSvATIARRv35Wa5kqk85svQbFU1SqV1YKyvjRaGPr6I/9jf5Rb1ljcfZxTbywff2hptItaUQIEyAGUQCeWSBPk7ZsESRLLnRZPT2wHXeytEhGFItzRvmBe6L0HhGjAPnaK61ezrD3qi71dpH8pESAHUEq9WcRtkbHdW7hDtNs73/oX8fVvETcGqtvs0VXftj57e3TdR6LdWXHmHRUHHyPaOcMDFDcK0t7EBMgBmLhzuo1qir9N9sW2e8Nb/4K16MckXmFyf/lO6L/3K6F26+6TXL+8yr77yG7TmdTQYiJQ9BdbWth8IUFYsYgng1/aKpJlQNVGxDsr568d4gbb3kXiUkRuaQQpa3WdgBf/O385UZ5bf9ZxfOLS9kb/gjku7O5Q5i4/6orKk84XysqTjZBk6Qabz/RPVnjadIPiUD5tFZTBJARKzQFoYwJjICKqLeLbOZ1OjihFEN+yZQv+RUhIbnFE1kVcYj7xTZs2ITIfVsN1wmOlGz5QGyEhI5EIghpyiMMEIKghfoimySeOVkMBj8eTWhxfe4mI9AIfYK8QLJVCmx8tQ+3ouKamps2bN0uIiJvlD+KABnF0fZaiO7Kjv6A8gvqCf6Z9J1rwv4p13wn/vlHxN0uOGu/hl1ZMPLZlWzOimGWlBvTHwEO9fPGc2bgFf4TkzFR5jX4QB3YMG76Bx5TXhmXVlE2HpiNQag4AwxcXLUYhSOMYBxjK+HGAhziTNSiOmPIc1yHTHLKoHZpkqz/EIQvjyy3O2s4KybZ25Gf1phMXleZNoiKLZQ5LbC+EWK+xHwQZf9aVO5Oz+K8RcXCDOHRAKzKq0mpTOvzS/Pvk5c8JsmLb69gfx51QP2yEFTHc0XdZdp86aKEA38hhbecWZ13PejCj5msysYGnSaBDUxMoNQeAQNjaWNjLly/v06cPd1B4RAaHOG4g+Pqwvb29trYW4bn5xDELNhIUHpcigoNzh/ZGQHO40vp6zrdWMP2E/mnE21ukiFOoclr77h639I/pJ+CnEU+OFRYc94Lou+RZ0pxhXZ+hA5DWrmj/52yp8XtFaqs891+Y+Lu++65vXR13UPi2tjbElMcdQBotk5wGfIy6mhrOWJKAD+fBPXK4fXaS1lByHgmUmgNIRMUxh1ILgWzqFQw1p+6BcXEjykNzM4sroaDU0hiL9NKjd5z1B0yDyqPh7MdnjCCbsQKK/7WngvNnCUKFdcjEqmk32PoNgf6KJBkZORnXrjvu8AqV0XGLEvSLziDVSMMzKJ6y5JJA6TuAXNKisnJFIBqJRXrBMp2r2lLFeYeUK124y5Fbt3rnXBtd/T84rPLjrnWdMA1vf3KXRoJEoOsJkAPoeuZUoyA1bxGkaGzpv0evIsWB/Zz9D89UQn5L9QDXRfPLho0p0oaQ2t2ZADmA7tz7hWk7tnvDI1Ms+1gQ54tF/iqMIpy1Iki97/kHQovvF8Kbyg75fdXp08VKN2dZJEYECkqAHEBB8Xe/ymW/N7bdmyBaEenFVlZ0AMI/LPU/c4e0+h3RWVd5/rzyAw7TvrxUdM0hhbs5AXIA3XwAdGnzsd0bpv94RGpx14qVRbYdJnaqCC5+KbjgRiUSsI05xTVlBnve26UEqTIikFMC5AByipMKS0FAUWIPfqNRsaLSgo9+i+onbdvse/xvkWXPiY6eFb+4qvL4cwRLZp8IFFUzSdnuRoAcQHfr8YK1V8aD32gI78lY8d5nUf06Pn7T/+QflMA2S/1Y13m3lA0bXVTqk7JEICkBcgBJ0dCJHBKQ21tkXxse+cY2+yyWdyU7v+9tf+rO8PtzEcjFcfg1rlMvFHfuVpRDOFQUESgUAXIAhSLfjepVohG5uTG29F9dLzoqiqPl9rLIquXeu6fJ/nbREnRdusix74Ti0Jy0JAIZE8h6k5mMS6aMRKCTgCJLW9biSMQ3X+4eRQFFkaM1Hy4M/HG03N5o2/Owmn/8QNa/KDqOlMyWAN0BZEuM8mdHIPbgV5bE4ln6lxrXex/8Q2XDF4pFcE69zXnElOwaTLmJQPEQIAdQPH1VdJriIy9fqxL0xyK99BpYFN98Bd9/JbDgz4pvs1QzvPqKhY7+sY196EcESpWA2R3ABx98MHDgwAEDBqADmpubX375ZWyQiWNs8oXd6idOnLjXXnuVat8Ud7vQQ0EfdstBK6y1fXAHYPLmyC1bfQv/Gf7fg4ISdhwze92QcbV9B5tcZ1KPCBgkYF4HgB1xP/roo0WLFs2YMYM1Etvbnn766Wx/R/w7b968wYPpEjU4APIlLkpRS3uLUO/BXm/m3ykhvOIL3+M3yxs/tvQ9oHLqdWWj9pe+XoaVK8Fi3gskXz1H5XYnAuYd3y0tLXvvvTfm/urusjD66gbrCxcuPOuss7g3+u9OXVyItmI34eYtohzFOz+iyd/6j0Z8Lz4aev12JeqzT7jEdcqF1pq62GbIiOVCPyJQ6gTM6wAGDRoE+Fu3bkV4o7heaGhogCfgDrQSV1rqP9kNR+o8JXnWSMOVtm1KKIBlH7GWMySLkdrV7khbSHRLg++fN0ZXvStW1Fee9Y+KiceossYP0tZuvApzltBtG27O7kitlXkdQAq9n3zyyeuuu043A+Zuselb5495DhZbQ03UldJNhDikjIirsny1Qxw/6Mb+1VUyWaKqPJ84ruHOyrOvXRQVX5vctl1WhGhVDyyhcCjPamfQOMRZkyGOeML6xki0YIIf/uRN/6OXC1jmGTTBdf4t9j4DEUAHsVQgDln8UDU7SAY5RToEUTufOHQ2Uju6HppzK6+KM4wp2qh7CuIcYZx1i6LELiBQfA5g48aNI0aMSIZmzZo1CGmNUYgM+Bex8ZYtW8YdFH716tVer5dbfP369Ygt7na7cT0nUzhZOpRHQxCUEfdAHOIwIghqiKDw0B+2IFktydIhDkE8hkFQ+yzEIRYN21u2ILavT7Rv9IUjFnsW4hptEJUQXYnAhHzikFq1ahXU0RS58xCF+lpr/vdc+bLHBWul74ALvQdOlrZ5hcavduaIOQB0PQxZhiEhVUH1AMoHg0Hcp/L13bp167AEWllZySeOawThJD0eD594U1MTGs498DDmEc9SRUEHZiZQfA5gyZIl48ePT8YUC0d4a4hd+RjEb7zxxqhRo/guJIjDCqM0XEscFxJky8vLWYxiPnF4DrzyhMC2HEaQ+Q9E5d1tt904xAEQ71zBheyxxx4Zi4uYVkuN65QelVZnVaDCU9HQAPgc80HUDs3x75gxY/jE4T8wQkaP3nXTHpRosYS+/bxj/o1y2wahsp9rxiPVI/bFHRab+KuDivUXXjCDA+boO5QD/kOGDOGbOkBN1Nu3b1/uqQPqxQIp4lFzKA/NscQKLwkFMu56lVys4Zs2bfrpbzoyN4EicwAY0JjRn3jiicmoYvzhp57FtcTsuJqS1QGKKqw4FGBNyEptlplpbkQcJWQlLm/dZJEioqPcWt/fGgiwjkAh3MpD0Ih4vKwi+/59T8frN2BjH/vPZrrP/Z1od8R00wwYpiqGmfGuhw1lBFiZWf0LzfEzIg5ZI+KoPauu17YODdf+ScdmJmD2rsIEUDuLwcRww4YNXQlUWztHvQbFOWpURQxWna247G2Wg77YN1/Y7q1zFVvVhOMg29p1q0AhsGLsVGTd976Hb5QaPhTK6lwXPlA+bpKuSA4TjTTBiGwOm8BXVFErz9fk4pUyuwMYO3asFi4WVefMmaNNoWMzEFBCQbkFkV4ES21vsazcDCppdFAC7ywMzv+j4vvOvt9019nX4kVPzVk6JALdl4DZHYA6g1O7CDen6jEdmIKALEvbNkIT0eWxVHpMoVKnEhg8SnuL9+GbI0sXYJ3Hee5858GTBavZv0k2D0DSpOQJmN0BlHwHlEADpa0bYnG+sPRfa5p3P7Dyg+e9yz7xP3yd3LzMOvQ411nX2gcNLwHa1AQikEMC5AByCLM7FiXjm6+OABb9Y5FeBL3XLgtBBSr1WDzft+w5JbzeMfnPrpN/Y76FqUJwoTqJwK4EyAHsyoP+yoaAEvDhmy+Y/c44X7FN+szww0fI/kf/XPnlU4JnUNWMRxyjDzCDVqQDETAhAXIAJuyU4lBJiYal5s34bFasqhGdVSZRWgmH2h+/PbLkMbn/wT1m3W3z1JpEMVKDCJiQwE+vzJtQOVLJvASwU8L2LQL2WiirsNb0MoueUrT9yTvC/5sj9h7fdMIMK1l/s3QM6WFSAuQATNoxJlcLcb5iS/9Wm7Wuv3lUbf/3PeHFd1pq96668t5ouSvu+17z6EmaEAGTECAHYJKOKCY1pKb1Suc3X7Glf9O8leubPyf01t9F91DPNQ9a6/sKUmw3CPoRASKQggA5gBRw6JQOgU7rH4vyaK0fIJY7dXIUIsm/6OGOV/8iVg7w/PYxa68BWJsqhBZUJxEoMgL0ELjIOqyw6kpNG5QOv2ARrXUmsv6Bt54LLvqjWO5xX/Word+QwiKi2olAERGgO4Ai6qwCq9pp/WO7/Vh79jfP3D/44RuBJy8RrWLVzCftg5PuE15gdlQ9ETAlAboDMGW3mE+p2FNfrPtj7t+zn1hRaRIFOz5b7H9gqiCGq658tWzYGJNoRWoQgWIhQA6gWHqqkHpK2zcrgfbOuT+sv6uQqmjqDi390Df3LEFqrpr1Udle4zRn6JAIEIGMCNASUEaYunMmuXkLojzGrH+diax/+JvPfPedK0QbK2e869j7wO7cQdR2IsBNoPTvABL3E80cFmS5o2qgFuPiRpSH5obFBcG7XfG3iVYLVn6E8ixWfnJRe9KdhcLff9V+73mCLFWe82zF/ockdigazn6JpzJJ2SmdVIG0haAEgyMHJaStJVmGwtZupOHJWkTpeSJQag6gtbXV7/eziwcbRyMoLiKU8oWExDhGWMSKigpucYRUREAb6MARIgO1b9u2DSEh0Qq+yHyI6Yr4OdCfQxwAQdK7Yc02v0dWZMVdJ/qCgtef4SiEOHoB9NAEvtqhOYtJGR8S0mqzb/hReegKIRIKH3i+PHZie9NWbEcRpxikII7YtnHpGf6J/oI4Rg53SEgmDvgcXQ96EEfViCrMIc5GDgAiLia3OKJ68Q08KA9uPXv2zBA1ZSssgVJzAKFQSOsAYH1wFWFQclDGhYSQ6AbFWQkc1yF0hiwMGeLa84lDc/gebvHw1s1S27aOSofiqZOtDsGfqfVnqFE7azuHA0AJ0BxdCeV3EbfahIbl4hPXWMJe4WfnRY46M9zertuz4MbE+boelUKcAdQtP20id7+jZNb1qJ0v9AXEoTy3AhBH1XAA3CMH4mn5UAaTECg1B9Cr86fC/fHHHwcNGoSJmJqS1QF8CYLCI8R2VlJqZszCEJgb4bnVlKwO4IFwB1BfX5+VlJoZIemhQP/+XFs1tLe0tjtbhV79R48TuJ76wnzAlAwYMEDVJ6sDzF59Pl+cuNLc2LrwNim41T7pYvevf5timze4TIij67OqVJuZifOZYJSDkYPaEcBOW2bmxzDfiMnucnE+b4f/w6irqanJvEZtTrQa9x9GBp62NDo2M4FScwCJrDmmz2ohmAnip/6Z7QGqNiLeWTl/7dzicutWBQF+obunjs/6AxR37QwyxOM6TvF7W/58lrxtRdlBv3Gfc03qvmC1owQ4odQ5dc+yjotTQDdnssScNz9ZRbrpTH/dU5kkGhRH2zOphfKYgUDpOwAzUC4iHRDgRfZuxyNMrPvLEbNcyUq4o+XWafK2H+z7n+Wedp0g0ttrRTSmSFXzEqALybx90/WaIbpLLMALi+1eVWOS3TSVSKgV1n/zl/axp7jPu56C+nb9wKAaS5UAOYBS7dms2yV7W+S22Bs1lh69RadbUMwx/Y+G2/5+sdTwqW3Po9wX3SKWca6qZ42DBIhANyBADqAbdHIGTZR9rXJLIzJaanpZXJxPrTOoJ5ssFosgR9vuvTb643vW3Q7yXP4PxJ3PRp7yEgEikIYAOYA0gLrDadnvxee+aKmlut6ClR8z/GD9wx2+R/8a+XqRWLOH55q5ooPzVS4ztIZ0IALmJEAPgc3ZL12nlexvl7dvwqdUluo6i7tH11WcsiYxGql9+/HQj2+LFbXVsx+3mGb7uZRa00kiUGQE6A6gyDost+pii7eY9cfc311rMU8EXVnqWPhQ5co3BIfb87t/W2vqcttqKo0IEAFGgBxA9x0JsP7SNlh/xVLVw2ImI+t76fHQ6zcJFbXuK/9JAV667wClluefADmA/DM2ZQ1KwLfD+rtqLDWcHxvno2WBN5/reP4Koaxn67Gz7ENH5aMKKpMIEAFGgJ4BdMeRsGPur2DuX2Pp0cs8CIL/ey3w+OmiY1DZ+fd5nSZSzDyISBMikEMCdAeQQ5jFUdSOub8Jrf8Hb/gfPlew1bgu/Vf5+EOFcKg4gJKWRKBoCdAdQNF2HZfiCOsobd+IT3zNNvfv+ORt/yOXCtIW1/T3HGMPCkciXO0jISJABLIgYHYHgI1t0Rrtroper3fDhg3YrnaPPfbIoqGUFU97g36E9oX1F13Vplr5CX3xX99DFwodqyqnv1U+flKsr5T4Lf6pA4kAEcg5AVMvAWFT3BkzZqxZs0ZtNrZ3fu6553r06NHW1jZ37lzad1Alk/ZA6QhIWzd0Wn+PtbZ32vxdliH0zZL2e38pBFc5L3i14sDDu6xeqogIEAHz3gEsX758yZIlJ598smrlcTfw0ksvXXnllei23r17Y6ffpUuX7rPPPtSLaQkooaDUtD5m/SvdprL+4R+Xtd85VZCC5b98wnnw5LQNoQxEgAjkkIB5HcDw4cNHjRoFN6A6AISamzRpEig6AOkAADUpSURBVEKFIFohloD2228/RB3JIYtSLWqH9RcU0Vll7dk3w2by7aSvFp6JeLRhZfsd0wQp6jj6JtdxZ6qyOMhEXJtf9zgnheiWTIkpCBD2FHDMdsq8DgAxiQAL1h+BsRi1tWvXIlLgggULRo4cuXr1aviAY489Ng4oYlngx4Yg+5f5DyTG5Uz7J8Q7C4sJFqs4VA93RBsbsLWn6Ki09OybYUPQdnBjmTMU0fJMLy6K0pb1bf+4QIiEscW/64xLGWW1ECO1s/6C2iiEjQG12AwPWO2qDhlKabOhdvyQwv7Vnkp7zAYeajcibrB2I+JM87TNpAxmIGBeB5BIB+v+K1eunDlzJk6NHj363Xff/frrr8eMGaPN2dDQgGDo7LKH50B40hUrViCmI8d1CHE8fkAJ3OJQBlHRPR4PX+2bNm1CSMiWlhYOcRDYtr1ZCvrC61bIUlSxOaQefYTt32pZpTiGOGjjh/stjusZ4rg527JlCxjqiFusdn9rj6euF4Mt4f77Nk6csuHbFXG7T6Ne+Hu+iPZoFyqFOOilaGOKUwAOcSjPHRKSTVYQlZOv7yDe3t6OkJB84nhLArJNTU184rjVRsO5Bx7EEZg1BV46ZR4CxeQAcDlNnvzTMvGhhx6Ku4E4B9CnT5+6uh1bx+AWAWNxyJAhlZWVHFcCrn88dUBYWr7rkNk+PK/Gj6926A8ThqcdHOKixeIqb4huXtend71SVmGpyy4yMCx4c+dv991317Hg6cYvxBFTF/rriMMlBPzBv/9BaN9kHXJgzWV31NjKME+OKxLRjPEKgI54XL4kfyIoLjw3uh6aJMmSKhlNhvjgwYPZbWiqrEnOYeRAnC8YNXSG/8NIdrvdPF0PeVHEtAPxqPnE8dId+o5z4IkidwztJCwpOY8EiskB4L1PGHQVBkwMjKz6JzuAxcRPTUQGuA3tW6TqqUwOIFhAcVSNtnAqL0XKfdvDdmu5u0bM0vozMqzhWpiZEFPzwITFekLTFztORSMtd85QGpdZB+7nmXWfaNcP8IIZqL64WkHKAzgAiPPZXxQMuwnsEIcdTFlP0pNMHAyT5kh5orADD7XD8+HflDomPcmNPWmJdCJvBOINaN4q4iwYlyJ+TBiT8ffeew9PgNmf8+bNO+SQQzjLzUxMW3tmEvG5VOXjT2TwN3ftSiQsb2lQolGhvJLP+kM7I5onE4diLbecK21YYu2/b/VvH0xm/ZOJZ8BslyxGmmBElulvpAQjsgyBwRKMiBuR3aX/6I/8E+Cc4ORfsR019OzZE/ez7A9M5/EA4LXXXsPaNKZmZ511FlZXukyTYqlIiYalpgZRlpSyctltoo2UoVjbbdOl9UssfcZ6rnmAArwUy4giPUuYgNkdQL9+/bT0YfePP/54TDFUr6A9S8dKNCI1rheiUYvTJdiqhEjULEwQ3PHuq6Or37f03Kv6mgdj6tGPCBCBQhMw+xKQLh+y/rpY8EI95v5CNCKWO8W6foLFqp+tAKlK27zro9/+R3T191w9z1LlLoAKVCURIAIJBIrSASS0ghLw5qMk4X3/SMz6W2NPfcXEV2sKhElpf+z2yGfPiOXVnqsftNaaKPZAgYBQtUTALATIAZilJwzp0Wn9lUgIC+vW+v5CwstRhgrnFu58eON74Z+hd24X7OXuqx629R3EXRgJEgEikHMC5AByjrTLC4T1b1qvhENiWbm110BBNEefwvqLYsc7CzuevxSLUe4rHrMPHtHlaKhCIkAEUhEwh7FIpSGdS0lAlqWmDUqoQ7SXWXsPgs1NmbvrTopWm2PJO/7HpgllfasufbJsxLiuq5tqIgJEIDMC5AAy42TOXIoibV2vhAICrH8fE1l/0BK//rj6rasUq6dy2hzHuEnm5EdaEYFuToAcQBEPALzzg13+BbvDFpv7m6grOz57P3DPIYKtznnmnRUTftq9o4hZk+pEoBQJmMhqlCLe/LVJkbas22n9dzPLU9/O5nZ8/r7v3p+LZQMjR9/gPOzE/CGgkokAETBIgByAQYAFEVfwtRd2+Y/N/fHU10Tv+wuhrz/23X+BoFTIx9zYvM8hBaFDlRIBIpAhAXIAGYIyUbaY9Y+t/JTZ6gcIVhN9yx1e8YXv/ulCx+qKKfOcx04VwrF4zvQjAkTAtATIAZi2a/QVi73xucP69xdssZg5JvlF1nzXft+liq/BccI9lcf+Ep8lm0QxUoMIEIFkBEw0f0ymIqWrBGLWP+iH3bfhW19so2+aH4KOeW8/V/FvLjv8t1UnTYNetCWkaTqHFCECSQnQHUBSNGY7IW3dqAR9gs0W+9bXbiLrL7dua/vzOUpwu/2As6rOmC5YTXRfYrZOJH2IgKkIlP4dgJGd4yBrEnF5+2Yh2C7a7LFvfTOb+3eN8krA3/rn3wiBbWUjj3afNxvPpdn4Nlg7tv42Tp67BKY8tzgIsBK4L/ViF+duOAl2MYFScwAIE4ZQguzSRVQpxKVCaFPEF+RYkYAN8nq9iD+DQrjFoQPK4RBHExD2AGGtEFVKaW6U/W1411Os7Sf6g4ISSDtKmDgazh1WMBYRuK0NBJKGhBRFC3afvvdqpXWNXDMwOvXa1mBIgHqd5g8dAdlU4inbEIlEEBSXWxwRwSCOrmcjIWVVOifRZCbOHREM4hg5fBHBoDOrHa3gHjkQxNjjFkfDYwNvZywmHUZJkqA8Gq6GZU2Si5LNQqDUHAAG3/bt22FzARj/wnYjRjyMEQdviMOCIDCe3+/nE4cmUAC2jO9C2t7cjMh8Fu82MdiO77zkHn0Ub7vQ5s1EGVyHqB0OAPrz1Q7rD5gILK7vAEA4HHI/+nu7d53grG6f9veoLyB4f+KMoPCgl1Q8XRsAzYg4TCcT53MAIMbEuR0AAiojEjVfVEXoDHGYb0xl+PoO4gDI7T8wclA71OCrHZ0+bNiwdD1M501BoNQcQP/On4p248aNw4cP55uIoRCY74EDB3IHuYYLQcyy6upqVZ+sDiqcTnv79vqqCkWps9QPFMqyi9GKsOBwAOCRVaVqZlh/2BHEVVdT4g68d8yMtK0RLNEet7xSmxDgBUHVgX3o0KFxUhn+CeMF+NzisFywgOj6DKtLzAZ0e+65J5tJJJ5NmwLlEdGezwGgcOiPUEhwIWkr0s2AeQN6v6amRvds2kTUC89XX8+5cTeqTlsFZTAJgdJ/CKw/gc0MP2QLKC62NAoBr6wIsbi+WVp/tC+PystS6z8uD3/zmiIKNbd9LCZYf9QOC24EnUFx1naOCSwbFxDMI70Mxl6x155BEymLKQiUvgMwBWYOJaLh2Ds/Fiuiu5grfK4stc2dHf32DdFRU/PX9yyVVRyNIxEiQATMQIAcgBl6QUcHxe9FkC+h0i04KnROFypJkb0P3xL58gWxot5z/Xyrh3ORoVDqU71EgAhoCZAD0NIwzTHWIAJ48CsKFeaaX3sfvz388VNiRU/3NY/YevUzDS9ShAgQAR4C5AB4qOVbJrbRmxRRrHYls1f+860PK7/9mXvDix8QHR73lQ/YB+zeNZVSLUSACOSPADmA/LHlLzn21j9uAhxO80T48i16NPT6X0S7s+qy++1D9uJvG0kSASJgGgLkAEzTFaoistz5+NeimGb9J/DW8x0Lpgk2p2v6g2XDx6qa0gERIAJFTYAcgOm6L2b9JUkscyLML+4DCq5f4IM3/A+fItj7Vl38T8eo/QuuDylABIhArgiQA8gVyZyVI/va8CWQiPd/ChvhHbXbbKHP3vffe7To6O264BHHvhNz1kgqiAgQARMQIAdggk7QqKBIUaXDHwvyha+rCjv9t9ktCO5418FiRb/Kc+aWH3ikRk06JAJEoBQIkAMwVy/GXv/HZmoVLvxTWM3sK770vDhNsfVxTr2r4hAK7VvY3qDaiUBeCJADyAtW7kIVXyvWfyyuAu+m0rH049Cc4wVbX+fUu50/P4W7OSRIBIiAmQmU2mZwZmadXrdISImEEFBFLOfcBSx9FRnkCC371D9nmmiplidf7zzi1AwkKAsRIAJFScDsDiAUCuGFeHU7T2wyvGnTJrbHLzbMwm6RfPv9mrOvZF9s/ceCx7+F+4W/+8I37xIh3Bo5/Pe+/Q6vK5wmVDMRIAL5JmDqJSBsKXzJJZesWbNGpbB48WLs8NwZaCQWqgW+QT1VAgdyAO//CAVc/4n8uLx9zqVK+zflJ/6t7LhfCdFICVClJhABIpCMgHnvAJYtW/bFF19MmTJFa+UR2mXy5MnJGlPU6UpHAAZXREjFnVEVu7g50YaV3rt+o3hXOo67s/IXZ4Vb8DSCfkSACJQyAfM6AITjGD169PLly7HUw3oANwSIswEfgGBVWBSqra0tpZ6Jbf+AV39c1QVplNS4oe1vUxX/ZscR11eddj50wJb4RjQxuDRnUJxpnpNCjEAgWSJgcgLmdQCIagR2sP5qVCZEyPvqq6969+7dt2/fpUuXIv3oo49Oy1d7A5E2c8EysO0f4AAK8QBA2ral7S9nCx3tZRMuqZo6HeEnjXMwiN2gONMfhfD5AFZ7TnTgJlnY2rnVJsHiImBeB5DIEaEZL7roosGDB+MUov29/PLLa9euHTRokDbnhg0btDGB4TNWrlzJF9MRDmbdunUIDcgtDvUQWhb6pLuYRbHDZ2lrUmwOOWwXOqfesFx43A0viPjg6cS1AHYcQxzBkBEWkT1F18nBkixWu6/V/cRswdcUGnjQ9gknbVm1BgpAHI9YcKeFXOodWNJC9E7gcf3mzZsR1J5PHJqvX7++qqqKTxwBxRoaGioqKrgdAGoHfO6YwKgdvcYXEhI6Y+DhfpcvJCTEobzL5UJET76Rg45Dw7kH3pYtW7jDSeoNJUrLI4FicgCwJsz6Mx7jx4///PPP4xwAYvBi6LMMGMS4Dvv06YMLieNKgAOAEcQNBwrkE0c0ecRlhUppxZVtGxV7naVH71gEmM4n26gd611oMhTgMIIQxw/eC3dLycRFq1Vq2RqZO1vwrrcOm+i+5G9VgkV1P1hkQ+2gl0w8xaiEDcJKHfTnFofmgA/lYcpTVKR7CrWjarheiOtmSJuI/oL1hPLwAWn7Trc0NnLggTjEoT+Mb69evdxuN4c4+h3zHshijZRPHD2Oa4d74EF/XSaUaEICxeQAFi5cOG7cOERpZxwxP1VtvUoWs3XthN1qteJKwHWoZsjqAJ4Dk1C+iRgqgnoQT1QyXgcpKtksgqvS2msXgwVBmGAjtcOMphBHzJnWObME7zrLbvtXX3WPYN1lMLC7hxTi8a3Y9W9YAchyi6N2yGq7ctfi0/wFEwZxwE+TL8lp2E0mzn0HwMTV15eT1JM0mYmnHzlJCjAuDs+HQpIUnyaZG3uacul0HgjkYLU3D1r9VCSuZHUWM2nSpOeee45NSGEgnn766QkTJvyUNcmRKp7kfKpkyHaBuOz3Koj+mLD5c15rx56jLbecJzV+bR10QPVv58VZf0Ax0nAmbqSEnLTdoAKpRka6c8b1T1dDmvNG2o6ijYgbkU3TKjqdawK7TPpyXXgOysNionpHibWUs88+Gz4AN8iYZVxzzTW4281BHYUtApcL9v8Rxa58/V8JdbTeMVPe8qV18EHVV94j2ssLy4BqJwJEoCAEzO4A4pZxsax52mmnwWaqXqEg1HJYKfZ+6Nz+wdZ12z9I0bb7rpN+fNs66Geey+8Use0o/YgAEeiWBIpyBl0y1h9DLjb9V2SLs6u2f5Dltrl/iC5fZB2wv+eyOy1VhfnsoFtea9RoImA6AkXpAExHkVshPNBA/C/R0lWv/yveB/8Y+XK+pffe7svvtlSX1Jd03J1AgkSg2xIgB1DIrldCASESRuhHsawrVuG9D98a/vRJi2eQ+4r7rD1on7dCdj3VTQTMQIAcQCF7Ae//YBFIrOyK3f/bn74r/P4DYpnbPeshW/0u75sWEgHVTQSIQOEImP0hcOHI5L/mWPTHAKI/ik7O19UzV9H33P2hN/4qiLL72mdsfQdlLkg5iQARKGEC5AAK1rlK0C9IUbGiUrTFdj3Ky6/zm0z/i490vPgHIbrVc9M39t2G5aUiKpQIEIEiJEAOoGCd1rn9p2LJ3/oPrL8oBl9/OrjgckFqd8/+zD5kr4K1liomAkTAfATIARSmT5RIWAl3CDY77gDypAG2+hFfeSzw/h2CtbLqmjfLRozLU0VULBEgAkVKgB4CF6bjlIBXkKXYx18Wa540cHz0uvODeQqs/2XPOkYdkKdaqFgiQASKlwA5gML0Xez7L7z/78rX+z+BdxZKC28QbE7Xhfc79plYmEZSrUSACJibAC0BFaB/lFBQ6Yz+mKfX/4P/ez3w9G8Fq0M59Y/l4w8rQAupSiJABIqBADmAAvSSguiPCLqCTXhyEXsrrgEdny32PzZTtNhDR8+Kjjww7iz9SQSIABFQCZADUFF02YEi+9ux/pOP779C3yzxP3ChEPVXTrs/sud4IRjsslZRRUSACBQdAXoGkKrLsOucwY3nEsWVYEBQJEu5EztApKo75iJiv9R5tGcjq7/z3XGSIIUrf3V3+cRjRGN7pmZVtVYNdtypexbKx5WQk9qNFGJEFm0pbPOZAnFIs/rTSPONyGalJGU2TqDU7gAQDA/hABkXhHPCZmsITosRyRGkAsEGEJcV4jjgE4cyEEdYqJ/ERYuydbMS7BDL3SKm553RH3V7kdWOtkCHn8R1syIRwbw2rgreerggVIlHXSkcdEysar8fYXNwwELoJBPVTQcxaA5ZI+LQnFscscwQzRg/jpCQaBGkWN/pti5tIoBDcxDgjgjGxFEOfmmri8sA+Ex51M4tjlhyfAEp2cDDyMlo4MWp3un5wC0hmRJMSqDUHAACUjc2NmIQgzcuJFiQ1atXI7Igx4WEQhBiHiaYWxyxuRG3EsFpf6pdUezbGqBZNCQqjdtTDAooz4KqI7DwT+K6ArATm9b0/PcMi1gdGvfL5tGTlJUrMfdGMHqYUfzSiOuVidq9Xi+UxwG3/0BfICo6hzg0AvaNGzci7A+fOKQ2bdrEHdQQxFA7bChCiurhSZ+GkYNCuIPCQxwuBCEh+foO4hh1iEvMJ44rCA1PP/D0MGDAQHy33XbTO0lppiNQag5gUOdPxYzQ3mPGjOEOzYrLAEOZO6QwBBHBxuPRvOsZbJdrHNj8R6xNvx1bdXU1bBBioqnN0T1QGte3zrtJUsrs+0+pveB6tdytW7fC//Xv319XKm0irD9cyJAhQ9Lm1M2A+SOiMe+55566Z9MmYgoPBzxy5Mi0OXUzwPDBEo0ePVr3bCaJEEftbCaRSf64PBg5Q4cORffFpWf4JwT79evH7cDgOTB48MuwurhsCL2HSUXagRcnpf6JcPbqMR2YnECpOYBE3HxTSFYOZPmWIJKJR9uaBVm2JoT/TVQbKagdP91TamJ009q2W05CiEfb6OPdF9yopuMAmqcV1+aPOy5q8U5ysWjSsONx7crkTwiyErgdAMQNjhwjfVfY2o00PJPeoTw5JEAPgXMIM01RSjQsIACAxZKr7T+jm9d6//5rbClqG3Vs9WV/TlM9nSYCRIAI7EqAHMCuPPL5lxJ7+1PI1duf0pYG750zZO9G+5gTPJfeLIicq9X5bDGVTQSIgKkJkAPouu6Jff8lCJZKzpVZraJS08a2e66St35nG3WC+4IbRHtXBBTTKkDHRIAIlAABcgBd1Imx7R8iIcFqFx1GjbW0vbHtnlnyli9tI491n3+9WOHqojZQNUSACJQWAXIAXdSfnbu/CRaX0em/7G323nu1vOkz2/Aj3OffkL/t5LqIC1VDBIhA4QiQA+gS9ngtIxjb/sHichupTwn6YP2l9Z9aB09wX3izxd3DSGkkSwSIQDcnQA6gKwaA0uEXsP1nmUOwcb4YHtMyEmq75+ro6g+tfUZ7Ztxm8dR2hepUBxEgAqVLgBxAV/StHNv9XxCravgrk6Jtc2dHv3/P4unjmTWPrD8/SZIkAkRgJ4HS/xBsZ0sL9F98iCRLSuz1f6vFwNNa70M3RZa+KNqV6j88S9a/QH1J1RKBUiNADiDfPSoKAZ8gRWPv6mDLtmx/nR+yeh+5NfzJk4K4vfqmbyzVPbMtg/ITASJABHQJZG+SdIuhxGQEsKeA3xd7/Mux/gPrryi+J24Pv/8gYod5Zi+x1qk7/SSrj9KJABEgApkSIAeQKSm+fEq4Q4ni9X9bLP57lj9sRCMvuDe09AWxvNo98zFb/6FZFkDZiQARIAKpCJj9ITA2xcUvsQXt7e3qvv+JZ82SgoiPgXZs6maJRX/Melcy57vzHV8vFOwO16UP2Ifyb2xpFhqkBxEgAiYjYGoHgMgSF1100Zo1axKhnXnmmdjxPDHdXCmKJOAFUOxIWanZETozFUPLPlEWzxHsLtf59zhGjs9MiHIRASJABLIgYN4loGXLln355Zcw9IlBLV599dVf/epX5t91VoyERTmK0I9Ywc+iTwQB+0YEnrtHUETrERc49jkoK1nKTASIABHIkIB5HcCIESMQ0GP58uVxG6OvWLGid+/euDlIdAwZtrnLslkw/bcpsc2fs1z/Cf73FXnjF9Gaocq+h3aZtqaqiG8ff1M1gZQhAuYnYN4lIBaONc76w+5//vnn++67LyIdZgjXiCmBrAFxRQwHRDz+zXL9R25pCr5yj2h1hCacIVXybx1hTPnYwhV+GUJOzIYn2EbEOyvnr52JcytgUBw0cqJ/ItUMUwpee4Z6UraCEzDvHYAumhdeeOGUU05hp3Qvb8TRRShUdgo2yHhMYCw0ITJf9ncbojUcaNy40dejZ2vF+izE7WXulx8Ug22RAfs11PQvW7064PVmIb6TGgggJCTcJB6V84kDI6JCIrRhnA/eWUOq/6J2RJRFXyAoJoc4ikZQckTlRVBDPnH0GsQRGVF3kKRSvfMciEEckUTZLCRt/sQMiMqL4WckJjCGLkIi8/VdQ0MDZNF9fOLoODSc7yYbwBELuq6uLpEJpZiQQDE5gA8//BAxZlmAX1zhsBEY33FXOK559aKF8frxxx9hRBDVnQM9LmBcRRDnCc0KtbYFvE5nVa9+npoeiDCYkQIWq7RuhfjZY4IoOKdc7onayqyWmpoavss4Fg8+HOYWR6UwvhDnM8F2u93n83GLo3MBn1scw4OJxw2PjHqhMxgnE1fHUoaCajbEQ0YsaL5o0tCZDTwUwtf1cN6Q5e56mH40nFscrkvlQAcmJ1BMDqBnz56Yl7399tuw7F999RVClo8bN27YsGFaxLhy8FNTMI4R4ZrvOkQhuIZxGfA4AElSAuUtsAK9+lRXZ/EKUOuCv0UtzvIjp7v2GF29fj2Cg3OH9sZ1CAfALQ7Tg98uEe1VrBkcYPKLV3W5xXHjwmxoBlXpZIHTgji6XudcZklMHCMts+zxuZg4m6zEn8vgb3CD8pjNZJBXJws6HSVwdz3u3nDhcIuTA9DpErMmmd0B4EpWJ3F7dP4YSdimkSNH1tfXpwWLnGnzJMsQM4Fc4rK/FUHZZbtDxqcAGf+CH74eXb9ELK9yTj4DQty1swoLK46O40PHlDcoztqOf9XBk3EnxDKq4llJaTOzErQpWR0Xu3hWjaXMBSRgdgcAE697DQ8YMIB7etUFuGVfm0VQ5Iqf7kXSVqoE/YGnfo/Hh86pfzUeNyZtdZSBCBABImB2B9C3r/7uN0OHDjVt5ynhkBAN491/EZv/Z3z74X9zPsIGWPvvWzHxWNM2jRQjAkSglAiY3QEUI2sF6z9YfMD2n34pQw8Q3bS2Y9ENQrjBdc78Ymwy6UwEiEAxEshihboYm1cQnVn4F8HlyXj6r/gXPSRYKssOmk17/hSky6hSItA9CZADyHG/Kx2B2O7/ZRWC1Z7h9D+0fElkyb9Ee3nlqZfkWBsqjggQASKQnAA5gORsuM7IvlbIiRl/wRt79vvCfZj+lx99mbVnH646SYgIEAEiwEOAHAAPtaQyiqxg/2cB9jzTLRyCH74hrXrbUj/CeeRpSYulE0SACBCBPBAgB5BLqJjO4+XPWOyXzKI/Ss2NwRf/IdgrK6fM4ogYk0vVqSwiQAS6HwFyALnsc7k9FqJAdFVnWKj/xUcVX5N95HGOsT/LUISyEQEiQARyRYAcQK5ICkI0ooQC2Acyw/Wf8A9Lwx8+jGe/zpMvgtfInR5UEhEgAkQgIwLkADLClEkmGav/eP3fmdnqvywFXpgnRKOOIy61D9xlO6NM6qI8RIAIEAHjBMgBGGe4owTF78VEPsNdHIIfvB5d+bal557OI0/PmQZUEBEgAkQgGwLkALKhlTwvgjgq4Q4BUVDK0289Lbe3Bl64Q5ACFSddbqnK9IFB8srpDBEgAkSAhwA5AB5qiTKdb38qGU7/A68+oXg32MacVr7fpMSiKIUIEAEi0DUEyAHkgjM2Lw76sPyTyfdf0Q2rQ+/MFZQO5/HniXZHLqqnMogAESACPARK3AFgX3UeKhoZ3c2oNedjh1j/iW3/aS/DDqBxpxLFfU/9TYmGHYfMKBs6Ki5zbv9MrDqr8otanLXUYBOywpWYubC1J+rTZSndtuFdRjiHFZXabqCIBagafYRzQlhHxJbSJmbODrIoKhNxqb1FkWSLq1KRJLwIxKqAOG4MdhG3WMPffxVau0ys2aPs6DOisizg/0l+TBwl8CmPixCCEGe/JJUkTYY4NN8pnVTJZPJMHPT4akexqB3/coszaIgryWeMUK+qQ7I2pk7PcOToFgKdDYpDfxBgEHSrSJGI2hl2bnHWdymqoFPmIRAL3r1u3brhw4fD4phHLW5NEAQYYSPVtsAHYDTjcuIrkJmP9OII+yWK+AZYtf6sOh1xqxU3CjgbixmQ3PonFc+mGTq1F484NIX+6cknb1Gxi6Nl3M0vbNfDARx66KHJe4bOmIhAqd0BIFCMNlbMRx99dOCBB3KH9v7iiy/gGnliAnd28XfffYeIZrW1tXwdvnr1akQ9SxYSJ22ZmzZt6ujoGDJkSNqcuhkQcrmpqWnEiBG6Z9MmYmLx/fff77vvvmlz6mbA9POTTz752c84P5CG6fz444//7//+T7fwTBJR+3777ccdE/jzzz9HyFLuoHVff/314MGDtdGtM9FZzQPyGHWIoa2mZHWAGSEumX79+mUlpWbGwMOsS52Eqel0YEICpTDrT4EVhgCmJEWG1KcgjqGcOk+Ks+xWOkWG1KeM144SUleR4qxB5SFupHYjvYZGseULbgVYpxvRwWDfQdxI7cb7jhFIMTxSnDIim6JYOpUPAiXuAPKBjMokAkSACJQGgRJ3AAYnI5iI4cfd0yRuBJ3xvuOuHYLGazc4cowob6Rq1Gtw3BrRnGS7mECpPQOIw9erVy/uZVwUhVVUux2BvTh/NTU13KvAqBJLwEZqx6OLsrLYA2e+HzSH/nyykILm3GvQEEevoe+4a8dTUIPieHhjZBUbbed+8oRW9+jRw0jfVVdXGxx4RtrudKb/GJ67Z0kwtwRK7S2g3NKh0ogAESACJUygxJeASrjnqGlEgAgQAYMEyAEYBEjiRIAIEIFiJUAOoFh7jvQmAkSACBgkQA7AIEASJwJEgAgUKwFyAEZ7Dp+86r51h3SjRedfHp8K635wBOUNvgeZle66ANMmQsNAIJBYEbYACofDieldmZJWeSjT3t6uq5LP59NNL3iitlEpRk7B9SQFMidQ4q+BakHMnTv3xBNP7NOnjzYRx7AjzzzzTCgUwh4meP3u5JNPZhkeeeQRvMsIU4jdFI444ghVCt/Zr1y58rjjjmMpl1122YwZM8aOHatmwAEynHnmmf/973+NvI2nLfCVV17Bp/lxtbAMixcvXrZsGT79b2lpueSSS1jiBx988OWXX7LEc889V1UDzbzttttmzZrlcMR2or7//vvxrufZZ5+trcvr9SIRH/QbeZNSW+CWLVsWLVp04YUXahPZMTZu+te//oUNJ0D1nHPOYb2zZs2a559/ftCgQatWrZo2bVpdXR0yY1+Ku+++e//998dGBfj3gAMOwFYZw4YN01olVubNN9980EEHHXnkkYnV8aXccMMN1157bUVFRaL4fffdB8jYNmPvvfeeMGECMsChgioSN2/ejG1I8ENiMBiE8tjdAQcul+uUU05BotvtbmhoGDBggLZY9Br2z0hslDZPVscvvfTSbrvtNmbMmESpN998E3tn4W1jjAqMAfbOMRIxej0eD5zr+eefDyko8+KLL7a1tSERu5tgwONFz8ceewyviv7mN7+JKxatQ040LS6d/jQngW7hALCzCoY1ZoW6+xQ+++yzhxxyCDM9MEOvv/760UcfjQv79NNPZ2+yf/bZZ++++y72t4JlXLhwIToSV4LanbjssWVQnGl+//33c7UfFmwELj9Ux+ygWi87wI43eN98+vTp+BMX3h133HHllVd+++23mAWzRMzU0JarrroKGd5++23YF5xS7QucCkvRfnCAlJkzZ8ZVxPcnLAuUh+3Wfa8c1vDJJ5/83e9+xwq/88474cDgcUGPKYx0WN6//OUvOHjooYdg2VnO+fPnDxw4EA2HLLTdZ599WDr+3bp1K96C1zZHPcVxgB2BAA3Kq8S0hdx6661XXHEFe+0dIwTjBNsHzZkzB04LdhA5MZaWL18+atSoJ554Av4PiiHxq6++WrJkyfjx46+77ro33ngjzoZi0qA2U1sXxzFcOMY2HJLurj7QAW7+4osvRskY2JjuXHDBBWgvEtk0orm5GYloC5oGD73HHnsg5wknnHDLLbfMnj0bDdywYQM6S7tTFpSHs9RlxaE/iXQBgW6xBIRJFgwKLrnEZQ0MfYxX9bYAW79hGy9wxwxO/Y4JVzVmnUiExZkyZQrmStplE0ygsP0cJq1qb7W2tuKS497ETS2HHcDZXHrppbjwtJWqeX744QfMdtWcmLOjjbgnYBNPpGPuj13J1q9fj2N4Kcz90Vj1EkWZuCv68MMP1QJxCoYDOdU86imOA3yQNXnyZOiv63pxm3LaaaepxR5zzDGw5vhp77dgH7GjH+4htC4W021ARkuBBa5dLQEHyHz44YfrstJmy/AYdxjwMQCYOHLgvfbcc0/1o6dJkyahL3ATBhFm/VEFZhJgC5JIYdYfiXvttReUxwFuwjDMMAJVZUAeZeaEPMpEpZdffjnuNnRpvPPOO+pWfZiw48M3QMZ8AqOd6YO7YQwerEfh/gaNYonoULQRx+jQo446Cj6PpbN/MROaOHFiIittHjo2FYFu4QAwqQF03csAE0bMJbVdMnr0aFzJcYmw8phf42pnaynaSxQTasxAMfdRC8GldfDBB+dqGRoOAFedrvJQCafUenEAVwRb37t3b3XNB4mwOJis4QDmBv9qr08Ui5WWpUuXIp39sJqE5uRqBo3PWZk11BLbWVXslgWrIuqf/fv3x70Cm8KriWgR1lLQIu2SDpwE/CsagmJhxSDF8mPVAnNS+HtdXGqZmR/ACCKzbmm4WQQ6tSjcJQAaVkjitl+Fn8BIOPXUU9WcuANgyz64OcM4we2OegreAqtbuRo5MOvJRg5qBHntbRnm+GvXrmXtVfXBtYAHFVAe5cDib9u27b333sOdMTLgT8yWMO/B+Gf5sZoE66/r6dUC6cBsBLqFA0gBHeOVuQc1D2wWbCgsvpqCA1zbuJK1KeoxzBAywzeoUzlcSFgv1rUaqlRODnDtxS1Mw+5jvha3CQGuavUqTawXmbGeDp/HTmFWjgu7ay7jOERgCEMDfeJ2QcBaBHRTG/Wf//wHxgt+EeShJ5basNLClP/mm29gQLUeLrG9uUrB1Fid6bMysX0FRo6qJ0vECgm8lOpQ4V9hNDHNx1koD5cMEWbxMX7QTRiNXQNfVYnpiUECNeIcAFqkTjJwf4CbLTgA7Xbr8BB4EsNKwBxo3LhxKUYay0b/mopAd3cAbIhruwQjGAs4cW9iIFE7p9bmZ8eY+2DxAcdYQcKtcWKGfKTA1sS5JUwqEy0IFhzirJJWGVhhGM1XX30ViY2NjZicwgp3jQ3VzkBRO9oCWwnzFzcFVhdPYIwefPBBzP2hMGsC8qMHMfFnInjYg1l51ygPd4W7jTiS0C3OfCMPayaG0IIFC3CzwmbQqiBW8PCcAH9i/QQLd2iReiqvB3F6YpDAJbC1KbVekIQDY2xxf3bGGWdcf/31eJrNMuAsFsfw2AB/QjDFrZ5aIB2YjUB3dwBY6McUTNsreIKK511szURNx2WsmiE1UXuANQq8O4EUzJJwVWhP5e8Ys+A4e4FnFbDgWEXR2lAsVug+BmSKoQRYfDxhhoXCcgQWJfKncFzJ0F8LH0s9UAOTUPVeCvlxO4XlaXaAZ79Y0cYz1bhysDqER5ooavfdd4c/iDubpz+xLI7n82rhwAiTitU29rhITcdCP25osJDy6KOPYoKM5xPqKXaA5mAJDvDRUtzZdI33QtXQUzt42LInbh+16mHk4E+8gaYmwpmp/KEqZkWYcEDq008/xRxIzUYHxUKg+zoADFlMfmHWcd3i9pZ1GIw4XpjDKMdcWJ0NYWEh7omu9spRjzGVe/nll9X37XJ7JaMWtSKoiqeFmEvCcGPCizVlpjysDNTGZB8TSbSOJWJ2jGUHvAjI/sS/2nLUYzyAxfuCsLzsBb5YZTmdimpLQy3MskBPEFMVw9oO6OFhL9YZ1ETMmgEWbxO99dZbeCAct0bBisWyO9qIN2pgYeMaqJZj5ECrPLqVvSaAdT+s+OOWi5UMDwTzjSUdGFb1zgBr+nj1ABleeOEFTJ+RQasGGyGsEzFvUA2rtjpt/pwcY6WePcUFVXXkYKhjRGGWgMc/uItiFcGs454M8wnceLFVOKRDNyalKsk6Ec5bfZMiJ3pSIV1DoFu8BspQ4g5Xu+aAF5lhazBhxIsoeBkRly4uSJhLXKjIj1fi8A4cLglMzZDOXtxWu0S7HKQe4+K5+uqrVYsW9xRBleU7gJ7aRVtcb0jB1BgTdrwyz4JHYnEZaqN8mFEYFMyXcf+OF+3xHYC2UqxdqH9CeRgg/Il5HO7lzzvvPHYK6VpWan7uA2iiysIBYLkMTxqA6LDDDgNn2A7M3zGRh2744RYKibipgj876aST0FIskuAYHxOwhQv8C7sDICoTmH7kYcyhuZquVmrkQKWEQlA1XBHzNHgHFEtSuO2AocRMn30agvfEnnrqKSyYYCzBGf/iF7/A9GLFihV4NQBjCSVgOGE+Afur3lNi5Nx0001/+tOfmJK4MTKibZwsaGgXANkNLrob8TLxhi7GDDSHhvhsBYKYxeObGAwn3LVgjE2dOhWJ+DIGibgWUBRmSGg1EvHwiY0c3LTBPajLnhBk6XFq0J/mJEDbQe/oF1ycGN+wNdp+wsQTpgTp2kQTHmM6BlVVV6RqCMMU95RYPWWqA9y7JPrLxKesptJZVUYXsm6iKmKeA92Ro5uIex2sK9IXXubpu5xoQg4gJxipECJABIhA8REw+9y2+IiSxkSACBCBIiFADqBIOorUJAJEgAjkmgA5gFwTpfKIABEgAkVCgBxAkXQUqUkEiAARyDUBcgC5JkrlEQEiQASKhAA5gCLpKFKTCBABIpBrAuQAck2UyiMCRIAIFAkBcgBF0lGkJhEgAkQg1wTIAeSaKJVHBIgAESgSAuQAiqSjSE0iQASIQK4JdKPN4DjQSYrw1pfBo1/1CyFF6Nw0LYtCsJtmmXhkT+uz53g8TnK0WZCjrESACHQNgaJ3ACff0/xCi1yepYHtaFOUv9elRbx+WxTWf7RNXBZOkhf1YjNNSe8sHIYirGyWHn7Xf+Wxu8QX08tNaUSACBCBriaQdwew9MfQlPntA8tjew6HZWFsD8td02IRetnvkNu3W0XBEjsp/BBU3pvmGdzHvuNcZv/xlAn7lYuOLB3AB8kM+q6VRqFxRDhtD9vXUzxyQqQmaB2RlO0+uU+1NfEsGrVmc2TI/W2NLbr+Ydea6C8iQASIQJcTyLsDgNn8ISLAPq/oUARJqXN0Gvud7VzcKseWVkRhjFNs2BFceue5TP8bm2nvUmimgtnlY14qTua1L4IPfdrx7AU1FWU6KsRElB3uLU6Q/iQCRIAIFJxA3h1ArIWycOFI+wVHuiJSvDVs/W0trH+ZTbxxftvXK6McOLZ2KEv8sLJZirbLWQroZP9qTfjElwJCVLl1kffGUzzWBB0S7hl0CqEkIkAEiEChCHSJA1CECrvo1Fun97h2WE3HLoFYsqDx6K89MtZfsnxCqyQu2WRR546sA3pat8ysxv2LrKA8xRp7GkA/IkAEiEDREMi7A5Bgav1KMJJmNtwejmWTs5+X9/Twug7DfVRbVbCqDetOBRABIkAEhLw7gD36l624zFOV7j3IK492XXSo0rdn1ib1mcU+X4eS5Q2AEAgr049zZ9v/322MjOiX/hl1w7Zo72or1rWyLZ/yEwEiQAS6kkDeHQBM//CBZWmbNLBXesOqW8jCb8IftMiVuo9odQU6E1d45enHJT+96xlmyF/9LHDs6wFlds9dT+r8tW5zZPbz7bef4clSKZ2iKIkIEAEikD8CeXcA+VOdlRyRhfWy0Atv22T1y+bNzHK78ObyjmOf8U/qlfCcV69Su1V8fF10/WOt9x7voucCeoQojQgQAVMQKHoHMOsw52XRrJeAwtEsHEYwIhwxqvzTC8T9n2nPpNNQ+FV7ld1yurupOZqtY8qkfMpDBIgAEcgJgbw7gG2t0rcN4V7V1tQLQUtXhdr88tjdHe7KjGbZauMP3KtcPc7rwfhhjsbLMsI1YnDZwWMqoIyU/TPtvDaBCicCRIAIaAlkZNG0AtkeNzRFJj3hu2uiI7UDePyD4O3LIz9eYcvWASxbHYpms57D9I9Kyvg9s/Yc9e6MnlHX0dtB2Y4Syk8EiEAhCOTdAVjwJLRCLE/3SowTz4ArRI6nprMW+f7TLMd/YJYWZaus3JW1A4grdU1jJBiURVHEnnG797XrfgwcJ0J/EgEiQATMQyDvDiDWVFHY5pc3NEXwJTA+l9W+8LN6UwRvcNptoje23SYPlrP2dpzUEbPCWQn7UZ3hH3zbyMfb8fnCvMMqRg1K/6aT4QqpACJABIhALgnk3QFgU2T8Hvoh8vtlbdgL6JTe1gXTe6gt2H1uK9sLaDim4wrPE9NfHeZSS+vig93qbItPcT3zecd5P9fXgeOGpoubQNURASLQnQnk3QE4HeLMIbbYCg/2BFKEXlW7POP93Qg7pu5s+u4LK450K0WJXbXdKzEfk3gqRQo+Oa6vSb+gHytZETqiSntATnxxCJqP6GufVWv1+SWsAsX9sMtpM3YcsghWHNGPCBABImA+AqLf71+3bt3w4cMtll1Ms/lU1ddo5r9a1nqVbB3H6oDy2TW1+iVqUpvapGMfbG0JK6tCmlTtYcx9xXa70//ZBIsi3DbeceVkigegT4hSiQARKCCBoncA+Wb36hfB55aG2nwyx3pOhyQcM8I+dUKlO91OGPluBZVPBIgAEUgkQA4gkQmlEAEiQAS6BYGiXPbpFj1DjSQCRIAI5JkAOYA8A6biiQARIAJmJUAOwKw9Q3oRASJABPJMgBxAngFT8USACBABsxIgB2DWniG9iAARIAJ5JkAOIM+AqXgiQASIgFkJkAMwa8+QXkSACBCBPBMgB5BnwFQ8ESACRMCsBMgBmLVnSC8iQASIQJ4JkAPIM2AqnggQASJgVgLkAMzaM6QXESACRCDPBMgB5BkwFU8EiAARMCsBcgBm7RnSiwgQASKQZwLkAPIMmIonAkSACJiVADkAs/YM6UUEiAARyDMBcgB5BkzFEwEiQATMSoAcgFl7hvQiAkSACOSZADmAPAOm4okAESACZiVADsCsPUN6EQEiQATyTIAcQJ4BU/FEgAgQAbMSIAdg1p4hvYgAESACeSZADiDPgKl4IkAEiIBZCZADMGvPkF5EgAgQgTwTIAeQZ8BUPBEgAkTArARsiYopipKYSClEgAgQASJQYgRsMPfqr8TaRs0hAkSACBCBFAQssP44HY1GU2SiU0SACBABIlB6BMS2trbW1tZQKGS1WpkzYP+WXlOpRUSACBABIqAlYJMkyel0wvrjgK0FaU/TMREgAkSACJQqgZgDqOr8qRN/9aBU20ztIgJEgAgQARCwYfXf4XAQCyJABIgAEehuBCy4A+hubab2EgEiQASIAAj8PyZw/gpzTm4lAAAAAElFTkSuQmCC"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Following is the screenshot from TensorBoard showing metrics such as loss, accuracy, evaluation metric and how they change as training progresses.\n",
    "\n",
    "![image.png](attachment:0a2e3dc9-cbdb-4f21-be20-d60b69c5cff3.png)![image.png](attachment:be2862b0-0e40-47f5-95f9-e07e451a4d8d.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Explore and log metrics\n",
    "\n",
    "After the job is completed, you can explore the evaluation metrics in the run\n",
    "directory of the job and log them to the experiment.\n",
    "\n",
    "**Note**: The T5X job might take a long time to run, depending on the task. If\n",
    "the job takes too long, you might\n",
    "[lose your Jupyter kernel or the kernel might become unresponsive](https://cloud.google.com/vertex-ai/docs/general/troubleshooting-workbench#notebook_is_unresponsive_2).\n",
    "If so, you can reimport the libraries and reset the details for the run in order\n",
    "to explore the metrics\n",
    "\n",
    "To explore jobs, get the list of experiment runs and the run directory from\n",
    "Vertex AI Experiments for the experiment."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get all experiment runs and run directories\n",
    "utils.get_all_experiment_run_directories(EXPERIMENT_NAME)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Select the run ID and run directory for your experiment from the list of runs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "RUN_ID = '<YOUR PREVIOUS RUN ID>' # Select a run id from the list of experiment runs\n",
    "RUN_DIR = '<YOUR RUN DIR>' # Select the run directory for your run from the list of experiment runs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(f\"RUN_ID = {RUN_ID}\")\n",
    "print(f\"RUN_DIR = {RUN_DIR}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set path to read inference eval metrics\n",
    "GCS_VAL_DIR = os.path.join(RUN_DIR, 'inference_eval/')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Call the `utils.parse_and_log_eval_metrics` function to log the fine-tuning job\n",
    "metrics to the experiment from the evaluation metrics that are reported in the\n",
    "logs.\n",
    "\n",
    "The results show a dataframe with the step that achieves the highest mean value \n",
    "of scores for a group of metrics. The columns of the dataframe start with `step` \n",
    "and rest are the group of metrics evaluated. For example, for the \n",
    "English-to-German translation task the metric evaluated is \n",
    "[BLEU score](https://en.wikipedia.org/wiki/BLEU) and shown as `WMT T2T En-De` in \n",
    "the results dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "results = utils.parse_and_log_eval_metrics(\n",
    "    summary_dir=GCS_VAL_DIR,\n",
    "    run_name=RUN_ID,\n",
    "    vertex_ai=vertex_ai\n",
    ")\n",
    "results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "toc-hr-collapsed": true
   },
   "source": [
    "## <a name=\"task2\"></a>Task 2: Run model evaluation on Vertex AI\n",
    "\n",
    "This section details how to evaluate the fine-tuned T5 base model on the\n",
    "translation task using a Vertex AI custom job. The evaluation job steps are\n",
    "similar to those that you used for the fine-tuning job, but the steps use an\n",
    "evaluation-specific Gin file.\n",
    "\n",
    "If lose your Jupyter kernel or the kernel might become unresponsive, please \n",
    "reimport the libraries and configure experiment settings from the section \n",
    "[Preparation: Imports and initialization](#preparation)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Configure a Gin file for the evaluation job\n",
    "\n",
    "For evaluating the fine-tuned T5 base model for translation task, you configure\n",
    "a Gin file by doing the following:\n",
    "\n",
    "1. Import the\n",
    "    [`eval.gin`](https://github.com/google-research/t5x/tree/main/t5x/configs/runs/eval.gin)\n",
    "    file and the\n",
    "    [Gin file for the model](https://github.com/google-research/t5x/blob/main/t5x/examples/t5/t5_1_1/base.gin)\n",
    "    from T5.\n",
    "1. Override the base configurations such as `MIXTURE_OR_TASK_NAME` that\n",
    "    refer to the pre-configured translation task in the T5X framework.\n",
    "\n",
    "The following listing shows the contents of the evaluation Gin file named\n",
    "[`eval_t511_base_wmt.gin`](http://configs/eval_t511_base_wmt.gin) in the repo:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "from __gin__ import dynamic_registration\n",
      "\n",
      "import __main__ as train_script\n",
      "from t5.data import mixtures\n",
      "from t5x import models\n",
      "from t5x import partitioning\n",
      "from t5x import utils\n",
      "\n",
      "include \"t5x/configs/runs/eval.gin\"\n",
      "include \"t5x/examples/t5/t5_1_1/base.gin\"\n",
      "\n",
      "# Configure the required values by the default eval.gin\n",
      "MIXTURE_OR_TASK_NAME = \"wmt_t2t_ende_v003\"\n",
      "\n",
      "TASK_FEATURE_LENGTHS = {\"inputs\": 256, \"targets\": 256}\n",
      "DROPOUT_RATE = 0.0\n",
      "LOSS_NORMALIZING_FACTOR = 233472\n",
      "BATCH_SIZE = 128\n"
     ]
    }
   ],
   "source": [
    "! cat ../configs/eval_t511_base_wmt.gin"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Specify the Gin file path(s) for the task\n",
    "GIN_FILES = ['../configs/eval_t511_base_wmt.gin']  # Change to your Gin file path"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Configure the Vertex AI custom job\n",
    "\n",
    "Configure the custom job run ID, name, and base output directory for the run. \n",
    "The `RUN_NAME` value can be arbitrary. We recommend setting it to a specific\n",
    "task or topology so that you can identify the run when you're exploring the\n",
    "experiment. For example, use `t5-eval-v2-8`, where `t5` is the model,\n",
    "`eval` is the task, and `v2-8` is the TPU topology."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "EVAL_RUN_NAME = f'<YOUR RUN NAME>' # Change to your run name for the custom job\n",
    "EVAL_RUN_ID = f'{EXPERIMENT_NAME}-{EVAL_RUN_NAME}-{datetime.now().strftime(\"%Y%m%d%H%M\")}'\n",
    "EVAL_RUN_DIR = f'{EXPERIMENT_RUNS}/{EVAL_RUN_ID}'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Specify the run mode as `eval` to run the T5X launch script in evaluation\n",
    "mode."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "RUN_MODE = 'eval'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Configure runtime parameters for evaluation\n",
    "\n",
    "Running an evaluation job requires you to set the location of the model\n",
    "checkpoint (`CHECKPOINT_PATH`) and to set the output path to the evaluation\n",
    "results (`EVAL_OUTPUT_DIR`). You can specify these parameters for the job by\n",
    "configuring `GIN_OVERWRITES`.\n",
    "\n",
    "Configure the path to the fine-tuned model checkpoint. You can get this path\n",
    "from Vertex AI Experiments using the experiment name and run ID. Run the\n",
    "following function to get the list of experiment runs and the run directory\n",
    "where checkpoints are located."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get all experiment runs and run directories\n",
    "utils.get_all_experiment_run_directories(EXPERIMENT_NAME)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The model checkpoint path is a subfolder under the `RUN_DIR` folder, with the\n",
    "prefix `checkpoint_`. There can be multiple checkpoint folders. Pick the model\n",
    "checkpoint that you want to evaluate the model with, as shown in the following\n",
    "example:\n",
    "\n",
    "``` shell\n",
    "! gsutil ls $RUN_DIR\n",
    "      gs://{BUCKET_NAME}/experiments/{EXPERIMENT_NAME}/runs/{RUN_ID}/checkpoint_1000000/\n",
    "      gs://{BUCKET_NAME}/experiments/{EXPERIMENT_NAME}/runs/{RUN_ID}/checkpoint_1005000/\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "! gsutil ls '<YOUR PREVIOUS RUN DIRECTORY>' # Change to the previous run directory"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "CHECKPOINT_PATH = '<YOUR MODEL CHECKPOINT PATH>' # Change to the path where model checkpoint exists"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Configure runtime parameters for evaluation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Specify parameters to be overwritten in Gin files\n",
    "GIN_OVERWRITES = [\n",
    "    'USE_CACHED_TASKS=False',\n",
    "    f'CHECKPOINT_PATH=\"{CHECKPOINT_PATH}\"',\n",
    "    f'EVAL_OUTPUT_DIR=\"{EVAL_RUN_DIR}\"'\n",
    "]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To help with troubleshooting, display the values of local variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for key in [\n",
    "    \"PROJECT_ID\", \"REGION\", \"BUCKET\", \"TENSORBOARD_NAME\", \"TENSORBOARD_ID\", \n",
    "    \"IMAGE_NAME\", \"IMAGE_URI\", \n",
    "    \"EXPERIMENT_NAME\", \"EXPERIMENT_WORKSPACE\", \"EXPERIMENT_RUNS\", \n",
    "    \"TFDS_DATA_DIR\", \"GIN_FILES\", \"GIN_OVERWRITES\", \n",
    "    \"EVAL_RUN_NAME\", \"EVAL_RUN_ID\", \"EVAL_RUN_DIR\", \"RUN_MODE\",\n",
    "    \"CHECKPOINT_PATH\"]:\n",
    "        print(f\"{key}={eval(key)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create the Vertex AI custom job\n",
    "\n",
    "Configure the worker pool specification based on TPU availability in the region.\n",
    "See\n",
    "[Vertex AI locations](https://cloud.google.com/vertex-ai/docs/general/locations#accelerators)\n",
    "for details."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define worker pool resource specification\n",
    "# Machine to run the custom job on. For TPUs, use `cloud-tpu`.\n",
    "MACHINE_TYPE = 'cloud-tpu'\n",
    "# Accelerator type to attach to the machine. For TPUs, use `TPU_V2`, `TPU_V3`.\n",
    "ACCELERATOR_TYPE = 'TPU_V2'\n",
    "# Number of accelerators to attach to the machine.\n",
    "# For TPUs, specify the number of cores to be provisioned.\n",
    "ACCELERATOR_COUNT = 8"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create the custom job spec."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "job = utils.create_t5x_custom_job(\n",
    "    display_name=EVAL_RUN_ID,\n",
    "    machine_type=MACHINE_TYPE,\n",
    "    accelerator_type=ACCELERATOR_TYPE,\n",
    "    accelerator_count=ACCELERATOR_COUNT,\n",
    "    image_uri=IMAGE_URI,\n",
    "    run_mode=RUN_MODE,\n",
    "    gin_files=GIN_FILES,\n",
    "    model_dir=CHECKPOINT_PATH,\n",
    "    tfds_data_dir=TFDS_DATA_DIR,\n",
    "    gin_overwrites=GIN_OVERWRITES\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "job.job_spec"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Submit the custom job to Vertex AI and track the experiment\n",
    "\n",
    "The `submit_and_track_t5x_vertex_job` function launches the T5X script in\n",
    "evaluation mode with the fine-tuned model checkpoint and writes evaluation\n",
    "metrics. To monitor progress of the job, click the URL in the cell output. The\n",
    "URL goes to the Vertex AI Custom Job console for the job."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "utils.submit_and_track_t5x_vertex_job(\n",
    "    custom_job=job,\n",
    "    job_display_name=EVAL_RUN_ID,\n",
    "    run_name=EVAL_RUN_ID,\n",
    "    experiment_name=EXPERIMENT_NAME,\n",
    "    execution_name=EVAL_RUN_ID,\n",
    "    tfds_data_dir=TFDS_DATA_DIR,\n",
    "    model_dir=EVAL_RUN_DIR,\n",
    "    vertex_ai=vertex_ai,\n",
    "    run_mode=RUN_MODE\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Explore and log metrics\n",
    "\n",
    "After the job is completed, you can explore the evaluation metrics. \n",
    "\n",
    "Call the `utils.parse_and_log_eval_metrics` function to parse the events in the\n",
    "run directory and log them to the experiment in Vertex AI Experiments.\n",
    "\n",
    "The results show a dataframe with the step that achieves the highest mean value \n",
    "of scores for a group of metrics. The columns of the dataframe start with `step` \n",
    "and rest are the group of metrics evaluated. For example, for the \n",
    "English-to-German translation task the metric evaluated is \n",
    "[BLEU score](https://en.wikipedia.org/wiki/BLEU) and shown as `WMT T2T En-De` in \n",
    "the results dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "GCS_EVAL_DIR = os.path.join(EVAL_RUN_DIR, 'inference_eval')\n",
    "\n",
    "results = utils.parse_and_log_eval_metrics(\n",
    "    summary_dir=GCS_EVAL_DIR,\n",
    "    run_name=EVAL_RUN_ID,\n",
    "    vertex_ai=vertex_ai\n",
    ")\n",
    "results"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": [],
    "toc-hr-collapsed": true
   },
   "source": [
    "## <a name=\"task3\"></a>Task 3: Run batch inference on Vertex AI\n",
    "\n",
    "This section details how to run a batch inference job with the fine-tuned T5X\n",
    "base model for the translation task on Vertex AI. The inference job steps are\n",
    "similar to those that you used for the evaluation job, but the steps use an\n",
    "inference-specific Gin file.\n",
    "\n",
    "If lose your Jupyter kernel or the kernel might become unresponsive, please \n",
    "reimport the libraries and configure experiment settings from the section \n",
    "[Preparation: Imports and initialization](#preparation)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Configure a Gin file for the batch inference job\n",
    "\n",
    "For running batch inference with the fine-tuned T5 base model for the\n",
    "translation task, you configure a Gin file by doing the following:\n",
    "\n",
    "1. Import the\n",
    "    [`infer.gin`](https://github.com/google-research/t5x/tree/main/t5x/configs/runs/infer.gin)\n",
    "    file and the\n",
    "    [Gin file for the model](https://github.com/google-research/t5x/blob/main/t5x/examples/t5/t5_1_1/base.gin)\n",
    "    from T5.\n",
    "1. Override the base configurations such as the `MIXTURE_OR_TASK_NAME` that\n",
    "    refer to the pre-configured translation task in the T5X framework.\n",
    "\n",
    "The following listing shows the contents of the evaluation Gin file named\n",
    "[`infer_t511_base_wmt.gin`](configs/infer_t511_base_wmt.gin) in the repo:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "from __gin__ import dynamic_registration\n",
      "\n",
      "import __main__ as train_script\n",
      "from t5.data import mixtures\n",
      "from t5x import models\n",
      "from t5x import partitioning\n",
      "from t5x import utils\n",
      "\n",
      "include \"t5x/configs/runs/infer.gin\"\n",
      "include \"t5x/examples/t5/t5_1_1/base.gin\"\n",
      "\n",
      "# Configure the required values by the default eval.gin\n",
      "MIXTURE_OR_TASK_NAME = \"wmt_t2t_ende_v003\"\n",
      "TASK_FEATURE_LENGTHS = {\"inputs\": 256, \"targets\": 256}\n"
     ]
    }
   ],
   "source": [
    "! cat ../configs/infer_t511_base_wmt.gin"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Specify the Gin file path(s) for the task\n",
    "GIN_FILES = ['../configs/infer_t511_base_wmt.gin']  # Change to your Gin file path"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Configure the Vertex AI custom job"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Configure the custom job run ID, name, and base output directory for the run.\n",
    "The `RUN_NAME` value can be arbitrary. We recommend setting it to a specific\n",
    "task or topology so that you can identify the run when you're exploring the\n",
    "experiment. For example, use `t5-infer-v2-8`, where `t5` is the model,\n",
    "`infer` is the task, and `v2-8` is the TPU topology."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "INFER_RUN_NAME = f'<YOUR RUN NAME>' # Change to your run name for the custom job\n",
    "INFER_RUN_ID = f'{EXPERIMENT_NAME}-{INFER_RUN_NAME}-{datetime.now().strftime(\"%Y%m%d%H%M\")}'\n",
    "INFER_RUN_DIR = f'{EXPERIMENT_RUNS}/{INFER_RUN_ID}'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Specify the run mode as `infer` to run the T5X launch script in inference\n",
    "mode."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "RUN_MODE = 'infer'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Configure runtime parameters for inference"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Running an inference job requires you to set the location of the model\n",
    "checkpoint (`CHECKPOINT_PATH`) and output path to the inference results\n",
    "(`INFER_OUTPUT_DIR`). You can specify these parameters for the job by\n",
    "configuring `GIN_OVERWRITES`."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Configure the path to the fine-tuned model checkpoint. You can get this path\n",
    "from Vertex AI Experiments by using the experiment name and run ID. Run the\n",
    "following function to get the list of experiment runs and the run directory\n",
    "where checkpoints are located."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get all experiment runs and run directories\n",
    "utils.get_all_experiment_run_directories(EXPERIMENT_NAME)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The model checkpoint path is a subfolder under the `RUN_DIR` folder, with the\n",
    "prefix `checkpoint_`. There can be multiple checkpoint folders. Pick the model\n",
    "checkpoint that you want to evaluate the model with, as shown in the following\n",
    "example:\n",
    "``` shell\n",
    "! gsutil ls $RUN_DIR\n",
    "      gs://{BUCKET_NAME}/experiments/{EXPERIMENT_NAME}/runs/{RUN_ID}/checkpoint_1000000/\n",
    "      gs://{BUCKET_NAME}/experiments/{EXPERIMENT_NAME}/runs/{RUN_ID}/checkpoint_1005000/\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "! gsutil ls '<YOUR PREVIOUS RUN DIRECTORY>' # Change to the previous run directory"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "CHECKPOINT_PATH = '<YOUR MODEL CHECKPOINT PATH>' # Change to the path where model checkpoint exists"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Configure runtime parameters for the inference."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Specify parameters to be overwritten in Gin files\n",
    "GIN_OVERWRITES = [\n",
    "    'USE_CACHED_TASKS=False',\n",
    "    f'CHECKPOINT_PATH=\"{CHECKPOINT_PATH}\"',\n",
    "    f'INFER_OUTPUT_DIR=\"{INFER_RUN_DIR}\"'\n",
    "]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To help with troubleshooting, display the values of local variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for key in [\n",
    "    \"PROJECT_ID\", \"REGION\", \"BUCKET\", \"TENSORBOARD_NAME\", \"TENSORBOARD_ID\", \n",
    "    \"IMAGE_NAME\", \"IMAGE_URI\", \n",
    "    \"EXPERIMENT_NAME\", \"EXPERIMENT_WORKSPACE\", \"EXPERIMENT_RUNS\", \n",
    "    \"TFDS_DATA_DIR\", \"GIN_FILES\", \"GIN_OVERWRITES\", \n",
    "    \"INFER_RUN_NAME\", \"INFER_RUN_ID\", \"INFER_RUN_DIR\", \"RUN_MODE\",\n",
    "    \"CHECKPOINT_PATH\"\n",
    "]:\n",
    "    print(f\"{key}={eval(key)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create the Vertex AI custom job"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Configure the worker pool specification based on TPU availability in the region.\n",
    "See\n",
    "[Vertex AI locations](https://cloud.google.com/vertex-ai/docs/general/locations#accelerators)\n",
    "for details."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define worker pool resource specification\n",
    "# Machine to run the custom job on. For TPUs, use `cloud-tpu`.\n",
    "MACHINE_TYPE = 'cloud-tpu'\n",
    "# Accelerator type to attach to the machine. For TPUs, use `TPU_V2`, `TPU_V3`.\n",
    "ACCELERATOR_TYPE = 'TPU_V2'\n",
    "# Number of accelerators to attach to the machine.\n",
    "# For TPUs, specify the number of cores to be provisioned.\n",
    "ACCELERATOR_COUNT = 8"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create the custom job spec."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "job = utils.create_t5x_custom_job(\n",
    "    display_name=INFER_RUN_ID,\n",
    "    machine_type=MACHINE_TYPE,\n",
    "    accelerator_type=ACCELERATOR_TYPE,\n",
    "    accelerator_count=ACCELERATOR_COUNT,\n",
    "    image_uri=IMAGE_URI,\n",
    "    run_mode=RUN_MODE,\n",
    "    gin_files=GIN_FILES,\n",
    "    model_dir=CHECKPOINT_PATH,\n",
    "    tfds_data_dir=TFDS_DATA_DIR,\n",
    "    gin_overwrites=GIN_OVERWRITES\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "job.job_spec"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Submit the custom job to Vertex AI and track the experiment\n",
    "\n",
    "The `submit_and_track_tx5_vertex_job` function launches the T5X script in\n",
    "inference mode with the fine-tuned model checkpoint specification. The function\n",
    "submits the job to Vertex AI, which generates inferences results and writes them\n",
    "to the Cloud Storage bucket. To monitor progress of the job, click the URL in\n",
    "the cell output. The URL goes to the Vertex AI Custom Job console for the job."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "utils.submit_and_track_t5x_vertex_job(\n",
    "    custom_job=job,\n",
    "    job_display_name=INFER_RUN_ID,\n",
    "    run_name=INFER_RUN_ID,\n",
    "    experiment_name=EXPERIMENT_NAME,\n",
    "    execution_name=INFER_RUN_ID,\n",
    "    tfds_data_dir=TFDS_DATA_DIR,\n",
    "    model_dir=INFER_RUN_DIR,\n",
    "    vertex_ai=vertex_ai,\n",
    "    run_mode=RUN_MODE\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Explore batch inference results\n",
    "\n",
    "The batch inference job writes inference results to the run directory that you\n",
    "configured. The output is written in JSON lines (`.jsonl`) format."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "! gsutil ls $INFER_RUN_DIR"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The following example is a snippet from the output of the batch inference job\n",
    "for the translation task from a fine-tuned\n",
    "[WMT English-to-German translation](https://www.tensorflow.org/datasets/catalog/wmt_t2t_translate)\n",
    "model.\n",
    "\n",
    "``` json\n",
    "{\n",
    "\t\"inputs\": {\n",
    "\t\t\"inputs_pretokenized\": \"translate English to German: As recently as last Tuesday, the Nasdaq indices were not calculated for one hour due to data transfer errors.\",\n",
    "\t\t\"targets_pretokenized\": \"Erst am Dienstag waren die Indizes der Nasdaq wegen Fehlern im Datentransport eine Stunde lang nicht berechnet worden.\"\n",
    "\t},\n",
    "\t\"prediction\": \"Erst am vergangenen Dienstag wurden die Nasdaq-Indexe wegen Daten\\u00fcbertragungen f\\u00fcr eine Stunde nicht berechnet.\"\n",
    "}\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## What's next?\n",
    "\n",
    "This notebook template covers how to run a T5X job for fine-tuning, evaluation,\n",
    "and inference tasks on Vertex AI.\n",
    "\n",
    "You can copy this notebook template for a specific task or dataset or\n",
    "configuration and then make changes as needed to run the T5X job on Vertex AI.\n",
    "Refer to other notebooks in the\n",
    "[notebooks directory](/) of this repo as\n",
    "inspiration."
   ]
  }
 ],
 "metadata": {
  "environment": {
   "kernel": "python3",
   "name": "common-cu110.m93",
   "type": "gcloud",
   "uri": "gcr.io/deeplearning-platform-release/base-cu110:m93"
  },
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.12"
  },
  "vscode": {
   "interpreter": {
    "hash": "802ea0518c7535cb908ec450b955e980eb8525a80af866d337675ca6fae56b98"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}