2. train-hyperparameter-tune-deploy-with-pytorch.ipynb

{
  "cells": [
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Copyright (c) Microsoft Corporation. All rights reserved.  \n\nLicensed under the MIT License."
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "# Train, hyperparameter tune, and deploy with PyTorch\n\nIn this tutorial, you will train, hyperparameter tune, and deploy a PyTorch model using the Azure Machine Learning (AML) Python SDK.\n\nThis tutorial will train an image classification model using transfer learning, based on PyTorch's [Transfer Learning tutorial](https://pytorch.org/tutorials/beginner/transfer_learning_tutorial.html). The model is trained to classify ants and bees by first using a pretrained ResNet18 model that has been trained on the [ImageNet](http://image-net.org/index) dataset."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# Check core SDK version number\nimport azureml.core\n\nprint(\"SDK version:\", azureml.core.VERSION)",
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": "SDK version: 1.0.2\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "## Diagnostics\nOpt-in diagnostics for better experience, quality, and security of future releases."
    },
    {
      "metadata": {
        "tags": [
          "Diagnostics"
        ],
        "trusted": true
      },
      "cell_type": "code",
      "source": "from azureml.telemetry import set_diagnostics_collection\n\nset_diagnostics_collection(send_diagnostics=True)",
      "execution_count": 2,
      "outputs": []
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "## Initialize workspace\nInitialize a [Workspace](https://docs.microsoft.com/azure/machine-learning/service/concept-azure-machine-learning-architecture#workspace) object from the existing workspace you created in the Prerequisites step. `Workspace.from_config()` creates a workspace object from the details stored in `config.json`."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from azureml.core.workspace import Workspace\n\nws = Workspace.from_config()\nprint('Workspace name: ' + ws.name, \n      'Azure region: ' + ws.location, \n      'Subscription id: ' + ws.subscription_id, \n      'Resource group: ' + ws.resource_group, sep = '\\n')",
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": "Found the config file in: /home/nbuser/library/aml_config/config.json\nWorkspace name: AMLSworkspace\nAzure region: westeurope\nSubscription id: 70b8f39e-8863-49f7-b6ba-34a80799550c\nResource group: resgrpAMLS\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "## Create a remote compute target\nYou will need to create a [compute target](https://docs.microsoft.com/azure/machine-learning/service/concept-azure-machine-learning-architecture#compute-target) to execute your training script on. In this tutorial, you create an `AmlCompute` cluster as your training compute resource. This code creates a cluster for you if it does not already exist in your workspace.\n\n**Creation of the cluster takes approximately 5 minutes.** If the cluster is already in your workspace this code will skip the cluster creation process."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from azureml.core.compute import ComputeTarget, AmlCompute\nfrom azureml.core.compute_target import ComputeTargetException\n\n# choose a name for your cluster\ncluster_name = \"gpucluster\"\n\ntry:\n    compute_target = ComputeTarget(workspace=ws, name=cluster_name)\n    print('Found existing compute target.')\nexcept ComputeTargetException:\n    print('Creating a new compute target...')\n    compute_config = AmlCompute.provisioning_configuration(vm_size='STANDARD_NC6', \n                                                           max_nodes=4)\n\n    # create the cluster\n    compute_target = ComputeTarget.create(ws, cluster_name, compute_config)\n\n    compute_target.wait_for_completion(show_output=True)\n\n# Use the 'status' property to get a detailed status for the current cluster. \nprint(compute_target.status.serialize())",
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "text": "Found existing compute target.\n{'allocationState': 'Resizing', 'allocationStateTransitionTime': '2018-12-26T13:27:24.128000+00:00', 'creationTime': '2018-12-10T16:09:58.703952+00:00', 'currentNodeCount': 0, 'errors': None, 'modifiedTime': '2018-12-10T16:11:24.446527+00:00', 'nodeStateCounts': {'idleNodeCount': 0, 'leavingNodeCount': 0, 'preemptedNodeCount': 0, 'preparingNodeCount': 0, 'runningNodeCount': 0, 'unusableNodeCount': 0}, 'provisioningState': 'Succeeded', 'provisioningStateTransitionTime': None, 'scaleSettings': {'minNodeCount': 0, 'maxNodeCount': 4, 'nodeIdleTimeBeforeScaleDown': 'PT120S'}, 'targetNodeCount': 1, 'vmPriority': 'Dedicated', 'vmSize': 'STANDARD_NC6'}\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "The above code creates a GPU cluster. If you instead want to create a CPU cluster, provide a different VM size to the `vm_size` parameter, such as `STANDARD_D2_V2`."
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "## Train model on the remote compute\nNow that you have your data and training script prepared, you are ready to train on your remote compute cluster. You can take advantage of Azure compute to leverage GPUs to cut down your training time. "
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### Create a project directory\nCreate a directory that will contain all the necessary code from your local machine that you will need access to on the remote resource. This includes the training script and any additional files your training script depends on."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "import os\n\nproject_folder = './pytorch-hymenoptera'\nos.makedirs(project_folder, exist_ok=True)",
      "execution_count": 5,
      "outputs": []
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### Download training data\nThe dataset we will use (located [here](https://download.pytorch.org/tutorial/hymenoptera_data.zip) as a zip file) consists of about 120 training images each for ants and bees, with 75 validation images for each class. [Hymenoptera](https://en.wikipedia.org/wiki/Hymenoptera) is the order of insects that includes ants and bees. We will download and extract the dataset as part of our training script `pytorch_train.py`"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### Prepare training script\nNow you will need to create your training script. In this tutorial, the training script is already provided for you at `pytorch_train.py`. In practice, you should be able to take any custom training script as is and run it with AML without having to modify your code.\n\nHowever, if you would like to use AML's [tracking and metrics](https://docs.microsoft.com/azure/machine-learning/service/concept-azure-machine-learning-architecture#metrics) capabilities, you will have to add a small amount of AML code inside your training script. \n\nIn `pytorch_train.py`, we will log some metrics to our AML run. To do so, we will access the AML run object within the script:\n```Python\nfrom azureml.core.run import Run\nrun = Run.get_context()\n```\nFurther within `pytorch_train.py`, we log the learning rate and momentum parameters, and the best validation accuracy the model achieves:\n```Python\nrun.log('lr', np.float(learning_rate))\nrun.log('momentum', np.float(momentum))\n\nrun.log('best_val_acc', np.float(best_acc))\n```\nThese run metrics will become particularly important when we begin hyperparameter tuning our model in the \"Tune model hyperparameters\" section."
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Once your script is ready, copy the training script `pytorch_train.py` into your project directory."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "import shutil\n\nshutil.copy('pytorch_train.py', project_folder)",
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 6,
          "data": {
            "text/plain": "'./pytorch-hymenoptera/pytorch_train.py'"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### Create an experiment\nCreate an [Experiment](https://docs.microsoft.com/azure/machine-learning/service/concept-azure-machine-learning-architecture#experiment) to track all the runs in your workspace for this transfer learning PyTorch tutorial. "
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from azureml.core import Experiment\n\nexperiment_name = 'pytorch-hymenoptera'\nexperiment = Experiment(ws, name=experiment_name)",
      "execution_count": 7,
      "outputs": []
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### Create a PyTorch estimator\nThe AML SDK's PyTorch estimator enables you to easily submit PyTorch training jobs for both single-node and distributed runs. For more information on the PyTorch estimator, refer [here](https://docs.microsoft.com/azure/machine-learning/service/how-to-train-pytorch). The following code will define a single-node PyTorch job."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from azureml.train.dnn import PyTorch\n\nscript_params = {\n    '--num_epochs': 30,\n    '--output_dir': './outputs'\n}\n\nestimator = PyTorch(source_directory=project_folder, \n                    script_params=script_params,\n                    compute_target=compute_target,\n                    entry_script='pytorch_train.py',\n                    use_gpu=True)",
      "execution_count": 8,
      "outputs": []
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "The `script_params` parameter is a dictionary containing the command-line arguments to your training script `entry_script`. Please note the following:\n- We passed our training data reference `ds_data` to our script's `--data_dir` argument. This will 1) mount our datastore on the remote compute and 2) provide the path to the training data `hymenoptera_data` on our datastore.\n- We specified the output directory as `./outputs`. The `outputs` directory is specially treated by AML in that all the content in this directory gets uploaded to your workspace as part of your run history. The files written to this directory are therefore accessible even once your remote run is over. In this tutorial, we will save our trained model to this output directory.\n\nTo leverage the Azure VM's GPU for training, we set `use_gpu=True`."
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### Submit job\nRun your experiment by submitting your estimator object. Note that this call is asynchronous."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "run = experiment.submit(estimator)\nprint(run)",
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "text": "Run(Experiment: pytorch-hymenoptera,\nId: pytorch-hymenoptera_1545830893450,\nType: azureml.scriptrun,\nStatus: Queued)\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# to get more details of your run\nprint(run.get_details())",
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "text": "{'runId': 'pytorch-hymenoptera_1545830893450', 'target': 'gpucluster', 'status': 'Queued', 'properties': {'azureml.runsource': 'experiment', 'ContentSnapshotId': 'a0f005c9-2bb3-4314-a092-665accae8c34'}, 'runDefinition': {'Script': 'pytorch_train.py', 'Arguments': ['--num_epochs', '30', '--output_dir', './outputs'], 'SourceDirectoryDataStore': None, 'Framework': 0, 'Communicator': 0, 'Target': 'gpucluster', 'DataReferences': {}, 'JobName': None, 'AutoPrepareEnvironment': True, 'MaxRunDurationSeconds': None, 'NodeCount': 1, 'Environment': {'Python': {'InterpreterPath': 'python', 'UserManagedDependencies': False, 'CondaDependencies': {'name': 'project_environment', 'dependencies': ['python=3.6.2', {'pip': ['azureml-defaults']}, 'pytorch=0.4.0', 'torchvision=0.2.1']}, 'CondaDependenciesFile': None}, 'EnvironmentVariables': {'EXAMPLE_ENV_VAR': 'EXAMPLE_VALUE', 'NCCL_SOCKET_IFNAME': '^docker0'}, 'Docker': {'BaseImage': 'mcr.microsoft.com/azureml/base-gpu:0.2.0', 'Enabled': True, 'SharedVolumes': True, 'Preparation': None, 'GpuSupport': True, 'Arguments': [], 'BaseImageRegistry': {'Address': None, 'Username': None, 'Password': None}}, 'Spark': {'Repositories': ['https://mmlspark.azureedge.net/maven'], 'Packages': [{'Group': 'com.microsoft.ml.spark', 'Artifact': 'mmlspark_2.11', 'Version': '0.12'}], 'PrecachePackages': True}}, 'History': {'OutputCollection': True}, 'Spark': {'Configuration': {'spark.app.name': 'Azure ML Experiment', 'spark.yarn.maxAppAttempts': '1'}}, 'BatchAi': {'NodeCount': 0}, 'AmlCompute': {'Name': None, 'VmSize': None, 'VmPriority': None, 'RetainCluster': False, 'ClusterMaxNodeCount': 1}, 'Tensorflow': {'WorkerCount': 1, 'ParameterServerCount': 1}, 'Mpi': {'ProcessCountPerNode': 1}, 'Hdi': {'YarnDeployMode': 2}, 'ContainerInstance': {'Region': None, 'CpuCores': 0, 'MemoryGb': 0}, 'ExposedPorts': None, 'PrepareEnvironment': None}, 'logFiles': {}}\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### Monitor your run\nYou can monitor the progress of the run with a Jupyter widget. Like the run submission, the widget is asynchronous and provides live updates every 10-15 seconds until the job completes."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from azureml.widgets import RunDetails\n\nRunDetails(run).show()",
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "d75a435425cd4269b92883cf03b03524",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": "_UserRunWidget(widget_settings={'childWidgetDisplay': 'popup', 'send_telemetry': True, 'log_level': 'INFO', 's…"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Alternatively, you can block until the script has completed training before running more code."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "run.wait_for_completion(show_output=True)",
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "text": "RunId: pytorch-hymenoptera_1545830893450\n\nStreaming azureml-logs/80_driver_log.txt\n========================================\n\nextracting files...\nfinished extracting\ndata directory is: hymenoptera_data/\nDownloading: \"https://download.pytorch.org/models/resnet18-5c106cde.pth\" to /root/.torch/models/resnet18-5c106cde.pth\n\n0.0%\n0.0%\n0.1%\n0.1%\n0.1%\n0.1%\n0.1%\n0.1%\n0.2%\n0.2%\n0.2%\n0.2%\n0.2%\n0.2%\n0.3%\n0.3%\n0.3%\n0.3%\n0.3%\n0.3%\n0.4%\n0.4%\n0.4%\n0.4%\n0.4%\n0.5%\n0.5%\n0.5%\n0.5%\n0.5%\n0.5%\n0.6%\n0.6%\n0.6%\n0.6%\n0.6%\n0.6%\n0.7%\n0.7%\n0.7%\n0.7%\n0.7%\n0.8%\n0.8%\n0.8%\n0.8%\n0.8%\n0.8%\n0.9%\n0.9%\n0.9%\n0.9%\n0.9%\n0.9%\n1.0%\n1.0%\n1.0%\n1.0%\n1.0%\n1.0%\n1.1%\n1.1%\n1.1%\n1.1%\n1.1%\n1.2%\n1.2%\n1.2%\n1.2%\n1.2%\n1.2%\n1.3%\n1.3%\n1.3%\n1.3%\n1.3%\n1.3%\n1.4%\n1.4%\n1.4%\n1.4%\n1.4%\n1.5%\n1.5%\n1.5%\n1.5%\n1.5%\n1.5%\n1.6%\n1.6%\n1.6%\n1.6%\n1.6%\n1.6%\n1.7%\n1.7%\n1.7%\n1.7%\n1.7%\n1.7%\n1.8%\n1.8%\n1.8%\n1.8%\n1.8%\n1.9%\n1.9%\n1.9%\n1.9%\n1.9%\n1.9%\n2.0%\n2.0%\n2.0%\n2.0%\n2.0%\n2.0%\n2.1%\n2.1%\n2.1%\n2.1%\n2.1%\n2.2%\n2.2%\n2.2%\n2.2%\n2.2%\n2.2%\n2.3%\n2.3%\n2.3%\n2.3%\n2.3%\n2.3%\n2.4%\n2.4%\n2.4%\n2.4%\n2.4%\n2.4%\n2.5%\n2.5%\n2.5%\n2.5%\n2.5%\n2.6%\n2.6%\n2.6%\n2.6%\n2.6%\n2.6%\n2.7%\n2.7%\n2.7%\n2.7%\n2.7%\n2.7%\n2.8%\n2.8%\n2.8%\n2.8%\n2.8%\n2.9%\n2.9%\n2.9%\n2.9%\n2.9%\n2.9%\n3.0%\n3.0%\n3.0%\n3.0%\n3.0%\n3.0%\n3.1%\n3.1%\n3.1%\n3.1%\n3.1%\n3.1%\n3.2%\n3.2%\n3.2%\n3.2%\n3.2%\n3.3%\n3.3%\n3.3%\n3.3%\n3.3%\n3.3%\n3.4%\n3.4%\n3.4%\n3.4%\n3.4%\n3.4%\n3.5%\n3.5%\n3.5%\n3.5%\n3.5%\n3.6%\n3.6%\n3.6%\n3.6%\n3.6%\n3.6%\n3.7%\n3.7%\n3.7%\n3.7%\n3.7%\n3.7%\n3.8%\n3.8%\n3.8%\n3.8%\n3.8%\n3.8%\n3.9%\n3.9%\n3.9%\n3.9%\n3.9%\n4.0%\n4.0%\n4.0%\n4.0%\n4.0%\n4.0%\n4.1%\n4.1%\n4.1%\n4.1%\n4.1%\n4.1%\n4.2%\n4.2%\n4.2%\n4.2%\n4.2%\n4.3%\n4.3%\n4.3%\n4.3%\n4.3%\n4.3%\n4.4%\n4.4%\n4.4%\n4.4%\n4.4%\n4.4%\n4.5%\n4.5%\n4.5%\n4.5%\n4.5%\n4.5%\n4.6%\n4.6%\n4.6%\n4.6%\n4.6%\n4.7%\n4.7%\n4.7%\n4.7%\n4.7%\n4.7%\n4.8%\n4.8%\n4.8%\n4.8%\n4.8%\n4.8%\n4.9%\n4.9%\n4.9%\n4.9%\n4.9%\n5.0%\n5.0%\n5.0%\n5.0%\n5.0%\n5.0%\n5.1%\n5.1%\n5.1%\n5.1%\n5.1%\n5.1%\n5.2%\n5.2%\n5.2%\n5.2%\n5.2%\n5.2%\n5.3%\n5.3%\n5.3%\n5.3%\n5.3%\n5.4%\n5.4%\n5.4%\n5.4%\n5.4%\n5.4%\n5.5%\n5.5%\n5.5%\n5.5%\n5.5%\n5.5%\n5.6%\n5.6%\n5.6%\n5.6%\n5.6%\n5.7%\n5.7%\n5.7%\n5.7%\n5.7%\n5.7%\n5.8%\n5.8%\n5.8%\n5.8%\n5.8%\n5.8%\n5.9%\n5.9%\n5.9%\n5.9%\n5.9%\n5.9%\n6.0%\n6.0%\n6.0%\n6.0%\n6.0%\n6.1%\n6.1%\n6.1%\n6.1%\n6.1%\n6.1%\n6.2%\n6.2%\n6.2%\n6.2%\n6.2%\n6.2%\n6.3%\n6.3%\n6.3%\n6.3%\n6.3%\n6.4%\n6.4%\n6.4%\n6.4%\n6.4%\n6.4%\n6.5%\n6.5%\n6.5%\n6.5%\n6.5%\n6.5%\n6.6%\n6.6%\n6.6%\n6.6%\n6.6%\n6.6%\n6.7%\n6.7%\n6.7%\n6.7%\n6.7%\n6.8%\n6.8%\n6.8%\n6.8%\n6.8%\n6.8%\n6.9%\n6.9%\n6.9%\n6.9%\n6.9%\n6.9%\n7.0%\n7.0%\n7.0%\n7.0%\n7.0%\n7.1%\n7.1%\n7.1%\n7.1%\n7.1%\n7.1%\n7.2%\n7.2%\n7.2%\n7.2%\n7.2%\n7.2%\n7.3%\n7.3%\n7.3%\n7.3%\n7.3%\n7.3%\n7.4%\n7.4%\n7.4%\n7.4%\n7.4%\n7.5%\n7.5%\n7.5%\n7.5%\n7.5%\n7.5%\n7.6%\n7.6%\n7.6%\n7.6%\n7.6%\n7.6%\n7.7%\n7.7%\n7.7%\n7.7%\n7.7%\n7.7%\n7.8%\n7.8%\n7.8%\n7.8%\n7.8%\n7.9%\n7.9%\n7.9%\n7.9%\n7.9%\n7.9%\n8.0%\n8.0%\n8.0%\n8.0%\n8.0%\n8.0%\n8.1%\n8.1%\n8.1%\n8.1%\n8.1%\n8.2%\n8.2%\n8.2%\n8.2%\n8.2%\n8.2%\n8.3%\n8.3%\n8.3%\n8.3%\n8.3%\n8.3%\n8.4%\n8.4%\n8.4%\n8.4%\n8.4%\n8.4%\n8.5%\n8.5%\n8.5%\n8.5%\n8.5%\n8.6%\n8.6%\n8.6%\n8.6%\n8.6%\n8.6%\n8.7%\n8.7%\n8.7%\n8.7%\n8.7%\n8.7%\n8.8%\n8.8%\n8.8%\n8.8%\n8.8%\n8.9%\n8.9%\n8.9%\n8.9%\n8.9%\n8.9%\n9.0%\n9.0%\n9.0%\n9.0%\n9.0%\n9.0%\n9.1%\n9.1%\n9.1%\n9.1%\n9.1%\n9.1%\n9.2%\n9.2%\n9.2%\n9.2%\n9.2%\n9.3%\n9.3%\n9.3%\n9.3%\n9.3%\n9.3%\n9.4%\n9.4%\n9.4%\n9.4%\n9.4%\n9.4%\n9.5%\n9.5%\n9.5%\n9.5%\n9.5%\n9.6%\n9.6%\n9.6%\n9.6%\n9.6%\n9.6%\n9.7%\n9.7%\n9.7%\n9.7%\n9.7%\n9.7%\n9.8%\n9.8%\n9.8%\n9.8%\n9.8%\n9.8%\n9.9%\n9.9%\n9.9%\n9.9%\n9.9%\n10.0%\n10.0%\n10.0%\n10.0%\n10.0%\n10.0%\n10.1%\n10.1%\n10.1%\n10.1%\n10.1%\n10.1%\n10.2%\n10.2%\n10.2%\n10.2%\n10.2%\n10.3%\n10.3%\n10.3%\n10.3%\n10.3%\n10.3%\n10.4%\n10.4%\n10.4%\n10.4%\n10.4%\n10.4%\n10.5%\n10.5%\n10.5%\n10.5%\n10.5%\n10.5%\n10.6%\n10.6%\n10.6%\n10.6%\n10.6%\n10.7%\n10.7%\n10.7%\n10.7%\n10.7%\n10.7%\n10.8%\n10.8%\n10.8%\n10.8%\n10.8%\n10.8%\n10.9%\n10.9%\n10.9%\n10.9%\n10.9%\n11.0%\n11.0%\n11.0%\n11.0%\n11.0%\n11.0%\n11.1%\n11.1%\n11.1%\n11.1%\n11.1%\n11.1%\n11.2%\n11.2%\n11.2%\n11.2%\n11.2%\n11.2%\n11.3%\n11.3%\n11.3%\n11.3%\n11.3%\n11.4%\n11.4%\n11.4%\n11.4%\n11.4%\n11.4%\n11.5%\n11.5%\n11.5%\n11.5%\n11.5%\n11.5%\n11.6%\n11.6%\n11.6%\n11.6%\n11.6%\n11.7%\n11.7%\n11.7%\n11.7%\n11.7%\n11.7%\n11.8%\n11.8%\n11.8%\n11.8%\n11.8%\n11.8%\n11.9%\n11.9%\n11.9%\n11.9%\n11.9%\n11.9%\n12.0%\n12.0%\n12.0%\n12.0%\n12.0%\n12.1%\n12.1%\n12.1%\n12.1%\n12.1%\n12.1%\n12.2%\n12.2%\n12.2%\n12.2%\n12.2%\n12.2%\n12.3%\n12.3%\n12.3%\n12.3%\n12.3%\n12.4%\n12.4%\n12.4%\n12.4%\n12.4%\n12.4%\n12.5%\n12.5%\n12.5%\n12.5%\n12.5%\n12.5%\n12.6%\n12.6%\n12.6%\n12.6%\n12.6%\n12.6%\n12.7%\n12.7%\n12.7%\n12.7%\n12.7%\n12.8%\n12.8%\n12.8%\n12.8%\n12.8%\n12.8%\n12.9%\n12.9%\n12.9%\n12.9%\n12.9%\n12.9%\n13.0%\n13.0%\n13.0%\n13.0%\n13.0%\n13.1%\n13.1%\n13.1%\n13.1%\n13.1%\n13.1%\n13.2%\n13.2%\n13.2%\n13.2%\n13.2%\n13.2%\n13.3%\n13.3%\n13.3%\n13.3%\n13.3%\n13.3%\n13.4%\n13.4%\n13.4%\n13.4%\n13.4%\n13.5%\n13.5%\n13.5%\n13.5%\n13.5%\n13.5%\n13.6%\n13.6%\n13.6%\n13.6%\n13.6%\n13.6%\n13.7%\n13.7%\n13.7%\n13.7%\n13.7%\n13.8%\n13.8%\n13.8%\n13.8%\n13.8%\n13.8%\n13.9%\n13.9%\n13.9%\n13.9%\n13.9%\n13.9%\n14.0%\n14.0%\n14.0%\n14.0%\n14.0%\n14.0%\n14.1%\n14.1%\n14.1%\n14.1%\n14.1%\n14.2%\n14.2%\n14.2%\n14.2%\n14.2%\n14.2%\n14.3%\n14.3%\n14.3%\n14.3%\n14.3%\n14.3%\n14.4%\n14.4%\n14.4%\n14.4%\n14.4%\n14.5%\n14.5%\n14.5%\n14.5%\n14.5%\n14.5%\n14.6%\n14.6%\n14.6%\n14.6%\n14.6%\n14.6%\n14.7%\n14.7%\n14.7%\n14.7%\n14.7%\n14.7%\n14.8%\n14.8%\n14.8%\n14.8%\n14.8%\n14.9%\n14.9%\n14.9%\n14.9%\n14.9%\n14.9%\n15.0%\n15.0%\n15.0%\n15.0%\n15.0%\n15.0%\n15.1%\n15.1%\n15.1%\n15.1%\n15.1%\n15.1%\n15.2%\n15.2%\n15.2%\n15.2%\n15.2%\n15.3%\n15.3%\n15.3%\n15.3%\n15.3%\n15.3%\n15.4%\n15.4%\n15.4%\n15.4%\n15.4%\n15.4%\n15.5%\n15.5%\n15.5%\n15.5%\n15.5%\n15.6%\n15.6%\n15.6%\n15.6%\n15.6%\n15.6%\n15.7%\n15.7%\n15.7%\n15.7%\n15.7%\n15.7%\n15.8%\n15.8%\n15.8%\n15.8%\n15.8%\n15.8%\n15.9%\n15.9%\n15.9%\n15.9%\n15.9%\n16.0%\n16.0%\n16.0%\n16.0%\n16.0%\n16.0%\n16.1%\n16.1%\n16.1%\n16.1%\n16.1%\n16.1%\n16.2%\n16.2%\n16.2%\n16.2%\n16.2%\n16.3%\n16.3%\n16.3%\n16.3%\n16.3%\n16.3%\n16.4%\n16.4%\n16.4%\n16.4%\n16.4%\n16.4%\n16.5%\n16.5%\n16.5%\n16.5%\n16.5%\n16.5%\n16.6%\n16.6%\n16.6%\n16.6%\n16.6%\n16.7%\n16.7%\n16.7%\n16.7%\n16.7%\n16.7%\n16.8%\n16.8%\n16.8%\n16.8%\n16.8%\n16.8%\n16.9%\n16.9%\n16.9%\n16.9%\n16.9%\n17.0%\n17.0%\n17.0%\n17.0%\n17.0%\n17.0%\n17.1%\n17.1%\n17.1%\n17.1%\n17.1%\n17.1%\n17.2%\n17.2%\n17.2%\n17.2%\n17.2%\n17.2%\n17.3%\n17.3%\n17.3%\n17.3%\n17.3%\n17.4%\n17.4%\n17.4%\n17.4%\n17.4%\n17.4%\n17.5%\n17.5%\n17.5%\n17.5%\n17.5%\n17.5%\n17.6%\n17.6%\n17.6%\n17.6%\n17.6%\n17.7%\n17.7%\n17.7%\n17.7%\n17.7%\n17.7%\n17.8%\n17.8%\n17.8%\n17.8%\n17.8%\n17.8%\n17.9%\n17.9%\n17.9%\n17.9%\n17.9%\n17.9%\n18.0%\n18.0%\n18.0%\n18.0%\n18.0%\n18.1%\n18.1%\n18.1%\n18.1%\n18.1%\n18.1%\n18.2%\n18.2%\n18.2%\n18.2%\n18.2%\n18.2%\n18.3%\n18.3%\n18.3%\n18.3%\n18.3%\n18.4%\n18.4%\n18.4%\n18.4%\n18.4%\n18.4%\n18.5%\n18.5%\n18.5%\n18.5%\n18.5%\n18.5%\n18.6%\n18.6%\n18.6%\n18.6%\n18.6%\n18.6%\n18.7%\n18.7%\n18.7%\n18.7%\n18.7%\n18.8%\n18.8%\n18.8%\n18.8%\n18.8%\n18.8%\n18.9%\n18.9%\n18.9%\n18.9%\n18.9%\n18.9%\n19.0%\n19.0%\n19.0%\n19.0%\n19.0%\n19.1%\n19.1%\n19.1%\n19.1%\n19.1%\n19.1%\n19.2%\n19.2%\n19.2%\n19.2%\n19.2%\n19.2%\n19.3%\n19.3%\n19.3%\n19.3%\n19.3%\n19.3%\n19.4%\n19.4%\n19.4%\n19.4%\n19.4%\n19.5%\n19.5%\n19.5%\n19.5%\n19.5%\n19.5%\n19.6%\n19.6%\n19.6%\n19.6%\n19.6%\n19.6%\n19.7%\n19.7%\n19.7%\n19.7%\n19.7%\n19.8%\n19.8%\n19.8%\n19.8%\n19.8%\n19.8%\n19.9%\n19.9%\n19.9%\n19.9%\n19.9%\n19.9%\n20.0%\n20.0%\n20.0%\n20.0%\n20.0%\n20.0%\n20.1%\n20.1%\n20.1%\n20.1%\n20.1%\n20.2%\n20.2%\n20.2%\n20.2%\n20.2%\n20.2%\n20.3%\n20.3%\n20.3%\n20.3%\n20.3%\n20.3%\n20.4%\n20.4%\n20.4%\n20.4%\n20.4%\n20.5%\n20.5%\n20.5%\n20.5%\n20.5%\n20.5%\n20.6%\n20.6%\n20.6%\n20.6%\n20.6%\n20.6%\n20.7%\n20.7%\n20.7%\n20.7%\n20.7%\n20.7%\n20.8%\n20.8%\n20.8%\n20.8%\n20.8%\n20.9%\n20.9%\n20.9%\n20.9%\n20.9%\n20.9%\n21.0%\n21.0%\n21.0%\n21.0%\n21.0%\n21.0%\n21.1%\n21.1%\n21.1%\n21.1%\n21.1%\n21.2%\n21.2%\n21.2%\n21.2%\n21.2%\n21.2%\n21.3%\n21.3%\n21.3%\n21.3%\n21.3%\n21.3%\n21.4%\n21.4%\n21.4%\n21.4%\n21.4%\n21.4%\n21.5%\n21.5%\n21.5%\n21.5%\n21.5%\n21.6%\n21.6%\n21.6%\n21.6%\n21.6%\n21.6%\n21.7%\n21.7%\n21.7%\n21.7%\n21.7%\n21.7%\n21.8%\n21.8%\n21.8%\n21.8%\n21.8%\n21.8%\n21.9%\n21.9%\n21.9%\n21.9%\n21.9%\n22.0%\n22.0%\n22.0%\n22.0%\n22.0%\n22.0%\n22.1%\n22.1%\n22.1%\n22.1%\n22.1%\n22.1%\n22.2%\n22.2%\n22.2%\n22.2%\n22.2%\n22.3%\n22.3%\n22.3%\n22.3%\n22.3%\n22.3%\n22.4%\n22.4%\n22.4%\n22.4%\n22.4%\n22.4%\n22.5%\n22.5%\n22.5%\n22.5%\n22.5%\n22.5%\n22.6%\n22.6%\n22.6%\n22.6%\n22.6%\n22.7%\n22.7%\n22.7%\n22.7%\n22.7%\n22.7%\n22.8%\n22.8%\n22.8%\n22.8%\n22.8%\n22.8%\n22.9%\n22.9%\n22.9%\n22.9%\n22.9%\n23.0%\n23.0%\n23.0%\n23.0%\n23.0%\n23.0%\n23.1%\n23.1%\n23.1%\n23.1%\n23.1%\n23.1%\n23.2%\n23.2%\n23.2%\n23.2%\n23.2%\n23.2%\n23.3%\n23.3%\n23.3%\n23.3%\n23.3%\n23.4%\n23.4%\n23.4%\n23.4%\n23.4%\n23.4%\n23.5%\n23.5%\n23.5%\n23.5%\n23.5%\n23.5%\n23.6%\n23.6%\n23.6%\n23.6%\n23.6%\n23.7%\n23.7%\n23.7%\n23.7%\n23.7%\n23.7%\n23.8%\n23.8%\n23.8%\n23.8%\n23.8%\n23.8%\n23.9%\n23.9%\n23.9%\n23.9%\n23.9%\n23.9%\n24.0%\n24.0%\n24.0%\n24.0%\n24.0%\n24.1%\n24.1%\n24.1%\n24.1%\n24.1%\n24.1%\n24.2%\n24.2%\n24.2%\n24.2%\n24.2%\n24.2%\n24.3%\n24.3%\n24.3%\n24.3%\n24.3%\n24.4%\n24.4%\n24.4%\n24.4%\n24.4%\n24.4%\n24.5%\n24.5%\n24.5%\n24.5%\n24.5%\n24.5%\n24.6%\n24.6%\n24.6%\n24.6%\n24.6%\n24.6%\n24.7%\n24.7%\n24.7%\n24.7%\n24.7%\n24.8%\n24.8%\n24.8%\n24.8%\n24.8%\n24.8%\n24.9%\n24.9%\n24.9%\n24.9%\n24.9%\n24.9%\n25.0%\n25.0%\n25.0%\n25.0%\n25.0%\n25.1%\n25.1%\n25.1%\n25.1%\n25.1%\n25.1%\n25.2%\n25.2%\n25.2%\n25.2%\n25.2%\n25.2%\n25.3%\n25.3%\n25.3%\n25.3%\n25.3%\n25.3%\n25.4%\n25.4%\n25.4%\n25.4%\n25.4%\n25.5%\n25.5%\n25.5%\n25.5%\n25.5%\n25.5%\n25.6%\n25.6%\n25.6%\n25.6%\n25.6%\n25.6%\n25.7%\n25.7%\n25.7%\n25.7%\n25.7%\n25.8%\n25.8%\n25.8%\n25.8%\n25.8%\n25.8%\n25.9%\n25.9%\n25.9%\n25.9%\n25.9%\n25.9%\n26.0%\n26.0%\n26.0%\n26.0%\n26.0%\n26.0%\n26.1%\n26.1%\n26.1%\n26.1%\n26.1%\n26.2%\n26.2%\n26.2%\n26.2%\n26.2%\n26.2%\n26.3%\n26.3%\n26.3%\n26.3%\n26.3%\n26.3%\n26.4%\n26.4%\n26.4%\n26.4%\n26.4%\n26.5%\n26.5%\n26.5%\n26.5%\n26.5%\n26.5%\n26.6%\n26.6%\n26.6%\n26.6%\n26.6%\n26.6%\n26.7%\n26.7%\n26.7%\n26.7%\n26.7%\n26.7%\n26.8%\n26.8%\n26.8%\n26.8%\n26.8%\n26.9%\n26.9%\n26.9%\n26.9%\n26.9%\n26.9%\n27.0%\n27.0%\n27.0%\n27.0%\n27.0%\n27.0%\n27.1%\n27.1%\n27.1%\n27.1%\n27.1%\n27.2%\n27.2%\n27.2%\n27.2%\n27.2%\n27.2%\n27.3%\n27.3%\n27.3%\n27.3%\n27.3%\n27.3%\n27.4%\n27.4%\n27.4%\n27.4%\n27.4%\n27.4%\n27.5%\n27.5%\n27.5%\n27.5%\n27.5%\n27.6%\n27.6%\n27.6%\n27.6%\n27.6%\n27.6%\n27.7%\n27.7%\n27.7%\n27.7%\n27.7%\n27.7%\n27.8%\n27.8%\n27.8%\n27.8%\n27.8%\n27.9%\n27.9%\n27.9%\n27.9%\n27.9%\n27.9%\n28.0%\n28.0%\n28.0%\n28.0%\n28.0%\n28.0%\n28.1%\n28.1%\n28.1%\n28.1%\n28.1%\n28.1%\n28.2%\n28.2%\n28.2%\n28.2%\n28.2%\n28.3%\n28.3%\n28.3%\n28.3%\n28.3%\n28.3%\n28.4%\n28.4%\n28.4%\n28.4%\n28.4%\n28.4%\n28.5%\n28.5%\n28.5%\n28.5%\n28.5%\n28.6%\n28.6%\n28.6%\n28.6%\n28.6%\n28.6%\n28.7%\n28.7%\n28.7%\n28.7%\n28.7%\n28.7%\n28.8%\n28.8%\n28.8%\n28.8%\n28.8%\n28.8%\n28.9%\n28.9%\n28.9%\n28.9%\n28.9%\n29.0%\n29.0%\n29.0%\n29.0%\n29.0%\n29.0%\n29.1%\n29.1%\n29.1%\n29.1%\n29.1%\n29.1%\n29.2%\n29.2%\n29.2%\n29.2%\n29.2%\n29.2%\n29.3%\n29.3%\n29.3%\n29.3%\n29.3%\n29.4%\n29.4%\n29.4%\n29.4%\n29.4%\n29.4%\n29.5%\n29.5%\n29.5%\n29.5%\n29.5%\n29.5%\n29.6%\n29.6%\n29.6%\n29.6%\n29.6%\n29.7%\n29.7%\n29.7%\n29.7%\n29.7%\n29.7%\n29.8%\n29.8%\n29.8%\n29.8%\n29.8%\n29.8%\n29.9%\n29.9%\n29.9%\n29.9%\n29.9%\n29.9%\n30.0%\n30.0%\n30.0%\n30.0%\n30.0%\n30.1%\n30.1%\n30.1%\n30.1%\n30.1%\n30.1%\n30.2%\n30.2%\n30.2%\n30.2%\n30.2%\n30.2%\n30.3%\n30.3%\n30.3%\n30.3%\n30.3%\n30.4%\n30.4%\n30.4%\n30.4%\n30.4%\n30.4%\n30.5%\n30.5%\n30.5%\n30.5%\n30.5%\n30.5%\n30.6%\n30.6%\n30.6%\n30.6%\n30.6%\n30.6%\n30.7%\n30.7%\n30.7%\n30.7%\n30.7%\n30.8%\n30.8%\n30.8%\n30.8%\n30.8%\n30.8%\n30.9%\n30.9%\n30.9%\n30.9%\n30.9%\n30.9%\n31.0%\n31.0%\n31.0%\n31.0%\n31.0%\n31.1%\n31.1%\n31.1%\n31.1%\n31.1%\n31.1%\n31.2%\n31.2%\n31.2%\n31.2%\n31.2%\n31.2%\n31.3%\n31.3%\n31.3%\n31.3%\n31.3%\n31.3%\n31.4%\n31.4%\n31.4%\n31.4%\n31.4%\n31.5%\n31.5%\n31.5%\n31.5%\n31.5%\n31.5%\n31.6%\n31.6%\n31.6%\n31.6%\n31.6%\n31.6%\n31.7%\n31.7%\n31.7%\n31.7%\n31.7%\n31.8%\n31.8%\n31.8%\n31.8%\n31.8%\n31.8%\n31.9%\n31.9%\n31.9%\n31.9%\n31.9%\n31.9%\n32.0%\n32.0%\n32.0%\n32.0%\n32.0%\n32.0%\n32.1%\n32.1%\n32.1%\n32.1%\n32.1%\n32.2%\n32.2%\n32.2%\n32.2%\n32.2%\n32.2%\n32.3%\n32.3%\n32.3%\n32.3%\n32.3%\n32.3%\n32.4%\n32.4%\n32.4%\n32.4%\n32.4%\n32.5%\n32.5%\n32.5%\n32.5%\n32.5%\n32.5%\n32.6%\n32.6%\n32.6%\n32.6%\n32.6%\n32.6%\n32.7%\n32.7%\n32.7%\n32.7%\n32.7%\n32.7%\n32.8%\n32.8%\n32.8%\n32.8%\n32.8%\n32.9%\n32.9%\n32.9%\n32.9%\n32.9%\n32.9%\n33.0%\n33.0%\n33.0%\n33.0%\n33.0%\n33.0%\n33.1%\n33.1%\n33.1%\n33.1%\n33.1%\n33.2%\n33.2%\n33.2%\n33.2%\n33.2%\n33.2%\n33.3%\n33.3%\n33.3%\n33.3%\n33.3%\n33.3%\n33.4%\n33.4%\n33.4%\n33.4%\n33.4%\n33.4%\n33.5%\n33.5%\n33.5%\n33.5%\n33.5%\n33.6%\n33.6%\n33.6%\n33.6%\n33.6%\n33.6%\n33.7%\n33.7%\n33.7%\n33.7%\n33.7%\n33.7%\n33.8%\n33.8%\n33.8%\n33.8%\n33.8%\n33.9%\n33.9%\n33.9%\n33.9%\n33.9%\n33.9%\n34.0%\n34.0%\n34.0%\n34.0%\n34.0%\n34.0%\n34.1%\n34.1%\n34.1%\n34.1%\n34.1%\n34.1%\n34.2%\n34.2%\n34.2%\n34.2%\n34.2%\n34.3%\n34.3%\n34.3%\n34.3%\n34.3%\n34.3%\n34.4%\n34.4%\n34.4%\n34.4%\n34.4%\n34.4%\n34.5%\n34.5%\n34.5%\n34.5%\n34.5%\n34.6%\n34.6%\n34.6%\n34.6%\n34.6%\n34.6%\n34.7%\n34.7%\n34.7%\n",
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": "34.7%\n34.7%\n34.7%\n34.8%\n34.8%\n34.8%\n34.8%\n34.8%\n34.8%\n34.9%\n34.9%\n34.9%\n34.9%\n34.9%\n35.0%\n35.0%\n35.0%\n35.0%\n35.0%\n35.0%\n35.1%\n35.1%\n35.1%\n35.1%\n35.1%\n35.1%\n35.2%\n35.2%\n35.2%\n35.2%\n35.2%\n35.3%\n35.3%\n35.3%\n35.3%\n35.3%\n35.3%\n35.4%\n35.4%\n35.4%\n35.4%\n35.4%\n35.4%\n35.5%\n35.5%\n35.5%\n35.5%\n35.5%\n35.5%\n35.6%\n35.6%\n35.6%\n35.6%\n35.6%\n35.7%\n35.7%\n35.7%\n35.7%\n35.7%\n35.7%\n35.8%\n35.8%\n35.8%\n35.8%\n35.8%\n35.8%\n35.9%\n35.9%\n35.9%\n35.9%\n35.9%\n36.0%\n36.0%\n36.0%\n36.0%\n36.0%\n36.0%\n36.1%\n36.1%\n36.1%\n36.1%\n36.1%\n36.1%\n36.2%\n36.2%\n36.2%\n36.2%\n36.2%\n36.2%\n36.3%\n36.3%\n36.3%\n36.3%\n36.3%\n36.4%\n36.4%\n36.4%\n36.4%\n36.4%\n36.4%\n36.5%\n36.5%\n36.5%\n36.5%\n36.5%\n36.5%\n36.6%\n36.6%\n36.6%\n36.6%\n36.6%\n36.6%\n36.7%\n36.7%\n36.7%\n36.7%\n36.7%\n36.8%\n36.8%\n36.8%\n36.8%\n36.8%\n36.8%\n36.9%\n36.9%\n36.9%\n36.9%\n36.9%\n36.9%\n37.0%\n37.0%\n37.0%\n37.0%\n37.0%\n37.1%\n37.1%\n37.1%\n37.1%\n37.1%\n37.1%\n37.2%\n37.2%\n37.2%\n37.2%\n37.2%\n37.2%\n37.3%\n37.3%\n37.3%\n37.3%\n37.3%\n37.3%\n37.4%\n37.4%\n37.4%\n37.4%\n37.4%\n37.5%\n37.5%\n37.5%\n37.5%\n37.5%\n37.5%\n37.6%\n37.6%\n37.6%\n37.6%\n37.6%\n37.6%\n37.7%\n37.7%\n37.7%\n37.7%\n37.7%\n37.8%\n37.8%\n37.8%\n37.8%\n37.8%\n37.8%\n37.9%\n37.9%\n37.9%\n37.9%\n37.9%\n37.9%\n38.0%\n38.0%\n38.0%\n38.0%\n38.0%\n38.0%\n38.1%\n38.1%\n38.1%\n38.1%\n38.1%\n38.2%\n38.2%\n38.2%\n38.2%\n38.2%\n38.2%\n38.3%\n38.3%\n38.3%\n38.3%\n38.3%\n38.3%\n38.4%\n38.4%\n38.4%\n38.4%\n38.4%\n38.5%\n38.5%\n38.5%\n38.5%\n38.5%\n38.5%\n38.6%\n38.6%\n38.6%\n38.6%\n38.6%\n38.6%\n38.7%\n38.7%\n38.7%\n38.7%\n38.7%\n38.7%\n38.8%\n38.8%\n38.8%\n38.8%\n38.8%\n38.9%\n38.9%\n38.9%\n38.9%\n38.9%\n38.9%\n39.0%\n39.0%\n39.0%\n39.0%\n39.0%\n39.0%\n39.1%\n39.1%\n39.1%\n39.1%\n39.1%\n39.2%\n39.2%\n39.2%\n39.2%\n39.2%\n39.2%\n39.3%\n39.3%\n39.3%\n39.3%\n39.3%\n39.3%\n39.4%\n39.4%\n39.4%\n39.4%\n39.4%\n39.4%\n39.5%\n39.5%\n39.5%\n39.5%\n39.5%\n39.6%\n39.6%\n39.6%\n39.6%\n39.6%\n39.6%\n39.7%\n39.7%\n39.7%\n39.7%\n39.7%\n39.7%\n39.8%\n39.8%\n39.8%\n39.8%\n39.8%\n39.9%\n39.9%\n39.9%\n39.9%\n39.9%\n39.9%\n40.0%\n40.0%\n40.0%\n40.0%\n40.0%\n40.0%\n40.1%\n40.1%\n40.1%\n40.1%\n40.1%\n40.1%\n40.2%\n40.2%\n40.2%\n40.2%\n40.2%\n40.3%\n40.3%\n40.3%\n40.3%\n40.3%\n40.3%\n40.4%\n40.4%\n40.4%\n40.4%\n40.4%\n40.4%\n40.5%\n40.5%\n40.5%\n40.5%\n40.5%\n40.6%\n40.6%\n40.6%\n40.6%\n40.6%\n40.6%\n40.7%\n40.7%\n40.7%\n40.7%\n40.7%\n40.7%\n40.8%\n40.8%\n40.8%\n40.8%\n40.8%\n40.8%\n40.9%\n40.9%\n40.9%\n40.9%\n40.9%\n41.0%\n41.0%\n41.0%\n41.0%\n41.0%\n41.0%\n41.1%\n41.1%\n41.1%\n41.1%\n41.1%\n41.1%\n41.2%\n41.2%\n41.2%\n41.2%\n41.2%\n41.3%\n41.3%\n41.3%\n41.3%\n41.3%\n41.3%\n41.4%\n41.4%\n41.4%\n41.4%\n41.4%\n41.4%\n41.5%\n41.5%\n41.5%\n41.5%\n41.5%\n41.5%\n41.6%\n41.6%\n41.6%\n41.6%\n41.6%\n41.7%\n41.7%\n41.7%\n41.7%\n41.7%\n41.7%\n41.8%\n41.8%\n41.8%\n41.8%\n41.8%\n41.8%\n41.9%\n41.9%\n41.9%\n41.9%\n41.9%\n42.0%\n42.0%\n42.0%\n42.0%\n42.0%\n42.0%\n42.1%\n42.1%\n42.1%\n42.1%\n42.1%\n42.1%\n42.2%\n42.2%\n42.2%\n42.2%\n42.2%\n42.2%\n42.3%\n42.3%\n42.3%\n42.3%\n42.3%\n42.4%\n42.4%\n42.4%\n42.4%\n42.4%\n42.4%\n42.5%\n42.5%\n42.5%\n42.5%\n42.5%\n42.5%\n42.6%\n42.6%\n42.6%\n42.6%\n42.6%\n42.7%\n42.7%\n42.7%\n42.7%\n42.7%\n42.7%\n42.8%\n42.8%\n42.8%\n42.8%\n42.8%\n42.8%\n42.9%\n42.9%\n42.9%\n42.9%\n42.9%\n42.9%\n43.0%\n43.0%\n43.0%\n43.0%\n43.0%\n43.1%\n43.1%\n43.1%\n43.1%\n43.1%\n43.1%\n43.2%\n43.2%\n43.2%\n43.2%\n43.2%\n43.2%\n43.3%\n43.3%\n43.3%\n43.3%\n43.3%\n43.4%\n43.4%\n43.4%\n43.4%\n43.4%\n43.4%\n43.5%\n43.5%\n43.5%\n43.5%\n43.5%\n43.5%\n43.6%\n43.6%\n43.6%\n43.6%\n43.6%\n43.6%\n43.7%\n43.7%\n43.7%\n43.7%\n43.7%\n43.8%\n43.8%\n43.8%\n43.8%\n43.8%\n43.8%\n43.9%\n43.9%\n43.9%\n43.9%\n43.9%\n43.9%\n44.0%\n44.0%\n44.0%\n44.0%\n44.0%\n44.0%\n44.1%\n44.1%\n44.1%\n44.1%\n44.1%\n44.2%\n44.2%\n44.2%\n44.2%\n44.2%\n44.2%\n44.3%\n44.3%\n44.3%\n44.3%\n44.3%\n44.3%\n44.4%\n44.4%\n44.4%\n44.4%\n44.4%\n44.5%\n44.5%\n44.5%\n44.5%\n44.5%\n44.5%\n44.6%\n44.6%\n44.6%\n44.6%\n44.6%\n44.6%\n44.7%\n44.7%\n44.7%\n44.7%\n44.7%\n44.7%\n44.8%\n44.8%\n44.8%\n44.8%\n44.8%\n44.9%\n44.9%\n44.9%\n44.9%\n44.9%\n44.9%\n45.0%\n45.0%\n45.0%\n45.0%\n45.0%\n45.0%\n45.1%\n45.1%\n45.1%\n45.1%\n45.1%\n45.2%\n45.2%\n45.2%\n45.2%\n45.2%\n45.2%\n45.3%\n45.3%\n45.3%\n45.3%\n45.3%\n45.3%\n45.4%\n45.4%\n45.4%\n45.4%\n45.4%\n45.4%\n45.5%\n45.5%\n45.5%\n45.5%\n45.5%\n45.6%\n45.6%\n45.6%\n45.6%\n45.6%\n45.6%\n45.7%\n45.7%\n45.7%\n45.7%\n45.7%\n45.7%\n45.8%\n45.8%\n45.8%\n45.8%\n45.8%\n45.9%\n45.9%\n45.9%\n45.9%\n45.9%\n45.9%\n46.0%\n46.0%\n46.0%\n46.0%\n46.0%\n46.0%\n46.1%\n46.1%\n46.1%\n46.1%\n46.1%\n46.1%\n46.2%\n46.2%\n46.2%\n46.2%\n46.2%\n46.3%\n46.3%\n46.3%\n46.3%\n46.3%\n46.3%\n46.4%\n46.4%\n46.4%\n46.4%\n46.4%\n46.4%\n46.5%\n46.5%\n46.5%\n46.5%\n46.5%\n46.6%\n46.6%\n46.6%\n46.6%\n46.6%\n46.6%\n46.7%\n46.7%\n46.7%\n46.7%\n46.7%\n46.7%\n46.8%\n46.8%\n46.8%\n46.8%\n46.8%\n46.8%\n46.9%\n46.9%\n46.9%\n46.9%\n46.9%\n47.0%\n47.0%\n47.0%\n47.0%\n47.0%\n47.0%\n47.1%\n47.1%\n47.1%\n47.1%\n47.1%\n47.1%\n47.2%\n47.2%\n47.2%\n47.2%\n47.2%\n47.3%\n47.3%\n47.3%\n47.3%\n47.3%\n47.3%\n47.4%\n47.4%\n47.4%\n47.4%\n47.4%\n47.4%\n47.5%\n47.5%\n47.5%\n47.5%\n47.5%\n47.5%\n47.6%\n47.6%\n47.6%\n47.6%\n47.6%\n47.7%\n47.7%\n47.7%\n47.7%\n47.7%\n47.7%\n47.8%\n47.8%\n47.8%\n47.8%\n47.8%\n47.8%\n47.9%\n47.9%\n47.9%\n47.9%\n47.9%\n48.0%\n48.0%\n48.0%\n48.0%\n48.0%\n48.0%\n48.1%\n48.1%\n48.1%\n48.1%\n48.1%\n48.1%\n48.2%\n48.2%\n48.2%\n48.2%\n48.2%\n48.2%\n48.3%\n48.3%\n48.3%\n48.3%\n48.3%\n48.4%\n48.4%\n48.4%\n48.4%\n48.4%\n48.4%\n48.5%\n48.5%\n48.5%\n48.5%\n48.5%\n48.5%\n48.6%\n48.6%\n48.6%\n48.6%\n48.6%\n48.7%\n48.7%\n48.7%\n48.7%\n48.7%\n48.7%\n48.8%\n48.8%\n48.8%\n48.8%\n48.8%\n48.8%\n48.9%\n48.9%\n48.9%\n48.9%\n48.9%\n48.9%\n49.0%\n49.0%\n49.0%\n49.0%\n49.0%\n49.1%\n49.1%\n49.1%\n49.1%\n49.1%\n49.1%\n49.2%\n49.2%\n49.2%\n49.2%\n49.2%\n49.2%\n49.3%\n49.3%\n49.3%\n49.3%\n49.3%\n49.4%\n49.4%\n49.4%\n49.4%\n49.4%\n49.4%\n49.5%\n49.5%\n49.5%\n49.5%\n49.5%\n49.5%\n49.6%\n49.6%\n49.6%\n49.6%\n49.6%\n49.6%\n49.7%\n49.7%\n49.7%\n49.7%\n49.7%\n49.8%\n49.8%\n49.8%\n49.8%\n49.8%\n49.8%\n49.9%\n49.9%\n49.9%\n49.9%\n49.9%\n49.9%\n50.0%\n50.0%\n50.0%\n50.0%\n50.0%\n50.1%\n50.1%\n50.1%\n50.1%\n50.1%\n50.1%\n50.2%\n50.2%\n50.2%\n50.2%\n50.2%\n50.2%\n50.3%\n50.3%\n50.3%\n50.3%\n50.3%\n50.3%\n50.4%\n50.4%\n50.4%\n50.4%\n50.4%\n50.5%\n50.5%\n50.5%\n50.5%\n50.5%\n50.5%\n50.6%\n50.6%\n50.6%\n50.6%\n50.6%\n50.6%\n50.7%\n50.7%\n50.7%\n50.7%\n50.7%\n50.8%\n50.8%\n50.8%\n50.8%\n50.8%\n50.8%\n50.9%\n50.9%\n50.9%\n50.9%\n50.9%\n50.9%\n51.0%\n51.0%\n51.0%\n51.0%\n51.0%\n51.0%\n51.1%\n51.1%\n51.1%\n51.1%\n51.1%\n51.2%\n51.2%\n51.2%\n51.2%\n51.2%\n51.2%\n51.3%\n51.3%\n51.3%\n51.3%\n51.3%\n51.3%\n51.4%\n51.4%\n51.4%\n51.4%\n51.4%\n51.4%\n51.5%\n51.5%\n51.5%\n51.5%\n51.5%\n51.6%\n51.6%\n51.6%\n51.6%\n51.6%\n51.6%\n51.7%\n51.7%\n51.7%\n51.7%\n51.7%\n51.7%\n51.8%\n51.8%\n51.8%\n51.8%\n51.8%\n51.9%\n51.9%\n51.9%\n51.9%\n51.9%\n51.9%\n52.0%\n52.0%\n52.0%\n52.0%\n52.0%\n52.0%\n52.1%\n52.1%\n52.1%\n52.1%\n52.1%\n52.1%\n52.2%\n52.2%\n52.2%\n52.2%\n52.2%\n52.3%\n52.3%\n52.3%\n52.3%\n52.3%\n52.3%\n52.4%\n52.4%\n52.4%\n52.4%\n52.4%\n52.4%\n52.5%\n52.5%\n52.5%\n52.5%\n52.5%\n52.6%\n52.6%\n52.6%\n52.6%\n52.6%\n52.6%\n52.7%\n52.7%\n52.7%\n52.7%\n52.7%\n52.7%\n52.8%\n52.8%\n52.8%\n52.8%\n52.8%\n52.8%\n52.9%\n52.9%\n52.9%\n52.9%\n52.9%\n53.0%\n53.0%\n53.0%\n53.0%\n53.0%\n53.0%\n53.1%\n53.1%\n53.1%\n53.1%\n53.1%\n53.1%\n53.2%\n53.2%\n53.2%\n53.2%\n53.2%\n53.3%\n53.3%\n53.3%\n53.3%\n53.3%\n53.3%\n53.4%\n53.4%\n53.4%\n53.4%\n53.4%\n53.4%\n53.5%\n53.5%\n53.5%\n53.5%\n53.5%\n53.5%\n53.6%\n53.6%\n53.6%\n53.6%\n53.6%\n53.7%\n53.7%\n53.7%\n53.7%\n53.7%\n53.7%\n53.8%\n53.8%\n53.8%\n53.8%\n53.8%\n53.8%\n53.9%\n53.9%\n53.9%\n53.9%\n53.9%\n54.0%\n54.0%\n54.0%\n54.0%\n54.0%\n54.0%\n54.1%\n54.1%\n54.1%\n54.1%\n54.1%\n54.1%\n54.2%\n54.2%\n54.2%\n54.2%\n54.2%\n54.2%\n54.3%\n54.3%\n54.3%\n54.3%\n54.3%\n54.4%\n54.4%\n54.4%\n54.4%\n54.4%\n54.4%\n54.5%\n54.5%\n54.5%\n54.5%\n54.5%\n54.5%\n54.6%\n54.6%\n54.6%\n54.6%\n54.6%\n54.7%\n54.7%\n54.7%\n54.7%\n54.7%\n54.7%\n54.8%\n54.8%\n54.8%\n54.8%\n54.8%\n54.8%\n54.9%\n54.9%\n54.9%\n54.9%\n54.9%\n54.9%\n55.0%\n55.0%\n55.0%\n55.0%\n55.0%\n55.1%\n55.1%\n55.1%\n55.1%\n55.1%\n55.1%\n55.2%\n55.2%\n55.2%\n55.2%\n55.2%\n55.2%\n55.3%\n55.3%\n55.3%\n55.3%\n55.3%\n55.4%\n55.4%\n55.4%\n55.4%\n55.4%\n55.4%\n55.5%\n55.5%\n55.5%\n55.5%\n55.5%\n55.5%\n55.6%\n55.6%\n55.6%\n55.6%\n55.6%\n55.6%\n55.7%\n55.7%\n55.7%\n55.7%\n55.7%\n55.8%\n55.8%\n55.8%\n55.8%\n55.8%\n55.8%\n55.9%\n55.9%\n55.9%\n55.9%\n55.9%\n55.9%\n56.0%\n56.0%\n56.0%\n56.0%\n56.0%\n56.1%\n56.1%\n56.1%\n56.1%\n56.1%\n56.1%\n56.2%\n56.2%\n56.2%\n56.2%\n56.2%\n56.2%\n56.3%\n56.3%\n56.3%\n56.3%\n56.3%\n56.3%\n56.4%\n56.4%\n56.4%\n56.4%\n56.4%\n56.5%\n56.5%\n56.5%\n56.5%\n56.5%\n56.5%\n56.6%\n56.6%\n56.6%\n56.6%\n56.6%\n56.6%\n56.7%\n56.7%\n56.7%\n56.7%\n56.7%\n56.8%\n56.8%\n56.8%\n56.8%\n56.8%\n56.8%\n56.9%\n56.9%\n56.9%\n56.9%\n56.9%\n56.9%\n57.0%\n57.0%\n57.0%\n57.0%\n57.0%\n57.0%\n57.1%\n57.1%\n57.1%\n57.1%\n57.1%\n57.2%\n57.2%\n57.2%\n57.2%\n57.2%\n57.2%\n57.3%\n57.3%\n57.3%\n57.3%\n57.3%\n57.3%\n57.4%\n57.4%\n57.4%\n57.4%\n57.4%\n57.5%\n57.5%\n57.5%\n57.5%\n57.5%\n57.5%\n57.6%\n57.6%\n57.6%\n57.6%\n57.6%\n57.6%\n57.7%\n57.7%\n57.7%\n57.7%\n57.7%\n57.7%\n57.8%\n57.8%\n57.8%\n57.8%\n57.8%\n57.9%\n57.9%\n57.9%\n57.9%\n57.9%\n57.9%\n58.0%\n58.0%\n58.0%\n58.0%\n58.0%\n58.0%\n58.1%\n58.1%\n58.1%\n58.1%\n58.1%\n58.2%\n58.2%\n58.2%\n58.2%\n58.2%\n58.2%\n58.3%\n58.3%\n58.3%\n58.3%\n58.3%\n58.3%\n58.4%\n58.4%\n58.4%\n58.4%\n58.4%\n58.4%\n58.5%\n58.5%\n58.5%\n58.5%\n58.5%\n58.6%\n58.6%\n58.6%\n58.6%\n58.6%\n58.6%\n58.7%\n58.7%\n58.7%\n58.7%\n58.7%\n58.7%\n58.8%\n58.8%\n58.8%\n58.8%\n58.8%\n58.8%\n58.9%\n58.9%\n58.9%\n58.9%\n58.9%\n59.0%\n59.0%\n59.0%\n59.0%\n59.0%\n59.0%\n59.1%\n59.1%\n59.1%\n59.1%\n59.1%\n59.1%\n59.2%\n59.2%\n59.2%\n59.2%\n59.2%\n59.3%\n59.3%\n59.3%\n59.3%\n59.3%\n59.3%\n59.4%\n59.4%\n59.4%\n59.4%\n59.4%\n59.4%\n59.5%\n59.5%\n59.5%\n59.5%\n59.5%\n59.5%\n59.6%\n59.6%\n59.6%\n59.6%\n59.6%\n59.7%\n59.7%\n59.7%\n59.7%\n59.7%\n59.7%\n59.8%\n59.8%\n59.8%\n59.8%\n59.8%\n59.8%\n59.9%\n59.9%\n59.9%\n59.9%\n59.9%\n60.0%\n60.0%\n60.0%\n60.0%\n60.0%\n60.0%\n60.1%\n60.1%\n60.1%\n60.1%\n60.1%\n60.1%\n60.2%\n60.2%\n60.2%\n60.2%\n60.2%\n60.2%\n60.3%\n60.3%\n60.3%\n60.3%\n60.3%\n60.4%\n60.4%\n60.4%\n60.4%\n60.4%\n60.4%\n60.5%\n60.5%\n60.5%\n60.5%\n60.5%\n60.5%\n60.6%\n60.6%\n60.6%\n60.6%\n60.6%\n60.7%\n60.7%\n60.7%\n60.7%\n60.7%\n60.7%\n60.8%\n60.8%\n60.8%\n60.8%\n60.8%\n60.8%\n60.9%\n60.9%\n60.9%\n60.9%\n60.9%\n60.9%\n61.0%\n61.0%\n61.0%\n61.0%\n61.0%\n61.1%\n61.1%\n61.1%\n61.1%\n61.1%\n61.1%\n61.2%\n61.2%\n61.2%\n61.2%\n61.2%\n61.2%\n61.3%\n61.3%\n61.3%\n61.3%\n61.3%\n61.4%\n61.4%\n61.4%\n61.4%\n61.4%\n61.4%\n61.5%\n61.5%\n61.5%\n61.5%\n61.5%\n61.5%\n61.6%\n61.6%\n61.6%\n61.6%\n61.6%\n61.6%\n61.7%\n61.7%\n61.7%\n61.7%\n61.7%\n61.8%\n61.8%\n61.8%\n61.8%\n61.8%\n61.8%\n61.9%\n61.9%\n61.9%\n61.9%\n61.9%\n61.9%\n62.0%\n62.0%\n62.0%\n62.0%\n62.0%\n62.1%\n62.1%\n62.1%\n62.1%\n62.1%\n62.1%\n62.2%\n62.2%\n62.2%\n62.2%\n62.2%\n62.2%\n62.3%\n62.3%\n62.3%\n62.3%\n62.3%\n62.3%\n62.4%\n62.4%\n62.4%\n62.4%\n62.4%\n62.5%\n62.5%\n62.5%\n62.5%\n62.5%\n62.5%\n62.6%\n62.6%\n62.6%\n62.6%\n62.6%\n62.6%\n62.7%\n62.7%\n62.7%\n62.7%\n62.7%\n62.8%\n62.8%\n62.8%\n62.8%\n62.8%\n62.8%\n62.9%\n62.9%\n62.9%\n62.9%\n62.9%\n62.9%\n63.0%\n63.0%\n63.0%\n63.0%\n63.0%\n63.0%\n63.1%\n63.1%\n63.1%\n63.1%\n63.1%\n63.2%\n63.2%\n63.2%\n63.2%\n63.2%\n63.2%\n63.3%\n63.3%\n63.3%\n63.3%\n63.3%\n63.3%\n63.4%\n63.4%\n63.4%\n63.4%\n63.4%\n63.5%\n63.5%\n63.5%\n63.5%\n63.5%\n63.5%\n63.6%\n",
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": "63.6%\n63.6%\n63.6%\n63.6%\n63.6%\n63.7%\n63.7%\n63.7%\n63.7%\n63.7%\n63.7%\n63.8%\n63.8%\n63.8%\n63.8%\n63.8%\n63.9%\n63.9%\n63.9%\n63.9%\n63.9%\n63.9%\n64.0%\n64.0%\n64.0%\n64.0%\n64.0%\n64.0%\n64.1%\n64.1%\n64.1%\n64.1%\n64.1%\n64.2%\n64.2%\n64.2%\n64.2%\n64.2%\n64.2%\n64.3%\n64.3%\n64.3%\n64.3%\n64.3%\n64.3%\n64.4%\n64.4%\n64.4%\n64.4%\n64.4%\n64.4%\n64.5%\n64.5%\n64.5%\n64.5%\n64.5%\n64.6%\n64.6%\n64.6%\n64.6%\n64.6%\n64.6%\n64.7%\n64.7%\n64.7%\n64.7%\n64.7%\n64.7%\n64.8%\n64.8%\n64.8%\n64.8%\n64.8%\n64.9%\n64.9%\n64.9%\n64.9%\n64.9%\n64.9%\n65.0%\n65.0%\n65.0%\n65.0%\n65.0%\n65.0%\n65.1%\n65.1%\n65.1%\n65.1%\n65.1%\n65.1%\n65.2%\n65.2%\n65.2%\n65.2%\n65.2%\n65.3%\n65.3%\n65.3%\n65.3%\n65.3%\n65.3%\n65.4%\n65.4%\n65.4%\n65.4%\n65.4%\n65.4%\n65.5%\n65.5%\n65.5%\n65.5%\n65.5%\n65.5%\n65.6%\n65.6%\n65.6%\n65.6%\n65.6%\n65.7%\n65.7%\n65.7%\n65.7%\n65.7%\n65.7%\n65.8%\n65.8%\n65.8%\n65.8%\n65.8%\n65.8%\n65.9%\n65.9%\n65.9%\n65.9%\n65.9%\n66.0%\n66.0%\n66.0%\n66.0%\n66.0%\n66.0%\n66.1%\n66.1%\n66.1%\n66.1%\n66.1%\n66.1%\n66.2%\n66.2%\n66.2%\n66.2%\n66.2%\n66.2%\n66.3%\n66.3%\n66.3%\n66.3%\n66.3%\n66.4%\n66.4%\n66.4%\n66.4%\n66.4%\n66.4%\n66.5%\n66.5%\n66.5%\n66.5%\n66.5%\n66.5%\n66.6%\n66.6%\n66.6%\n66.6%\n66.6%\n66.7%\n66.7%\n66.7%\n66.7%\n66.7%\n66.7%\n66.8%\n66.8%\n66.8%\n66.8%\n66.8%\n66.8%\n66.9%\n66.9%\n66.9%\n66.9%\n66.9%\n66.9%\n67.0%\n67.0%\n67.0%\n67.0%\n67.0%\n67.1%\n67.1%\n67.1%\n67.1%\n67.1%\n67.1%\n67.2%\n67.2%\n67.2%\n67.2%\n67.2%\n67.2%\n67.3%\n67.3%\n67.3%\n67.3%\n67.3%\n67.4%\n67.4%\n67.4%\n67.4%\n67.4%\n67.4%\n67.5%\n67.5%\n67.5%\n67.5%\n67.5%\n67.5%\n67.6%\n67.6%\n67.6%\n67.6%\n67.6%\n67.6%\n67.7%\n67.7%\n67.7%\n67.7%\n67.7%\n67.8%\n67.8%\n67.8%\n67.8%\n67.8%\n67.8%\n67.9%\n67.9%\n67.9%\n67.9%\n67.9%\n67.9%\n68.0%\n68.0%\n68.0%\n68.0%\n68.0%\n68.1%\n68.1%\n68.1%\n68.1%\n68.1%\n68.1%\n68.2%\n68.2%\n68.2%\n68.2%\n68.2%\n68.2%\n68.3%\n68.3%\n68.3%\n68.3%\n68.3%\n68.3%\n68.4%\n68.4%\n68.4%\n68.4%\n68.4%\n68.5%\n68.5%\n68.5%\n68.5%\n68.5%\n68.5%\n68.6%\n68.6%\n68.6%\n68.6%\n68.6%\n68.6%\n68.7%\n68.7%\n68.7%\n68.7%\n68.7%\n68.8%\n68.8%\n68.8%\n68.8%\n68.8%\n68.8%\n68.9%\n68.9%\n68.9%\n68.9%\n68.9%\n68.9%\n69.0%\n69.0%\n69.0%\n69.0%\n69.0%\n69.0%\n69.1%\n69.1%\n69.1%\n69.1%\n69.1%\n69.2%\n69.2%\n69.2%\n69.2%\n69.2%\n69.2%\n69.3%\n69.3%\n69.3%\n69.3%\n69.3%\n69.3%\n69.4%\n69.4%\n69.4%\n69.4%\n69.4%\n69.5%\n69.5%\n69.5%\n69.5%\n69.5%\n69.5%\n69.6%\n69.6%\n69.6%\n69.6%\n69.6%\n69.6%\n69.7%\n69.7%\n69.7%\n69.7%\n69.7%\n69.7%\n69.8%\n69.8%\n69.8%\n69.8%\n69.8%\n69.9%\n69.9%\n69.9%\n69.9%\n69.9%\n69.9%\n70.0%\n70.0%\n70.0%\n70.0%\n70.0%\n70.0%\n70.1%\n70.1%\n70.1%\n70.1%\n70.1%\n70.2%\n70.2%\n70.2%\n70.2%\n70.2%\n70.2%\n70.3%\n70.3%\n70.3%\n70.3%\n70.3%\n70.3%\n70.4%\n70.4%\n70.4%\n70.4%\n70.4%\n70.4%\n70.5%\n70.5%\n70.5%\n70.5%\n70.5%\n70.6%\n70.6%\n70.6%\n70.6%\n70.6%\n70.6%\n70.7%\n70.7%\n70.7%\n70.7%\n70.7%\n70.7%\n70.8%\n70.8%\n70.8%\n70.8%\n70.8%\n70.9%\n70.9%\n70.9%\n70.9%\n70.9%\n70.9%\n71.0%\n71.0%\n71.0%\n71.0%\n71.0%\n71.0%\n71.1%\n71.1%\n71.1%\n71.1%\n71.1%\n71.1%\n71.2%\n71.2%\n71.2%\n71.2%\n71.2%\n71.3%\n71.3%\n71.3%\n71.3%\n71.3%\n71.3%\n71.4%\n71.4%\n71.4%\n71.4%\n71.4%\n71.4%\n71.5%\n71.5%\n71.5%\n71.5%\n71.5%\n71.6%\n71.6%\n71.6%\n71.6%\n71.6%\n71.6%\n71.7%\n71.7%\n71.7%\n71.7%\n71.7%\n71.7%\n71.8%\n71.8%\n71.8%\n71.8%\n71.8%\n71.8%\n71.9%\n71.9%\n71.9%\n71.9%\n71.9%\n72.0%\n72.0%\n72.0%\n72.0%\n72.0%\n72.0%\n72.1%\n72.1%\n72.1%\n72.1%\n72.1%\n72.1%\n72.2%\n72.2%\n72.2%\n72.2%\n72.2%\n72.3%\n72.3%\n72.3%\n72.3%\n72.3%\n72.3%\n72.4%\n72.4%\n72.4%\n72.4%\n72.4%\n72.4%\n72.5%\n72.5%\n72.5%\n72.5%\n72.5%\n72.5%\n72.6%\n72.6%\n72.6%\n72.6%\n72.6%\n72.7%\n72.7%\n72.7%\n72.7%\n72.7%\n72.7%\n72.8%\n72.8%\n72.8%\n72.8%\n72.8%\n72.8%\n72.9%\n72.9%\n72.9%\n72.9%\n72.9%\n72.9%\n73.0%\n73.0%\n73.0%\n73.0%\n73.0%\n73.1%\n73.1%\n73.1%\n73.1%\n73.1%\n73.1%\n73.2%\n73.2%\n73.2%\n73.2%\n73.2%\n73.2%\n73.3%\n73.3%\n73.3%\n73.3%\n73.3%\n73.4%\n73.4%\n73.4%\n73.4%\n73.4%\n73.4%\n73.5%\n73.5%\n73.5%\n73.5%\n73.5%\n73.5%\n73.6%\n73.6%\n73.6%\n73.6%\n73.6%\n73.6%\n73.7%\n73.7%\n73.7%\n73.7%\n73.7%\n73.8%\n73.8%\n73.8%\n73.8%\n73.8%\n73.8%\n73.9%\n73.9%\n73.9%\n73.9%\n73.9%\n73.9%\n74.0%\n74.0%\n74.0%\n74.0%\n74.0%\n74.1%\n74.1%\n74.1%\n74.1%\n74.1%\n74.1%\n74.2%\n74.2%\n74.2%\n74.2%\n74.2%\n74.2%\n74.3%\n74.3%\n74.3%\n74.3%\n74.3%\n74.3%\n74.4%\n74.4%\n74.4%\n74.4%\n74.4%\n74.5%\n74.5%\n74.5%\n74.5%\n74.5%\n74.5%\n74.6%\n74.6%\n74.6%\n74.6%\n74.6%\n74.6%\n74.7%\n74.7%\n74.7%\n74.7%\n74.7%\n74.8%\n74.8%\n74.8%\n74.8%\n74.8%\n74.8%\n74.9%\n74.9%\n74.9%\n74.9%\n74.9%\n74.9%\n75.0%\n75.0%\n75.0%\n75.0%\n75.0%\n75.0%\n75.1%\n75.1%\n75.1%\n75.1%\n75.1%\n75.2%\n75.2%\n75.2%\n75.2%\n75.2%\n75.2%\n75.3%\n75.3%\n75.3%\n75.3%\n75.3%\n75.3%\n75.4%\n75.4%\n75.4%\n75.4%\n75.4%\n75.5%\n75.5%\n75.5%\n75.5%\n75.5%\n75.5%\n75.6%\n75.6%\n75.6%\n75.6%\n75.6%\n75.6%\n75.7%\n75.7%\n75.7%\n75.7%\n75.7%\n75.7%\n75.8%\n75.8%\n75.8%\n75.8%\n75.8%\n75.9%\n75.9%\n75.9%\n75.9%\n75.9%\n75.9%\n76.0%\n76.0%\n76.0%\n76.0%\n76.0%\n76.0%\n76.1%\n76.1%\n76.1%\n76.1%\n76.1%\n76.2%\n76.2%\n76.2%\n76.2%\n76.2%\n76.2%\n76.3%\n76.3%\n76.3%\n76.3%\n76.3%\n76.3%\n76.4%\n76.4%\n76.4%\n76.4%\n76.4%\n76.4%\n76.5%\n76.5%\n76.5%\n76.5%\n76.5%\n76.6%\n76.6%\n76.6%\n76.6%\n76.6%\n76.6%\n76.7%\n76.7%\n76.7%\n76.7%\n76.7%\n76.7%\n76.8%\n76.8%\n76.8%\n76.8%\n76.8%\n76.9%\n76.9%\n76.9%\n76.9%\n76.9%\n76.9%\n77.0%\n77.0%\n77.0%\n77.0%\n77.0%\n77.0%\n77.1%\n77.1%\n77.1%\n77.1%\n77.1%\n77.1%\n77.2%\n77.2%\n77.2%\n77.2%\n77.2%\n77.3%\n77.3%\n77.3%\n77.3%\n77.3%\n77.3%\n77.4%\n77.4%\n77.4%\n77.4%\n77.4%\n77.4%\n77.5%\n77.5%\n77.5%\n77.5%\n77.5%\n77.6%\n77.6%\n77.6%\n77.6%\n77.6%\n77.6%\n77.7%\n77.7%\n77.7%\n77.7%\n77.7%\n77.7%\n77.8%\n77.8%\n77.8%\n77.8%\n77.8%\n77.8%\n77.9%\n77.9%\n77.9%\n77.9%\n77.9%\n78.0%\n78.0%\n78.0%\n78.0%\n78.0%\n78.0%\n78.1%\n78.1%\n78.1%\n78.1%\n78.1%\n78.1%\n78.2%\n78.2%\n78.2%\n78.2%\n78.2%\n78.3%\n78.3%\n78.3%\n78.3%\n78.3%\n78.3%\n78.4%\n78.4%\n78.4%\n78.4%\n78.4%\n78.4%\n78.5%\n78.5%\n78.5%\n78.5%\n78.5%\n78.5%\n78.6%\n78.6%\n78.6%\n78.6%\n78.6%\n78.7%\n78.7%\n78.7%\n78.7%\n78.7%\n78.7%\n78.8%\n78.8%\n78.8%\n78.8%\n78.8%\n78.8%\n78.9%\n78.9%\n78.9%\n78.9%\n78.9%\n79.0%\n79.0%\n79.0%\n79.0%\n79.0%\n79.0%\n79.1%\n79.1%\n79.1%\n79.1%\n79.1%\n79.1%\n79.2%\n79.2%\n79.2%\n79.2%\n79.2%\n79.2%\n79.3%\n79.3%\n79.3%\n79.3%\n79.3%\n79.4%\n79.4%\n79.4%\n79.4%\n79.4%\n79.4%\n79.5%\n79.5%\n79.5%\n79.5%\n79.5%\n79.5%\n79.6%\n79.6%\n79.6%\n79.6%\n79.6%\n79.7%\n79.7%\n79.7%\n79.7%\n79.7%\n79.7%\n79.8%\n79.8%\n79.8%\n79.8%\n79.8%\n79.8%\n79.9%\n79.9%\n79.9%\n79.9%\n79.9%\n79.9%\n80.0%\n80.0%\n80.0%\n80.0%\n80.0%\n80.1%\n80.1%\n80.1%\n80.1%\n80.1%\n80.1%\n80.2%\n80.2%\n80.2%\n80.2%\n80.2%\n80.2%\n80.3%\n80.3%\n80.3%\n80.3%\n80.3%\n80.3%\n80.4%\n80.4%\n80.4%\n80.4%\n80.4%\n80.5%\n80.5%\n80.5%\n80.5%\n80.5%\n80.5%\n80.6%\n80.6%\n80.6%\n80.6%\n80.6%\n80.6%\n80.7%\n80.7%\n80.7%\n80.7%\n80.7%\n80.8%\n80.8%\n80.8%\n80.8%\n80.8%\n80.8%\n80.9%\n80.9%\n80.9%\n80.9%\n80.9%\n80.9%\n81.0%\n81.0%\n81.0%\n81.0%\n81.0%\n81.0%\n81.1%\n81.1%\n81.1%\n81.1%\n81.1%\n81.2%\n81.2%\n81.2%\n81.2%\n81.2%\n81.2%\n81.3%\n81.3%\n81.3%\n81.3%\n81.3%\n81.3%\n81.4%\n81.4%\n81.4%\n81.4%\n81.4%\n81.5%\n81.5%\n81.5%\n81.5%\n81.5%\n81.5%\n81.6%\n81.6%\n81.6%\n81.6%\n81.6%\n81.6%\n81.7%\n81.7%\n81.7%\n81.7%\n81.7%\n81.7%\n81.8%\n81.8%\n81.8%\n81.8%\n81.8%\n81.9%\n81.9%\n81.9%\n81.9%\n81.9%\n81.9%\n82.0%\n82.0%\n82.0%\n82.0%\n82.0%\n82.0%\n82.1%\n82.1%\n82.1%\n82.1%\n82.1%\n82.2%\n82.2%\n82.2%\n82.2%\n82.2%\n82.2%\n82.3%\n82.3%\n82.3%\n82.3%\n82.3%\n82.3%\n82.4%\n82.4%\n82.4%\n82.4%\n82.4%\n82.4%\n82.5%\n82.5%\n82.5%\n82.5%\n82.5%\n82.6%\n82.6%\n82.6%\n82.6%\n82.6%\n82.6%\n82.7%\n82.7%\n82.7%\n82.7%\n82.7%\n82.7%\n82.8%\n82.8%\n82.8%\n82.8%\n82.8%\n82.9%\n82.9%\n82.9%\n82.9%\n82.9%\n82.9%\n83.0%\n83.0%\n83.0%\n83.0%\n83.0%\n83.0%\n83.1%\n83.1%\n83.1%\n83.1%\n83.1%\n83.1%\n83.2%\n83.2%\n83.2%\n83.2%\n83.2%\n83.3%\n83.3%\n83.3%\n83.3%\n83.3%\n83.3%\n83.4%\n83.4%\n83.4%\n83.4%\n83.4%\n83.4%\n83.5%\n83.5%\n83.5%\n83.5%\n83.5%\n83.6%\n83.6%\n83.6%\n83.6%\n83.6%\n83.6%\n83.7%\n83.7%\n83.7%\n83.7%\n83.7%\n83.7%\n83.8%\n83.8%\n83.8%\n83.8%\n83.8%\n83.8%\n83.9%\n83.9%\n83.9%\n83.9%\n83.9%\n84.0%\n84.0%\n84.0%\n84.0%\n84.0%\n84.0%\n84.1%\n84.1%\n84.1%\n84.1%\n84.1%\n84.1%\n84.2%\n84.2%\n84.2%\n84.2%\n84.2%\n84.3%\n84.3%\n84.3%\n84.3%\n84.3%\n84.3%\n84.4%\n84.4%\n84.4%\n84.4%\n84.4%\n84.4%\n84.5%\n84.5%\n84.5%\n84.5%\n84.5%\n84.5%\n84.6%\n84.6%\n84.6%\n84.6%\n84.6%\n84.7%\n84.7%\n84.7%\n84.7%\n84.7%\n84.7%\n84.8%\n84.8%\n84.8%\n84.8%\n84.8%\n84.8%\n84.9%\n84.9%\n84.9%\n84.9%\n84.9%\n85.0%\n85.0%\n85.0%\n85.0%\n85.0%\n85.0%\n85.1%\n85.1%\n85.1%\n85.1%\n85.1%\n85.1%\n85.2%\n85.2%\n85.2%\n85.2%\n85.2%\n85.2%\n85.3%\n85.3%\n85.3%\n85.3%\n85.3%\n85.4%\n85.4%\n85.4%\n85.4%\n85.4%\n85.4%\n85.5%\n85.5%\n85.5%\n85.5%\n85.5%\n85.5%\n85.6%\n85.6%\n85.6%\n85.6%\n85.6%\n85.7%\n85.7%\n85.7%\n85.7%\n85.7%\n85.7%\n85.8%\n85.8%\n85.8%\n85.8%\n85.8%\n85.8%\n85.9%\n85.9%\n85.9%\n85.9%\n85.9%\n85.9%\n86.0%\n86.0%\n86.0%\n86.0%\n86.0%\n86.1%\n86.1%\n86.1%\n86.1%\n86.1%\n86.1%\n86.2%\n86.2%\n86.2%\n86.2%\n86.2%\n86.2%\n86.3%\n86.3%\n86.3%\n86.3%\n86.3%\n86.4%\n86.4%\n86.4%\n86.4%\n86.4%\n86.4%\n86.5%\n86.5%\n86.5%\n86.5%\n86.5%\n86.5%\n86.6%\n86.6%\n86.6%\n86.6%\n86.6%\n86.6%\n86.7%\n86.7%\n86.7%\n86.7%\n86.7%\n86.8%\n86.8%\n86.8%\n86.8%\n86.8%\n86.8%\n86.9%\n86.9%\n86.9%\n86.9%\n86.9%\n86.9%\n87.0%\n87.0%\n87.0%\n87.0%\n87.0%\n87.1%\n87.1%\n87.1%\n87.1%\n87.1%\n87.1%\n87.2%\n87.2%\n87.2%\n87.2%\n87.2%\n87.2%\n87.3%\n87.3%\n87.3%\n87.3%\n87.3%\n87.3%\n87.4%\n87.4%\n87.4%\n87.4%\n87.4%\n87.5%\n87.5%\n87.5%\n87.5%\n87.5%\n87.5%\n87.6%\n87.6%\n87.6%\n87.6%\n87.6%\n87.6%\n87.7%\n87.7%\n87.7%\n87.7%\n87.7%\n87.7%\n87.8%\n87.8%\n87.8%\n87.8%\n87.8%\n87.9%\n87.9%\n87.9%\n87.9%\n87.9%\n87.9%\n88.0%\n88.0%\n88.0%\n88.0%\n88.0%\n88.0%\n88.1%\n88.1%\n88.1%\n88.1%\n88.1%\n88.2%\n88.2%\n88.2%\n88.2%\n88.2%\n88.2%\n88.3%\n88.3%\n88.3%\n88.3%\n88.3%\n88.3%\n88.4%\n88.4%\n88.4%\n88.4%\n88.4%\n88.4%\n88.5%\n88.5%\n88.5%\n88.5%\n88.5%\n88.6%\n88.6%\n88.6%\n88.6%\n88.6%\n88.6%\n88.7%\n88.7%\n88.7%\n88.7%\n88.7%\n88.7%\n88.8%\n88.8%\n88.8%\n88.8%\n88.8%\n88.9%\n88.9%\n88.9%\n88.9%\n88.9%\n88.9%\n89.0%\n89.0%\n89.0%\n89.0%\n89.0%\n89.0%\n89.1%\n89.1%\n89.1%\n89.1%\n89.1%\n89.1%\n89.2%\n89.2%\n89.2%\n89.2%\n89.2%\n89.3%\n89.3%\n89.3%\n89.3%\n89.3%\n89.3%\n89.4%\n89.4%\n89.4%\n89.4%\n89.4%\n89.4%\n89.5%\n89.5%\n89.5%\n89.5%\n89.5%\n89.6%\n89.6%\n89.6%\n89.6%\n89.6%\n89.6%\n89.7%\n89.7%\n89.7%\n89.7%\n89.7%\n89.7%\n89.8%\n89.8%\n89.8%\n89.8%\n89.8%\n89.8%\n89.9%\n89.9%\n89.9%\n89.9%\n89.9%\n90.0%\n90.0%\n90.0%\n90.0%\n90.0%\n90.0%\n90.1%\n90.1%\n90.1%\n90.1%\n90.1%\n90.1%\n90.2%\n90.2%\n90.2%\n90.2%\n90.2%\n90.3%\n90.3%\n90.3%\n90.3%\n90.3%\n90.3%\n90.4%\n90.4%\n90.4%\n90.4%\n90.4%\n90.4%\n90.5%\n90.5%\n90.5%\n90.5%\n90.5%\n90.5%\n90.6%\n90.6%\n90.6%\n90.6%\n90.6%\n90.7%\n90.7%\n90.7%\n",
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": "90.7%\n90.7%\n90.7%\n90.8%\n90.8%\n90.8%\n90.8%\n90.8%\n90.8%\n90.9%\n90.9%\n90.9%\n90.9%\n90.9%\n91.0%\n91.0%\n91.0%\n91.0%\n91.0%\n91.0%\n91.1%\n91.1%\n91.1%\n91.1%\n91.1%\n91.1%\n91.2%\n91.2%\n91.2%\n91.2%\n91.2%\n91.2%\n91.3%\n91.3%\n91.3%\n91.3%\n91.3%\n91.4%\n91.4%\n91.4%\n91.4%\n91.4%\n91.4%\n91.5%\n91.5%\n91.5%\n91.5%\n91.5%\n91.5%\n91.6%\n91.6%\n91.6%\n91.6%\n91.6%\n91.7%\n91.7%\n91.7%\n91.7%\n91.7%\n91.7%\n91.8%\n91.8%\n91.8%\n91.8%\n91.8%\n91.8%\n91.9%\n91.9%\n91.9%\n91.9%\n91.9%\n91.9%\n92.0%\n92.0%\n92.0%\n92.0%\n92.0%\n92.1%\n92.1%\n92.1%\n92.1%\n92.1%\n92.1%\n92.2%\n92.2%\n92.2%\n92.2%\n92.2%\n92.2%\n92.3%\n92.3%\n92.3%\n92.3%\n92.3%\n92.4%\n92.4%\n92.4%\n92.4%\n92.4%\n92.4%\n92.5%\n92.5%\n92.5%\n92.5%\n92.5%\n92.5%\n92.6%\n92.6%\n92.6%\n92.6%\n92.6%\n92.6%\n92.7%\n92.7%\n92.7%\n92.7%\n92.7%\n92.8%\n92.8%\n92.8%\n92.8%\n92.8%\n92.8%\n92.9%\n92.9%\n92.9%\n92.9%\n92.9%\n92.9%\n93.0%\n93.0%\n93.0%\n93.0%\n93.0%\n93.1%\n93.1%\n93.1%\n93.1%\n93.1%\n93.1%\n93.2%\n93.2%\n93.2%\n93.2%\n93.2%\n93.2%\n93.3%\n93.3%\n93.3%\n93.3%\n93.3%\n93.3%\n93.4%\n93.4%\n93.4%\n93.4%\n93.4%\n93.5%\n93.5%\n93.5%\n93.5%\n93.5%\n93.5%\n93.6%\n93.6%\n93.6%\n93.6%\n93.6%\n93.6%\n93.7%\n93.7%\n93.7%\n93.7%\n93.7%\n93.8%\n93.8%\n93.8%\n93.8%\n93.8%\n93.8%\n93.9%\n93.9%\n93.9%\n93.9%\n93.9%\n93.9%\n94.0%\n94.0%\n94.0%\n94.0%\n94.0%\n94.0%\n94.1%\n94.1%\n94.1%\n94.1%\n94.1%\n94.2%\n94.2%\n94.2%\n94.2%\n94.2%\n94.2%\n94.3%\n94.3%\n94.3%\n94.3%\n94.3%\n94.3%\n94.4%\n94.4%\n94.4%\n94.4%\n94.4%\n94.5%\n94.5%\n94.5%\n94.5%\n94.5%\n94.5%\n94.6%\n94.6%\n94.6%\n94.6%\n94.6%\n94.6%\n94.7%\n94.7%\n94.7%\n94.7%\n94.7%\n94.7%\n94.8%\n94.8%\n94.8%\n94.8%\n94.8%\n94.9%\n94.9%\n94.9%\n94.9%\n94.9%\n94.9%\n95.0%\n95.0%\n95.0%\n95.0%\n95.0%\n95.0%\n95.1%\n95.1%\n95.1%\n95.1%\n95.1%\n95.1%\n95.2%\n95.2%\n95.2%\n95.2%\n95.2%\n95.3%\n95.3%\n95.3%\n95.3%\n95.3%\n95.3%\n95.4%\n95.4%\n95.4%\n95.4%\n95.4%\n95.4%\n95.5%\n95.5%\n95.5%\n95.5%\n95.5%\n95.6%\n95.6%\n95.6%\n95.6%\n95.6%\n95.6%\n95.7%\n95.7%\n95.7%\n95.7%\n95.7%\n95.7%\n95.8%\n95.8%\n95.8%\n95.8%\n95.8%\n95.8%\n95.9%\n95.9%\n95.9%\n95.9%\n95.9%\n96.0%\n96.0%\n96.0%\n96.0%\n96.0%\n96.0%\n96.1%\n96.1%\n96.1%\n96.1%\n96.1%\n96.1%\n96.2%\n96.2%\n96.2%\n96.2%\n96.2%\n96.3%\n96.3%\n96.3%\n96.3%\n96.3%\n96.3%\n96.4%\n96.4%\n96.4%\n96.4%\n96.4%\n96.4%\n96.5%\n96.5%\n96.5%\n96.5%\n96.5%\n96.5%\n96.6%\n96.6%\n96.6%\n96.6%\n96.6%\n96.7%\n96.7%\n96.7%\n96.7%\n96.7%\n96.7%\n96.8%\n96.8%\n96.8%\n96.8%\n96.8%\n96.8%\n96.9%\n96.9%\n96.9%\n96.9%\n96.9%\n97.0%\n97.0%\n97.0%\n97.0%\n97.0%\n97.0%\n97.1%\n97.1%\n97.1%\n97.1%\n97.1%\n97.1%\n97.2%\n97.2%\n97.2%\n97.2%\n97.2%\n97.2%\n97.3%\n97.3%\n97.3%\n97.3%\n97.3%\n97.4%\n97.4%\n97.4%\n97.4%\n97.4%\n97.4%\n97.5%\n97.5%\n97.5%\n97.5%\n97.5%\n97.5%\n97.6%\n97.6%\n97.6%\n97.6%\n97.6%\n97.7%\n97.7%\n97.7%\n97.7%\n97.7%\n97.7%\n97.8%\n97.8%\n97.8%\n97.8%\n97.8%\n97.8%\n97.9%\n97.9%\n97.9%\n97.9%\n97.9%\n97.9%\n98.0%\n98.0%\n98.0%\n98.0%\n98.0%\n98.1%\n98.1%\n98.1%\n98.1%\n98.1%\n98.1%\n98.2%\n98.2%\n98.2%\n98.2%\n98.2%\n98.2%\n98.3%\n98.3%\n98.3%\n98.3%\n98.3%\n98.4%\n98.4%\n98.4%\n98.4%\n98.4%\n98.4%\n98.5%\n98.5%\n98.5%\n98.5%\n98.5%\n98.5%\n98.6%\n98.6%\n98.6%\n98.6%\n98.6%\n98.6%\n98.7%\n98.7%\n98.7%\n98.7%\n98.7%\n98.8%\n98.8%\n98.8%\n98.8%\n98.8%\n98.8%\n98.9%\n98.9%\n98.9%\n98.9%\n98.9%\n98.9%\n99.0%\n99.0%\n99.0%\n99.0%\n99.0%\n99.1%\n99.1%\n99.1%\n99.1%\n99.1%\n99.1%\n99.2%\n99.2%\n99.2%\n99.2%\n99.2%\n99.2%\n99.3%\n99.3%\n99.3%\n99.3%\n99.3%\n99.3%\n99.4%\n99.4%\n99.4%\n99.4%\n99.4%\n99.5%\n99.5%\n99.5%\n99.5%\n99.5%\n99.5%\n99.6%\n99.6%\n99.6%\n99.6%\n99.6%\n99.6%\n99.7%\n99.7%\n99.7%\n99.7%\n99.7%\n99.8%\n99.8%\n99.8%\n99.8%\n99.8%\n99.8%\n99.9%\n99.9%\n99.9%\n99.9%\n99.9%\n99.9%\n100.0%\n100.0%\n100.0%\n100.0%\nEpoch 0/29\n----------\ntrain Loss: 0.7832 Acc: 0.6803\nval Loss: 0.2227 Acc: 0.9346\n\nEpoch 1/29\n----------\ntrain Loss: 0.3055 Acc: 0.8852\nval Loss: 0.2022 Acc: 0.9150\n\nEpoch 2/29\n----------\ntrain Loss: 0.5734 Acc: 0.7705\nval Loss: 0.2507 Acc: 0.8954\n\nEpoch 3/29\n----------\ntrain Loss: 0.4898 Acc: 0.7992\nval Loss: 0.3335 Acc: 0.8627\n\nEpoch 4/29\n----------\ntrain Loss: 0.5010 Acc: 0.7828\nval Loss: 0.2959 Acc: 0.9085\n\nEpoch 5/29\n----------\ntrain Loss: 0.5820 Acc: 0.7910\nval Loss: 0.6116 Acc: 0.7778\n\nEpoch 6/29\n----------\ntrain Loss: 0.3672 Acc: 0.8566\nval Loss: 0.4672 Acc: 0.8366\n\nEpoch 7/29\n----------\ntrain Loss: 0.3972 Acc: 0.8443\nval Loss: 0.3588 Acc: 0.8758\n\nEpoch 8/29\n----------\ntrain Loss: 0.3726 Acc: 0.8443\nval Loss: 0.3315 Acc: 0.8954\n\nEpoch 9/29\n----------\ntrain Loss: 0.3574 Acc: 0.8689\nval Loss: 0.2835 Acc: 0.9150\n\nEpoch 10/29\n----------\ntrain Loss: 0.3757 Acc: 0.8361\nval Loss: 0.2860 Acc: 0.9020\n\nEpoch 11/29\n----------\ntrain Loss: 0.2755 Acc: 0.8975\nval Loss: 0.2708 Acc: 0.9020\n\nEpoch 12/29\n----------\ntrain Loss: 0.3720 Acc: 0.8607\nval Loss: 0.2505 Acc: 0.9020\n\nEpoch 13/29\n----------\ntrain Loss: 0.3254 Acc: 0.8770\nval Loss: 0.2675 Acc: 0.9020\n\nEpoch 14/29\n----------\ntrain Loss: 0.1930 Acc: 0.9098\nval Loss: 0.2733 Acc: 0.8954\n\nEpoch 15/29\n----------\ntrain Loss: 0.2900 Acc: 0.8893\nval Loss: 0.2781 Acc: 0.8824\n\nEpoch 16/29\n----------\ntrain Loss: 0.3090 Acc: 0.8648\nval Loss: 0.2833 Acc: 0.8889\n\nEpoch 17/29\n----------\ntrain Loss: 0.2820 Acc: 0.8730\nval Loss: 0.2643 Acc: 0.9020\n\nEpoch 18/29\n----------\ntrain Loss: 0.2425 Acc: 0.8811\nval Loss: 0.2617 Acc: 0.9020\n\nEpoch 19/29\n----------\ntrain Loss: 0.2675 Acc: 0.8934\nval Loss: 0.2466 Acc: 0.9085\n\nEpoch 20/29\n----------\ntrain Loss: 0.2575 Acc: 0.8852\nval Loss: 0.2523 Acc: 0.9150\n\nEpoch 21/29\n----------\ntrain Loss: 0.2574 Acc: 0.8893\nval Loss: 0.2754 Acc: 0.9085\n\nEpoch 22/29\n----------\ntrain Loss: 0.3119 Acc: 0.8689\nval Loss: 0.2503 Acc: 0.9085\n\nEpoch 23/29\n----------\ntrain Loss: 0.3033 Acc: 0.8852\nval Loss: 0.2536 Acc: 0.9150\n\nEpoch 24/29\n----------\ntrain Loss: 0.4267 Acc: 0.8197\nval Loss: 0.2589 Acc: 0.9020\n\nEpoch 25/29\n----------\ntrain Loss: 0.2823 Acc: 0.8689\nval Loss: 0.2556 Acc: 0.9085\n\nEpoch 26/29\n----------\ntrain Loss: 0.2692 Acc: 0.8893\nval Loss: 0.2717 Acc: 0.9020\n\nEpoch 27/29\n----------\ntrain Loss: 0.3246 Acc: 0.8525\nval Loss: 0.2644 Acc: 0.9085\n\nEpoch 28/29\n----------\ntrain Loss: 0.2452 Acc: 0.8975\nval Loss: 0.2547 Acc: 0.9085\n\nEpoch 29/29\n----------\ntrain Loss: 0.2668 Acc: 0.8975\nval Loss: 0.2566 Acc: 0.9150\n\nTraining complete in 3m 33s\nBest val Acc: 0.934641\n\n\nThe experiment completed successfully. Finalizing run...\nLogging experiment finalizing status in history service\nCleaning up all outstanding Run operations, waiting 300.0 seconds\n2 items cleaning up...\nCleanup took 0.20651507377624512 seconds\n\nExecution Summary\n=================\nRunId: pytorch-hymenoptera_1545830893450\n\n",
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "execution_count": 14,
          "data": {
            "text/plain": "{'runId': 'pytorch-hymenoptera_1545830893450',\n 'target': 'gpucluster',\n 'status': 'Finalizing',\n 'startTimeUtc': '2018-12-26T13:33:35.616575Z',\n 'properties': {'azureml.runsource': 'experiment',\n  'ContentSnapshotId': 'a0f005c9-2bb3-4314-a092-665accae8c34'},\n 'runDefinition': {'Script': 'pytorch_train.py',\n  'Arguments': ['--num_epochs', '30', '--output_dir', './outputs'],\n  'SourceDirectoryDataStore': None,\n  'Framework': 0,\n  'Communicator': 0,\n  'Target': 'gpucluster',\n  'DataReferences': {},\n  'JobName': None,\n  'AutoPrepareEnvironment': True,\n  'MaxRunDurationSeconds': None,\n  'NodeCount': 1,\n  'Environment': {'Python': {'InterpreterPath': 'python',\n    'UserManagedDependencies': False,\n    'CondaDependencies': {'name': 'project_environment',\n     'dependencies': ['python=3.6.2',\n      {'pip': ['azureml-defaults']},\n      'pytorch=0.4.0',\n      'torchvision=0.2.1']},\n    'CondaDependenciesFile': None},\n   'EnvironmentVariables': {'EXAMPLE_ENV_VAR': 'EXAMPLE_VALUE',\n    'NCCL_SOCKET_IFNAME': '^docker0'},\n   'Docker': {'BaseImage': 'mcr.microsoft.com/azureml/base-gpu:0.2.0',\n    'Enabled': True,\n    'SharedVolumes': True,\n    'Preparation': None,\n    'GpuSupport': True,\n    'Arguments': [],\n    'BaseImageRegistry': {'Address': None,\n     'Username': None,\n     'Password': None}},\n   'Spark': {'Repositories': ['https://mmlspark.azureedge.net/maven'],\n    'Packages': [{'Group': 'com.microsoft.ml.spark',\n      'Artifact': 'mmlspark_2.11',\n      'Version': '0.12'}],\n    'PrecachePackages': True}},\n  'History': {'OutputCollection': True},\n  'Spark': {'Configuration': {'spark.app.name': 'Azure ML Experiment',\n    'spark.yarn.maxAppAttempts': '1'}},\n  'BatchAi': {'NodeCount': 0},\n  'AmlCompute': {'Name': None,\n   'VmSize': None,\n   'VmPriority': None,\n   'RetainCluster': False,\n   'ClusterMaxNodeCount': 1},\n  'Tensorflow': {'WorkerCount': 1, 'ParameterServerCount': 1},\n  'Mpi': {'ProcessCountPerNode': 1},\n  'Hdi': {'YarnDeployMode': 2},\n  'ContainerInstance': {'Region': None, 'CpuCores': 0, 'MemoryGb': 0},\n  'ExposedPorts': None,\n  'PrepareEnvironment': None},\n 'logFiles': {'azureml-logs/60_control_log.txt': 'https://amlsworkspace9663571855.blob.core.windows.net/azureml/ExperimentRun/pytorch-hymenoptera_1545830893450/azureml-logs/60_control_log.txt?sv=2018-03-28&sr=b&sig=X1ZzCrvTfSqh23be601wGDH1N15VUujtEpXbVgGOPoI%3D&st=2018-12-26T13%3A31%3A10Z&se=2018-12-26T21%3A41%3A10Z&sp=r',\n  'azureml-logs/80_driver_log.txt': 'https://amlsworkspace9663571855.blob.core.windows.net/azureml/ExperimentRun/pytorch-hymenoptera_1545830893450/azureml-logs/80_driver_log.txt?sv=2018-03-28&sr=b&sig=LL%2Bl3yS0KwLH0IjGRY6Cehb%2FZmAjgXCs9xn42w2JlM8%3D&st=2018-12-26T13%3A31%3A10Z&se=2018-12-26T21%3A41%3A10Z&sp=r',\n  'azureml-logs/azureml.log': 'https://amlsworkspace9663571855.blob.core.windows.net/azureml/ExperimentRun/pytorch-hymenoptera_1545830893450/azureml-logs/azureml.log?sv=2018-03-28&sr=b&sig=Rc6F%2BX1pXwvVaMl%2FKgcYd%2FvjOzTUqSIFhuP0XlZHPFg%3D&st=2018-12-26T13%3A31%3A10Z&se=2018-12-26T21%3A41%3A10Z&sp=r'}}"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "## Tune model hyperparameters\nNow that we've seen how to do a simple PyTorch training run using the SDK, let's see if we can further improve the accuracy of our model. We can optimize our model's hyperparameters using Azure Machine Learning's hyperparameter tuning capabilities."
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### Start a hyperparameter sweep\nFirst, we will define the hyperparameter space to sweep over. Since our training script uses a learning rate schedule to decay the learning rate every several epochs, let's tune the initial learning rate and the momentum parameters. In this example we will use random sampling to try different configuration sets of hyperparameters to maximize our primary metric, the best validation accuracy (`best_val_acc`).\n\nThen, we specify the early termination policy to use to early terminate poorly performing runs. Here we use the `BanditPolicy`, which will terminate any run that doesn't fall within the slack factor of our primary evaluation metric. In this tutorial, we will apply this policy every epoch (since we report our `best_val_acc` metric every epoch and `evaluation_interval=1`). Notice we will delay the first policy evaluation until after the first `10` epochs (`delay_evaluation=10`).\nRefer [here](https://docs.microsoft.com/azure/machine-learning/service/how-to-tune-hyperparameters#specify-an-early-termination-policy) for more information on the BanditPolicy and other policies available."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from azureml.train.hyperdrive import *\n\nparam_sampling = RandomParameterSampling( {\n        'learning_rate': uniform(0.0005, 0.005),\n        'momentum': uniform(0.9, 0.99)\n    }\n)\n\nearly_termination_policy = BanditPolicy(slack_factor=0.15, evaluation_interval=1, delay_evaluation=10)\n\nhyperdrive_run_config = HyperDriveRunConfig(estimator=estimator,\n                                            hyperparameter_sampling=param_sampling, \n                                            policy=early_termination_policy,\n                                            primary_metric_name='best_val_acc',\n                                            primary_metric_goal=PrimaryMetricGoal.MAXIMIZE,\n                                            max_total_runs=8,\n                                            max_concurrent_runs=4)",
      "execution_count": 15,
      "outputs": []
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Finally, lauch the hyperparameter tuning job."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "# start the HyperDrive run\nhyperdrive_run = experiment.submit(hyperdrive_run_config)",
      "execution_count": 16,
      "outputs": []
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### Monitor HyperDrive runs\nYou can monitor the progress of the runs with the following Jupyter widget. "
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from azureml.widgets import RunDetails\n\nRunDetails(hyperdrive_run).show()",
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "ceef04a491044a45ae4a699501376c9a",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": "_HyperDriveWidget(widget_settings={'childWidgetDisplay': 'popup', 'send_telemetry': True, 'log_level': 'INFO',…"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Or block until the HyperDrive sweep has completed:"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "hyperdrive_run.wait_for_completion(show_output=True)",
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "text": "RunId: pytorch-hymenoptera_1545831695451\n\nExecution Summary\n=================\nRunId: pytorch-hymenoptera_1545831695451\n\n",
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "execution_count": 19,
          "data": {
            "text/plain": "{'runId': 'pytorch-hymenoptera_1545831695451',\n 'target': 'gpucluster',\n 'status': 'Completed',\n 'endTimeUtc': '2018-12-26T13:57:22.000Z',\n 'properties': {'primary_metric_config': '{\"name\": \"best_val_acc\", \"goal\": \"maximize\"}',\n  'runTemplate': 'HyperDrive',\n  'azureml.runsource': 'hyperdrive'},\n 'logFiles': {'azureml-logs/hyperdrive.txt': 'https://amlsworkspace9663571855.blob.core.windows.net/azureml/ExperimentRun/pytorch-hymenoptera_1545831695451/azureml-logs/hyperdrive.txt?sv=2018-03-28&sr=b&sig=mZrMi3OkAN1kTKD4SypN5BoPD9mEJwtctUjoCMnLGoI%3D&st=2018-12-26T13%3A48%3A24Z&se=2018-12-26T21%3A58%3A24Z&sp=r'}}"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### Find and register the best model\nOnce all the runs complete, we can find the run that produced the model with the highest accuracy."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "best_run = hyperdrive_run.get_best_run_by_primary_metric()\nbest_run_metrics = best_run.get_metrics()\nprint(best_run)",
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "text": "Run(Experiment: pytorch-hymenoptera,\nId: pytorch-hymenoptera_1545831695451_4,\nType: azureml.scriptrun,\nStatus: Completed)\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "print('Best Run is:\\n  Validation accuracy: {0:.5f} \\n  Learning rate: {1:.5f} \\n  Momentum: {2:.5f}'.format(\n        best_run_metrics['best_val_acc'][-1],\n        best_run_metrics['lr'],\n        best_run_metrics['momentum'])\n     )",
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "stream",
          "text": "Best Run is:\n  Validation accuracy: 0.85621 \n  Learning rate: 0.00123 \n  Momentum: 0.94703\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Finally, register the model from your best-performing run to your workspace. The `model_path` parameter takes in the relative path on the remote VM to the model file in your `outputs` directory. In the next section, we will deploy this registered model as a web service."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "model = best_run.register_model(model_name = 'pytorch-hymenoptera', model_path = 'outputs/model.pt')\nprint(model.name, model.id, model.version, sep = '\\t')",
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "text": "Deprecated, use RunHistoryFacade.assets instead.\n",
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": "pytorch-hymenoptera\tpytorch-hymenoptera:1\t1\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "## Deploy model as web service\nOnce you have your trained model, you can deploy the model on Azure. In this tutorial, we will deploy the model as a web service in [Azure Container Instances](https://docs.microsoft.com/en-us/azure/container-instances/) (ACI). For more information on deploying models using Azure ML, refer [here](https://docs.microsoft.com/azure/machine-learning/service/how-to-deploy-and-where)."
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### Create scoring script\n\nFirst, we will create a scoring script that will be invoked by the web service call. Note that the scoring script must have two required functions:\n* `init()`: In this function, you typically load the model into a `global` object. This function is executed only once when the Docker container is started. \n* `run(input_data)`: In this function, the model is used to predict a value based on the input data. The input and output typically use JSON as serialization and deserialization format, but you are not limited to that.\n\nRefer to the scoring script `pytorch_score.py` for this tutorial. Our web service will use this file to predict whether an image is an ant or a bee. When writing your own scoring script, don't forget to test it locally first before you go and deploy the web service."
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### Create environment file\nThen, we will need to create an environment file (`myenv.yml`) that specifies all of the scoring script's package dependencies. This file is used to ensure that all of those dependencies are installed in the Docker image by AML. In this case, we need to specify `azureml-core`, `torch` and `torchvision`."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from azureml.core.conda_dependencies import CondaDependencies \n\nmyenv = CondaDependencies.create(pip_packages=['azureml-defaults', 'torch', 'torchvision'])\n\nwith open(\"myenv.yml\",\"w\") as f:\n    f.write(myenv.serialize_to_string())\n    \nprint(myenv.serialize_to_string())",
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "text": "# Conda environment specification. The dependencies defined in this file will\r\n# be automatically provisioned for runs with userManagedDependencies=False.\r\n\n# Details about the Conda environment file format:\r\n# https://conda.io/docs/user-guide/tasks/manage-environments.html#create-env-file-manually\r\n\nname: project_environment\ndependencies:\n  # The python interpreter version.\r\n  # Currently Azure ML only supports 3.5.2 and later.\r\n- python=3.6.2\n\n- pip:\n  - azureml-defaults==1.0.2\n  - torch\n  - torchvision\n\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### Configure the container image\nNow configure the Docker image that you will use to build your ACI container."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from azureml.core.image import ContainerImage\n\nimage_config = ContainerImage.image_configuration(execution_script='pytorch_score.py', \n                                                  runtime='python', \n                                                  conda_file='myenv.yml',\n                                                  description='Image with hymenoptera model')",
      "execution_count": 24,
      "outputs": []
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### Configure the ACI container\nWe are almost ready to deploy. Create a deployment configuration file to specify the number of CPUs and gigabytes of RAM needed for your ACI container. While it depends on your model, the default of `1` core and `1` gigabyte of RAM is usually sufficient for many models."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from azureml.core.webservice import AciWebservice\n\naciconfig = AciWebservice.deploy_configuration(cpu_cores=1, \n                                               memory_gb=1, \n                                               tags={'data': 'hymenoptera',  'method':'transfer learning', 'framework':'pytorch'},\n                                               description='Classify ants/bees using transfer learning with PyTorch')",
      "execution_count": 25,
      "outputs": []
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### Deploy the registered model\nFinally, let's deploy a web service from our registered model. Deploy the web service using the ACI config and image config files created in the previous steps. We pass the `model` object in a list to the `models` parameter. If you would like to deploy more than one registered model, append the additional models to this list."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "%%time\nfrom azureml.core.webservice import Webservice\n\nservice_name = 'aci-hymenoptera'\nservice = Webservice.deploy_from_model(workspace=ws,\n                                       name=service_name,\n                                       models=[model],\n                                       image_config=image_config,\n                                       deployment_config=aciconfig,)\n\nservice.wait_for_deployment(show_output=True)\nprint(service.state)",
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "stream",
          "text": "Creating image\nImage creation operation finished for image aci-hymenoptera:1, operation \"Succeeded\"\nCreating service\nRunning.........................\nSucceededACI service creation operation finished, operation \"Succeeded\"\nHealthy\nCPU times: user 21.6 s, sys: 12.3 s, total: 33.8 s\nWall time: 7min 25s\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "If your deployment fails for any reason and you need to redeploy, make sure to delete the service before you do so: `service.delete()`"
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "**Tip: If something goes wrong with the deployment, the first thing to look at is the logs from the service by running the following command:**"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "service.get_logs()",
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 27,
          "data": {
            "text/plain": "'2018-12-26T14:15:32,629850207+00:00 - rsyslog/run \\n2018-12-26T14:15:32,629191783+00:00 - nginx/run \\nok: run: rsyslog: (pid 13) 0s\\n2018-12-26T14:15:32,630198319+00:00 - gunicorn/run \\n2018-12-26T14:15:32,636091429+00:00 - iot-server/run \\nok: run: gunicorn: (pid 14) 0s\\nok: run: nginx: (pid 12) 0s\\nok: run: rsyslog: (pid 13) 0s\\nok: run: rsyslog: (pid 13) 0s\\nEdgeHubConnectionString and IOTEDGE_IOTHUBHOSTNAME are not set. Exiting...\\n2018-12-26T14:15:32,767782413+00:00 - iot-server/finish 1 0\\n2018-12-26T14:15:32,769534175+00:00 - Exit code 1 is normal. Not restarting iot-server.\\n{\"timestamp\": \"2018-12-26T14:15:33.038425Z\", \"message\": \"Starting gunicorn 19.6.0\", \"host\": \"wk-caas-d9227ef224fa45efb0240d09c4ce3c9e-8eab1c3e0606d3902d519e\", \"path\": \"/opt/miniconda/lib/python3.6/site-packages/gunicorn/glogging.py\", \"tags\": \"%(module)s, %(asctime)s, %(levelname)s, %(message)s\", \"level\": \"INFO\", \"logger\": \"gunicorn.error\", \"msg\": \"Starting gunicorn %s\", \"stack_info\": null}\\n{\"timestamp\": \"2018-12-26T14:15:33.044105Z\", \"message\": \"Listening at: http://127.0.0.1:9090 (14)\", \"host\": \"wk-caas-d9227ef224fa45efb0240d09c4ce3c9e-8eab1c3e0606d3902d519e\", \"path\": \"/opt/miniconda/lib/python3.6/site-packages/gunicorn/glogging.py\", \"tags\": \"%(module)s, %(asctime)s, %(levelname)s, %(message)s\", \"level\": \"INFO\", \"logger\": \"gunicorn.error\", \"msg\": \"Listening at: %s (%s)\", \"stack_info\": null}\\n{\"timestamp\": \"2018-12-26T14:15:33.044217Z\", \"message\": \"Using worker: sync\", \"host\": \"wk-caas-d9227ef224fa45efb0240d09c4ce3c9e-8eab1c3e0606d3902d519e\", \"path\": \"/opt/miniconda/lib/python3.6/site-packages/gunicorn/glogging.py\", \"tags\": \"%(module)s, %(asctime)s, %(levelname)s, %(message)s\", \"level\": \"INFO\", \"logger\": \"gunicorn.error\", \"msg\": \"Using worker: %s\", \"stack_info\": null}\\n{\"timestamp\": \"2018-12-26T14:15:33.044789Z\", \"message\": \"worker timeout is set to 300\", \"host\": \"wk-caas-d9227ef224fa45efb0240d09c4ce3c9e-8eab1c3e0606d3902d519e\", \"path\": \"/opt/miniconda/lib/python3.6/site-packages/gunicorn/glogging.py\", \"tags\": \"%(module)s, %(asctime)s, %(levelname)s, %(message)s\", \"level\": \"INFO\", \"logger\": \"gunicorn.error\", \"stack_info\": null}\\n{\"timestamp\": \"2018-12-26T14:15:33.046073Z\", \"message\": \"Booting worker with pid: 40\", \"host\": \"wk-caas-d9227ef224fa45efb0240d09c4ce3c9e-8eab1c3e0606d3902d519e\", \"path\": \"/opt/miniconda/lib/python3.6/site-packages/gunicorn/glogging.py\", \"tags\": \"%(module)s, %(asctime)s, %(levelname)s, %(message)s\", \"level\": \"INFO\", \"logger\": \"gunicorn.error\", \"msg\": \"Booting worker with pid: %s\", \"stack_info\": null}\\nInitializing logger\\n{\"timestamp\": \"2018-12-26T14:15:36.517054Z\", \"message\": \"{\\\\\"requestId\\\\\": \\\\\"00000000-0000-0000-0000-000000000000\\\\\", \\\\\"message\\\\\": \\\\\"Starting up app insights client\\\\\", \\\\\"apiName\\\\\": \\\\\"\\\\\"}\", \"host\": \"wk-caas-d9227ef224fa45efb0240d09c4ce3c9e-8eab1c3e0606d3902d519e\", \"path\": \"/var/azureml-app/aml_logger.py\", \"tags\": \"%(module)s, %(asctime)s, %(levelname)s, %(message)s\", \"level\": \"INFO\", \"logger\": \"root\", \"stack_info\": null}\\n{\"timestamp\": \"2018-12-26T14:15:36.517275Z\", \"message\": \"{\\\\\"requestId\\\\\": \\\\\"00000000-0000-0000-0000-000000000000\\\\\", \\\\\"message\\\\\": \\\\\"Starting up request id generator\\\\\", \\\\\"apiName\\\\\": \\\\\"\\\\\"}\", \"host\": \"wk-caas-d9227ef224fa45efb0240d09c4ce3c9e-8eab1c3e0606d3902d519e\", \"path\": \"/var/azureml-app/aml_logger.py\", \"tags\": \"%(module)s, %(asctime)s, %(levelname)s, %(message)s\", \"level\": \"INFO\", \"logger\": \"root\", \"stack_info\": null}\\n{\"timestamp\": \"2018-12-26T14:15:36.517380Z\", \"message\": \"{\\\\\"requestId\\\\\": \\\\\"00000000-0000-0000-0000-000000000000\\\\\", \\\\\"message\\\\\": \\\\\"Starting up app insight hooks\\\\\", \\\\\"apiName\\\\\": \\\\\"\\\\\"}\", \"host\": \"wk-caas-d9227ef224fa45efb0240d09c4ce3c9e-8eab1c3e0606d3902d519e\", \"path\": \"/var/azureml-app/aml_logger.py\", \"tags\": \"%(module)s, %(asctime)s, %(levelname)s, %(message)s\", \"level\": \"INFO\", \"logger\": \"root\", \"stack_info\": null}\\n{\"timestamp\": \"2018-12-26T14:15:36.517506Z\", \"message\": \"{\\\\\"requestId\\\\\": \\\\\"00000000-0000-0000-0000-000000000000\\\\\", \\\\\"message\\\\\": \\\\\"Invoking user\\'s init function\\\\\", \\\\\"apiName\\\\\": \\\\\"\\\\\"}\", \"host\": \"wk-caas-d9227ef224fa45efb0240d09c4ce3c9e-8eab1c3e0606d3902d519e\", \"path\": \"/var/azureml-app/aml_logger.py\", \"tags\": \"%(module)s, %(asctime)s, %(levelname)s, %(message)s\", \"level\": \"INFO\", \"logger\": \"root\", \"stack_info\": null}\\n2018-12-26 14:15:36,518 | azureml.core.run | DEBUG | Could not load run context Failed to load a submitted run, if outside of an execution context, use project.start_run to initialize an azureml.core.Run., switching offline: False\\n2018-12-26 14:15:36,518 | azureml.core.run | DEBUG | Could not load the run context and allow_offline set to False\\n2018-12-26 14:15:36,518 | azureml.core.model | DEBUG | RunEnvironmentException: Failed to load a submitted run, if outside of an execution context, use project.start_run to initialize an azureml.core.Run.\\n2018-12-26 14:15:36,518 | azureml.core.model | DEBUG | version is None. Latest version is 1\\n2018-12-26 14:15:36,518 | azureml.core.model | DEBUG | Found model path at azureml-models/pytorch-hymenoptera/1/model.pt\\n{\"timestamp\": \"2018-12-26T14:15:36.571691Z\", \"message\": \"{\\\\\"requestId\\\\\": \\\\\"00000000-0000-0000-0000-000000000000\\\\\", \\\\\"message\\\\\": \\\\\"Users\\'s init has completed successfully\\\\\", \\\\\"apiName\\\\\": \\\\\"\\\\\"}\", \"host\": \"wk-caas-d9227ef224fa45efb0240d09c4ce3c9e-8eab1c3e0606d3902d519e\", \"path\": \"/var/azureml-app/aml_logger.py\", \"tags\": \"%(module)s, %(asctime)s, %(levelname)s, %(message)s\", \"level\": \"INFO\", \"logger\": \"root\", \"stack_info\": null}\\n/opt/miniconda/lib/python3.6/site-packages/torch/serialization.py:434: SourceChangeWarning: source code of class \\'torch.nn.modules.conv.Conv2d\\' has changed. you can retrieve the original source code by accessing the object\\'s source attribute or set `torch.nn.Module.dump_patches = True` and use the patch tool to revert the changes.\\n  warnings.warn(msg, SourceChangeWarning)\\n{\"logger\": \"logger_stderr\", \"request_id\": \"no request id\", \"message\": \"/opt/miniconda/lib/python3.6/site-packages/torch/serialization.py:434: SourceChangeWarning: source code of class \\'torch.nn.modules.conv.Conv2d\\' has changed. you can retrieve the original source code by accessing the object\\'s source attribute or set `torch.nn.Module.dump_patches = True` and use the patch tool to revert the changes.\\\\n  warnings.warn(msg, SourceChangeWarning)\\\\n\", \"timestamp\": \"2018-12-26T14:15:36.532612\"}\\n/opt/miniconda/lib/python3.6/site-packages/torch/serialization.py:434: SourceChangeWarning: source code of class \\'torch.nn.modules.batchnorm.BatchNorm2d\\' has changed. you can retrieve the original source code by accessing the object\\'s source attribute or set `torch.nn.Module.dump_patches = True` and use the patch tool to revert the changes.\\n  warnings.warn(msg, SourceChangeWarning)\\n{\"logger\": \"logger_stderr\", \"request_id\": \"no request id\", \"message\": \"/opt/miniconda/lib/python3.6/site-packages/torch/serialization.py:434: SourceChangeWarning: source code of class \\'torch.nn.modules.batchnorm.BatchNorm2d\\' has changed. you can retrieve the original source code by accessing the object\\'s source attribute or set `torch.nn.Module.dump_patches = True` and use the patch tool to revert the changes.\\\\n  warnings.warn(msg, SourceChangeWarning)\\\\n\", \"timestamp\": \"2018-12-26T14:15:36.534278\"}\\n/opt/miniconda/lib/python3.6/site-packages/torch/serialization.py:434: SourceChangeWarning: source code of class \\'torch.nn.modules.activation.ReLU\\' has changed. you can retrieve the original source code by accessing the object\\'s source attribute or set `torch.nn.Module.dump_patches = True` and use the patch tool to revert the changes.\\n  warnings.warn(msg, SourceChangeWarning)\\n{\"logger\": \"logger_stderr\", \"request_id\": \"no request id\", \"message\": \"/opt/miniconda/lib/python3.6/site-packages/torch/serialization.py:434: SourceChangeWarning: source code of class \\'torch.nn.modules.activation.ReLU\\' has changed. you can retrieve the original source code by accessing the object\\'s source attribute or set `torch.nn.Module.dump_patches = True` and use the patch tool to revert the changes.\\\\n  warnings.warn(msg, SourceChangeWarning)\\\\n\", \"timestamp\": \"2018-12-26T14:15:36.535394\"}\\n/opt/miniconda/lib/python3.6/site-packages/torch/serialization.py:434: SourceChangeWarning: source code of class \\'torch.nn.modules.pooling.MaxPool2d\\' has changed. you can retrieve the original source code by accessing the object\\'s source attribute or set `torch.nn.Module.dump_patches = True` and use the patch tool to revert the changes.\\n  warnings.warn(msg, SourceChangeWarning)\\n{\"logger\": \"logger_stderr\", \"request_id\": \"no request id\", \"message\": \"/opt/miniconda/lib/python3.6/site-packages/torch/serialization.py:434: SourceChangeWarning: source code of class \\'torch.nn.modules.pooling.MaxPool2d\\' has changed. you can retrieve the original source code by accessing the object\\'s source attribute or set `torch.nn.Module.dump_patches = True` and use the patch tool to revert the changes.\\\\n  warnings.warn(msg, SourceChangeWarning)\\\\n\", \"timestamp\": \"2018-12-26T14:15:36.536695\"}\\n/opt/miniconda/lib/python3.6/site-packages/torch/serialization.py:434: SourceChangeWarning: source code of class \\'torch.nn.modules.container.Sequential\\' has changed. you can retrieve the original source code by accessing the object\\'s source attribute or set `torch.nn.Module.dump_patches = True` and use the patch tool to revert the changes.\\n  warnings.warn(msg, SourceChangeWarning)\\n{\"logger\": \"logger_stderr\", \"request_id\": \"no request id\", \"message\": \"/opt/miniconda/lib/python3.6/site-packages/torch/serialization.py:434: SourceChangeWarning: source code of class \\'torch.nn.modules.container.Sequential\\' has changed. you can retrieve the original source code by accessing the object\\'s source attribute or set `torch.nn.Module.dump_patches = True` and use the patch tool to revert the changes.\\\\n  warnings.warn(msg, SourceChangeWarning)\\\\n\", \"timestamp\": \"2018-12-26T14:15:36.539857\"}\\n/opt/miniconda/lib/python3.6/site-packages/torch/serialization.py:434: SourceChangeWarning: source code of class \\'torch.nn.modules.pooling.AvgPool2d\\' has changed. you can retrieve the original source code by accessing the object\\'s source attribute or set `torch.nn.Module.dump_patches = True` and use the patch tool to revert the changes.\\n  warnings.warn(msg, SourceChangeWarning)\\n{\"logger\": \"logger_stderr\", \"request_id\": \"no request id\", \"message\": \"/opt/miniconda/lib/python3.6/site-packages/torch/serialization.py:434: SourceChangeWarning: source code of class \\'torch.nn.modules.pooling.AvgPool2d\\' has changed. you can retrieve the original source code by accessing the object\\'s source attribute or set `torch.nn.Module.dump_patches = True` and use the patch tool to revert the changes.\\\\n  warnings.warn(msg, SourceChangeWarning)\\\\n\", \"timestamp\": \"2018-12-26T14:15:36.548115\"}\\n/opt/miniconda/lib/python3.6/site-packages/torch/serialization.py:434: SourceChangeWarning: source code of class \\'torch.nn.modules.linear.Linear\\' has changed. you can retrieve the original source code by accessing the object\\'s source attribute or set `torch.nn.Module.dump_patches = True` and use the patch tool to revert the changes.\\n  warnings.warn(msg, SourceChangeWarning)\\n{\"logger\": \"logger_stderr\", \"request_id\": \"no request id\", \"message\": \"/opt/miniconda/lib/python3.6/site-packages/torch/serialization.py:434: SourceChangeWarning: source code of class \\'torch.nn.modules.linear.Linear\\' has changed. you can retrieve the original source code by accessing the object\\'s source attribute or set `torch.nn.Module.dump_patches = True` and use the patch tool to revert the changes.\\\\n  warnings.warn(msg, SourceChangeWarning)\\\\n\", \"timestamp\": \"2018-12-26T14:15:36.550197\"}\\n{\"timestamp\": \"2018-12-26T14:15:36.572192Z\", \"message\": \"{\\\\\"requestId\\\\\": \\\\\"00000000-0000-0000-0000-000000000000\\\\\", \\\\\"message\\\\\": \\\\\"Scoring timeout setting is not found. Use default timeout: 3600000 ms\\\\\", \\\\\"apiName\\\\\": \\\\\"\\\\\"}\", \"host\": \"wk-caas-d9227ef224fa45efb0240d09c4ce3c9e-8eab1c3e0606d3902d519e\", \"path\": \"/var/azureml-app/aml_logger.py\", \"tags\": \"%(module)s, %(asctime)s, %(levelname)s, %(message)s\", \"level\": \"INFO\", \"logger\": \"root\", \"stack_info\": null}\\n{\"timestamp\": \"2018-12-26T14:15:45.763708Z\", \"message\": \"127.0.0.1 - - [26/Dec/2018:14:15:45 +0000] \\\\\"GET / HTTP/1.0\\\\\" 200 7 \\\\\"-\\\\\" \\\\\"Go-http-client/1.1\\\\\"\", \"host\": \"wk-caas-d9227ef224fa45efb0240d09c4ce3c9e-8eab1c3e0606d3902d519e\", \"path\": \"/opt/miniconda/lib/python3.6/site-packages/gunicorn/glogging.py\", \"tags\": \"%(module)s, %(asctime)s, %(levelname)s, %(message)s\", \"level\": \"INFO\", \"logger\": \"gunicorn.access\", \"stack_info\": null}\\n{\"timestamp\": \"2018-12-26T14:15:53.684134Z\", \"message\": \"127.0.0.1 - - [26/Dec/2018:14:15:53 +0000] \\\\\"GET / HTTP/1.0\\\\\" 200 7 \\\\\"-\\\\\" \\\\\"Go-http-client/1.1\\\\\"\", \"host\": \"wk-caas-d9227ef224fa45efb0240d09c4ce3c9e-8eab1c3e0606d3902d519e\", \"path\": \"/opt/miniconda/lib/python3.6/site-packages/gunicorn/glogging.py\", \"tags\": \"%(module)s, %(asctime)s, %(levelname)s, %(message)s\", \"level\": \"INFO\", \"logger\": \"gunicorn.access\", \"stack_info\": null}\\n'"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Get the web service's HTTP endpoint, which accepts REST client calls. This endpoint can be shared with anyone who wants to test the web service or integrate it into an application."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "print(service.scoring_uri)",
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "text": "http://52.236.154.243:80/score\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "### Test the web service\nFinally, let's test our deployed web service. We will send the data as a JSON string to the web service hosted in ACI and use the SDK's `run` API to invoke the service. Here we will take an image from our validation data to predict on."
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "import os, json\nfrom PIL import Image\nimport matplotlib.pyplot as plt\n\nplt.imshow(Image.open('test_img.jpg'))",
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "execute_result",
          "execution_count": 37,
          "data": {
            "text/plain": "<matplotlib.image.AxesImage at 0x7f302e93ac50>"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAD8CAYAAAB9y7/cAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4wLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvqOYd8AAAIABJREFUeJzsvc+yJUmS5vVTVTP3c+6NiMwqqqSBnkGYBWs2LTNPgAg7RFgBDzArHgDhUViwZs1iRHgCFr0eYJCeFpmhqO7prurMjIh7znE3M1UWan7uiawqhK7qFFJKrnVHZcS9fo67m5vpn+/7VF0igrfxNt7G23gbf3xD//++gLfxNt7G23gbP8x4M/Bv4228jbfxRzreDPzbeBtv4238kY43A/823sbbeBt/pOPNwL+Nt/E23sYf6Xgz8G/jbbyNt/FHOn4wAy8i/7mI/CsR+QsR+e9+qPO8jbfxNt7G2/jtQ34IHbyIGPB/Av8Z8Avgz4H/OiL+t3/wk72Nt/E23sbb+K3jh4rg/ynwFxHxlxGxA/8T8F/8QOd6G2/jbbyNt/FbRvmBvvdPgf/r4d+/AP7Z7zr4Zz8t8R//o3kpDhGOiOS/QwgRCEfc5yfi+H/yMPneNwbcMxPJf3/vrw9/efiFvP7zdx4W9x+LyMPHXj8fX3yF8Nv++jvPH4DEbz/sd4z7nR5zEXH/yvzW474e5+R3DHmci+8f//9+MTE/cn92v/nl86vk9d/E984TXxxOSF7T46Q+ztFxr4+fj/lzj4ePPV6THhf68H3zv+6vP//i0h9/JvNevj8fAjiIfnFPuVY0fx867+e477yOmPP2euxxaQ9XL0FEPNzu8e85RwhfxGz3+ZvTFJ1AEHeCwAd0lDYqHc1ziQKKCCiRxx+3ez9HoPO+Yk5DkB89/p73lMcRQYgQBBrgMmclgtfd9LBlgYggwuejnMcGeDggeIw80PNngeOeD9I97UeEEJLHSwiBQ4CK5v2oPqzXPFZVCc/nKKZI5P3nETptk97XgyKEBKKGu2Ovk/WwdAQ9noE82ojXZ68K4cfxMVdG8Jf/+1/9KiJ+zu85figD/9t2+Be7QUT+OfDPAf6jP638+f/8p4y2IVRojo4BOCwL2Am/7dh2w0cn2o6q4poPQvVwDp0YHYkBUQgGeIAeG7nMBz9AGiI693MgIvN38YWBOhYLQKig97vQdDwyuG9cAI3XjecGUgk13Pv9HGIQvt+vKSdnzIkzJFqe+3uzKJLn9+nopCriQnguVtf58+8Z4nBADdX8rGqhe6OoEcIX9zzoEBUTxxFUnsAaMXpuWM37dBwJQ0zv1+ZqCI0ICBe01NycsiEsIBWk43POjs/CgKhIFIJ23xyhgogRLphV3HNec/ZzjiJGnlsc9Zj+QKAFcd3nWlgQqUS0vAf7gLc2jUigtUxnLcS257MQQxGcQW7jQgiYCYNAiwGBY4Qoon0+JM1r1o7KcxoYLYCiskIofRyxQn7GllMaG97NtVlB8lmGCKjl9xIQ2zwmQPPawgXVd7iAqqXRoOfeiMoYisRA+gUZO9v2Lb45/8ff/QlNf86/bF+zbgt2qogpURyRoIyC1YJK8PX6xG3bOL1biQhK5PMbAhYgZa4hNaQ7TQ23CzUEojIIRldcPyG+oqHEcJDOdc5DjLzn3nOt7XxGxGgvgkXn8mmjl8bLdqHdOhLK5+0zYUa5XbjtztY7n64XTmdjENz6ijOo6uAn1lOlDLjpTsP4qjyDOYPBuSwwFuwE4pVSg9UFrxUzIVxRVZZSQE5zLyvnp4qwMHpgGqis2ALm0CQ41Qp9IMuJ8JbrFSiqtLl+ixra08HsmsGJKfyXf/bf/xv+gPFDQTS/AP7xw7//EfDLxwMi4n+IiD+LiD/7+b+X3g83zAVlIHQiBqODj5xIJw2MmTDGuBumY6M+/kE8owdye6oo4Me54fF32ON1zT8ZCRwGJSMlm1Omr8eSGx0ZuSk9FzSRxlM0kNhAdtQ6WoIYTrihsrx+9xERGIjY3Zg/nh/S6OX9vUaw4Z1wT2cG0yjyxfV/ee+kodbcZO4joxTh1VnO+RmjffG9kGZN6Cjz+12I+Yewu7OLaECHWOdcbfN7MkKUiDw2KoKDbHmdMzpLZ+qIDjx2RDsSO0ojYqRx13zWr0553q/pNNpKxKD3C4QiLPT9U64Nn9/hwRgzItTIc+IEI6+Lh++XQNSJEMDy/PR83jP7zIjzDLIjMu7Oe4zBGI5iCEaEpkGeZ4zIY4OBSDAYKK9Z6+HMIgLRiojhXkAWXHeQ/rCe5lr1QrBDKGN8IobTJbjEwuf2gX/7XcdGMIqz+wVngxZoK5wqLDI4a2AxOJe8cnfHiemA06hHF3qHMa/fYyO8EH0lWkeaERlpYBKIBF0GTlBQFi1YwKLGakq0nVM/MT45rV35vN9o553t84WFSqyGn4KznmDcuLLj6vQnxd8tdC30MMowTsOwHUoPdm+8XL7hnaycQ2H7xAjB5IlaV5a6MSIoakgoISWNtwuLVk5iFOBkHWNDi/Hx86fcazOI29qO9wGmuAmo8LSeaK1lhC+OzfUprVHFiBGEac5J3PJPv/GHjh8qgv9z4D8RkX8C/N/AfwX8N7/zaM8Izorj+2d8D4wV0UAtUAYuKy6fCe8wBqXUTGN8zM0yUA/AcvFF3Pf6PeLV3Ih6RwaONFlmNHFACHAYxGNDzS/6Ii2PkBltL4SkIREGkMZeKNMQCSaGDwXvGEFIT4OA3TN+CNwjAYTD6B9RZoB4Xm9mAkDPFC+jP0ccZJom8XRqmRoLxKB3MFPGaKgqozVUBUPvTi98oWjL5wF42YioKJH3FZJJkS5Av89ROrkA74jZTD8sr10MqBA3QnbUzxArrhmFIzuBpnOYRkoe5li9gPg95U844PX5jJnQquVvIwIUoiYsEEPQciJ8J2KnlPeMsSPqaEQ6X7E0hFUQN2QfGRHLMiGOIEiHpGKZTehAxjKN9IZYIJyAjsuORyWGYVbm53J9+fgOgGJGiIFYzj3pcBSbRh5cG8LA51odAWorHgKyUhbL1P8OX/Z0kGOgIoQPLAbChRidMXauf3Pjl/sH/k37im9a8FUx3AKTNGpVch1vtiLhSCl8My6oVc7uqBitCBpBFaGLo7qjVhGvIIJKUNqGawdZEXfMVxRj+I2gYQU8FO+dUgzvQusNcNzgZk73oPQzY/tEbEE5V27jirWg7c7NdqQPNIwXYFxvLK1ztYr4QugLolDlhJVBuSn9+R0fCzyPQluNFUPC6U2QeGbRwVqCXjt9dLRWDCMMPt1uVDtztnfc2s4pgvfP76gIlMo2NkpdsLXQ94YtlTY63/SNVSu4M0ZHi1Jrwajs205ZKvvI9Vfa8hCY/WHjBzHwEdFF5L8F/hfAgP8xIv7l7/yACKaALfjyM5ALvl0ywmwl09MlMLPEG1Tv0alIEDIe4Nlp8A/I49FAo2mUJaGd3zaJckDhEb/xeVEIj1eOQAQFfG7+A8sTXzI1vzuDPi2WYwdOGus8fkINDrnJv0TwjXn+iT3eEdiQjJxVGDPClAwhJ246v2ka5AinWCU87vDgIxQVE5IWHYwxsAOKGYJZgFcSXx4oCRGMiQff52em4zIS2pBoE64yRPeMfuOURpeXOaEF9WUagp5zM6PmiIQ70Fte6/A7p5D3+IChA10CSVcxH7cgRcEDjwtFn3MO44rKilieZwwndMJPUWbUn7gu2hO6YeSzkTpxW4OohOzTiRRMG+IFl8isQK/U8iEhEx949BkEJHQloXiAkUHM4DLPofM6BVxB7J5JHDCXxHOeN9aZ8bR0FihCA/0EsRO+oL0DO9Kdvjm/6id+3U7covPu6RmncuqOqFMMdhGCwtKCYoLvznk5I1Yysxgdd8WsoD6gKD0WREpmnToQC8JXVDt927AyiKURreBmGAXGYMQgBuy3Rm+gWhPesgW5fGQRw+lU+0Brn7hev6U+r2z7FUxpn3aeTwmB6GiEeDqreMJ9Z5WKu3Ezw6zyvkLUirfGPgZyeiIIVoJmK8WAsdK70HyhUCj1hOggWuf9+QnTBR/OV++eE3pNioDeG6fnE7We6D33tVXD98bzemLrG8Uqi51oEez7jp4qitNHZxFo152yVoYzA5c/bPxQETwR8S+Af/H/6di5UEe8UOqfwLIw/JrESNzANiS+IuwJHxsqgYoTveXmLmXCDEr0QfiOyimNkDdkOoDX7Z9pkqgSYwbycyOhgxjjgQx5MPTuQKBqOIqI4wO0ZKSexI+ilkYhxzT0oSDKACRD7fzmO6JwEG8AFWfPa4wxo8uJdd/nLK9FYhI4M7tI9MYIeYV2VASxNeEWQIvdo+4RifvlIhXUnFAlNGGqzI7kjm27p5O0mHCMxx0vlpEpuOMZrTqoCCP2NFoiiPSEuhwYgujOEEGmwXaPjKqLT743wFdCGx6CSEZB2GD0CdnphEdkICpEz+entdBFkb0j0emR2H/hic5OicSvTQoRCu55n8UIT6MRnhCcqFBEGX2uhzIQ8SThItlFGZXQz4gUIgqVpzv0k9mY3aNzM52BSf7cPdeeARI7YwRmp4nXJiQCgokwOGFWJ+mXTnDceaiWa9Q1uQdvtP1vkOHsn4zvLhv/bvtP+euXnzAUttH4ydMzt/3KImd8BCIbaz0lx6IVetA0YOw8S8XUeNEgfKdSGCGUEJYGW4lc230gtuOyEgW8PNPkhWURvDXayGxsLZU2nctyCgLncrkh6tT1a1rb2GPDZcNtUGslbmBSwTu1Vj6OxghJwz2cxQrs3yKlUuSJ1jZO1mj9xk3OnIawnE80b1SLzGxPJywUH3BSp6lztsLzWhLeE2N7XoguvNNCUaFtn1nKCl3R9YTKDl4Yt44uipgRt85pWWm+Q32m0cAT4rSlwK1hJ2XIYPRBVWO/OHVJPusPHT+Ygf/7DcPrt0h0to//Cm2Vos9TPLHPqHkj6AQ7PhLLVUBUGQdr7hnlh6RjEC1IBMKap/GOSkIN4UJEQUpL46+eUbIEqBND76To90dM5YZMeMS9341kigMSh/aYRJcFER3jnJfxgIfLkS0c3AEgjMQ6Jc2Rz9/fj2XqYu7EcJ5X9MC++fLaVV8xZl6JYxFJeMHTgCExlQjpMA7MXSPjy4SH8gozg5LfyAa+r4TI01tG3QfMITFhpJJ8hV3m07TjzlLpMR1XcANf5omuSWoSE383DicakWsg/YJDZCaT6yAdYmLdDaRNeCwIrwkrISCWgI9wJ5RVFWek87GCFGX4JNdhniedUzqynqSugGCpzIhgeBKDjMGwQExRJg9CYFIRCZAGvhDREV3S+XkwH0cGMsdD18ySROpUqKSTjxBiBNpeYBsUa/x6rPx6f8+3l2c2r4hWqhpt3+Z+EIopxU5EBuZE73RTnkO4jc4uwkrhXSl4CVoEzTvvxWiW0GCoQa1EdyI6RCH8BfGBi6GxouGoZ5ZXykL4Tm8J7xVdiRi8bB9ZloXqQfeBjCRCt21DuiMBp7rg+5XbtlPryhZKYKg+oQbb/plSC9v+DFrxIsANv92I9Wu+bcpPa+d2HYg0zk8rPpQiFVfj423nw9N7xhg87QpW6KpYBOv5mWgjuZDoWNUMkEIoUujjSlVjtB0fgi4vuaZEGd3Yh1Ct0FpHgForexu0ooQqNupvM5Z/r/GjMPBpM06IO6sl5ulNCGmZZrMChvSOuRFMjEoLqCJjT8PgU47lAjbxXy13SCV8meoCmaqWyI2EozJmhA+CgvodVo67YdUHg+yMGAgFkSTQUsEAwpY4NQW04JHEbGYZcUduBAhJijceDEbIDmHIeMXGYTw4m2kEVadj8LsKJH/huWHvuLgTWtB7lJxSPhdwBibKGB00VQIy5zC/LQlwrQevcTgOuZNm8AhpHf/D/Lkg3nBNA5ILnIRB8GmIaz4TLqifEXRGko5GY0hFNQgq+AlkzLnMLCTxcM3nieSxEfiY81Jqkp37QNWJgGIHxFczPp5RJ/rqHEMi4R2PmRFNqd0YKHPu9S5ChZI8jrAkfONyXxsigkkKD7XkehhJaSA2Ju8yHaTEVP8oMYlgNWO4U8p83vbqbCM64T2zw8mViKSh9VBM4PZt4d9984Ff+9f8dQ9YbmT4UHDZEEsivG1OOZXE9H1wrgtdjU++o0th6UErzuhOHbCKUV24PBfUHcbOSgZdmagJURvWTpxCuFnHzBihIM7wQYSjOGs19tFQGfTRcAo+FI0FS3qcq2/0ItxGA1Wul87mgiwLPgYlGmPcsLrgGLdNeDoZtVw4PT8xLg2rK00GtXziawRtz5hVluUdPjroTi2DGx1bKn1yarutmBlLd6oY+77z9M7Yry8UnmBAH8Z6ekLGxqoLYEk8241lrHQ3sHxmRTs7FUOoemJvV1wHRkO8YfGHm+cfhYEHI2xFZCduW6LokunaYTDCB9H3DLBr4hu9b9ADlZJG6MDLdUILJDyRwzPqk4xEmbCA+z4NxjRWehhUfYhKD1Y2x6se+IjiAxmZrqdm/wzheEw1kM1osMc9uv7i3n8D79eHc2a0r1KmikJntJUG8JFHiLvi4uE67xF2zEzoOH7CTUzsPXRey/Edh574IF+PWz54jC9VSa9zNcltfb1HmUb9MIWpAgHRLR2ZnAjpiP8M5BPENNYShA90ygFTLbPOyHZCVIeTE8+oOTJKP86ukRlTkt3MCF9hfOCQG0Z0xG85EVMIJZoS1MzUXp/+wdHcn9cx9xF3R5VsdyKoqoKMzOgSBuv5bMVRXe6R/yH3//LZyd3ZDHfUZGaLTxDlVbIvinAoaAZyKIxw3K/YXtn3K9/IM3/tP6GU92yto0XYt41ST9hyJagsy4p7Q8WRcuIlBksX5LwgzalI6s5noNNNGALrpRNF0GVJcrvt6CioCp0OOujuLC64vBBkZuACGoUxBj72zJyKggtPS2beH6+fcWA3iJ4S1TIKty0lh0tNVdLL/oLqmb3tRHsm2PjZV0+0a1DKmdgGi5yIj59ZzyvhlVJPrMugxyfCGtUWigl9F56f3mWg4UbIksFN9yRLRTjXymhKXVZOa6G3K2ZB37+j1HcEbWbxCZfd/Eo9v8PdqFVp241aboxa+fZ2wxyelgXxLTVc/wAaxx+HgdeByte4fKQ8/ZwyBuPTL2a0V6ZqYuCa8jV6Z0y5GofqhFfjNQiKHIUPJcksnyoXaZOAXJIoKjE3XBpO7zvglHIQnqmGPoLjNIAHLpuRWPTtbpxHdNANtTLVKYJFJZqA7PM7DhlbGqDg0bhPpEgyOjwkin4P7xIe0ggi9AuHcdhmNZnRYH554Kil4O5+rJAYeUnCOgnSQRp5coMLhASqM+on/55GfmB36Ege7mF+sQcHsTwmdKKawlWRMa/NEo6INaEN3dJnzBTHmRJWF4KEL0IGULK4xGNGyclfBLPGgYxk7XCeR/Zl5N/7FUxIaeukbAMY6XgCQ9QQI5/nkXFEfrOgaeQiJnwzA4boBEbECdFcI9FJmPCuAkrp3Z3k1wmTSerpc/2mUkVmdkIIYpPTKGU6SJlG9oZ46sYjBkSuX42d1Bg/881145eXn/OLl3/Cr+3GySsWHV0M9R1X8F7BVrar8nQ6IxZc4kYpBV0K2ga1KB8Z1FL4OgwXZxuBduf67KxidK1oE3xUTqXTGGgsOM5gpXAjQlnLib13FKVFwoEy9f/7vqNxosdnVJVSg+6O0YlNGG2gm2MjeFbjup3ZekuIxnfKSSkmbLcT18+O2YWuG80L7wt8YuX903tOYox9Z+iK2omiGUT1fUUV2u48L09oKFoLxR2pwqYtJZQOaz3T3XnZbtRSoRXW5QODF/a9U8sJANMNs/dEd2oJbu0CZTB2MEkOQ6bcV3zlECT/oePHYeBDCN2xEHxpGZnsSt8atZYkJUog51N6VJLwUb2CQW97GnlNQrBqIcah9vA7UWbqU2kAoIhGGjcc6FhNParGQzQqrwVKOeHlboxjBOg1SU2XqcoAxjKjbAAhSla2HJCRCzAOR3OamGrLzKVkAVHvneJZiAWaKiMyUwk6qqSBk8S1D/hlCMQ4ik6mM0Hw8Uq6HjBNUbuD5lm1l45EROjdk8/wgRRNJ8mg95ZGauLbDEeKIREzH8gCKeDgXtNYSkbPTKjM3bEyQQXZk1id1zceJKpixpCEVvBXeAsfpIQ8HYVGJGQmhlHBBy49ifQ+iKhYa4gUWv1MyDPqcy2YIyPxax8NswFyxjWzw7lIU/8+TylTWjoiKyNTQlnTOZbE0W2S0qoJZ/WW8ITHhlmdsMyg+QtlWXF/Tl5E5txyZFZr1gHcISrAbxPfKekAxpb1FrESvjP8G3QYXK7829uf8q9v72ks1PHEJo0wZd1aQpgiSL3RTFiWBeKKXwfyLpARbA3WYjQRTn4imvPRMk9aMawqdQRBo0hm4F0qfay4bey+Y1JYys4WJfXiIyilMMIRdnSSuTEaQsexudcz7tAhyJ52YLtuaBHOttJuRvNfE3WkhPYljW4b37CeVjYvrO3Eppn5Rwh/cnomtp29LnBeMyIncf6lLJhVRnQqJQO+U6fLCalPdBEWCaQ+I/2SYoQR1PXM7o11gdavxOisFhCdsIKXwR5Xwiv71lE9EdIIFZpb1gbE4NZuFD1K636T//v7jh+HgYeMaKVBfMBUkHMg4cTYcS/oOBPxgpnRfUdcCTlnyhpZICN+4NCeEI1ODFbHEUxOfJlJfKWBUVJnzdSrp+FNtU3+X+DRJ/4+HojE+ML4iygW4DPaFEky2KNCKGoJuWho6uaZe9Ud1dStp8MJTI4y7cTfvXecLERSsRlRdg6y93D2R6VthM/Cm7jj6q9zLXfSc4beD9E+D/eUn89ryspO5Mga7uDEw1M8IpBH3gKYktV8CIboQCcpCpLzEhPWOLB/PWCwgUx4A5GZSUVCFOOAkCaZa4GGw1QghY+sB6AkcR4d3DE7QWy8FmRNZ+yDsljCBf1CsYWwkdDKRPYOzfoB1WT26Kms8oFaneuqAoaqMUZCCbXW6dge9Pqz+Eo86z4i0sHfITJVRPqdnMtisJpwUByOJYhIuabGJ8J3ZDjRNy63E5+3r2n2gZsmbGQKMgaNQZTKQkFJjXoVJSzoOjA94VZYoyAjWMPw2rPOhMaprsToyTH1AXai+85QZ6HSuabEsRo+kg8o9gmNQoygtSSaxTUN4UiYSVXTgRejjY1Snuj0XHd+4amc+fjpEyi8eKfYM9frR5alslphEEj9mu6N5xD6IhjGsglU5aUoxoKEUKWwh+IWfLAMvjo7T+WJwcBOC30IQqf3lOv2UnG/chZjGSBm+BisA4o4YQk5NwbaCwtG2RbkaeXjuFCWivSG7jAsaCP3gzKoZowZRfgfHsD/OAx80DOl9ycifgX6Aep/QCyf4PIxYRUMs69SGrUCMYjWKPVE9D4x95p7Vnpq0x+VKZHG0XtDiyTxOIksVLIISYXUrE8MHp1GIIA2o7fXWT8w/wQHSJ4AQF8IDJ/qHZEGuhNxFP4EaKpCErKQiZ1KYpYtqzfTaSSue3dCJEko6L2SUIBxyERhcg3580Am/PIKKcQ0UCoDZ0r3fEw4JlUpSjBGw8yQg4TlcDiOqcwFGHd5YYikk0XTGegBtQzcU00TsxBMTfFh93tPYvAwjAdK4/m5h94sh+MRCsNfZWQRAzxlqAf8r2FkK4ieWeF4QnTPjIvrJBZn/QCGsNJv6QCsCKnfv04DmjCPTuQpnyNggTu46yTPEjayYowRRDR0UiEuPjOY+Zwke8OoK6NtaF0QK3gIEakzR0dWL2OILPQwVOp0WLMPjRd0EnI+LrDf0N2JHf6mL/zlRXjRFQ9De2c/QcjgjLFqYQ8wdeg3RjQWO2N6gjFYQ5Al0N4ZujK8ZiAjBRk2My4oWnDbgEKRUwofbM3M8ZApq+H+TItGaIAmWbzeTvTe2TwzXPoNYVAkgMrunbbvSHFizzqEpZxp0Wh+QfuJup5o7il/VU0FjwR6eodebizF2d8rvCwIlWVJyKWUJdteuHCV4P36ntvnDVmN4iM79AylnAxzqKZcPfjaFoaUzAzckRF4EdpSiOZUEnsfdfBxXIjziWVcOdUn9lu249i4Ic1QWyBAFUYbFD9NRd6r8u33HT8KA5/WoRFyQboDG16S8BRPgs3R1LerZAl3+JQopnHRmAZPhZGllhNKmHI8wEdPLLkrwdFbhYRPHkjAJPcaop5qm6NFwSxGeh2zV8rDT1SysjIhiUMDfcAPNiP7mD1OmDj6xLanISslwJ9BBI8Lifk/wEYZ23MUV722JMhGSI/R9+GP7jLFiNcw/5j+R6I2/H48HrkR/TCsh7Obc+Ax8f5ZVDVJyXu/mhmCWJZrgb+qeOKQA4oAJed4YtVzJkknO5tG+UHo5rWodphR3X3u76qi1NL73CSHQU64Ix922InQhIrUZ9sLS5VUH7mx1SQdxBeE9sE2y93A3vkTT5I6Wwh0EJ0wzeucegQyt929OdfUzTMDioT3PCu5syQ3SWQclSc8Ap2Xkbj+dPzs0BTpSoxBG8qny1e4/BShoA61rEjfCDGWYkQL1IxoO75kTcXmDemNYivbHpSqqC1c9sHTKXsa6SSnc30LQ5PrEhpDEuLSyXG4g0tW6MroFOmUqoxNGDfhVq60EXQL0I0WG0UKY6/s3NijoE8L1+tninQaNzb7hJlRW2GUxuIgN6fWZ17GjVaErzhxaTvLutDHhnZH3hvuO1vf8cW4RuOJM1KUn0ilXT7x1WmljVvyA1vj/PQOH5W2DmILTu8qn0dFY1CGoEthf7nyvjyxv+xQU+BhIsTYeLaCidDIVgleU8JqruyxY5GV6h3Hx4BTPtcSj7bm9xs/CgMv4oh/xqNgfgFZ8NBUjsxLDL+mxtQj5WYaDCcr3qxlCwOSJMwPzCrIozcIGeWZVcZoBI5pvafdqiONUNi9qZXEoXCR2SVvVog+WBWZkEIWpOSxxgLhDJkR5ihANm0iIERfm4aVgsZgeE9Dov7qvSOhq1Q6jmyopdkbQySLlPIaUlmgCIwDA5+whR4GaDbsGqmuEBVi6JT/jQlZkQZ3aq5NNOsMRn8lhTWj+EOCeDgNEZ16e8fMZ6O0VMPESJgLfEYqRoynAbJWAAAgAElEQVQksROCqPffHeMo7FLJplHCAanNit6wh745AE54NjrTOPogdkQFHwtFFdcbPguVTFOloqopCRQY4wrh1LUmT8wOcbpnbQmspMw2/V+gbqmz18BxTMt0eIkJj9GzKvhhzrS0lPK6McIxy/zzINJLqXMOBj5FAnrAgHZBvKbh5wgAdsSvRGwoF2Dj06a8XFf+Jt5xk4XWd6QWbmPnSQ3Ds+jGhNUqHgV1xUoqX1Th5M4ujrx07FT4oEq4sWhFgNY2hglSjJN0ugzGeI9IwfmOJorIilghvOEuYAttOD2CqMHQTh1ZRHYUP7UwxtjpMjCtCDvehT5ud0FDLSuXbcfKiWu/4L2z1IIPQ6NSh7KrYNPh9vKB1VbKeEH0lA3SxiBEeC6GunKrCd10FIuG1IpYndye816EF+vghTWypubFZLZzcDYNxIwawqUHN9kIqxRXTgxawBJ7VtyGEFoJL+goGMIWDasLOhIG/odoFPajMPCEZpsT3WEo2J4bsTc4YAVPjD07QI4Z9VpGETEVFdHBuGPcB/abKhjPRl4IRZV9VnW65//khszzDI7mZLkxE3s/2g1wV3mIHtj3EdkecMskZo/pndCKBIwQRJKo1YmzJmGoCInlZSCc0amhEI7qCaRNfbvP7oVHFDmx8oiswi1lErf+0Ienvh43ZhFT8BurKH8/4CETeMTvj+9DMyoPIqPrqQdXLRztcg98/N5J8+FcClnHwGNWkvPkh1pnZje/lWsaei8yy4vsROjEtBMHj4fWvxn5z3oEbrM/2JSjRsoUTZLQpyd0lUtzwiATr78Tx6XzqqiasJnaxP5zHrIa1ebadRhBWMl7lbivC/TCCBDeTRz+0Wluk0I/E14QBhGFYJvnarg3rCeha95gdLbtmctu/Gp7YliDlpDaqgOXyh7OElAxbj4Q6yxRKF5YumJVGUWpGL0Ie4GtKM9j0Ho62rpmW4YRO7Qz5SQMrogqz/LEBvQwfHh2WYydcKPomdadcKeIcO0basI+bhQ7M9TZR0O2E4zGEgs9NhaCWAxxobTKWYQr8FQWLmOnGYy+UUUINq4GP+GJSx943SECF0X7hCJNKWultQVXoXbPWKwGSynUMDSEbQv0qfPSBnUxnkZhqxv0hZ9GZbvd0Fpga0g98bltrFVYNOE4QrhuL5TlzB6KWKGq09uNUFh0YYxB943WNlhLBln/AOb5R2Lgs2mXtp0oFQlDW2O0DaPjPYtX3JNsExXQksj3jGxFHCUllZiinl4SMZwtMfooU0csmGU7YbOSEj6brW3VE/5h8n7yCrvLEWVOmCDGUZmZ6adOrFondMQBoTCPlzor9mN2R5z4eELkQM9NfUTmMjUpKkS/zQ0PMOh9TOJNQBU7GBnLZmsMfyArM6Ues4Q/WwUElFcY6yjGQWT2/Dn6obxmP0wi+dDgj30Dg3L0TpEB3undsxncvTnaXGYHLzKVSdlqGdSz6VZkHJsR4JjQlhlKwWNDRkkZYwRhs3sfC+gVouJxQ2LgPnu++Ckx7N4m1JdGu4xKJ7C9p5FEk5+JwIunzR2JGcdIKNAWI8aYFb6ZEclsICeSUk6fhGdyPYGK0ttOnSB8SjAHI44qaZtS25RNigfSBCwdDqVNFY4CN9xfMDHAsi1GZCCguiNcYXxEbjsvfeWvbj/jb/sHNnlPj4Vr2XgncOaZa6SaJ2ZPuNUK1xisttJ84GtgI1hkQQxOZtAHhUKvgg6jagU1qnZKzFL7Hdb1BNK5xZ57a9uopeQ1A9IVP2dmJ35LCqckSV2rsW8XhIEMMrrXymV8wqqy9nfs2ydaC3qGQOBZ4n/Wyrh2Bo2+VN5dTwhw1eB0rg+1MYNTLYgppVSsQdPGWgpmSmuNEz+h33aaCetaWbxjY6FIYb91Xk5CvxlrtSyCKpUdgVVYxuAsuY4rlSFKd0ftmegDE2eYcH25Ulfj1IJrzXBdDZ5Z2C+BVqNPCOwPGf8QWcAfPIKUMvpYiUiJUjYNC4gV4QTyDDBhkKPNakZ9JnLXPLvPqNUT1xf2CXm0jNIf7jj121NKeeDTLrM/zYzIeQ0gH9Uov62FwXH8XXoNE4I4osgDYkjlRMrqDmORxjVREkVD7n8ijnO/Xncpdv/7kQncWwMcDige//5QjDRliMdcfQE5zb9mzcBh3FO3/1oiPzFCWye8UghPgqi17DPv/eAfUqZ4RNoyYRE1y3ayPgumpIPsd/z+OEd20Twalx3z5yg76jWNgWd2YyUwW1EbiFSCLcnzGIRnZek9m7q3QH5tPnd/3srrvx/aMud/D85krsY7j7NiUWfx0ozkJbJUfYy7A1XVCXO9FtdlhH8smDTuQYeRPVdkOL5v6AjYC+Y75qC9oeMz0m/EfkHHjWuHb2/Kt/1n/NXtieswGM77siKivIyG1oJC9jpvnWsMnsvCXiZ/1AZrXZCx064XaMEiJ2jC2GdggbDvyWN5z+6kSxV67/QGxQJ1x4T7/YISJYi2pRMuldCFkyr0Ri0pBxYR1rpSrdD7fPeDBy2uFDOY3FYPx5bKaVWsPfH09AFbfsJpX/lUbxRzqhW27mgv0LLVuDMYrXO7XdnVs3solTGc0+mcTiI61OCmO7sGF2mMZ6OXzMae3j3R1Okz4Fi6cPZ0YrvC2Bv7fmP4BR879AVkY1mMtt0AZb85YZIqq95Qh7131nBOI1jGHwsGD8AJqZ/R5eu5hz/NiHuqvGMSF0a2SnV7JewGyOydckijUSHGmHj6CdWWxNvRhwW7Y75pwI7oKwuFxoMC564++QJ7D47yQ5mSPEcmMce9L4yPBymlLrmxY7/DFRoTChiv5GdwfPWrYc4ukEfvl4zQy4SuPA71yzFiXlfOrngk7DOdyKEMKtNgO/LF/R2OoPc+iUdJKZscOLmCDDpHZW1uONOCacFHNkg7lExWk0D1Yz49VUyqr8RpeMmIOAaCYsf1TO28xFMa6n6aznhL2MwVShAj21moV9wbRL5XIKKlSkh2QixfKuLZLfN4wcSQjoYhekp9ORm9i0cqY7wz5hpAU+Ey9oRhyjKNt27EWIA9W+IzUIdOxTQrNkcERCd6KrS0nmch01yvs9tktMjagv4V3ToqS2YF3hHpjG0GKt4TjuojM4i98O31H/PXH9/xF/uf0KJhyxn8xjYaT1YhhJNVvrld8DWltl/LGUMZexrq7KwhwIIAIyDKoPcrdZnr1Res2B1q8wkfllNmwn2vSRKXSvMO5nS/YXXFPVJf1htbOF2MsMLNnTAYW8OKQjG8OXiSxqsZ22hwVso1+GBnvus3VhFu753rdbDuG6UaMox1KXwnZNA4BKmGjkGtC6Pkmi+i1LXSe6MugGZ9zDtWgqBguBbqSEmjS7ZV2G8bLQZPhYQrtdBGvv9AIhhhqK20uBC2Qe+YvOPz1vEwXHaKOZtnf9mbwnPPoi83Y0T/Xl367zd+FAaeKbFzCjoysrFphNQ6ETtoYuwxuy6GZuQqtqcx9SRCsxf88RaeGfUKEIWI4y1Ar9rwx8Zb+cuEVuAh+oW7sVLNlx4ghz4ewrMjo0pWOOoBWRA8Rs4+3+KUlaGvGcJ9+NED5fXR3svh5/k94rWF8BdzmPOS95PH3+EUjp6YD+eL13tXslhDJ1F8ZDNmuZmHOyb+vdNJ6vEn+ZfyQOfoVZNvjFrScM0sIXX8KaEMYTptZvTeQPpUPMUX+HmiXX0WkM0XcXQIFqgbMhawHY06//uUOPt4QqRlO+VYIOyemYR7QntqqCW8Fn22q4C7M43REStwXyt6jyhzHvP6anU8NtKhZuaFS8pJ59SZZhbUmW99itc+NeJCGYnVud/yjVHS5mciFTL+yu2E7BnYROLwervyaVd+/Wvjb6l8bjc6g9MNhu8c4tc+NtpLY1kLpQilFFpruGeTqyrO+2H4bcdtUGxBFMYIluWES59tEyBlrLNViARFC703HOe5PtN90CaRaWIsqyB9cMPZEU5q1KEMT1jw5fZCKUqU+fpAoCe9Dh74mNnu7ozwSZAubHu+0cvtBV/f83EPpAYN0K2zns+4N94LXMuSGdLorHpKAYA2zrVilkHNsmRFrCCzDUrS3fHtJ2oEelrQ+TybdFY1fEu4T4vSe77Y59ou2f3SJd8INW60sbHUQhsDM7A9xQMOSF0oe6PdGwN+b4//HuNHYeATorhkBeL4iLowvDM8qAbiPYnBsnBopsEYXZBYcLklLt+TIFRj9kZfsgpWLklSSbLhZpZkBlAnDuk+i4wOAtRe3/SjqmjowR3yOvFTYXPf/MyqxcwGnGkkhVnFOWEVFNgnnvyKt2eh1iQVPe7nURHGvQK3H5N2dx4iwuj9DgFEeL5WTibkonp/J2SOhEUUmTg5r/CL2Eypg957VlxqwUd/gJqmfG/q/kPlHtwf11M4eIjZn0eOCs2Hoipp81rS+UYMzIOQo+Ww4kPyOe5n6KB+nmvgglrHR0u1RVXQNQultGNWGHHJdst2ytJysglWsZkRTbgl692CkH2aldmhdMJ/MXoaqVqAhCHUtoTX+oliwehJgB7rRcKgZVamU2LrfU/HucsdTkzNtrDvO5/6GYmGjxv7WNjIl5As9YwL7O3Kumbzt8IZ0U5vf4cCf3vdaJ9+yv/66Zd86zd+dfs7nssTF17Q04ltOKdFeX5aef98hpvzkw8fAKWcn+i3zyynZ9ydZk5Y4WxnZHIyvbdJoD9n5qOald+zTYL6gnNNFYgoMb5Dy5m1rowReAy8OzcNzrtQS+Fm2Z+msTEI1lK5XD6zrmfGCFr/TLUTfVxwhb2tqA/W84nSnMvlwgs7a1VOu9OJJFP7zqJf49HQNYUYa1lwlGcWelyRInzXr1n1SqdIYLEQ0hnjHabGyWAMYalPXP3G2Qvn8xnUeLneQCWNvAdRsq1E6Ur1hU061QJGR12yYG4ouKE7QGG0Tpiz2aA2J9urca/kNvkjIVlTivSMREPGng3EdCoc9ppkJFekjyxKomYP73uPFF5fjzdTqCzv3pKMjeytPNxRam5CG1mm3bOda3OfxjgY0ZJk4wF3F9KgRWREF/lARS3hgNn2YIzAqszmZ7kxSj1a3aZTiVhQKn5AIMGUxE0n4anwgEO1EndyUUZmKWqFx8ZgqpqKhknK6pRKph4/u7qEZs96tZTh9Z6Z+GF0XRTxNLrH6xGVYOwNKfWBd8h7jSpJQnqSxUe7WiB7ysckew1AUVkSphkTBrIHhxFJHmYP+nx5Rh8D/AxxgWZonBKPNsX3rC5UFpBGhBL+OXkMMbxFchxaZsbXoeyoZDSqChIj+4JZZi5NV8wblWdUPwNOGwnZ4eAjJXpFPuBINs96UsJTfaMOrKcszhqfGVtNdct+AQ/2pngTtvmSi+3S6WFsu7Dfdhb7yPXlli+toHDbFm7Xz+ls6pmyGlffWJbCSU50PiNlpW+dl+vCL9sv+fXHwq9G8Kk4f+fBal9R/Ve4OS/Pz+xPK2Vd+fD+a5q/IE8VORWiNsJ3Fj2hkh1Wo0DfO+vE76vWdMolu3GKd5Byj9BDnhkkf7bXM7Jn07Lh+fo6VHiHcFs7jZ4qlVnFPMaVMZTz+Zk+rgSd6AvizkmVVrJVxrV0tr3jY0EW5Vkq/Xam8S3r+p7rdsO0stpnbq60bixV8NroYbQ+eDo90Vrj3//6a7bthtnC7k7VAlagvvBuPOFR2MeVq/4alcIohUvfqaUgdiVCUE4wLpx05eqDsTrDldqyotXJNsKYcG2T1yqF1jYWqfThnFmRIkRv3EbDlrQfZn8sEbzvRP8Otf8Q928JXTFdULsg6453x/sTItfsWQF0z1eTaYCWlfCCj0b2XemIeVbuWcBIY2iypLxsgOmJfd8pOvuXFO79YxS+iFZfG1YlXBJHCmUKbHC0NZj2z8d836blq8sOjFoj9d6DTsJPU9str60Dkud9Dbd1ZgzZBTGje5la9e/LB+/1S8kV5+diOoaq0FJXPHqqxF+LlGbrWTlURfnmKYfMiFTxo1RcEvBR1XzZUUhWWgrZAKzHvaJTj3YD9xbDc97KLIw6OH45iody8kfckLbOeoLPjLZg+0/oLqi/x+2GyUZ4Y9xSg6zVcZuVCJ4a/hBNeawnQOGz4ycT/RmR128lr/nkntmDX9H5Eg6Lhup7YGf0gdZnRv+W0EG1in8Eic9YNHp7Qm4fUV/55tNg+3xhk5X4f7h7s1jb1uy+6ze+Zs65mt2d9p5z+3urL8rlpmJjHJcqsWQVGIkg24kRDyAi+4UHSxBBkGhMpEQRIoKHCIglJCAIKbw4NpIFFmBHkcCxndjlW42rUs2te+uec+9pd7OaOefXDB7GXGvvU2WVoypLlLKko92cteaae661xje+//g3g1BTJZcNi3lL6W1n2M0bnFdmVYk64oaWPIxkX2mixRdqs+RiM1L7yLYvFDxOOu7XgVKgDsJYHWd1pK7nrIeeQKWfgk/W4ynbUQgaqPKAZha5t4gQN9x9/Q6Hx9d4Pd9FQyReK2RdIzLQNq1ZgznoveKcEiTRupaiSp5M6lxRvCgprgniaMVTCuRi/vY7awkhGnSp1li5CqSMCoR6gnOZ3l8wlpGgRoXEVaooOZnS3LlA248EPEl6xsHjYyTJGT401DySa8JFx9rNLEfWB1yxhiVF5Wi2oCbbjZ5t13TzGbKtnMQlF7KmUjkqS9Y+4wViNBvlLnRYMHzBpcFsljGRWaJlo4kYPSF3FN3QOyODaLV2U8tIdY5QIG96nBZGNWsCdCTEyKDmT9WFhnHI5OG7R+G/Jwq8SDAhh/bGcfcJyaNhkTqJhNxodqmquDCbCtMW5wplFGTCjFOqROcoNeN9h2LGTkYt6/Eyw3rzgei9DeSQSydGvsmC98r3O4aHdbuTISGmnN0XP63mHuh2HPVJPQns7He9iPmXe2OYPHstvnVyLoB4U2mWYhFU++J45TF7zrvIs2JVqex43DsWD7gr8JAVYM807N3x62WiaOqVx+9GFTrxw69oAAy3N9pE0boPGXHO73cJNk/cncsOFroyq6iCea1MUY3SQokGu+Qj1CUbOlYz+kIcVBMreW/cK6vgBovtBr472ufltZp45hPlUdUi9kqKNG1A+4RznpwjdbjATaEpXgaLbBttYc6bEa+OtJqRN8KFDiiCjI6kjjBkarH37ay9xuqiZxlaYmNBEMFFSh5o/JxBMvPrC8b1gBbhuD2hSM98HlmtwKH4MKMMa0JxBB/ZlA15AMnXeZTXnLo1KRW2Y2DpGyoD2UXUF9qcaXNHe5pxPkC35uJpz0M/YzmD2fwOoVvQMKP2hRIjsxCgZAu6z0JfhDZ6QpoCUJo5iEOdedLXWnFhcuiUaIPKNhoUUxVflRZHLpnRG8Mm57UxSdQM8HK2ZC4RoXENqRo9tQyVISuuMd1BnHes1lu8jyzyAeu85TAmHDPGYYtr50yeosTYEsbCWdlwtDhAUzJW0cWKebfgVEda9TSxQ3NDk20e4EMg5Qltkw2lZqI04OZUTTYLAKKKhZe7xIjSqKnlq5j+IZdC4yupZtvM1IZIZGSDoOQhmVmGa1itB8Puv3sSzfdGgdeaDO/V+3hdGkNmzEgp1LxFCIia+X8FfLCwDe8iqhHXFqRktE8mgR4T3tlk3Hl3OciSSM0bg2ZqRHzFlYaaAVV8dFOhvxqucTl021nRgpt41dYKyq7AB2/DM7LtJCZTM9MBBUTStGAEcsLCgg0reAZPf3bwO7FPrmDscAnVGzvlMhZuR4/c/Qxgvi2jRdupLTBVdL9rsEEZTPSfywGrGBSDd+hE2fJ7uwahmBwXN1kZCFBK3YuJnb9S2FVhFzCuFtgibrrOKiimpHWa8ESqT5O6tyOII/ce544mwYhH1pas5GYjuGxxfskUvTjzF/fYdddSbYjsHCVle0V3M5U6WfricXFus4jRYxPMQJAMOqd/VBhHx8NVpOYWTQuia+jXj2w3ELZIyFxrFgRpYOEpY8e6DhSNlLEyDpVZnHM+DDRqEncHTGeExsigmbBYEn1HKYUoldA0NGGg74UxJTa+YxYhJUePNwm8F+qmUNYH6OGco/GCzeBZhhmwZchKdcLFuEKDI42J1buAE8azz3N8+xoXw8Dy5IBr128xn89xPjOUAdd6sli8nrAliSBtRyng2dDEhpSKceudYywZcZGkw+TmaZQHXyrqHZua8FGIo1KLGhxZPX5sjbrrtqivoIlhvMBJtJbMrYkBNHe4ocO3A02TWfeeLl5Ys5eXRqdslpSSyCEzqhCqIjNhURqjWVYlBse8neOCp4yFUh0lB4qDEA8orlgAiXcIhaAdokbRDT6aarY4svQkVYrzUAaUyKgFxRGZIVLp2i0pDTbHSAUlcaE9B958eErNSPCMUolxQHXEu/a7rq3fEwVeFLx2uHJBdglXTfGm1YE0BnfgccE6MgkdoRQsyV6o/SRA8Q6dtl9pHIltQy0Y97Vect1rtYmg0eTMi6MUUyAKnokMAlzpivEWEXcFF9l17TvqogUtyJ7iuOtsRattN/fCH4hhwthVjdYrl1RMN9HmYDfAdNaRat0Lbb4Fn9mf0yWktBNYVYEg0VhIFCwb1LFzbHR71pBM0M5u53J1G3DpEbM7vsFPTF2wWsxYEHZ2wdRK1el1EXDOuBGi6XKOTJ0YJfZcRZUdd8K71haWOiKlUHiMajt5CmXD/qs3F8JqplyiOxbVlZ2Is/l2HY3tpCXv6Z2VatejQk4rs6UNB5StJw8dq23D2Cc2p4EhKefr6xwff4TV5h6LcE6nDcE7WhcojESNCJ4xDwwDBJQsim8bcB0Pnj6lkwhZOF54pA6EkMwTrbR4p8i4QWvmbDKQk4m33/hKHjvEw6ZcMJZK6JZsi6PvoQ8tbuGo4zkuRuZtR96saQrk6BHvmPlM3/e0sSPVQKGwPXtCLpGw6Bi2iUV7SONn9DIwny9Jg9DOJotnZsa8yaO9d2TG2IOPhUJB1LjnAKE6LO1m4seXQnSOJlmq1cY7c71Mk0FYq5CFoB05DdNutbWX0ivBLfElQRgo40DNgShzjprKUM5YeMdq3OJComSDP5NzlCKEYpBJH5XoFIkCTqz5C57SwJF2BGkYZU3O1ixFH5DS4BXoiokwpSUNG4oqMcxxzuO0MuaEF0fnA5KE4kZyNUPDUgohL7hwK5w3o8SjZsF2a0yb4DxejPG0w3rz+N0Lnb4nCrxR+lYQrMNzKUJYGx80zayS9j1OTR2oaYu4YMO0qhBBYkC2NrAtWfFtoeadkjSZb3qNOD+bbFgnT5tqHYalOlmMn8NRa5oohBO0EiabXN1h4Ze0yipyKewJFjFWS546RDFJfU7mfDDVzDINdaVWwyXFuOE23PTsuXXPsHbYe6CbV7bbY+JXYRLAPNjZ4fWm+lQxwZK5E+quwtpAdirQog7nodSMFpvk11ytcGuZRDiXEJF6zO+lVjzVWC9hwuBl0ipMg9jqjH1hXnAB1O13KzDtdmSyBsYsG9wmwRgYyoxx6OjaGVKNGukdUM8ZHHSMkOcgBZVMzZEcE60qFKO9NXRoWpHF47NHZ0IY19AD28B7DxObYcaQMtoHmnCd06po9Mh7Hm1bbj93E9cu2OhXcW2Hbw7w7QuM5+9Rhg33+opvtgQ/UNOS2h+wah2L5RFpfEpshaWfMgVqIY0jXhx1XdmUFaTCZhCG0DOPyvHBdXwoPM49rnek7UPoTujLAVsS/UVPdjMeaSbOO9p+pPiG/qJwwyVWjXJOomtb/ChIieTOozEQaqZzlRxmcHHK194458bRAc/fWTIEpcoLOE10rc1pYoiIjKgYDh59Y+ZfLuMmY7Ukijoz4RrUdrfeOIUEHym1MjpzGmrFkxH73BeD9KqDNFyAKq1CEaEykJLF2CUdiDI3GwW/RtSzKgnvZmxzoU4KXGJiLA3HRfACG4FZo6j3tDkSvKeIQiNQHDOZM/qe6hJBZwyixCmn1je2mw1ayQG2ZaBtW8owEEIlFyZlb8F7YZNWZq6mHk+gqrmZEtZ0OVAlIK7hbFMJ0bOuW+Pk5w5PNA2FCM0/P0NWJUVPyBXfZ3IEF4/MhnP7CMmVQjQ1oir4Kd3JBdvWi0dLYpfz6cNohRfM42bqsEOw8FuKQSsmSBmQyQOeK/ityGQwNuWI1roLungWSrFFQ40JpIWawbliXG1k+rorxEzmX4poolQB2in9x+yLd+yXS5GVcfl3hXA3K/C7Dn9nqwD7c9Np/dnRFJ2TPbYOTKpJw9Z3tEozV8MybYth7zItJubgqcZgkl1vbB2/VLX7wyWrqbiJQ7+7Xh7chLGLaQBqzSaEmRKjdtRJcS0iGWpj+Proyec9ORe2acTHY4JrGPTrzAOMztPllo3PzFxPLxALeB/I/hDiDRi+QUsgqSPWY2I5Brasz46o7wY+ex8yLbUu+dgHjjkv99g+fYvt+gxXW5aHhSfLiI4P+dL9L7OQFWdPBMKWeTpEFu/w/HPH3Ow66uyYJ6uHrPM5XQTVNe0IaXvG8vAF5Po18uYelZFN7xi2gdxvGdbCWS/0ZfKwbyLrmHmwecKd2ZJ5t2BdE3V5m3Ta20JZhLEENn0mbB1DrGxG81log7Ce4Mgwthwt5nQ+MQsd98/OCeOIazpWeQQXqd2IlMimL7zzxprF88LxjYFw/RbdoqX0gWbmGDTTtQ1IJFXBhwAFkma885OfEya9F9OjWC6yUAsEhSpx0h1ki/aLFix+6bYRQQuDjqSaqKkgXmnqHA1LShkZ0hNUj1nLBXM/kLYzui7yOK8ZfEOjjhCUVVYaFyYGkrCs0AeziOhQSj/ArEPJ+BjNxsFtQIUQG8Q7qgyTq2zDmBIhRpishrMvjCWbpUQUtqNRtmO2LIB+2CBOybnHqbCdspZjKZTWEcdA4zwef+nQmRTnIf8pVOfviQIPQIHiPHGKraoYU0Qm1z12MnPBaHEY9GG2sIJoBBlBMkUdropJ3J2b8H1VvqYAACAASURBVHA/Kf6cvUCaJuHRNylAmcI0sPAJKz5TYd5v+S+HrXabxEXsREpX7AnsjkyrxTTYA6rJssvEcqmTi+T++GrH28ExFgpi8I6bCuV+O8BUuif3Ry3lCu59efvmAe5ukQL2bpJGNZ0M3QT27loTG2aPDIlaBu3uyhn9hxAj5arJmZ3wfsMjE83F/saJwvhMssE085i6e4oj+Ia8LXhZMmwdya3x/pDt+DYz71D3gNkUNxeW4PQIoRLy87B+B02JPDtCs4fuefRiwHFIWL3OU/kqH/nBD3LR/Yusv/Z3GV/4KX7/N/8+w6MzToLSb5+if+Q46ITqKzcODhjqyJ0bd+jqAe2tIxZty7XlnHLgcGnN7EQI4Q4PT1es3TnD0yXz5pCcV2z6ns16TdEN47Zh2AyghfUFlKrMQ0s3D7hlQLJZYffR3vfDZuD8dECaJU/OBlwTKWREAlEaSoxQE7F6iqvQOcYsHLSBWaOksbKqhRotwfWirPFtQ1ccbe9Zt1BL4uz8Aaer+3T+dbbNkuYIlmFBrZl5jNSkttmukLOpTot3hEmrUUpBvHntAORqnbUTMY/8iWYbvCeEwGDYKMVbI5GyDVFdmOFyCy5RMkhYU3rTx0iMiNsyqy3bmljOI5vNyLLzDGULpaXJinNzUim0FIJrbe6mPV49sV2CCH4ybmvcAf22ELtArlYLxiHbvF8cfeqZzWaklM0RNkTmxdTK2zTiCMQYybUQYySlZAlZYlTh9TAQqTiFGGeE5BnrymgKqmgxnYSGaonS/9ywaBSaTaIezsiuGpVoPMd528IpieBGlMW07Y8GlUx4uln9JmoJpoSUBK5QNEymUZXgjBddat7zzlXBe/Pf3hs+OTX4RhuDdXRKqK82DMWZqAhMmm9hEtaFOsGK7uRwqf7SPwYmtLsyeaK0+27fSNZ6pQA79pTC6RaCnd/e/lj1kvkz8clrqZMIRZ4dyFKQacB6dQeyX3B2p631chHbUzcnVs4OMtqtsaIm398NcqcdSslXzNa8MWjq5La432GwS4naDZndXqFqgSIGXUmtMCRSL5yveiQm5m1irFsWRXGyhuuvsx3+DNG/QPf8p+DJb8D6i/Qv/hTdvc+Q5jeJN99PfOtt8myB1hZtBlK7pJV/yp3X/hznT+6xPP2fufmhH6ThAX/2+ob4k7/E4ZPHlLuv8fTNX+EwPWV2+3X6/Hma5kOsv/EmuVzgZ8oiZfRmsjBodwTjCSU95Ln5Ajfc5HPrd3jjrTf40N07XAuRky6gruPBtsdN/umlKbjWM5vPWT1a42tris2ZUvpTngyOwILVdktJiRJbNn3i8aYQmxnp0DOOAx2e7mBOr0r/9IIsI6E94N0nK2at6TU0Q+cbjto5Y7XnLeo5mvesNhk59eiB8vXPf5712X1i84Ms7x5TikLbEoIjUQghEsQzjNt9wIn33hSagFbIogQfUK+kNNlnOwsRF+epTnBqBZNiVF/PFNOoI6ZE9kgs5LGAtIhuOHBLcp+pweFrR0EZy4ZQbqB5MBoiFXwmkHGhY1SHSoeTkVQL0mdiaO36l4E4KLNmSS6ZEBNQje3ExAgTR54+pzlnHMpWwp4O7Kvgi9Dg2JQ1WqcFbcox7nBs1WxRct6SfE8cw2RbXu1vrQWhMW+nfBlo853evqsCLyJvAhdYNcqq+gkRuQb8PeAV4E3gL6rq029/JEWHhGxNvm2qSXP1M6Op3bA1sePUWrHKNtisHmqLaI9g/HdqQJyFdjinaJKpoJunt5Zi7JVqhU+8FVqddgN2/HrZsaqzABC1AurwE3umGl9yGjbu4JVdERYmqXnVKTDa4ZwpMF3wU0F7htP4TGcO7Nk7ohhdEqbcVfv/S0vgHSNl97irxdyu81V+/74Zl8lkajfknaT4qnUaQJe9EvaqZ43zJnDa0R93xxIR6+pQo6vGxhakHRVy4s5/SyaZOpPeMw1KK0ipPHmayfTktGIhB+RxzTfOHvP63R8lXf8ws+ZV8vgyOb1DPXgNug8gs4+xXX6N2dljhvQQV05BjpHjT+FO/wEaZ+jtj1H6+ywOr+NvvUwOT3DhFeTVFzl58cfpz34b985vcPv4Nmn9RWp5jm4z4+L09xjrEUeukucLXJmRHz1F4obcf5lxnTg9b+HwmEbO2V6ccTy7w2w2p7Khiy1n5wPH1+ccXj+kaCbVQrdY8uRsA3pAUmXcVJYxEiI8qlv6nJjFJY/7DdocEWNHCpnz9YY4HhAWR9Tcsz19whg8bhY4qEZpPGgPGCTRjyMaA2dpwNWR0AQKmcO0YJ0FOs95zsy1Q2phOKtsL96lH67TtYfUmhkSZimgpo8IbWDn71Rkoss6c89vnDfefLbEMgl+EgXae9gsK7ACB/iq1FpMHOUmXyEg4MmlpQmRtfasxy1tM8PnkVmj9P2a2fyY1fCUrvPkYjt/X5UkarBf6vFSaQg2wPeTO2nqcQ6qGyiSyU5pJVC0mGamiTQ+ID7Q9z3ee9q2pVKJRRj6ntBF1ilRXKFSCNOMrRQTUA15oAszxnKKqhK0gY3Hzwt5rFPinMPHhjFtAGjb7w0WzZ9T1UdXfv6rwP+lqn9TRP7q9PN/+G2PoIKkAJuKu3FMHRKyGSZYw+GDI2dH8N0l5KHVirSCSKXmMm3ud/4zI+KtwJINCnG+TnS/gLgp2b6am6QJRp2l+7DFlJdCqVYwfYggGZf1kvURWhgHxHvE29bUFcU1U4JUrdQyGh7txLxXRFHs/obFGzURrOMRF8jZ3kRWSK908tPU3yT/xrO183R7DH4y5JnWiKmAVrfvyO1y2++riPnqsMPlJ65/mZgnIpMKFmwwMR3R2dxEq04zA9kXcB8t5MLJFAWoStnZA9dpIZgwWJEpBnAaegt2XFfsQ0cK9H1LDg7djmzP/4hw1HHz5Ee4/Zd+ifi4Zbw+wIPPoc1T3NGryFd/n3j4CdLT3yG0L1DmS5w7QJYvos5Ru/u4xV3ypiCbt3HNIRpbfA6s3zxn+ZHA4Ud+hvL412nuKqvT91Fe/SRt81fY/pP/BjnwdIefJn/21wnv+wt88bf/NneP7/LowRrtz6mjEtqOOt5nfnFBD9yugZNeeeeNd7n+guJmc4TA8dGchR7w7uo9/KCsL7Y0F1vWW1ilnudvXifpBTFGjt2MvjrWGgk+8nY/0o+J1diypdA2mbBdUXOLjwsWbWFLoV+NlOhYNh1bZ2ZZPUYmWKZCFxpSUi66FV0J+BqJThgvNqSDBXFUvvzFP+TGzTvUoHQcGeRZFNFMCeYtFLJCKJRiDpokG5iqTtoNh4kUk1BLTwyBQQWngb4ILjqqjraT9dE+U8WyXL1mi4+MkVU6p+SRZYzkNNpMKMyhhTxuWLQdm3GgykgjM2I9wJVC4zLVFYpWsofqEq0zzrzEFhFljDYfa/sKs0ozNTOpFCMgDCu66KgaqDKimti6ltBFEhmhELOyRaFx5CETNVHGLV4axrpGdUEoAxVPbUZcMs3BfD5nm3soSsn2uczle5NF868Bn5q+/x+B3+JPKvBYmn0dwaUMZWfhu/tfN0HlE6Wu7krOZBIlvSXelzx1z+Z7su9RRfBSL4eItk2YOs+CaJygg2S4obYoyeYAYvfLaZgCQ+w8VCs1jwbxaKFWYWdjWnfKVfE7SPmyk3aGtVoyFFZ0VfeZplAIMZqB2RQWYcf8VjxO92EkelngmY65S4/Sie++t7+dLJJhCi7X/e+v7iQu8ftLcdAl/D6pWXfFff94qCViGlg1PYATxDXTumPWDLuh7jQanqibMl2jAqGlblrQLTFkZH1ITYVXfvDvEt9/QnvcUPIthsU38CkC58TVaAXpzr/CWD6LDkfU1YZWb6PtDF19HT97njrOyPKA7enbNOEaPsxpYsuTh085E88iBZqSqPIiuduwfGHE3f9V9Lmfprn+IvF0Szl+Ho1njMPXeN9Hn0fXr3H7Y69T/uBXkYPnOH/8iCY0XJwKkZ7sMqdUzr5yQbs94aE+5CIpzc0P80hOOdo8Qhe3OJENa5fRbWGph4SZ0sicIkJsCjXNmSv0AQ6H0Qy0qrJsWh6f9RQRFksP4lltA352yPF8S66J7Zg5CZHGL7hIW/oCKQgzLUhIHLg5eBhrJoSGg/mCNlZkveLl2zeoh47a1j3sF6OVDvP2cRDNvtc13lSswaHeds3BRbRa86OSJ2tpo7bWbGKmmjMz76lqswbfOIIKc/WstEeiJ2+2Fv7hHbkmshqhdtj0NCoEbZDsaZ0ycx2bfkttTNA4Os8oM+aTHmJMieoFdKQFmhBxGYpafliqCUeYGHQNpUIbjLU35MKsE8ZRad0Oriz7DOO28chFJbuGUyk4J4TiwA1QPbkmJIupprXQHG453z7B+ROCBJtvqF75xH3nt++2wCvwG2IUlL+jqr8M3FbV+wCqel9Ebv1xDxSRXwB+AeClWyAl451CyQYV7PJMfTOxNgpV04RBu8kjxk+hvnHq6i0JxSiH2Zg0UtkFH+eJNumMJmPpOWWGo9ibTSPFKSqTxawWEItjFgdVHVUKUqtxrjEXO6ll3+WWSalai9rAVSzFyVXzoXZqW1eZ2mHzbClTR2uLQ82GZdZyKbjyEzSz8yO3Yazfs18s6/Wq7/uzP1fdG8/sO3mZmDT7UbGAn+7vlClb9opBmL2oO3b/M7uE/QIhZtVbpwBlrUoZTAQjzlPrzhfdT1TPycVyOn4dI75L+MZT3nxC+IH/npsfPOPu63+Jkh7j/UgqN4jDBdv127S6AW4xLr9C4IicHiNnT/EO9OWPwL23COf/lNLPGFql2Tzi/PQRR/NXWcUNh8e32D495+ToGge3X8WNj5DFEX5+i9XX/wmH8xfR5SfN1K538Kl/j8e/9l9z/af/Gl/6736eo/lLnPzQz/HOb/1PXP/gT7D98mfYnra81f4MXfqH+IXyR1/4HKvthlt338+Xy0M2py3na8enf+bv8MMfXvLu+X3+h//o57hz8A7Xj1ueP2zwR+e0ecnB8W0uzp5CChQS1R1zPWY4Tix85uGysAGOfLYwDh3Ig6ORjtj3bJqCrzDrFkhWVv3A2JlR2TVn9ruBAzLJLG2dzWwuhp7SdPjW4Z9/iSM5JmqHwwajuZi/TBMitVQbuqPoWPaDdc02+6lq9N/qwA8BCVuGMeP1CNEBDRXJyphHNAhlSm3KsiIFR6cBVz3eOy5yQqJHkhkRRnEoA0mm91vK5GFDzQc4aalJca0wb5UyrJHlCSVVlsulYfXedsVFKkOtOBXmTWuGa2p+TZ1vGPqewoLYFqIzLYmQSV4ImBtojJFYlLKtDC7jFI7ETPucKwypxZWR4sHLyFgz4gvjGKi0aN3YDsh3QCXX7x6Dd3/yXb7t7cdU9QeBfxn4d0Xkk/+sD1TVX1bVT6jqJ24eYRF9U9CGNEaTAia2yG4lMyqhaJm6xDp1z6MxYiQjrthX8bjgrwwup8i+qas1e1sBt52GewXcFnGTJ7bu/GPM9KxWo6YLlgXqJ0HU5R90Zdg4KTt3zxPclKk5BSo/UzDtWZ65NjvaoPPmHHj1n7kZX+Lo+69XdM3fzKCx743jrlX2/2p5FgPfwTW7x1792bruyz/YXRma7jt8EaS2yBQgrWSQQohuiqLbZb8KJV/uTPa7E3VIB2yWjG8/xn/g32b96g8RlpFB7jDyHDx5j6KPyOe/ifdfI2/uweCoeYETiP6QePfjyPyjuP5rbNbvsPIt49334eOIe7xiHl9Hj27ShEN8MiZR7TrK7Ih1vAXuFoTI0exFNk3PSGbY3CeF5ylfeYPKAf3vf47QvEx8+cN0H70J1x0PLp7QXr/Nl1zmK196g1s//BeQo09yY/EKEjryoeML74zQD3zstUzzhU/y+I3/lKNmxY9+Yg5FuHdauL/esi03ycs5oyusY8P53LH2gfPNY8bjQHTCcul46WDBnRi41d2iTS2phzyMDHrKRh/QoIx+xNUtKfa4eWKRRqI4HpaBLYW1u6CoEEKDE08aRjrfEqtj6ZRGEsULTF7/u2HpDn4RETPg8x4XGoNYQqDFEdUKoFNACzX2iF4juCPUrwkhUDcDMStt9bhs5mIzCfgCvjqSQhJlEwqRhKuJQqG4yqCZmVvic0MdPLG0dLKkOGhjQxCliLLZjMxlSeyVWJT+fEUIgVIGfIyMtdA0DU3TMIwjsUSqBOLBgj6vWc49ocmMY0IIbIf3aDuh0RHSCnRg1J5V6RkCqFMGyayksCWT8kAIkV2mc8oWIdgnBV9QrcTa0Gqglg219IZkfJc3+WYq3Xd8IJFfAlbAzwOfmrr3O8BvqeoHv91jP/F+0d/+25EQlXG5IBzMTc5bK17jVNgyToPVMW/F3fzTDe/bDexEzFhIBLRUqmZcLZM82TBrrWKDW5jYMQklgYZpa5TxbraHMuDSwWXnNQ9W7BE7ZsXhJmxz18+betPtoRucN5Nev7Mi2OHs9ZlOWUX3TBTn/LOFvO6KbtmzZ77ltp9c1f0uZ2f/slt8rt7VBsPPDnF38qm6EyGpTk8t+w6/1mKd+JXTUM04Z5mwuVhXBSbDRnQPm9ncYLcwWLRCUaE0H6Qtr8GH/zr9o9+gaZak2YLw9f8H195lDIHm8Hny+dvE2fsofY88XaMnnrHeZ+Zuw+qrMCj6+JS0nDEmZXntFbbpMXU4YvH+1yhf/iK9Psa3M7rrr7N6ep9lasiv/zjp67+DCwPN0UuM9z+PHp4QKpzW+yweHzE7GTh7csomO7pZy6PTr+LlLq5JjKt7PHrQEniblz/+01x/3wlPf/0/Y370A/Rnn+WgS1PK0j1myxOKu6BtF/TbgflxBzeeZzh/xIOH9/nDz2w4u59Rd0IpkXZm70sXl1SvtI3j7dWGvgT6i8hQKxfDmugSpVRWyRqUjSp5U4ntAagways19xw0nnXOpGqOnGHuKFIRLcxD4OjohNwNfODjH+eF1z7EfHYI1YyxygRN7IbuGqeYOh8vaZKpXL7GkzivhoCm0WiwBaoUhjwg0pFSRUiUkhhzwWlEx0JxSp/WuFrImF1DzhNTbaykBsaxR1NGq8d3gX4caX2gjIkYWygNjUSGWaFk8xOKrqWbCTlXpFSieHwMDGI0RpxSKXRtoGy3oIGqI85Hs9kIns3oWYbAOA6UYmZsWSuuDIxjj4sNZ5sV2thnN6cNgRYlM0sdtTkjp5YxKU0DDY5VTbYo1sIv/pt//x+r6ie+Xf38drfvGKIRkQXgVPVi+v4ngb8G/BrwbwF/c/r6q/8MB8NrNZpUzrjJbzrXijobGlItXEP8LlgjTwXGIZpMXTlR9RAQ1+4tCUQE2SfiyMQVB3BYDFww8yXJuOLQ2pBSxjnwwQpbnZgksqOC6wTBTJbBuyxV1YwZm+XLHcTuOZnCvCc4ZWc+9s035yw0YC+m0p117+StDogEduHe+wVg9/ip4O4671r1SniJAV570dQVFalMQrHdY2UawtaJ/umm9eUZI7O97GliTxB2IUw2F2E0LxG94lq5+97FyWI5E5oOjyPc+Tl0Oyf1C2RxAx5/jja/j/Fgjk8LAjBulNi8wCgDsZtTblTCk68S44J69ogiM7RsYHmARo9vBsaH52z1Hp17l6d/8DU4PGGePeU8sXZvUYoyzlqa86+TDgLtdgnH1yiPbtAV6POak2HG6mDAlyPKScAPSq2PmTXvZ7F0VtCeOyY2Hbfnt8nHS/p7SvP8p1n39ygnH+PR+g85bM6YXfs+cv+IbnkbITCL30Cvv8A2DzQHc148fIXby1M+80W49yTy1v1z4sYxr1ucbjm8s2D0kYNQ6YaRs65w2CzJ64FaRoZecQdz6irRDT1jA61fs9WM1MjR0QFn66d4qbQdqG8odUvTeELT4J2Q3ZoXn7vNwck1utZR0prZbEGptpvzPux3X356L2tvmLdOBVOdY8iJ7BT1AozUXAgiBG+5Ba0LDMVmZAFPLYNFBg6V3hV0VHwNbFRYVEeuGPSKor4QkilDBz9Fdybokll8OLdAgBJGBj8Sh8Bi1jAMSkoFcQZ/1ugZSiWqOamuhzVRDI5a9Rua4GmbiiIE19GnC8iOxlfGzTlKIjaBIZ2RVMlS8Z3AMNDGwJh7vC5wUZHsGGQk1Q0yHuO4IHpl6EGbhuotHY78/y8P/jbwK1M3GID/RVX/dxH5XeB/FZG/DLwF/OyfeKRpMKq5ElDzheki4gIyJqM+yo79USdoQfdQigkqrnC88cZV38XcqceUVHHipNapCd/5x2Rqschnq0xmWmbbKW9vPNl16ZVSbZhS1EzNqo57XBtzebFFqKopYJ0zfnjw03DXo243sHp2SClimKCf/Dy0TIvADisX68J3NEfzetmh4pZitXeJnF7eEI0lVCe8HnYY+eV9LUHJFoRLx0XbKe0WKdjJwnbX2lH2f8G0gEwL4L67n/juRdXOpnpqNOWx1AC+J5TC9vZfIdYX8FsH127R+AvSgwtc80Hq4gi9+DDoFtfdItx/B6Kn3vpRxAf04NAMpTYP6Ltjtk+/QNc40mPHmXvK+btf5eSF1/CLD3Ceem6+8nEefP7XeGu94sVXfoyQC66s2NZCiJk2dvDqbfLZ48lAq2OzeY8yu0mzfpNNv6LOT2CTGf0Jm1nD5t03OXnpJeKw5toLI7m+QNnco2tPqDduM++O0bKg6d/H5uJt/Es/RHr038JLvwiP/kvUfxiXz4kzJWwDxC16Ihy/fIfhpZd4+JHnSDlzLp5yccHF+XuUcYWuz0huRdc0jGXDrWsVLYc8eWyY8OnyKZvYME+J6sGtHQ+2W87P1tyYzyG0aHPEOJxRxSO1ofEdiUzTOvoB2sPWVNfBMeYeiZEqLcNYaIOChyxC8J6s5gjfhGAce8EammyfrSxKbALkRE3tJGRMtMGz0jXZV4pALMoQPUMeCU0mD4UjJ5xVC+to1SjGq5RQtoR6RObMwrGdJ88SuWZqyrRTalXU1uyP1yMhtiwaU+DWGgkThOkm4sLcObIvpDwyCzPEC5uabWCcMr5Y9GB15vce6FhvNtRg1M6GSnSRXM+o3tPpIS6tWZWRSjJbEomI21CymHFbFLblgliC+cfrdw/RfMcFXlW/Cnz8j/n9Y+AnvoPjTbbgHh8Ceb5EtwNsq3l1SrDucxrmPNu52jDU/NWVWke8b9GJ1151gFoQl69A1ZZqpMVPOLnJ6Usy++CrxQ91UyH15qsiE/4uBr+4iWe+O7adm8NpneCLq+EdDvGXvPG9edc+bNrMr2zrMdn1VtnvUESEMnXuTrxN/Sc7Ba1M1+jymgJ7QdQ3bxbsPK8ob0WvLFSX99m/PvvvgUkQ5cVNBmE8e80Ape6hIa8WxVcFam4I6ijeI+WIYf4BwvGfJ2zOyDwlDIHkvow/fI5S17A+x2mBYSC7Ne6kUB5U2qCkINQakMcXJLlJPniNuPoKD9+9z3zecP+tN3nx6HkuHj2mvPMOf/T5z/LRT3wf7917A9fc4Pfe+Qw35yfMFy1Hd2/w+Ol1Xjl+PzlHQhpJMfPOo8/Thdu0BwfQ3GJ4/BW69iVGnnK9uUMeBvydE2qMbM8GDl+4RjprOT29x/zm9zO+u2aeZ7jrCt0RYfUAdXO6xccZD+5S79/Bn7yMe/w7wJqN7/HZodtbbOtznG3nbKvScMDoPH6xQNqWoIWVPmCMj6nrU0KNyJgYstLMIk42NP2ImwVyDeTasmxH+sEjOIqDw4MFjzenxE5ZHswY+4HteEazaEniGeuaVueUUlDvCVjAjbiC+kquihdPwFKrXBXjlE+5qjlnU3eKKVC9zNG6JRfwmibtRkCzIq5FtRCCknKmpExwFgxu6KTlK9SqaLV4QEmFsQ00Zc1cI7XZkKRgCW3QNA1pTHRxRsGU5vPZjOQdfarUKpbI5ZSaMh6lc0bsCCFACPZ3OQe1p6mt5RKEAl6R3hG7yJgHaCO1DEQPQ4GLVc/x/JjtsKGUDZ0DN4KEAC7TC3Qpgq4RV0i5EJqIaDX75Pq9SZP8jm4em7yXvqJtxo86kdeF6vyU+FOmcOqpo1RwXHbvtez43oWUt5NFLxjp3O3ACXb+31ZcxwkKSWgpE2fdTaEe1sUjxdR1ueCDn7BxB7UimkBsqChTkd+FOZu5mEzK1Yj303nuPGDqrgCqQVA7CGPnSqa6L6ayq9tyKTaqtVxSLK/QqnTfqU83a/Wfud4ifoqe2w2GdU+5vHzYJb/+2cHwswPdvf5AHVeXLia8f5RKI2HKvywEN4AEhv6A2Z0fQW7968Tzr6CH78Ofr9G5J+o54+qc2BwgZUNtP4COX0BCg9+8gLt1hq4jIl9G/R3aazd58OCMbvUHjOGI9975R/zIX/xbfPHrf53xxbv83//Ff8L3f/InGPov8oGf/V3e/Rs/zN3DC+TGkjK8Szu/TTl/i0X7Fg8f/B4Hfcu99+6RhpdhDDTf/y/Rv7elXdxB0ylDviANa8ZFQZ9s6ObP09SBi+Ymw2NF0paXrn0ffjgnHc05S4nDeozfOuRD3wfn75J5hebRO+RrP4/rAyV8Bjk9Yh6eUupDHvSFN8cTzuSEMJuzWp8SfUQKyGxOHrcsX71LkFepYyL3a/qLM84efoMubwl6He8apF/RXFtQzjKjbrnzXItWT/CCzDLLdqQL1/Gh4sKISKRsB05efR/PvfY8xQtd8PhRcF026KA0RGcOqx7zOa/OE6Oj5ERsDX6LsaXve8SbMCjkkVILjZtTAox5i4YVUSN1mD7bI9RJ3enE4yQjnWccL5hpayKwnOlcMFriGFiTEDJVGmoqNDjmsbVs4BCncBsb/btcmHtHdUIvQjdBl9I2FK34rqGMI01WxpIpscFTcbmh94KWc5rZHJcd6myHO6QpaC8LLnqWWQhuzWo4RZqGWmFTN+TsGSsspEGHLTWAd/Z8QSLDujcxpsKfRnn+blk0f2o340IrmguuFiT1oAO1bs0d68bYzgAAIABJREFUUMZnuskdm2RXqOvkU26FyDpa53kmpMOSmu2NIGREjLboJtql9xGqUIb8zGCz1rq33t39vtaJV+8u2TLfUljhkrOvlZzN5tYj+zAM9JsLryCkaaWIU7fE/jmverzs/jlMUbo7X4c88/875hBc2WVwOejdHe+bz2MP5v9xr5c+29kbjFQu/2Z1xorBdAWI6dDFNZQaGFgwf/kX0YOfQOQ2QyOk07cQf4SoJ6824C9sBsEBmv4AjSd4EcaDLXk9kA8KbrvFSWLMI5/5R79JK0qqnm+88xVq84T+/BFVB3S+oBxc4+joo/y9//xnOSs/QFcj33jjH/N7X7jLH331Szy4eJOL/hscBU+QQ26++jrXX0qswhuMX/qHfPGN/4PaLnF+TlBHd/sudLdo795ieeOI4hzHy+uMaU2fRrZuoFBZlBssaCiaqbNC2CbKLBJe/TSJBe74VerTP8Bt/jzl4M9C3rIuLX1akhXO05Z68ZTDJtAGZS4t6804xSgq45hQybBc0j13i+de+QDt8nnOnGPpIyEuaL3CtYKfz2iio104qrcQ8Zm7Bm7LZjhlcWOBO2xwM8+NF++y7G7QRvP572MyAkFwiIdEpioUjCnjFUJVgirkS0l/05infwhm0YDzjHWDlJFIQBSKVBpfkDqAW+PYUjmj6IpCZhwzrnSsdSD5jM4c6zKS3JzkN8zcaOy2uiX63gRNtZAsi4xNSRAtQEZdYFum1AbXIGK0zForIQS2wwDVHue9pyn21p7HOa2bsfBHbIeeTYXRJdbbFbFMLKCmZZMzGy+MvsO3c3Ss1E2LSzMa52kdbIcLYtuQq2PMxt5LeaTxHVJGAgX/p0CT/B7p4KdC4hw6JHQYGE+WRAK6qfjQUXIBHadit/NqmZzrtOBimDJWE2hBJE4862n4400qza6jZgKLZcK2dcLmVacV1JsCdAeTSwQ3olr20IQNCgNlsg2wQjopY9Ww/+AEXEGpE9d9CtRw8bLrVgUfEC3UXPZmZrvbLt/VTTsXLRbuLBOOLs4ZL5kd+wZA8bgJIrqKlDOpd2UKR67TOdgf4PY+9exFVlfdNncQjPOBnLINfrXuBVN2KeXK8zmiGJZfZQ5cwOxfoD35NFz/caS+h168Q3AtOVTYXlDmQognUISUGqSd0bwL9fqcOj5EfUOTsjmInhzy+K3Pc33WcvvaDTbNIa6e8YlP/ztcfP0rvPqRj1KfnPGv/hv/Pi994if5lf/t/+Rz/++v8h/8rb/Bf/Uf/2V+9qd+getv/y7b4ZDV0yO0H7jfZ9rhIWO7YK4L7tx4gRJm3Ll1ja/8g1+mLo65PzT86I/9DBdPP4vPPevumLZ5lZV7xM3nfggdvk6IDYOb48dHcHxMN2SYHTK4d2mHQ3jvAe7gOdyjL6Kv/Rlk85h6+oeM44biA+ngU7z34D1iN8cFYZO2zDcHbIOwbA9w24EhQJoV4tYShgiVeNwyP36Nw/whnpy/zcmDe2y3G27khrUObDTTSiT2a1oS7sjh1TOPJyxu3uKlazeYh0B38xbqGzS3+CaixURbroix2rzDOXBaKdJQBCoFFUcKM2LqCQ40Z5rGMQwDmtUSuTykEiklEdyMIWcYMrPg2QKtevpqBTbXtSUc1UnoNFZqLUQnFFmhuWPQNR2erQ8kKq4axz1WM+ebu5ZxW5DW06ceaYRZaNiURGrmaMpTwlsldi3nw4ZrvqOMidE5gkApa0KtrIKn8QtIa5zagpeisFn3LLBwkTKxgbQxFptrn6KrAk3LTD1OIl1f2XqbjZVaIHr+P+7ePMjW/Kzv+/yWdzl779339t3vzNyZ0YykWZGQEGiEBBJYIDDEhgAplYsEyg4ELxWDK7HjgEmZCgkKOI4J2DimQMHIKimgZYRAYqSRZkYzc2e7d+bO3fouvZ7TZ3vX35I/3vf0vYOdVMqiKiqdf7ru9OnT7+nueX7P+32+z+eb5FOUKUBKsm8aiUZ4cFV+plIe4SzKWKwARQR5WXWlM6ufv9X9egdWeLSo7I3OFxV1UVIdCqJirEirKhnC11q9o1qLVwpbOkSd9Yosa+2+IkqKetAoZ9q2DyoLoQdwVZyZCCpvtzco6UAEB2hhX3+ZULJa76dCHbi6g54NbKv3VPErPKYqpDNtu+6yXY0N8N6jhcRrVd35uCq9yM06ek+1lj+TVkTFzZEzyUX5WmLi4G7H1j/TKnO7PuxEtYRUBYWImmJZy0NldUc18+p7UfFILGXNwa5+LwDCFRQ6IiwSCtlBLP0YMlwiL/cRviQSKRAhfEAuSsJkG5t5ykaHuBxhTYgNPSJOMTsDZNShbDWRg3NkkaQdxYQ+RUdNgsAShxHR4jLBZMKRYw+S2ymTkWWaBaw2It77oZ+gu3A/r11b5d7HPszPf8+vo5c7uHiJ77p7SoMWuc0RYUYqLdYoCEcEhaGxNKQsxrz9fX+fvfOvsVO+zKnoGFk6JlpO6RqHiRZBjyEvUUED73tIGVKIfYRqooeQrR5FuQyd9qHVhEKSl5Ds5sRmnjS6Bx1ITCQwqiQs59DKkzYLpPOMrUFGIJ1HZ5KuFjgVMsozZCNm4nNcmLIyv0IaSNgfkG1vEbUCmrqFEZ6hSLECOqaDbEsKOUEGJZ2lORABOpqjyGsZwzm8KnGZo6GhCJqELscJifESpRwxAmGgFA7JFCcEpQAhY4p8SozGRjFlPkYA2gusySmlJZAxRRBTugIpDXmZEnuLMGHVsXuPswIhCwJrcU5RlBKrQgJCPJJceoRVBFZg0KQOmkJiJNjSQ9RE+xLZCjEWSlcSqxiTpKhA42tkuUwdPR1TliU6gLIYE6kuia5cQ22tKdOkSlaONRQFwjqiyGKzAG1CCqY0tGY3zyCSdFNF0o4oJjuIoI2IFPuuxGdjQtGDUjFxGVEc4qQF65Aq/rpL6zdEgfe+cr44X5HufO7xyqDjCFcqfGaroaWp0lGq7ttRRevJatPUVXY+KUK8twdboJ4CR4QQATLWYF1VnKhJh6bKehW42rY4GxLWfJSablg5dyo8cbVFWv0br/HS1vJM9Qdmpam3S2vPifcIK2pBbKZY1+fFbYtYBz74WUc9e45zBy6WSpIXVPcd1fe9RXSspRhPnR41k1BuWSaFEDhZy06mOlBmB4cQAl8jiasbnNm0trpzcNa/QdR7A26YypJaBT/M/lsMsoQiJiSGuccID/8XkNzEhIuosk8wHWKCHN89BsNtIt3AZiVqOsVLi8n66KiBMSVm8wYhinTqCaVFtDSNUYFpN8BqxvmUWK6T6j5NP0B21hhPtwiDLl21gzKXuPeBk4jew/gCfvBHfogvfvpxXkhO8Hf/3u/w1Kf+Df7kMr/58X/Bu97zIbZe+SR3rR2GuRbbl/c4vFSixhaX9ei+8mm2d4csrz7IlcE2d54+ymuXXuPIsQVaexsIPUEEhzAU6HiC0CuUQqHLBCNDArONok2ZZYjVUyijiUSKWroH3X0vjfU7eOKjv4bLJTKIyfV29fdUhLjQ0pQSXTrK0mBbIRMswllaUZUXK40ljDWGlHa7SbPbJVy5g9QUTPvXkUXKYmApy5w0S2hHcyzdeR+NVgutOmgVot0+kegimi0Su02YaIRu0veGKBmRRRFhUR30WjuEDJkqgVIhyntsUKJyTSAzSieYyCm6nKBEp+LOCEMUtfDWkJcZVQ4ASBcS+ohMGryUaJPjrSOnxNDAh47UJahYEDuLCwv8JCXSEbnzWCWIhEMYmJITCkWpLYGzSNXEpimRDpBRA585GnGE8gKT5ySqpOl9JQsKByKi1YwxJqNpqAAdhScKYoQrmaYJQShxpUU7yGVBGhqkqVjyvaJ6vgkFYTKBoEHqHMKWqCxARvPVAREYGhiKLKNRb7+6N/qR/6Me3xAFflbirLVoVVn4RCFAUy0eOY+qAzxmnnBqO6FAoOp/3+YER2lXScKuWdkIRYlzQVWYlcSXtvaRV9JDVackQlQTbDkjSlLdJglZkSed4CCUwyFr3G+12COcw4sKcFwNcOulKFFr2rV04mV1qbOlJnjjcPN2OYXZ4PW2+QP1O709vm/Go6m66LrY3kZ+nOGBq7It69e9zYk0e566/ed4y5J5mznn1lX629w1vrprUsKDi/DGU8ZVaLrpCFSWE0w3wDyLWb4DnYWU0TzoBvgBwXQLlMBlQ1QUUlKiAo0YTvGtm+ggwKVjrF4jn2REjTbGO0I7JZ+kiHiRkwsddE/QnnRJxrvEKymNviZNp7RbbfJU02uE2PYC1y5/jnd/4L08/dJFfuzHPsSRvOTVuM1Su81Tj2+Qp4/zHW+bZ6eY0jKnybINjDjK62ef49ij82xs3qAZN7h04WlW59/E3qBkZe0Ig7SPDAu6zRPkNkNMS2i0KEYZcdQCWyBVhMgD0H2C+QZWHaUcncd3NGp3gSGHMMMFvvWDP8vFT/w6RlmMCrHWV1iNwpEoSxRotBWo3FR/WtJTYBDeE0iNLRxSeUwgKMdTgnabpoeotQzekiVzOOGYJvuEGua7CyAjpIxxMscXXUoUYrpPNim4cvNVyqSgtbhALAxpLaHYEsJOSBBELHcXccoz32vhphbVWCKxBq0lbRFSugBTpEgR48jITIlSATJSOCOQQUhmc4xPUUWItQ6X2wopHnaR+QhrLE0JTngK43GZQMkIVwqKwhDEEcbkBEFcb5xX1yk9CFciI0WSjwidQAYtCmeIQklgLBEBZX1AaR1TlgXWpQgvMTJAShAuAw9GSdo6prRA0MDYPWLfxE5TRFDinCLVUySCIs2IAoFPFaFQWJFSNAqs9URCUliLsA7lBYkQVaZF8f88A/v/+viGKPDgcd4QBFWykS0LtGjgvMJTolT1+Zkro/oSia47Yl9v33kqH7bEY0yBogHCYK1EyrAqXsYitapSYoQAF1Yv52u9y+u6YzV1VyuphrhVh+Ktw1IvR0mJN1UWpZQVT8Z7XXUi3Frhd9Q6utQVu0aIWga6DdFbe81vxwwIeVup/YuFGA6Kvr2t+t4egVc9pSrYM396NaCuEQpBcMBvnx0S1TT5tkNh9tHf+uio71aUAHc7pwYwAqErvkhoWnhVIlOHbH6A4s2/QbjzHHrzNVy3RXDpJuZwF72zTyIaNNe+Hbf/NDLLCJpLuO09bG+9imybWFRnHWEEMlKkZoRyAWlmkErgopg0CMgKhUwnqN4K490BnZUF8v4IV+Y0ex2S4UWa8QbFuVeYu+s7ePKP/yn/4L/5VS6c/wx/9MnPcXLu+/hfP3mW/+Ufvp8vfDnh6PoSreF1Ts29hc99/Is88NgPsn3pJv3xWRaiksGww2R5k/uX3sunPv1/8c6HHuPiza9y8u53sNCZx+gpIlgliAvKscDHGTIf4uZbeHkP2ZUXiboXCYMBfrqKn3d01QuI8HWuni3IxpbFQxGDst5BECWhaBDZEi0kVgusBJsVEAicEhSlIRIKVxjCZkRoS3wUYjKPDAK8L0DHRAsCaS3z3Xm8DCjSiEYUU5IgIo8VGSGenhTcNVdw7uWrnBdd+hdhc+syrVyQxzllIeiFh9DScSHICChYP7rAZDKBbptSRiysLNFp9eiEkJshQnZJc03hSkymkcqRurzKUHaSKK7iNRElRVUVUCohDmIKW5CbDKkgdAqDZ1JmyECjtQRTEoiI1FiUlagoRCeWIqg6el0omlGbXICxhi5N/MRSxjHKWrTKKE11Vy+8JJAdlDAkeYJoRFgjiaWEsiBXIApFhGHqAgpdoiNNhiAXBbGLUU5CCHmxj9eNKnu1jCldSVsqKj3BkcspVlkaSYR1kOlvFg0egQxjjKlWdIMgRKhqSUh5A8LjrUAKV4VsGwHe1l2sp4pav53xIhEqwpuKPzNLTRLWVVLMXwDpV2v/tSvGebyz+DpXVNSHinAC4St+hJSyCrFwDikdwisEGi8dzhXIOo1IiYqYI6WEGRhMqOpAqvVyd3ANtd5NleDODB9Qb+fOirOaDWCtO4CBzerr7c6ZGSdE1xAzU4dp1ATm6vO3bctKWWvxNQZY1alSBy4ZWaEhqgO0GjBXOGRfuY9q3d7XuwUoBb4ihLL6o7DyfoKtfwdzpyiy+9H5JcqOIth1WNkgjOYpB1/BtLqUgymhFOSui51cp9Posrk/4YhdYWjGtLqCkdM0U0vqBHEnYt8KltcfwozOY4lI0pKGirlxdYNe8wjZUqdCB5fHKN0eR+95GIbb6L0dhjc/zx1Hj/PzP//TnFy5g1/53/8JwdxdbO/ukJ+f8Ni7H+H5jYv8d594hRuf/12e7j/M3VcLXnz1OUwr5Nz2NV762Md44OGTvPDaixyZu499tUgQKHS8hOhP0F0IVo9jkgStGhR0UMOU1sIiLpBY16WM1on3drCNFUqzzMef+Kf8xI/9HL/z+/8KtxaBUYhAkKsAOa6cFjaUuKKERoi3hqYTUDp8pHCRJg88eVl50gUJSta6uHXENGk0I5TylFgCAaUb45MBp6JVvnL2q1zeuEl/qw+DITZfYaB3uONUiwfXe2ze3OOvf9/b+Ae/+LvszBlkYcmSCZkzdJ65QakVzeYCeTElsBJtUuZXFUHcJC9LRFMhVEAhU4RokLs2URQQCUuWpKyuzdOOFNdHBaMsY7W9iGiF+LKiVuauJGh3UNKjnaAZNHBeQBCQFYMqWFx52B8zDSQ298RBgRVdkklOYUp8wxDLAhMGBLmuhtu5x/qizoktES6mcJpAtTCpRTeauLLEOElsSgyeVGZV6puzlN4RupJ2aSmdxWhBXiR0o0UGxYjSl/g8QASWsc0pnam4/5mrFgFlgXN51dR+nY9vkAIPmAJpfJXOVxrENEE1g2rQKkXFSJGSLM+JdRVCIXzlIKkazmqz01hTJzHVREYv6tuqarKNqTZjvXcoqRBC4201VBUSHK7eOJ0NFakJl7bW22etcHXZ3okqVJi8SmgSEuEEqu7IZ4UZ66rCXicVOQFK64rxwu3oglsuFClmNMhq01eICs86Aw5UXTeoGpNgnK01dYv0Neu9gsxX/AtR6XrqwEdfuYEQqkY/zNAJ9Z2BuwU+E0iQrvbu13cb+OrQpQQHpdXIsN4GtA5Mgph/D271Q9hgH5kJbP98lT8pW5idbegdwpNjipIwigkycO0l8jTB9bcI23eQqZzJdJfhoQCdK8aTDjrcZFymtIIAzzL2ylNEZ76DJJujEUxR1rCTRiwdOoYWMTs3zrGwHNNQJcgFRDRlZ2/M3NFjiAKYZNxzao3W2hzR5j7v/8m/yd/72Z/lF3/mv2LsHcWLn+Wn33eEv/Orf8iaeZVPPfUlzjz0fp59/kssnLyD8dY+47LHuLjKzvWvcuRJS/ddD1MMSqJmQSjvhLKNMDHMacJhRtZ0uNQSRAFZIknbIdMgoJtsES1+Bx989zv5g499FLVcASSOHl5ka+MKaTygG7cocs+4yGg0NYEBb0fkQQgNhS0toTKoXDDNHUGkKVFILyt7HoK2KWgUUy5uXKPIDFfTASsqYrc/ZHuhy42dm+z3MzZfkbxGzI9863Hu7hSszQW4aMrKsR6f+rOzHF0+ws7NKaku6DYFq2WLDEujldDtLFKkTRp5igwa5KagEwSMfINmJyTuaAabAhsojBjRoInNcuIAyiIn8Z7+YI+mWmJzd4vOqMlUglLVUmIc5ZjC0Oo02Cyu0gyaOClQukUsEpyoNmp1W5BRsV5Ct4UzljgO0QVMlcOHIWPniYIQm5VkoaFZVpz5MhkRtVokmSAONKmTkHk6cZupy4hCSZqCiCXeTNBSY9CYuMFkukukQwI5zyDPiVyJFBF5aFBCIAswPiRJM6RukWcTSlsQBAHK3R4n+h/3+MYo8LUkILWoF2scvvQoEaFChcuKOnhCVhqvAKgAVRUDxmLr5SOt6yUhyrpQeoQIEcKjMNU2nJOgBN5UzhvqHFKkR9SkRXmgzZvbxqI1RdG5g8Gmh4OBrpz5xqWs7Zim7uD1gVd8Zq2sIGLgqYK2XY0BVqrCGM+Glwf6t6iBYZ4qGed2T37tqde1c0bWn7e1dVIrVW2/ykqBn4VxK12xdLyrGDdC3NLmAWS92SpE5a5RKkBS203xeFdgbE3tdNWqui0sBB3S6Rj9wP+BbDbRo+uUUQvvMjSScroHakS4eop88xIyXCCWgkJbfJphksu8+tIF/Il3c9ruEE4S1k7ehc8H7GQjpFaEgxxrBEXTU44v8fJre3zniStsXdzl2t6EtzzyELsXn2CwkbN291uYmsMci1fY3dtkrl0w2sq5fuMc3/m+76bVidm8eYWFw+t85YsvECwt0xj1efJTv814mvI7//qjfOdf/7tMpinDi9s8//RX+Jlf+wr/8p/9T8x11xlcTxGlpKua/PCHf4Hf+t9+Ad0+ySsv7aBbi5xsJLS7Q1x0EbGzzzA7RbR2kricYueamElGKVv0LDjm0cfewfDiJyk5zJEzexzurrK91+XpS8/SUZKWb5NbgwgEbdNCFZaCBNGcIx4XiKbAl5a0zvaMA0PkLXHqmQ73mAy28N6zYzL2Ryk698ytrPBth97Csy+9zN5OxqUbBQt2kd5Cj/f8Nc2RrudmkfLaixe5cF2TIPFZSjNYZ7Hd5f5vO4GTKauLi/z50+c5v5PTUsuMOhoTNgnyQ4TdedTkJs1gkeHeeYK5Q+gsxAWOlpqn10xIizGjJEB6x9ClrHbbLAVNwkCTJ4LJ5oi43WLiC0oMg7LPctHl0uZlpG4SdDw+yzgd7hJHms20YJSWBLqBajSZFBmZCIlUhimDKvIvirEuIZA5OIktYyLtKaRAlhN6C6BkhKSDDTTNOEZouJxPKAqNbkiWV3p0TFSF/egI5wq8yxBBiDWeuEiqhTDRYeILtPcUWHzoKfIJaI8zBiU0XhtKb7F8E3XwDnuQAKQQYPKKgS0Bqug9Wzi0rpaRBAKpBWWZESj9Bt250rHD+iBwUGNrhahS4G2WVpF7BwWshnrZytHja2+jc4YZc+UNXfvt257/gUElM4Sq0EjBQbFWqsLmVtqLuzWg5Jb1UwiNv/1zs9eUQL3ROivQM+KkUrVbx7nKl17LJ1qrmbWmKtKuCvwVszfjq+F0da2z9zOjWMqaiVN9rAbbtSTmJaBQFqT2mMyiIhAmxgYaX+wig2UCsQCEeDdCDLfRzUNYmSMihUKS7exA8xB2PEDFE2TRwuR9MgSrZx4lKwwex96gQyY2WZvvkWxn9NZDynab3a0RS5nGWcM4n7KznZI4WO2E3HjxKY4cOslzXzvL0UdbRFnBTrpHkqd0xorCpexkVzh06H2MRlt0WwuY/g6rPcH7vusdnLljlZ1rV3CNHm976E1I0ef+txzn2o0dlB+y+dorkBVc3O5jTMY/+sjv87H/8wle/sVfohxNeOC7DqNjmHM5Pi3IrCXsz+GyCdFqm7gcY8IeZu8qUjVoNk5hQ3CDLcr9dZ554Uv4xjLvecd7+cLnP41qTliJwPou1u7QFE2U12SMKbWjJZuMpvuMZItpOeGQ9szpJkWxz1x/j73JHpMkwZTgTEakGsSZ5fBcF5C0F+e5sjnkhQtXiNoNOrHmoUMlBHusN07x5ItP41zIJL6DcjpludtmTEY070gmOV+4dh1fFgyfOIcpFcZodgaGjVduouOIpbbCBhKdGnbmjzFIUw6rnEbksTpiazpAtiSTiUHkhny8h+tIJtEO2IKMmywuHEK0YwpR0m0ETMYZYSmZZgPmIsnpw6s8t/E6Wkj2c4OXHaQVaK8olCIbJyx0G6hCgJJoLzFlRiQiponBBdXffl6M8Y0OqrD4qMV06Gg6xz4jjDFoJ1HNBkVhaEQQNgN2cYyjgHavSRRXG+ImcNUil/MkUY62AbIoCcoc1Z7HT7ZIvSXQEcYWCDHF4JmWE6SUNIPG111XvzEKvAdRob6AWkseTUEFqPkOWI+r5vkIHNZlgMRbhXKyHnJWtMbZhqgQM9ufrrCfswxTYxEqQliP8QLhS6rcVPGGglsNSf0tcqL31SzAvZF/fuDqecOj5ty4Sievwi5mP+oqgORWQlP9WnX377w5cMAIOFh6mokyfnY+iEp2UgiMswe0yNs7/5kH3xiDDoJKZhG3rlsQVIeWcAfcCyduHRyzbFYpVZVdK2omvKgOvURr4kITTIGhw6Ylqsjx7z9LFDco3KuEfogPAugew+9cQSU5hC1s2EQ0uhTZBp07v53JxjPsXtyhoTo0F3vYfp+paDFuPoC/+nlM09KTK5h9x/lzn2Ll1IPcHG6wXzYZWMOx5ft4/oXLJIMdVld6bO2OOTyVDDLLlbPXWTu6yuTKK+wPUuZX38fGzac51TnFsg2Qc2tsXTzH5sZFjt95J607j+JcgpzrESYlb77vNNa3mY52uPOhB5nrQmt+m3e8fY2NS5KhPsOLX36O/u6X+Ee/8t/yBx/5DcJsl6U4RDaWefbJP+Jb1pfxekoq1+lYi8syyDeRvRPY4DChFBTbfYJj34Xf+Fe886G346N1hldf4L4zJ/j0E5/hrjvv4MKNDZRp0BKObmNAzw4I0fTy12iHivnDizz/0nme+swzHHrTEjDHpizp+Tm2p21WOh0+/uWzZKJkOhxh0yF3HT/MdP81vvdvfDcfWH8rbrzF6V6DcjPnhd0NPnttSGpajFOJ9ppkZ8JTgwuoFggfc88dpzns9mnJgN2jPcgNJm6DCFgfFOiwQ7vbZDzYJ1kMUfk2ZTlhs3+WaZ6w2J1nsL9NO+yCLVChQske++f3eG1U4tuae48skJUJU7GLKxxBb475uEFDW4zW3HO8TRxtUSzDNJfsbJaMJkOmCsJWg6iwBDYnNRLlJM40KYSoOPPpELylSAoaYYyUGptPsDoi8pCmOWWpyYNqsW8qNPFehnSW7dygHDjVx8eSaA5WV5ssrR2i3ZonD6YEsaYxiREqZZRbTBxRZgNyH1ZhN2aCKxROR+SK78fnAAAgAElEQVSyIAhaCOuQ5TeNRFOHPs/oj0KANDhZIEMqRm3hEIGlCiZqVp2sTZBa1jmrHBSlAywBiqpfDarhJimVv91ha1SotPWqPrX1sAaXgao3OG3dmdfFX7jqMLqti7/diuhqzV5KiVCq3oCzOJMRRCEH7HcvK6u7V3U3P+vyq+uxzh44bMAjhTpwv1QHSjUMnT2cmJEea3ukFGCra9d1UMOtYewMZlYtV92Oj1GIKt9Wa4w1dXCwrQeuMyeNwruQZtDGlyOSQZPpdMozz0y5+/6HOLogMcV1bL6D25wgO238tUu4+ftxc03E7p/gr+8QtRYIT76L6QtPk0lDY+EUMtlm88JrdFSbyHts/ixhd4FrN19i0LhER495xWxxcnWRRjZAMsDnKedff4Hre1u884H7Kcsey+ttjtxxHy/d2GayP6B98hibaPol6EtnCZ0iOH6Y4fgGzeGE6XiLvhxxVyMmHw3AFESHFiCd4vOYrNymKSEfvkzbplw8P+LQyhwPPfIIa8fPsHHhHHcvLFNc/GPe+rbTXL60jZm06B6aIzABduAYpgKrBa0yQ8+dQpsERBOLJB1so3tdVP86g3GLxfm78ItnSPde4rXXnuQ7Hv52xltXORo2WVN7tNSY6XSP0KeEToEqCFcX+Jf/4+/R7TT44I++j0Z7nusXztNWh3n65R0+8fxVDjUOc+lySul2eM+Dy7zl/kcwNmNcrnD2i39Cb+UYxd4Ok0ARdE+w3DtDr5nyuZdKpskeun+dWCtOLDfpdh2NaIG1Oc0gS7mRJMgsRjYDIlGSjHKiKCJqttgd7mJ9SbKTg1dopVhpBARxh618yvriPDpTFBIKA2WeE85pwsJzdH6dO6Iu151id3tI7AydMMUrRV54ogCu3txnoR0jjETrgvZixCj1xK4acDaaEY3SY1ODlJ4ChfMpkYnYT1Os9XSigOl0Stxsgg3IVYYxjqiMcNJiXFElS5WKxBbgc0LhKbyl8OASgUgE/f2MsgA5v8+JI0eRE0eppjQSSzuKKPKCLLR0fJMkndCgSZ4XCDXFZlO0aKNUiDXF119a/7ICP76ex8N3Kf/VX6vX4m3ArPtlroU4toQrDeyPEU4iZjZCVx0IbwBhKV0XxarQikAghabS0i0yHyNQmEygSBHe4W1lk0S4SrN3tS5dKeMw87kIBz5EHuSJzhwu9o1hGiiMBy2qbrxy3EikDvF1wUV6UA4ZKJypowdFJdMoYbCyHpxyy8XiqDZRZzZI7z2yDi2ZpcR4X+W4OlvhgaWcXadG1EtRt/vphRZ1prc4eN1qy/fWZq+tA7FlCTYsUWmPshkQXDuEC0teubjJYPc4z2yOeez73se9b383ZfcoUbKPG51DmRwnBd5IbAkiq9wL4fIJti68SLFTsJlkdOMmxWRCvPImiFukN86Ss4/1K3zbh36Y8eYun/njx9ELAQ3RJtkd8NAHP8SffOyjNE7dRf/K6+z1N2nF66A97/nev8oTf/gHvPt7/hMmW9cwQUE2uYnMSrorK0xKeMtDd7N7ZRsXtVjQmv3xZXrtAC9ztG8RxU2yrI/GE8wtMdnZJGiEJJOCxSN7MB3z2//6LHd/y49z4p4Oo/OvsrziaDbbtE8u8fJXNrjnnR/g5usvcrgXsbO1z/L9j0C/jwtipB9jGsto3cVnY4iXcPsZricIbEB+7WlSv89csMzFva8Ry4zHf+uXUFHIZO0UbqVHm0WkLbjyic/zXR96lEU9pV9kkGk+/dQGv/EHe/zAOzz3nl7knjffRWIDNjZH6MASuG2sSokX76adGk6trbLSUly9scmrL435zc/d4PXrI+KuZnenZP3IGt/73h5HTkTsb04prKQTC4rUMUr3kcU+b3vwMC+8toXWh8l8wcY1x9LCMpOdTWQQMnGOMMxRXuC0pwQ6KSRFTmELTBCiJg1ypZkiaesWFwdbLPfmKadTiv2MbgSBbBK0FWvHTrA/2SLoCYwvWeq1AYdymlYj4GZZpT9FQiFy2ByNiE1MJqeVvEmHtCgJwwwZLmJNRhS3yV1G3AgJRUkoJVvjkp7qkJU5k0mKdlDaFApDSJNxUBAIT8+EZFpiltq4UPHImxdZ6LXZiwyN3GG0xuRVWLjIc0QhKCLLOE9pl5rUeyZyTCgVIfArf/P5/38CP/5yHzNdFw46XOkrtoZ1KCFxMsTbsmaa+1taBbX3W6uqmNbLPAoBQtd2PY2kAB8CRRUIbFwV8FRvd0pZ5YIKbi0IVVJ5XcSFONDGD666tmS+wQsuK+gSQtbXWSWwemtBVEsueIF1FptDvexaB/aKW5LIwbcRs/2k6l9v8KbPrmdWtCs8wy275G3XOfuRiZkMVS13SW7ZKwGcre4mBGklA8mw2jKOHcZHCOeQk3V295e5mQZcvBSThCXWBpidMWr1ncj+6wg7xu3t4ZXDuZgwEHg1z/7+Fvm0ixyeJzEriMlVwqgi7KnWCjc2znL4yJs4//rzmBLaKmfj6S9QDDpomfBt64/y5y9+ia3Nkp3z1wHYeuU8zoekO4LV4x189io9t00hp0zFBBBkWY4M5sj8gGCUsnBqjY3rV9m5cpmHHn0PW7uvUwYB4wJacRuk4cbGS/S687iyJIwb2DSl0Qjw6Qit2rDoOLG+zJ3HmrTzXYJFhVKCdnuBjReucXL1DDsbF2kJmO7tEgrIzp9HrvQw+R5eNGnJbcxwB929EzMd4hZWCIsCVBPXgI5cwoVNjo2XeOnqWQ4tLpI0V5k252hmHfqvnKdXXuU//4UfZ/vaDa7t9OnuSH7on32NdMvzqx8+wWM/eJSL244vvnSFV14c8taTBXPzCyx2OjTaRzk014JDji99+SxffjXjd/98B3J4111NfvpD63zfD/wcP/4zf5tr/U0e/1LBWzdbnLq7zUrU4drNqzib0zi6zv6eIy3u5Pjdb2V4+SXcEHoaQjZpdwWpyDncPsH+3g1ElDMZeaJIsW9LtI4J3AKFyfFqSmEVgYzIkm0OLzbo5CNYXGQip6hGA5+NUbLBqN/HCEiHGXEcUmSCZrtDMTVcSRMarspDyNIcH8yzHM6xlw6woaUBpN7QnNMo18O4AqQicUPmGotMshRjPRkFRWmJdIzDYIRB+CYeRxkZpt4hAoVSin7QItKOXinJiwTRXqUvc5ouAAlymhJHAUmakmlIVYbKIQwUpR2hZMR81sQK/lJYNN8gHbzwT/1aBFCF8LrKlWEjjTzSQwQKO02gVNUWqa0LmxO1t8VUyz4HXHNVddphhA8ipNB4V+CTAdhqOUnVIasOibflAbTL21pqseWB/v4GbV6IAwgXSlf0y1pfd662U1LPFOrXQno8FufDuhu3KF0ByBARKH8LB4+stf6K0+5nFyGrDVpbo4ilp5JYvKxzMeVB3unBXQe36e1eHkQx/b/9ziupp7ZXAl5X700WgM3Jr0QM+kf4w8unuLx1iWQEnXjKP/mdf47r3IEc7mHlRVQWQD7ATxPyzZdJhWPzfEnUajKcpCyurrBx/hzTYkqjdQdi8TjFKOHLL23x2OmSaWuZycXLHH/fD/H47/0G3/PDH+Kpj32Ru956Apft019+FHXtVeKTh3j5ied4dTxlrXuIpg5ZPrxMMLdExw0Jveb4g2foX91kx0zoHV1nPfMM8z0Wlt6EXOlQTAeYy7ssHm1hspsEaJIypddus1ca4ixHRIZec56bl88RdkLWj3coNq6R+BP0jWU+3CdSAloLQIcoDrhwbZeTZx6G7CLIPUw2j8knlK7F4soSZZmgW4uYtQdQgx1kdwmfxfhYUwwuEEz6iHwPuXY3/+6X/jH3PbDMPle4MZKEkeCZT/8hf+Wn/xqjKyPOf+55Pv+K5dNPDvnwB2P+9o8e5WZuuLDZYGPzNSK9zno05IPf/xCjvTninmVy/QU+/qeX+Me/mZAknu9+bJWH7l6C8Q6FjLh4/jo0muxlBSePLhA0W0zyKXv9jGs7nmRzn7/1X/4mv/Qvfp7vfqRFp6k4cbjL4OY+l4d7GBmSeU8YNlkIQ2xaMp1oimafTqNDNvGosMnm7i7YhOW10+RZyaN3rLNx8SpXpgaMZAx0VzuIpCDP9smKgqPHj7G40GO8NWJnMkQpQRiGgKQwJc3OAj3luZgNaYddXJoileLM0aP8+dee4D3vvI9DjYzPvvA6gTpNfzjFFxA3QspsQBkETIVBCYX3CqQljNso7SnTFApAe3QYMB8p9n3JODF0A4HDsXL8MM3YcOLUauWTJ8ehSG1JWN+Jh6XDG8c0SxBCkDvJyOY0KNFeIazjf/g7z34zdPCVDCyFOAingGqIKCqsIbOYO0SNFXB1Vzr7anELDVwNDn0F9MLNvI0IGYPLcaIApfC+Dvf2spZlXLUOPivKfqbnK6yrLGdv2CitPeqzgnorfs/XWn69pCRFnfVaIHUITtY6tgNdUSGRsr4pcXWxl/XQlJp7w7939yCFqoamVD8Pf+DouVXkD4alB/bG+vr87E7BH3DsK1RxhViY2UCdcyAUpdXQyOlPBTvplGn/AkIoxt5yT9zg2tknOPTWHgQh6FVM/zV0npPphMb8KaTT6NZZEpmwc3UbUyga7VUmgz6jy9cY7Yx4z2Pv50tPfZQbo0fJzz+HXWxw6anf5Mw9byEfOLLpHmX4IP2dEXYOVJaz+dxL+GJMkIxZPXKSC1deR957D9/3vT/E5cd/l8H2LkUaMxYLHDt+PzubzzJwBbmcp7Xa4pVnz9Fqhiy0IVxcJtvcpLAFjWbM9v4QSkV37Th+cA7fcuTWMJ+WlMUYQo2aDlnqNgltQBjF2CAgS3IKEg6vdQhbGaSW6ShFKUlXt0gCiZtuIZd6uBzs7nlEHCHGAa7dxQyuQDpBBYfJ4hVMa5XvfOwRgvUuxfmYM4fG/PFnfp93Pfxunvjo5/jKl0d89DWHHJX82X+/xptPHeHJm5bddMRentNWK7TDAN2a58a4xVJ4mXPPbvLhn7/IztjxwD2S937LIi09z1a6zcu7BZvZFifvOkN3SXPKTzl3fQ857dO0HZqLcG8Y0V4/xT//7Z9i0QpefG5KoTWf1Rc50usRxW1k1OH0qS6NUGOmU0zTkMYjTkTrlFYT6ZRm1EFFELVDxOIiu9f3+JOvXCA3ZQX70pqdcZ/hJhQiQihBUYyZTqeoIGRtZRkRd+j3BzSbEUVeMplMaI00F9WUhbFlyIikXRXj5YHBlVO+8sKrLIeS61sGbTdxukDkIekUAppM5BDXdISNNqb+fyQf9hFBQFqCtAKZ5shmynowT2olo9LQXmmR9GBtOaYZCNIgoBSepnEUuadwAuMdpTUgBHmRYaNKKs1tRivWpC6qcmLzr1+D/wYp8JUk4/F4ESJxeGXw3uDyFFSMEBG1VaQqRAoQ7kA7FnDgoJFC4Gd5qN6CFZVeL1MEMUpWmrOTDmHEgWSBF2g9O0Rqfb8OwqDOLHWuymqtQjJsXYSrW6nKdSKQqhpqSi0qj7zTOKFRNWIXWTFFlJIgJF5UnXl1QDmckAcOmIOHu/Xeq4fHzOSY2h1TyVswQ5lJqetMV1N73OuDoTKHHgSTeKgOIy/wptocdrrW7029jJVm6D1FtlcSLkacOrlFvBuy3JP85E/+Cv7t76Ds72Oz64TtBvbwg4y++ikYX+Dy6xv0kyE3L19n4d1/i73kzxj3RyQ5nD52hi9d2MBNdvnox3+P+9bezOagT3z33Wy+fIVoELF56Y84vXackYNB/1VevXyOtThle7PPy197mjP3PcCV1zd5y0nDogopz32VP33ta9xz+g5euHCeE/d/G8fvXIThDtE4YfHB7+bciy8wytq4oENP9dndTSil5PSxOzj3yvOItsAWOSbWbO/dpOkM6fUN7j5t8bJgsN2gJeaQnRw32cG3NHJljvRaQiOK8AQENsL1C1LdoNU9jcv6ZE4SOss0t3SSgNKnxME8xoYIHUH+CkpKrC9IA8vZz32C/Guf5KH/9GeZ7m2SbZ3l8tVn2Hqlxy9/8lme39c0spwLH30QpiGf+8qTPDfKibv3srud0MhzOodarC0u8NzlDf7hj/0+Z/fhxGLAz33PETbVBNtqMQpCik4Xuad55E3zyM48r9+8yHsf/H6efelVHnn0Bmfu/yD/9qMf4VTnEDbaId1VvO3eZXTiGJUZSBhPepxozXN1vEMxnDIsljm//TrXJwPKUGA7Db793nUubt6g4U7QGI8oZYnvp7RcA9/PeORNd7O3P2QzGVBMSxYbx9BzkiNBh2vTDQoTQNlBqoCLF6bs2BGBt+hgSqgM3ThCmj4rKsYvWNbMmKN3nebVi9vYnV0evfM4F67usi0UotXAJAWRCkjFFOElWb257saKZD8jo9qCL40jzfaqcPgAgpbg2NoRxktd7jgkeGvnCDTXcKEhDnN8ZgnQ5GWCERIXCZqpw3pHL4jZT8fIVhOZCEyREzRA5w7KlMJbbPD1x3V8YxR4IXCiCrutCWPgqjCDyuThUAqcuaXTVwJGzRmvQzVmC0XOVlmqB892rhrOOgkSrMvQOqoyWp0BJw8CuX39fH3ARa81aVGdvFrKA398hfm93VpJLZHo2hWj6iJdhV84V1TWSxWA0/UyVIGXCiElbraJehtyoH7Vf+/7QLV4VLFl6uzUv4ADqyiRM/xw9fUS9QZf/u0BIrPOvwpTrncSPFWA8lCTJYJ+v2Cot4hZpelv8Oa7H+XTn/0TvvdtD2OL66hul9GNs4z2DS9+6Uk+//hn6UiJnOtxdX/AsfJx+qNd9LalIRucf/0KvWaXqLuEcCM2im18uEiwsc/asRWS/ojm4mlGoWT3+k1+9Ed+ivGvv8T+xogjq4d4aj9BGsckvUarF3H1qzd4aP0QyWCbLeDmdJ+NCxdYmq4zmJwjGaXYCy+SDQZ88U//jAfuPUXSXkbIMUM74OlXriHLjHSroC0CwjzDMGAcNolaXZx6Cbd4hGhfkAuLyD16rUfkR7hkn1i0kK6AVoPSOXSUErCACywyz/BZxMQMaa7EjEcjOusnQa0i2h43ynANjUinSAfRSo9zj3+Cn/jlX6bYucD+KGV7+zLPfHXKHz6XI7tt3rmyw9//8F1ceCrhhUtfo3X8DNYLpAVEwKH75nH9kj9+4hX+zWeGJEnAT9+pOXFXF3dao91ROk6yuHya/t4YPWfY35pgh9fI84TPffrf8q0f+NsMrn6RV85v06LN9vY2eSbZ2dnjeG+ea+mIoAjI85xD3R6T8ZRm3qDIR3zgr34/n338k4TROUbbHV69eY0vJBucPHKSe+88ztbGFZTvEoYJS+snWTh6P9g9el3FNPOM9JTMJpCHPDPo420L6QS5HyLCNomeoPOQhV6bOBJEKqMbSfI0YeIy4kTTl4Lxi5dJC8WOKujtFCTeMUgKdNimKDNyX6VGpeMUgaPUOVki6OiY/TIFY/CtFl7EBHOSO8/Mc/TIGuuLXVSvgwkKokaMEQUNCcYITCBJszFxpPFTT2BzykDijGe3mKKlZOKGyFAQ65BsalDOoSINiUGV3zQuGumf/IiuwgEsB+kqFo9cbCDacaVFJ3UHam8tAVVxeDOnSz2ABDyygi9JhZIhzpcIWyCMRGCwtspgFLZaq5e2llXqdCfni/oACcB7jDdV8PbsjuEgcu/WgPgAu6t17ZipN04pKzQxFXJASnkrG1ZBxUi9lUTlnKuxvbccL1LoatjrZrMGz1/83c06cik1ggDnM6QSYBSWstpkrWWwGaJZKIOwYGu7piDAihTtNAiLJULlBZsXJNuXDZdHmqLRpddyDPuK1sll4vm389hP/RYXn/gEv/A3Psi33v0uXtj+GsM8obe6zM4kYKVxlL4ZUsSKeC9gfWmOTT+m2VtnMO5zKJBMkgHt7jqb0w0Wu46G8HR0QKsoOHXfhyj6F/nCc69z/PQa7OzTWFrhta1rPHTqDjZ2PUu9Bk++/BQnFzqcPHaSCxtXYDzmyNEzNBYaXLtwljtW7uTt//VHiM0EtOULH/mfOf4t91K2oN3qogYDymzK/PGjZJMh5eQGh5cOM97foBM3WX6sQ7L/NP71u1HWk6bXmT/0Jkxi0UIwFvu0WxHCNDDNLra5gLOCYNCHw8dg9yJeNvDWMtxLobXA83/+JOSbHLnrLobJkIf/yn/GtRvXefGXfoHHfuAH/m/u3jtKzvQs8/494Q0VOkd1K+eRNB5NHtsz9jjggG2MjY0xrDEmGFjDWbMsB8PuxxoOHFj4bGAJBpPWeXHAAafx2BM8Y3uCxpNHmpE00qilbqlzd4U3POn7463WyMBZ+PAX9lDn9OmuUnXV26Wq572f677u38WKyjh59A6eeeg8X8hfyR1f/Shv3L7M77z7lbzn/d9i9/QgZW2QbneeXVe9krJzip0TCpOu8M5fOMKjLUEjDrzjxWNMXzvFcm0Y0+kwkQxTBoPN1hiuD3Fh+Ty2DCRxnVaRUSwV9DVHmDu3gDclrfUCFQ8SC0NR9lNPoGifYMVIpkb7SRs1ypZC6cBVV11GKXLml1oo7fj28dMcO5rR0EOsKEu7zFlaXmRkUx8TtVHwloYQKK/IjWa1OEMjGaDrDefaCpFXAd8tUeJNSdCCGNib9BHXctKGJgoOIwKlCCSFwWWCTEaoXJDrLklUp9VyRJHAopj3HZK4RrAQqRhjHF3TQTdjbFHJ7AMDKZOTEePbRrnssin6+mP6BoZIm3VKV8mZjkAkBMJZCtPFK1CZJophLWQkISLkgUK0kcZADKU1dAuJC6pCJ1MpECZzFCqvCjIHf/ofH/k3oMFfrEx77pWeni5ClUS0gdtyoqgohlL2eDKi97s9u+DFB/S9RT+pEleEowq2ixGygBChVNGzXILwrmeDDBelIgDCsyHSegNlsLEyX8QEXPqHVOIHPUkkBE8IlWTEBmcG0fPkmyqcmEsiBbm0cpe9QJINx8w/ftkugsDCd+rz1lqU7BFrRCX7gET6jWPvISGEQ/oezwaFdxW3R6JwMkJt2FFLxeKMpe1jhjbXqUV1TrccRdRg+duL/NSfv4A7/+bdfPSPf5cwNsRnH/k6b/zxn+HWz3yYYKcgrNOShoEoJarXsDqmnQ4wWm+wsrhE05SU6Tg6zWik/TQWJWPxAWzfEIsL3+LkYk7YfprTT55lfPQMI5NTPHy2zf6+zVw2spvMGQanG+weHeLY6ZQjT57itW/+cZLb7+G+1SPMzM7BnGFo6zjH5mYYePBr3Hv7Z3nd2/8PFkTG/oE+Dm3bxe13fpVxAfUBxfyZ0wjpEPUmC6tLDEb9nGu1GTsToReGWbOe2Hm8l7iVGfSmnSyfPUraGEakNYr1DkkU4ct1mG/Brueg83mC6cePbaW7fIbGriHOP3KBbz98G3a9jU4HmTzwHBZPnuHrf/BX3Pyq61kTkyCO8diDEX/0BcWK+0uG1ix/8MV/zx/91ZfYvLUfa5pkqyVTBw+wNXUMTm/h9lu/wkc/vsZTpsELxnNeeuNW4msmKEQNuiVlmWPDCr7sYNwgHVUyNTJCkIKJsb3c/tnbsCbj7HyHLDcM1GOGBxrktsu1117Pg0dOImxOMjBM0glcKGFu9gwj9S3krYxvn/wm532X1Ee0spyCklIM022fQ0lBn0zI4oTZxS4X1ByNSFSfJV3n5de+iPQZx0JrhqUAVkAZNPMCNrmSOPFMDdbos46gV0niGDKLLgOZ89hIcN7ENKhSpBIkK1ngeTs1R093WMuGyK3BRDGhgKwMWEqk8jTqdfprUN8xyO7Lp+mvW/ZO14kbgzRHJzCmIFL9FDbHB1vJts5VyAOhqyxZrQj1wFqnSyzSCmusumgZo/0AhSkJroU3UI8D686T511quoZONLJTLcvW/Vup4PfJcN/7KodMKDesf70FvhlDPSEkCgpD8FWKTdXkrFC+wge8kBcXxyo2T4FqgnKIqKxOBlYhfIlzorJNWkfIRaVvW3Oxeq+ECtMLsdiQVwxBBJSSz+asiiqJSUjZGyTq5aMqX/WFL+HVCNlj5PR6Bwj1rOOlZ5UMYqOx6tmw1Wx47SusgP8OZw/wHYu7EKpStzZQZL0qXwaJE7Y6yfXSnSpJKnq2yboxT2ADQToESZVVvpyztqy491uOzNdZcIGD103yyGOnmZ+f4sZrd/CEuYnbv/x+6m6AZLiJCbNENmJV7aAtO9SiCGlq6Ml+svYSwwOTLC61GQwxXgqszCk6DlMvSOUkEzu2M3PqGeoKYhnIyg59A0Mszz5DfW2Jvk0DrJUOHSxxoui0LEFp6HY4fP01yFaHpC7oLq9hQ416AosX5skGaqytt/mBt/0X/uJd/5HXf/+LiL1nZMduiqif+advZff+FxKnCQcPX8ajD91Frd7AtCTaL9OfLjOwo0brKcPEPkXalSSuwO7fjjtzAmESQjJMkvbj4xpe1FH9A9DNEJ0FaO6ExGELge7fQ95+lJm1Cb78kb9hrN9y9fU3cP7EKt3lo2zavZ+9Y0N8+dYn+NtvPM6xtuDxh7/JuT99LYt9ls995BaevqA5cGAnL3n5tfhaCcvH+dBnT/GVe9fJnOK6XV3e9iPXcM4rZNqh3W1Qs11qqo43HeLg8WXg4SfnWFwOHJuzJKJGPQr4ZiCyEevrbaxI6OSGwhSc7QgUBd0oxuSCQSkpFDRDiRKKCzoiDZbEFxgzhO7zyMIzNpLSOb1At1ZHipxO8KSihk0dUWERukZhPcpZRlxKojOunBzi5Nwaj6kSH8UMes1ALaC8R3pFbh3dOKL0QCSIogi33gXvKURAihqFDoTSce3eGifX1hEWpibGSGqeTdu2kDZy+rYMUYsTRgYGqSUp9Xo/1kqCy4kTSbExgCkrEGHXZNRVWmUrG0fQkk6ZIUOVUlVYA9KhJJi8QAlNW1u08WAtnbJLIQRDJKwaT2lygvC0ypxURehCooBut8sf/+d/Iy4auETiEOKi7h2cJ1gHiUAdw9QAACAASURBVEaEmMoX7qtFj2qIKYheKMdF7VpX2NBLnCSVJBGQMkWR40qN8BIoe37vyuaIr6iV8hLvee/BUVL1KukN+qPtNS03nru3oeiRH/0lA1GXGNsvVuvBi4plvmEduujEDD3o2sYiHi5KOhdtjNCzR4rea1Zp5uEfPIf3obcDUoRQzQmoQOXiAXDVQUvROymGUCU9EUBofC5ZWw6YWJEXEa21LksnVxiM+pjvXiBpdzhy2zkuf80bmH3sLNd/3w9z7GN/winXohXOEZcxmWmwd+cO7n/qCUbSJiuNhH3XXcHJ+x9gKK5D8IxOD7KULUOmaM09TV9fH8PN6vljPcJILSaq1chbBXkqMBfOYXUfYiBiYf40opYyEEV8+u5vcWDnFawut7h+KGZt4RxqYJpN+w5Qn97JLX/9aR448mV0TXDs5JNctns/x48/iWKZiandJCUslgvMPW5piqga588N/QNjlLZkZX4V6Wv0TQv86gp+YB96MEI+vUBZTJMOSMosJ67FeGXJ82US28aP7UCurSPsGN04Js/msGfPMHHVS9ADjhtfvIc7b7uN1nrJzsmdtFeWeFIM8JmnSoiaxBfu45afvIzTYpljdx/l1hnLDTsO89zv2UXcWubxbzzGZ+6d486ncgaC53fe1kDv28vTQtNnBWjJlhDo1mLypWVmF0rmz7fJuo7TK02QEdtGJYsr5+jSh1qHc3lGJzN0vWE992hVQ0UaKQLBCqJYEGtLsW7QQ3Vaq10GGjUyW+CIadYzMuOItKU1V/C8/WN88dQCw6lGe0E3kr1YQ4mIDMYluDSwEEGeK3ZrT70puHLLLk7Mr9CMIPN11k1GKjXW2orz0i6Qpo1PBFEt4FLBpoE6SklGp/rxUeD1L9jG/ecXafZvZmRIkSb96LRGIgNR2kB4QxQLDIKVYNEU9NUC1jmETBEelKr6Zn1JhDWBli1QQSOdI8Ej8Xjl8KIgpk6WlziZoEIgcpbEJbSkI9YNlBOV7JtbQnDEjQQpIet0K9ChEHhZfJcr6r+gghdC/DXwamA+hHCod9sw8LfAduA08IMhhBVRlax/CHwv0AV+LITw7X/uIK7Zr8K9f1RF720MDVVWPYnXAl+P0QP9uG7ZO2jYkB1AIqwHYQmit7A5RdCgompKVfQsiBCqyD5vK5ytcQgTLgZ1SykrRi69ivZiuVyN/rvSVcNIoZJ8pFIE37MV9hwwwVXgtGoqtLdgekBJpDc4T9VUVdVCXr1kHlRVXVvvkDJUjVZUT0Zy1SQsPY+9/079fYPzvrFjcM6htcRaj0IRvMAHixK9k4UJ4CqeTHDxRZjYxgSwaucYJcBJllc9zzylObVWke9WloewUYPRaImBTSPc9sB5nDeU6ynrCZQix0ZNhBgCP4aoD2D7EoIfYf+BLax2F8lmChZbK8RJP0kcyLsL2KRBSZc+N0Yax3hREso6U1OD+ChmrbsGqsmW0YjlszOs5I6B5hDL2Soq1Nl92WYevfc++usDlOQ4Y5lbWqIvEtQcYBwHv2cXnfOPcvZohxt37qOsR3z7yJNcvnsL67PzbD+0g1gnKFHSv38Ca2Mu2z6B9G3y9XUunHqa/m3ThKULbNs+yYXl8+zrv0Dz2tdy9qHj1PpHGervYPIGSaMBahNoj4s0yqUUViBq49h6A1mu8qu/8ccszB5jy+QYUXmWmRA49VSHn3nNm5iI6jyqR/j0J77EztTyjhsXWWx7bjlWcOXhl1CTM1Cs856vncbaPibsOV50YCvTOzuM7tjF0soq/QM11nNJxgwrSyPc+c2TsBqxajVxn2C9o9HBciEvWc5LlCkJUQ0TRcjS0Se7uLjBemZIvMFLBcLSlwd0EFCT9OmAjhMoC0SSMB8kWjjW2zmtDLQBFyQ0FL5pmBhLGRsdor8umZxMSdOEljMYW5Kve9aW18haMZ1sjZt3DNDRjlNGYFbbDI+MUR/SDDX7GJoYpN7foNlQJLUGOqQYL6mplG7eIYiArAd8qRFRtQOOXZNIdDE+wqWVHBzZglxXhoU0qlW75+CoJTmCYTCeWFTJbFYU+BBhRKcCjkmPsB4nocRXORBuYzg8B+tohSoGVJIjRITveCJgTeaYIiExJUIGMmfoek8wHm0LnIVgFb//rvv+X6/g/wfwx8AHL7ntXcDXQgi/I4R4V+/6LwOvBPb0vq4H3tf7/s9epBYE4aDUiI0oPe8JKoFYV4W1iggBvA2I4Kv4Mqpg7covXskgQnuU0LieMwYcPlBhC6RAonGmi1TVlilIEE5etCJW5bLrSTO90G2x4dIB2ZMznDG9Kdio97tV81P0NBdXOpACoSvZyAmJjKgiBJG96joC4QnOYbFValTvzRNwFXk4PNtgrRZhiRRVAIevIPQgFQSPDxbtU0IZEFLgQgVMqzY54mKghw6CkHkcJVpH+MKgohhZenxcR3Yc7cxDGeFUSVH3yM4QtbpB02Xf7hrHjsyQWYj6QYmSITHNcpmj/SALhWJ0eopg2+TdGCHgkSNPIXWHbmedw4cPc/rpRYpcUfg6w8kone4cZxZP8Prvey0PP3mW52zfw/zcKYKrE0cD2GKVrhlhPuT0Z5u54sY93PvgA2RI1lbPEtcmWWjN0hwaQtdqjGZtBpt9iOFpjp9/HHHHLNI6jBA8trDKS2/cz1NnJafPLzG8dRuff+hhrhndRq0BeucUcSpZOPEMbS3Yu0ngBxJW11cIrkBdCJiBPtbaIxTLLUZHhljmHNKOEEUCs95GD3cRhcRIhZIFSW2QLCpJk0ny1jwJNQbHb+Dv77mb0aZk6fxZhrZsYmVtlsHaVg694AY+/cHf510/9ka+/LW/4OQZwVvf+U4eefARbr/vJPXJQ2TlMlvMPL/y/ZM8fn6WvoE9NDuGtbVFpqZ2sfu5z+WdP/xexqc69NtJ3IBhvW2YX8pZK3LWSkdNBVJhiKRCakOR+yphyUrKsstQpOlLJSY4NvdNcPb8eZ53aCenZ85wNhLgDS2TsLZU4EJAJ4HJHZp9E2PsuLzO+GjE1YcvZ2h8D6SGkaRRvWe1QmhF8P34kNFurRG6grxoUYYWnUKjdIKyhkgGcuER1KuJc1/F4OXeUubrBAqUS2mbFlEq6HQ61EwNaatFM8SOXHQprCetOZwv0V7Snw6gyVHBo8gpg0OIFGsTlGvTqNUJDiwGE+dIk0OogZUYl0OcUBY5WgYiAu0ksJp1SAMIpREGFDEhJBQipxCOWGsiEyFESRbFeNMGAX0qZqEzTxRiIiko9f8HFTyAEGI78PlLKvgngZtDCHNCiE3AHSGEfUKIP+/9/LF/eL//1eNffZkKR/6shvBV+nkgwzuNFpJQT5GNFNFXh267qpA90IvTIogqX+hSbIEQ1Rvnoje814iVocd6F7i8JHiLLEokCqyrOqYb+IEqH/Di4NXFIaYQqjeYqGBkYPG2N4gkuFiRX2pzfNbiKPHikh5Dz42DqII2HKGXjPRsNb5x8Rf1d3nRIiq97tmORI+WGSqLVwAZJMpT7YhyXyEXHNXJxFews9JUu6Gi9Ii4ifceKxJiqVlrrdNeLDk9p+i2UtaLjGRkkISMkDaYf2SRbDrixBmPE3X6ByNCo8XoyB7m2ptpdwraYoqVPLB4fo4XveAmvn3Pfezft5eZ82cx1pNoTyz7kfWErlNMb9rPwtl72HrwIPnCBWZbSzTRXLdrD61kGxfKBaZMytj4CLNtz+i2jJnTGXZpEcoVFqwHGyiKDsIEbFagBmJkN2Ni5xSnzy3RdR3ac2sQl2y167RMm5sOb+PM3Aq7Nk8xdzpj5cICQ9v7GRpKGBtuUIsKrto2zrxtIRhBqC4NnVJPa4T2Kluv3MvMA7ez79DNLC8/xMBzDxKe8KhI45sD6P7DBG0oFUR+mHPnzrLeWub4ccc3jj/N3PIcX/3830NaUK/XSUNJ/6advCJ1vPpNr+eWW/+OnZM30SpLGqVmduksR0/dx8FNm9m/aYEHjy3jqRESzwlf4G3EwnKHdRc481TMez/yRt767z7O8DgEEaFESeEEZaxJLEQhoy5gU72GsIK+VFPHU7g2upGy2imYXYd2oWl5Q9IH+/dvZ9OWBje/8RCbN02xaXSQer1Jt2WRPqEoMrRM6PaSkTAZLpTIWpOQWWQtQZQltTSlyCo2i+kVb865aubFZHjrKYXDhYrkuN6pwnIaSpN323SFw5Wg4wjjC4qiqHpbvkSpCIVHoEHFBJWj5QDeWsCTJFGVsex7kmga4YylJjWCyiptnSGO415Bl+CVR4mywmlbSxkEIRaEVhsXK5JMkntPqStNPrd5Ne+SlchE4WxlgkB5CucJrSqSUCsBQZK1M3LRJoSKEvu+X3z0/xcNfmJj0e4t8uO926eBmUvud7Z32/9ygRdC4VRAqGqxFU7hREAGg+sUONdB6hIZaiACUm7IKD3OC5KA78kU1darcnVXEg7SV/yX4NFRhLUOHcXgFaE0PDv7GXqSSM8T76uhIy1khQjwFclRqt74aLC9adaN+L1nNfeLJ5qeRk4IeEqkkJUlsvoLCBu7DMVFmQfh8EEipb7Y0FW6GtZyLvSGkgJBmt4JqEIaiAAEiS6qqMGLYDIrCBZsEfAISu9wpaOba9JY44IGWSdt9pHbglWTAw3mjWVp3dGx6yidsrKyzHOmR1haa9EZd+TLcGhnymMnVlheqnHZyAjnF5Zptbey7vvwcZtG0HSjQGv9HD4umV1boGss1gZuuO4mfAgMDtb51p33M55awuRutk9s54u3PcErX/syyvU5ZLPG9GDMdr2ZQinuuvM2xOgg2/11aHOCkR0pUlyHXj3HuXMzbBrZTTtbY2psE8ceOoVLPXMzlsE4Znf/NMf1Kn3KYheWOLf8DH/19WcYH454au5hQsg5sGsbLl/l7BmJagUOXv08jp6+jR2Xb+fo40v4/Axjo5NkQRL5jPjMMFm0kydn5qjFTRrHLEL0oeIG5JJcnyKWw5AG5FBK/9Au6tOX89jco0yMeyLb5Ibr3sK9j3yYbscRj0d055b5RNFl+LbP8/Pv/HW++bkPsHfv1Tz5ib/l0JRjz44GXbXM7ecV60WDV77sJt730c9yrozIXUnX11FKkG4NvOPnPsL0Zk1DWDYlCWuFZaqvH2/a+DjBx5I1oVhZlbRNh2fmI1SIGN0uueqyzdx442Z279nK5Ng43VqC7jq0qdhQaaoplGZ52dNZ83hVADmx1xS+hQ0VQkCqOtrFeCNBRggLmpgsr8It8BoihfMZITi0iLCRxiqH6HgSldKlgjcp6chDTqk8kdDI2GHLSi5NVEyRG0ycUpiSWNVQQlQauuin9AVRUDghKbwljRPy4NBBoQtNkiTIyKFzRWELVKNGXpTUVIKMM5wWiBDRKjKUkETeIY1Ex3U6wtHWHZCCNNRQwSKkp3SWEGmUFcQ6pWszyrKk7iJCrKs1SHiUsFhKmvRjnSPL//cL/BD/xG3/5BZBCPF24O0AWzdppI6QxvTse6BKQEcIDLIIYCUhysHRq1YFQsZIGeGdufgsFRSsygwVG5V4kBW9DVXJKLLStYWHIKv//EsboqGXWHRpFa6Uqtgx0KuEXVXVC3Vxcd/YRUjBd1TvG9+lvoQGeTHYeoNF76vjENWZXKqepr/xInp6HviNoa7KhnkxAFuIitGDrKL2nEAYCC4gcshDoOxWGrsRElNCp+3pqhKhFKbIaIp+an07kK7FQjaHMhGdvEMtFYzv1MwtZ9TijGIlZ8tYk2JQ8sDDOR09RpIOsbK2idDsg6mDyKUOujWLiWOMz7j+hhs5+0yGaUcgM+KGYGCkSTNtsmXPTr5895d56MQxXv2G6xkQMLEl4e677mB43yA7mvsZGdhCZ/Yob3jTD3Ls0SfZNrKdpflzTI1P0V1+htBcYrQ5ylx+gtq45Py5VUYP7oYRh+gqxrc0OLj5EGdm5ggX1njzmw/wuY89g5cxI92YATfMskhRZc5I/yFmVh4hiRo85ZZQp55kvG+I2Zmc/PwpfvRXbmD55DOcnW2RdZqsL8wz3Bxkaf08cstBypUW6c5+5LrDyBWiQhNYZ2nhLPN3rvO1Zxrsvfw5rLbW2Ld7F9+YmWV9eYaYgI9KxHqNSFlGxrbQXhmjsBF9o9soiya3njecuONptoxE7BirMZ85TtqCo5+4n5OrCl9PMMoQNxRYjfDrXNaMGBY1iGF5psvIVMSyWqdVxiyUGa1nIrwwMGwZHFT80Bt2c9mBafYe2kaZxgy6fjrBs7BeUGtHONFGN5qAIiPHBMGArINrY7wnjuogCiISgsuJvMdZiw4KhyUoQYnF+mo3G6UxzgWMq3JIpQCEp/SymiKVikQmFEWBiqDwAVlCI6lzfn2JmmwQyBEix5aeWtIklhk2gPBV4I8UPRm2LsnLQCQkXkpWTcaQqlefLeXo5jlaxRB7pAhYUyCkr4op58FJSjwqBLTwCK/Ig0VhyDE0RYLxjix0iBCYLCdNIlwEq2VOf2aJQsAJT64cFFDKQKwTVlst0nQQmy8jlayMD9/l5V+7wF8QQmy6RKKZ791+Fthyyf02A7P/1AOEEN4PvB/gmkNpkH2jYAvCeht8p9cUtbBhGYwEQnt82GCmywplG8zFKcwND4nSmuBsz1NfOVWkrEb6vQAZaXCmmiINrmKt8+yiLKUAYSuPeqiypoSzVWO010jZ4LNbb6vuOs+y3H0PXraxOAt6w7lGVCwaqFDFISB0pd1XJxXR26FofKji9pSKwAm8N70TzsZ8bkA6EEFXbHlrKqyOC8jSEozAlQJTBlypKUpD3klYaWWIFKRIaa0YSpUSa0n/6ATdchu1+h6K6ALl0hKLrqS0Mavtknq7zRUDg/ikZGyX4Mv3GOabDcTkldQG9lKomGccaDnAkBD0jY4zffhyvnHrXYz0j/DNux6s8KqiQ+ITGlHKrV/4NN7kDPZvYbuus754ikfv6mfT5h0cvO557JiY4r7Zx3nJa27kT3/xv/EXn/wgXzn2EG/+sVezeOoZSjPKDZfvhv7n88TxBVaWYwb7xun6ReJmk7NPL3Hl1r2cmrlAkgxz3sPptRal6XL7kZMs6X7W155iaPce7j/2ANBmangXn7j7ftJRy5U7dvOcxjZa611mFzJOnzjG6162hZBJPvepr7AurmHPpgaZvsCmPbvgsTupdUaobd7M7PFHSZsRE2PjrJazrF/oUht7GY0dMdcPzBNFKdtuupYP/s6f8cTTd/PM+Tm8tOhQMNvVbBkWLC/N8vtHH+MN/+Ht/N4HPsnxZ5Z4etXxl3/yc7z9l/6Yka7GFAkiMtSyZdK6wvsujRCjO4ZctvE1OFF4TLvOcmeNh7/0c7zmVX/CC18+yWXXXIdLZ9n/vJuJrGE1tIl8zFq7RU1FdLOYOIP1eosoSJT3uKRL7FOcjxBB0gQKleMjgemWqFqCDZ44SIyr5kHyIiOuxWitMAZs6Yh1DZzFKygygxKBWCi8SzG+Km10KKk5g4sTSm+JlaRwBRKBUhF56dCigaFEiAIZKlSwCQ4p4+oEQoYOMVrUQRiiUpJIXQ1ROkUqa4ggscFVzjfhiX1v6DBA0BEiFXTznKasIQpHFjlSFaGCpi3bRJGia6ApanR6J6TgLCWgfIp0Cau+oIbvMbACRVkgnSBO+sizVawtiYXGiAIXKYQTOL77Bf5fCzv4HPDW3s9vBT57ye0/KqrLDcDaP6e/AyAgKEGIari40s+hkmCkiAlBIZzFq4QQBE6p6iyqqoqgik8VFznrwRcgHIiNhbZqNrpaH04J0AERRwRV7Ry8AHExj/VSe6JD4lHCVI3R3oSt1BURMghPpCpgmaDi4lRwLoF1rvfyapz3BBUQqudU8RYvwKMJvUEkIRziYgBHhSDGabyT+J411HuPMxZnykpbx1d9A1e5d6SVKOMppUaWAdeKKE1E5ix5oVhey9GdGssdmF8qObPu6YSCFVVnNoe1QtIxa4z07UKrMSK9GaUFpi04fU5hGnVYK7nnqKU2cgXDm19M/+S1mGiSTllVXEW2Rp5DIzgeuu9+rr3mKiYHLqOzLmk0U3bu2I+NLe3Ccc+5Nd5751dprJ6iEdao1zzDw/upD8aMDtZ4/Oj9DOzewWdueYifeMcP87W77iQ5s8bYnOXU0+fo3zZBNlZjYb3DC553DcN1w/COScZHtjI4vYnRqZ1cKDRRPdDNzrF3eoJYxExO1zl+dI048gxoj1lZo2y36WuOU5aGTneRzMQc+eYjnF/LOb54jpmVda5/0au4/8l1HnogJ/cvIoo0L/zBw9z42jG66glsupvV1Vkyt0pfohi96jDz844omyGNzvPMsUfIyi4jI46lxad44qGzPF3MoEVKwNC2BdDHNmlptFN8KIjiOq9404/y4LkOk/uvZs+BHXzwlluo10CLiFh00TZBhho+0qxLyWpsMTWBjhJmVwbwLdg3Yfil79vBVz/4YZ5Z/ALv+413cOXlk+zef5C5k+c4vbgGa23a3XXqSpNIz2BT4ET+7A43UgjjsMLibIZ1XTpC4EITkztkmvZ2kQajq92vlZJU1whloN3KcNago0DwXYwPOHSF3PARTkhCZEnrDu/bVXaxqlWTnj7HRVUhFfUWSWtzEgxaaHIXcOSQdCgo6ZYFSVrHhwKDpwgtSh8ogNzkRDrBK0cmOnRDSYg1SmiaUQ3pSmLpQEGiQOcGjaKgxKaeJJIULtC2BpTGGWj4CGc82BLpSxoioV/WqTXAskJTSqRNqtdRGPoTTS2p07aWWEck9RpORZhMYFxB7nNE5P/xWvl/8/LPVvBCiI8BNwOjQoizwH8Ffgf4uBDiJ4AzwBt7d/8ilUXyBJVN8m3/oqMQEIJBINFRUk2LRlkFvvJUDpS8hERXrBbj0U6BC2iV4qmq3cpNY9FS4YMG2a3cJ0ojlSb4DkpQLY4bUgmSICr//YZcAx7vNBuo3Y3q3vcmVL0NPT5O5Yuv7JLfiei9qKfzrLZeDR5tPC/VwFMvUMnTM/DIyukiJNUkpaqoY8FXuxY2OgbO44wCB9I48AFfOnwZCMKSFSlrtqSbe7yrUq06KDomY22xjlrJaY9qVCl5/hsPc+KWBzg/D4v5AZw9R9lWmLyJ7ldM1Wrsm4pIVxb40rEG8c4X0q5voyEHOTXfJV86jdN1JgaH2HrZLt7yE2/mq5/8IuPjExx57DS/8JNvYN52mbrySt76htfzwoNX0zKGT/3WT/LOy69ncvsB9tz0KlaMoNU6yg/8/K9y5pvfZNPwJm563Yt54Stew1TyNq4c38zfPXAbty6vcvX3HeYrn/wAV+65kocfu5PfeO9fccX1O5k5vkxLllym9nI0nEXIDscXBQODm3jksYcYracsdrfy6stP8BO/fh2//rNf4799+Czv/f3f5jd+8Vc4+CM/zYkjD/Cya67nU5/9CF868SjhbMbNV09y1xc/x+mFfuZnjzFyaJC9O6+lePTL9L/++4mPfp6Z04bXveun4akv8nRnC+V9p/ChpLvqWbczXPHKQ3zpv/8qa/oasu48z3uJ4VjtHH/9+GPUh8ZJco8SBXmqCe0l/tNv/R6T2Xkeu+NbbHruK5i48UZ+6+0/yfLD6xwYmwZV57rdKeuuZG6xYGVxlnY3RjRLrtg+zI27L+Olb7iKzCT4ImLL7lHa3YLPfvi/U6+nnJvPOHPyKP0vOIzuLlMOTkCZYftislJR6xSVBdJXXS5pPXFSgwBRj/ukZBUsk3tDFASRDQQtSJQkx1QVvKgCY9AJGodxQNAEHbB0cFQNaWMq2VMpjdYxWTA4HypESBQTlRYvPKVpYX0/XiaUodLs+2Qdg6XMI7QqsDZQGoeQMcYFoihGWXBWIYTGGIOMLAhJEjRlVhIiBQqCDOi8oBY1sA6MrnpVpStRQiKdRwVHEkUYqfHCk9uKJa+MoUbEmltHSI9H4WU/3nvqGgofo1SNzHQI3jEiAp2gsK0CH9uKxpWlSBQmfPcL/D9bwYcQ3hxC2BRCiEIIm0MIfxVCWAohvCSEsKf3fbl33xBCeEcIYVcI4fIQwpF/0VGEgHMC5yoJImhfOVKUqirljaaphKCrgSQpPE5YPAXVeEK1yOErfnPwBYI6IsQElxKsRISimtp0la3yIiJXiB533eNF1aW/5O//DhfMxvWNCdcN2NcGMgD4B7mol7QgxAYczfdCv591zGy4cgSqyp7t/Zv3VQjHPzwefKg0RRsQlgpDYCsrQquAjsmxKhB5TZFJTCdDZgbtBmmHLuM/cjX5guE5r30TrZWYgwdez8te9wts37yN80+c5eF7HmKpO09jYJL1Vh+jk3XOLBoam28gj0YoxRCLpaPoZnjVTz11bN+ygzf+wPcytanOxPYqAOHB47dy5JFj/NrP/CZPH7+Po3PnGNq0hZULS3z6PR/hiiuvpF04/v1/+CGmWWG8r8n8Xd9ies8UT62e4RN3PsCW2ib6opIHz32LxXPHOdu9wC0f/RJ9gxMcnW+x2BrEScgKR70vZWLvLlaiBbbu3sL84gmGBmts27adRI9B1EDJLvumS7pPf4W1b3cZrsP3vvoqNg3XaOYZT598lExI5ldW2comHjt7lkdnM86Viq1jM5yaPcGhPVfz95+5g3se6GPms7cz8txrWJt9lAuf+jPs5H6yM6fJMoHo1Bh65Y0sdjTR+sPs3Fpn1+YBdu1+DrfcdiuZrTE5vZP1pVnS2KIbGlYDPiR8+gMf5O8+9Ldc/9w9zLbPUK7MMGw6zEUrlN11+hHkoYQ80FmdZ9049uxUvOr5O/jx197Anpfu4eypZeZOHieKWqwvzHHqxKNsm95PHBIu3zrJ4cuu5Mx987iyj3ZnhbRvDMqYmrSVF9tLoihBKlCRxAaPlZALjyVgZSDD0qzVqzCYqJI/WraoDAnuktQ150EWoBxBuYqa6hVaWHww1XUZURRVtCYKhBZ46XA+VvRqCgAAIABJREFUxwVPbjSxHiBWlkQKUmFRiSU3FlkmhMRgo4goUohEQRwTJRpjClRdUZplAjlSRkjRIFhBQQnKIiiJcdQ8RHWFCxoZpygMPs+ryVVbNT6VUpTOkPjqpJYgEcET12sYPFGkkUIjjCYNkpq1OAFKGZBZZfEOkq5wCBkhdELiQBhHKiVRCITyu7dJqne/+93f9YN8t5f3v+833/2zP9RHFStX9BqYFZBL+oCSlQNG1PqRUVoxWkQ1pamUIrhKWpHBI4WvMAM+xYt2D+nb89WHGt62ERTI4CFUg0hIwLmqmva9BmtvMErK6noIHiVUNRUqfNXADaGHLd4Isa4mVpWsMk8r7X3DV0+v+O6xaHzVxHlWp+8FYktJ8L1c2KCQIqo0et9j5XhRAdKsQpYeYQKu4wlGYoykKAXdhZTOisB0HStLnrlZTWtVMPXCtzB4zStoDnh26gQzMcXqbI2vfHuVv7/rNr7xqS8ixxLe9+EvMp8lPL3q+fQ98wgc/XaFNbmPfPRKuvEgrULj1wyzq/PUabJl+y5++b++g9GpPrrdNu959+/yoh94Kfd/4TaStTY/8ytv4W8/9BXe/qa38pm/+xCjtZJTi4/QWfoWMhnjbT/2Fu558kEmB6boyiEOb72KSTvEXU99net2X82Fc7Pccdc9yLhDf+7oDg7jVnNu+cv3M7FzN9mSoC8p6axndItVprcc4sjXb2PP9n2M7tyOMW3WWi1MUvCjVy5y/dsPcOL2+/jcLRd47395C9e+7JfZe8X38KLrz/D5P381v/hLf4IeGudCZxafSrLlRQZUzh/+9ZuYO3WC2//mTn7qZ17I5hufzyN3PUxt5QI3/fS1DBzcSvtbf8/UD72Gs/fN0l56gskDe/nGh/+akYM/x/EH7yAdfylbt44T+wE++en7GDQlE0N9DI00GKxpFktDLj0L6ysM9cHtjzzO9OZ93Pa+9/PN1WXe8sJX8tTJ43jXZW2pQ0O22XZwjOsPTXJwch9xLaae1OiurjI2MkHfpiaEiObgGElzmKnpbeza/zwunD1NLfJ87At3U7SWOXz1NeRr54n7EogTGlFMaUqiJKr0bKGQuvosKqU2oNNoKTH0uEsegvdooSqbg5I4PF5Wn1VX1rDCY12JlFX2gXUVwbTKQvAEGVEKhTAWERzGBRIiCixxlFDkFmtzhM7IugqthvB2HRN3ieQAvswRTiOsJwopAUdNN5A2ELwmqaUIUWJsmzhSvcAeTyzAG4f0klTF5GQoXQHQnPREMiKOqh2BDQ6VRuS+JISAFiC8oYedQvkELVOCshTBUAaFkJbSaYyROKlxIpAJT2QCzppqhgAFUhG0IjeG+782N/fud7/7/f/atfW7Bw7/P3AJIWBNF2fbVENFuprsVPIiAz6EADYjhBwhDVAgpEF4A8IQsFSALgU+qTR1m1YgMZcibEpwXYKV4BMqpK+pZJlLKmmFQAZZLfC+ctuAr5AC/pLIPlE5XNxGziob0oze0Fr+kZPmWVZOjyjZm3i99HUQoTco1SNUOmeqpm3vuEII1XG4ADYQSrAF5FkgLwS5BeMU3TwwlwGuwct/4S2Mvvh1pPuew7c/fxv3fv0xPvap+7n3jgf4+G2f4dTJu3nqiUDabfI/P34vA5u3cMHlPHHcMN0c4vnDKTPLmixRtEOTju0jDwmWFjUJo+M1jB+gVo9wmeHMiRWm5Ta27TrE9TfdxPN/4FXMn17i6OP38OiTt6HNIkWpGbQRnRY8/MTX2b51J5NNz/CeqxFJH8fPnWV2eYmXX3EQW66w5NbRypNE/cx0OvzQ617GTW/5fkTa5ZGHbmXJzeP7J0hGFYenD6HOPM3kjgOsKU1nxRB8P32NKequyaHLY+SFh1H1nJXZGh9/+Ag3HL6C06tH2NG/zkOPH+F3f/Y5jD/TImlHTCJZXA2IPXs4+8TDXHnFOBMHB3i62Mr2vfDYsRnuuuc0Uf8g97/3I8ytLdCZX0avn2V662bu+rPf4rKDe1m4628Y3Xozm3eMcaHleeTEOZaKDDEU2DnS5NDIJkZzQegW1BPI1gxHZwzbp/fw4f/zLzh59DTv/9j/5JEnP8fLNje5cXqUyw8P0H9gjCSus7AesVLOUI9z1opF4lgSdEJfNMlA2kcNRb8KNHzC2twpBCVZ1/Hq6/ZjveWO2x/DMUKR5WgfKAtPpGsE52mkCZGGKARU7z0bI6uJVkA5iKQiQlPTKcoKdKh2o1HP1h4DQWUY230Wx+0dkRIYZ/C6gnW5skB7i+1V+UJajMwx1iLaOTUhCVhkPEARB5zvIkVKFBJ80UYqMFhMDCUGl7cR3mF9QZrGGFPlAmtVw1kNRhAR42VE1OhDNwfoBkWcKEyZE6sa0lSf3zzPQQqSKIXcEkuFFAJjLSrSCCdIZYyVOcbnWAs40EkX5aqdd5CKyMCAkfR5RTctcAMCncTgHKb0eAONJP2u19b/LRb4aqw/QkYNUHWCSAhxHZnWq2aoDATvkBhCYXqMMYEgwfXSjyoyY2WR9MERgu0RG2MIJSFkSKlRSQo6wkqNTBoIHeOIQGqCDPhQIvCVY6YHAdugQcqoCvEQPYxAD/WICBthIL7Hp7HVz67XvO1V+b4n50jhCMpU209vcc4AvV0GvZOIsHjRm5QNEhdprPdI61FSEUpPJw9IW2dhVXKh61lZM6zNKs490Wa9NcVK/WXc0bmc9/3B3Xzg05/jD3/9D7jj5DHObn89R/oP8Y0zu9l5/U38+VcWuW9ulQ+sSO54/DzkyzTiJi+9eZIfv2aVXQdL+pqjrBcDrJlVIt+iZhcpvSaK60yPTXDNzrga5q1p7NIxfvm338yZ8+e564GnuP+JR1gMgxw8sJ2jR+6hNrKZ0T3TzNoaWW0nWdngV37tN5HpZq6bnOTBr36ZD33kL/naXZ/i137v9zifnefYnXfzqhcd4NRdH8Uc+wavecXNPPSNI3SWlnjBZS/me25+KUurHS7bt59spGT6uVcxsW2EgdEGtlzh7BMPMzJUsLZ0jMUZy8zdx5ke3c5XvvoyFp+IObyjweahhJ/97YyHH9T0HdjGlW+LGB+fRAxPsm1a8bVPPMx7/kdCPH+Kn//1azgw9kE+8aGnyM5v4jlv/mVeP/YBjpye5v77axz/6p3UhgMzF1aw6krWyl1ktXEWFk5z6uQTDA+kbLniCn7r/e9h875reHjFcM0bf4J7Ty7xY9/7IrYNadRazlte/gJecGCYd/7iz/GJmW/zyf/0U7xi3wTje+ocvHKAfZvaxNki9dgxFOd4HbGYgSoF6/OnWLxwjNBtoUNM3C9oDE+Qk4GQTIxOEyeSe75+P089tszqzAx33HovfV5RrHShLrDKkruS0gW8U1iRgJNEQVI6ixMCU1Zob4/HyhLnc0xk6ZLR8QajFU5o2k5ghCCKEkIIGFdVv4UPKKmR1uKKLmiwElKVYkuD8QpjNUNRjSKWdGOLjFJMt6Qu/i/u3jPM0qs8071X+L5vp8pVHaqjulvdLRRBQhJgASKInEw2ZmyMmfEBbBNsg43HYITH2GDwGWzBAAacMBibHCSLIAkJISS1Ukvdrc7VVV25atdOX1phfqxdDTPnp851HZ3Z11U/qkNV7ara73rX+z7P/TSoeojQOAelNGghSBQoR9ibyYi0bEMiyHo5jhIvPanrUcguVW+InCFSHuNC8U+8RJZ1IlGlJEXHUV/J5sAbMtdDVBWx0HhjiKMIlysKa8mUIZIK5SW2tAgtGPUNhPdIChJdokVBFhmkiEhMlTiPsN2SUoZlbikNhX3sOvjHRYH/+aPPRVdy/T1QGnRYrtoysGrwCkdY8uDX81B/vtyU63REz7mue31Kcu7fyMCXkVKgdB9PLEQfHRA+5v8+Q/c2XD9/rrLpz95VyHNd/xjrn0ecG91AQAmHEZD3PohfrEMIhVIRQmnW2TqIEBAufYTUYZvvTJ93IxRlXoKFegFLnQ4xIX2m6IGrS8Zf8Rs86T1/wbHb/4PyxD0crexGiXHS0d2MPOPXODPX49isxW0Y5puf/hGfvv69FAKuuGQff/lXH+Wq5/wqs4s5//r9E+zfv5eFM116Oia1OS6FnjEsdzTdsgdxgVMxeX0TXjj+/kMfYdP2rfz5X34HzqTs2rCL2Jfc+rPP8ZyX/CYnp9pEQjI8ELN1cjuDE1tojG3jvZ/4KGO7d/Ldu76OmKzx6J0/oNlM2Sq20y2HKMYStuyu0FuC3FW59tIXc+eXvsf4rvP4zb/6ffZu38L9B0/x8MGYhhrk2KkVOtNr+FMZ4xs20IoGmFqdYWq1yzdud9zzyACNib2c8l0WFhY4OjeLzR3nn1/l9qPLnHj4DEcO9OgKR3VtlZWmQ9Y1Tm3ms/cPMDg8ya4LN/Hqt7+SzZseYMjcxvs/u5vuyjxveNsnWTw+i+9VqRSeoVggk4xEDqJrA4wNVFmZmqY9P89XP/WPzN59M9Hcce78wRe5+nzF0EjOS6+4kC9+9WPs3lPh3gem2XX1NXz1i9/mbKfLf9yqmJoVfP/2Iyys7cGmI7RTx9nZNpGLGB9Q1Mdidpy/j00bRpENkIM9Isaw+Qqm16MiYwrbozQxz37us9k7XOfYowucfnSaB45M0xhTrHZ7+EjilMc4QmB7WSK1onSGShShpERVdJjHE15nWmskkMiQpeD7GcrGlSFwpr8Dk1IjpCLyGpyiRFGqiNJDaR29Xoc41kQ6uL3X2i1wJoxisUhpwKUs4VFJTEVKEhmR23DbNrYg1hGxqhKJKj6DpCYZkJJGAUNuiEE/Tld7yhjKQuAzSVyUID2OLlI5ylwGBo2LcKKC9IrECcoyp1d2kEmElRpDiRISbRXGKYwE1RAIHTFvDaiYwikiUSUXCmSCLjOMlJjUEqkYrSXjqkbNSQrTe8wV9fExg7/h+g+8+VUJQqlz4xipdRhZRxKvBMp5nDXhh6oUznukL8JcW4WFXhia9HnxIuSKCgQi0gitcUojI91f4MowSXEFOIcr8lCIg36RdQWM6M8SPSLMw31QtQSTU0AWeG+Dbl14pOordPp0YLFusPLi5wcD4XBhHd+77pTtUyjDYVCGA8ErvLBh8Wo83kgUEeSeZtdSZIreimCtGZP3NEvLjpu/eQ8/OrLEzMhz8RsvoDm0jWzDk9mx+RruuO97rB6dZkRU6Z65l93Pegq3f/s/uHh8K8ncSb7wr1/HLR1hJXIsrgqeubnHSmZYIaJwAqcG8LmmbVaJVYIqYGhiI3sv2svb3v4n7G/U+dg/38h//cQHOXzoKL/3e29myXb4h3/6CtqtESUVTs4vsmfX5dx+403sPf9JuMkq3/rbv2P21BRf/rd/YeHR+9h02UWkeYvZ+aM8OvMQ48bx3a/9hA1P3ElX5qydPETh53nZO36NITFB2pnkSTtKPvWBP+Kr3zhAGS+x7+kX8KPv3Mam+j6ecOk4n//sP7Bz2y6W9RBT+Sbuvn2OzuI+TnRjFltwem6RYWc5erbN1248gbajXLmnyue/8Hvce9tBFtor2MUOr/31HezfV8e2EuZPa97+Bz/kgt2O1uwaQyNbSd0Mm4eH0TrGDsLg0AADMiLPBGMyYmDbJkTWY+OWId7w6heyb/9+fvnN7+TBu7/O61/zIjY0YOrEIlO33sjOX7qMSy/fw3vf+yGm77uD3//4h3ney17DwGTEzPIyt9/yAG1riRFc+sQtmNVpXKbJbIESDqkLxoZ201prUxuIsc7R7s6CiKjEY9TqDaz1TM11uH96jpW5KktrLRZmlnniZRdQdjNqaohYerQQeCnQ/Xm5UqqPbxIoofsucEHhPErHuPXOFoe1JUL6EMdJ3/fhBcY5SikxzjEYaRLvsKYgjiTWWby1aKFwxmKcJVJhZBSEBxJnDYkQWOspvMXICGcEygeJsRA6KHPidZWcIPcxPkogtuRFC2ljlK9gIk8hSgqpQm4EoR5J7VGRAiPRkaK0XRJdQVMjsoEmq2RQ1+EDAiE3OYmKkYUmLgMbCwoiPEJaeiLDmQIjNbr0xJWE0hlcntPCYqUkdgl33jz1mGbwjxtcsBS6z3ahX+SDksZLHQqvsaiSoBrRMdKvF90YL2zIQu3P001RAj+fFXoBQiqEVOECQB+5qwThh9j/nAKEdGHhyrrx6eeF2dlghljHDwRcrz9navL8IncmPML7nPt43gdcgHCuf4MAWA/2WP9fJgCNIKgOkAjrccbibHhe1jpcUUOaHislSGmY6W1HqpIzxSqN+hWsnLiR2W6FXVe/lMufciXHjhxmixjgvnyGSPRo90oO3XwXlbGIqyanqW92TG6vcWzVwbxjT71EtErSAUXR9pS+JPc9nKwilUMWKS4eISu6tM4cY4vq8vx3/wZXnl3jz977Xla7loeO/Ri1uMze8VE2j+ylXt3OPSe+xXs+8vs8dPAnLHVOMnPfw3zkrz7Lk598PmPGMLF5B3NTi0zu3IgtHXvHRzh14hive/3VPHjvCqt5zqYLN5C0h/nep+/k2O4F4u3f5YrzzmdmtOTtv/4imukM995wE2PFKJWNDaabbXaOnc+gLxmSQ7SSGpkb4+FFqDfGmTCG5tmYrulgsFQ3DhF7z+69Q3jR5eDdp4nGNavzc5w+mqMHLqI38yBLj57g+usv5WPveYDfeteLafTuZPHegu7QBFv2DCKWMiaeOsDi8RUGooSulYiFaeLGTgYnRjh+7EHue/AYr9t3Pk+45DoWV4+icsXwIFx01avxacq9d93IDX/3KQZHKsw/2GLndUP87Xv/lQePzrHpkj1Yo2i21vjpg2cQ0lKtdNk/FrOw2GRsYBNTJ2+hWhmkmShaaylj27dSH4qZnlphcWmaIluj3Z5FOYOop8gFy0zSozW/xPBYA2MzfKJCHkIhqFTjEApjHUpGeC+IlMB7ibUhdtIUNnT63oVmTIa/s4Swd0Q/CB6P6+U0ahXKPChlSGJaRcpgtYYpSkwZTHxxpYpzNsRs9vHYQknwJU7Y0FzZDINDRiIkJ5UOLTS26NcWQl0pXRBxOCSRVjhRUNUNjItJYo91HrygNDnVOAmySqHRXiNVg8xneJviazJgi60irlbpFAWp90QyISvTYKCqKVxPoJzGGouxDqFjFKERLRNF0U0pE4XSVQZaJV4oMv3YjU6Pjw7+U//tA2/51Y1BQSLD9Uh6109uCsoSZwpwnT610YCweClBGSThquexOO+QKrwPLhxhUQw6QlbigD9Xso8C9ngTTErSmXCoWNdnx8h+MQe8C8QbEa6Kdj3TVEehmLuQAiX7wdweH+byYr0jX4eRORChP5daBlW7NzgfOn7Rn9F7n+BkFr4ProIwIqCNnUQVYHqGXkfS7Qo6C4bzXvExoqf8Gjf/49/z3cWEq97wVkRjHJvvZSYbpVZJ+dlN3+TIj77NUG+ZuamTjFIypAWDmw0vP38no0mPsWSMJ1yxi/P9PPnZkt+6biMrWYdSwFqZkHlPpywwuoftgjMgKoNUqwPccttdbBsV3H/gESaHh9i8ZRSiNm9609s4dPABms0Fmi3L1l17EKun+LeP/wMrS6d4yVUvZU16nnbN07j/wKMsza0xMj7Jlp0v5o9+98/4ztfu5KnPegpJNeOq3Zs5eMaQ2Q6XXXUtU7sjLti5nc6x+3jOtVcy32xydn6NL37pK2xWl7Dr2U/gkpc8ndVDx9mxfRdHHrmXzRu3sWX3XuoqRiURpU7RRYGTiqnFEyxnwXdRrZWYqGDh4SWWV6e49cwyaWHZtm+S5z/vfbijn2Tyuhfyx9d+h6dds8zECMwttWn1OtgU3MgE4xOjtCWIeeiVOaYboXsn2P+KZzA6XGctTRCl4oK9e6ht3sNlz3gBUVnlsgtWUbpBq5JjZI2L915Nnh5j3+7L2XFehft/cCv33/0oV153JXk7RaguJk+JVYKL63QzS7rcRYsIL9vIpE7hCyBDaUdvbY25hWmieIRt2/YzvmmS51x5FR/5xPdBOZbbKbYpSTs9nF1j955NSBKEM9QqVQwWL0FZh7OWpJJg+q8fHUc460i8wvSdmOseEaUiBGCsxWNDWIZzSB2ROxOgYjJClB5tCItXGYQPRZkHxEHRRitH2F16jM3JhEQYG2B8zqGjGG91H7Bn8a5ECovwBSQ5iSjRzlNRFWIRISKFlhGeHoICbxVSVohUID9iY8AjKybUCiq4fnqbtAnKVkh8AnmKy9s0Yk2JDyo4BJoIaToY7/AVBcKjhEfpBO3AiP44xxiEtbiojXU5vpvxkx/M/B+goiGggNERQiuECqe9d+Z/0ZSDRIronMoluFwDitc7h5Qy8NqdCwW/3217KfjFrekvdthOOCw2IAykPLc8/X/o3nEBDKZU37gE3jicMefm/SFU438nS/a/dhHyW4UQrOOBrTHgZVC+e9lX10QobZEiRqDwLsXbsu+i9eEa6EIAVdoz9ARU98LMwe9TTF7Ky5/xTg594zC3/evneN/vP4O3vWqc4//wCaqdNaqJ4eKrdrElidk+VFIb1pynJPNTZ1lIO6z0zpKfOUHPleyciKjUHOlggpENbOYwmSY1OXQLKnkLWGOjNqyuHWVkS8LzfvkNdLoFX/rO/Wzefw133XQvd/z4Zop2hbSds3NXwp23fpe1tE1TnKYuPd/+1udpd2d5xhU7GfdLjOQFjYmtXLW9yYUXFbxgX8k//+mfcO0LX8b1X7ibnS+6Ghv1sGvzvHJwG3ufcCXdHRM8ePgII0nGdVedz+tf+Fze/qevwxWKm//Hv/Ds5z8F016hFg+xYWgjT37WtezefwlJUqVarZJqxYLPGXLDKJdRyRIiW0FnBamu8tnPHaEUgs1DA2zfPkk00mbHnlGK1jD/6Q+HefJr/paVI5LBTptICoZ2DSHdGU7PzdJb69G2JVEyjvMnGX3iJfQO/xS3lLF44hRC9OipnG6RYWcPML5tF2laoVnMs7myn93bz2dgaJT9Vz6b1aWz3HDDDfz1Jz7GfPss9/7sAHHWo2y16ZiStZ7l+PFTtNs9Zue7nJhdY/FsB9np0bAxA9EwVT3C8OAQ5+3YyubJDdSqmqoaYOrwEWpjmslGwmBd0lKedqo5dHQOJyp0ej2cE2SupCwLpLPkriCOI0xfI65EWLyWWIz24Zbc93CovrnI00dw+/VdlUJ7gzYlujCIIgt5Df0bbZ5n5GXexxwUOKsQJPhSgFOAYNxX0KJK4RIsiiLLifKSqDTUdIxCoaIEr2KMGyCTYxRxQq4MpYDSGwrbCUYrH6GFwxiLyTVlAUIVxJHApwllrmmbHpF0JEqi8h6xLMiKLjoZoFYbx1hJjEHKghKLwNJTCbXKAGXpcSpGmojCeGQhcWVBTUlcJDAAvYTINyAZesy19XHRwX/mk3/2gTe/vop1OQKN1CK0hwT8rfM2wIJ0JXQDgqAVlf2irDVO9dnqKITuHwpK4nWETKqIJEbEDo8IGl4nAI0v87DbtRrh1wM2PN6YoGEXYZQjJQgRhW5+3eHaJwsLGa6nPzdBcY4guU57DKe57Gvhf0H/LgIAbB1w5oVHxhqTl311jg4Jf5nCZ4KysHS7kK9FTB1VNC58De96y18z3RZc8JxXcsuBg/zy9e/kBW/5XV767Ddy8NFpXnvdFQyKWUbUGm+47hqm5g9RqY5hREYpBKpaZ7DWoFj0rM63SOIafkxTCAs9QypLGtUBVLTCrq2WMdlh26bLeMW73slfffiDNOiwdvYUdx44gK1VECeP86Nvf4M3/sa7eWjxJAd/9BXiwUkePnQPaXcR2SmRaNqywo5d26n4MVqbdzK0aZyjJw/QO3oHVu7lhr/5BksHf8zcyCCVDRN89qMf51PXf46No3sZ3beDhw/cxejICC991Us4eMdhehODNDaPcWKqxWWXXsRX//tnGD1/K5dc/kTu+dm9LM20OP8Jm7jmabvZsnmcEwceplt62s0FZE8z1z5G3BXkro0wlku2bePMyiJ6aJjYRtQao3zto5tYXvwh+ZLigR9/nahQzE3dxO4tJS+//rdZvvunDI6OM7swRiddY2xjlWSkwWBNs+eqp3L2oVvYdOEb6Bz/LmLgAk4dOcLG7duZHHacPHwHiVlh5FnPYqgOM49M46VDVgY5dnKVQ2em8XmFWxYEer5FY2CMu87Os1pKui1Hy0cMbNnHmZ5judNmrtNh8ewaR88ukbUtadSkbM8Tj2yjKAfQjQqtNGNx6SwrWZt/+Y978FqSoUk6GcutBRKh6C6e4OW/dBXzzTZRtYYUoalKkgShEkob3KzGBLqq9LKvGgtqNOM9hTcobxH9TIWyLHHOobUi9YZcgVMSZxxRHFP0bweyb/Sz1gUarFJIpbCuJIrCzNtQkltLXvQoTU6tXqXwgswJhMipCYHP9bmbvfQCYT2VKOC+wzpMIGxErdLA+NBJZzKEdWg9QK/MqSnQ0qGdQwvAOVQtIS3T/nllMKogx+GQaBeBFqgoIuo5WqpACU2RFXjh+9LKCGktWVkyEI3gS0HmCiwe6wruvOmxzeAfFx08BC66VklgtnuPkDWErAQpIhJEJahroj6nZj3Kz/bDt71GiAi/zlkX6pxaJgRiCCwJQvd16pHA+TTcHESYzTvh8NIjo742fv1rccG56vG4/rJ0XY0TTEviXHEPlEvf18nLn2st+QWUgdTI/ts57b3QSBEjRUyZOyKdoLQGF7oJH+Xk5OSpI0rBZoax576DT37pdnY85/lkI5fynVsPcMvBGSordT7/wT9HiQdJOitctn2Yetxjz+7NHDqxQl0Hnk2SxGRZxqp1mGiIY1lMqSssLHfJV3rYpgEKdm4fpbWyRjKxnWJpCJ3s56JX/xZXP/laWrQ5evIQD08dZnpqikcO3EZ97yBbn7aNL37m4xz5xg9442veSklBbCuYlkVvqFHWRhAYuisZ6dmf8YxnXk1t62U8aeLNvPiMgeUTAAAgAElEQVQ3f4/Ts7cizO2cZo23vu3PqDYH6S7GfPILH+EpL7qYY9/8ERdu2MZPv38n3//KzURqjCdu20nr1CLN47N892s/oNwyyfT0EtOnmiw2O+h6j+ENo1hZpTJUpWvXiFyPwoCL2sRRRLs0AZhlcx6YXkBXBAN4fAwrvbM0tm7n6c/ex+yJ0+zYuJ9D9zme8roP8d8+o1mRL+LRxYRvfn+Z3K4RF4ZuRyBMg3ZPs3p0lsbQfmZO30yWT9BqnmF4rMn8wjHm50+TDE0yM9vj7MGfMfvgIyT1CqMTF1KtD7N74wh7hke47PId/PELLyQbSDm5NMtir8XZ1RZNGbO0ljK1sEzWE+SNzazJcaYLiUkanOos05xeIRUVYtfB9ZZYnV8AegwM7OCKJz+L7bFmtV1Q5I6sWqeRV2gudnn0TMGJ6UdpxEl/bi0QcYXcQct0sUmIzgu7KRlkxd7jUTgvET6YhZRTYQ7fp8Fqrel2uwjrqMsIbcNrqpum/T2TDdAxHaO1JlaatGyR0qOQlp5x2FJDHubb9UqdRlKlSFMKeqgqlJGipTxFpSSjjfI9JCVSedp5SuYMtu+2TaIEVxoQhiiOGSRhEI+wBpygIyw95aESkzmHkZJuXiC9pKorGEdIUTOCSEaUrsTmHUzWRVUjBowkLhzDcRXtJdornLAkFUWkwKYpWkBFR2gHUfHYy/PjooP/9Kc+9IE3v77eJ0gKhEjAh/Dc0GkLcBlO6jA/9w7hXN8IBV7HgWMgA8PFeYkUEi8VUkf4SCGUQCQSqSWmDMoZ72UI+nagvQ9/ZkzfgRqYL77PmBfrV0voo4hVXy3TV+6I/mkpfH9xuj4W6vPghWOdVX8OAezXV7P0DV0uGKu0Ck5bE36xhPHY1JP2wDdjlluGnW/97/zu//UnyHrJaHQ1my89n20bNnHdK5/N2990HVdtHeL6d76W0TM/Y7ac4cyZKWbXDEdO3ccZO85dU4tUxCaqIzuYnZpnYMgi8hI3MoLbuIXT2SDz8Sj3rME9RxMOdoY5cLzLBVsv5czJw3zvo3/HDz73cRZ0D2cUJoko0jWaaYufHHyAR+97CF8pmZs/zsnOHBtkzOG5k5i6I10xuGwFn64w310mkjHbW5Zdu+FX3vMGbjxyDDubcfL4Mpe97AXsGxxjNUu5/i/fR2Oiyj994nPMtVc5udzkvHqV2VPHWFRdbrv5dhIjKX0Lk6VML80yXlPs3DdEs9vh7LEWr/pPL6GiHUJFPHDnwyy3LIszh+gudljuNWn2lih9QVf0Za16gIkk5sKN8Dtvv46Lt51h4aEWG85TVDZv57L9Lep79/Gn77+Nf/noP3IozxDRMOPDVSpDjkP33M+Vz3szQ9VBmr6LySt051dIbJMD953Cz6X4Ss7pU0foPHw/3bUj+NU6jcHdVEZ2sdI6S6Esg/URhnZt4tDdB+j0UlZWLCemFzBlTG4LjAtJvdb0IO3R6nXwhSFOYorFJlG9wc6xCQoR3JhSVajWqjgnKIqMNGsxOWG45dYZvCgQhWMZT1FGDEdVjq0ssdBaYfuWTUgJ1uTEOqJCFVGqPjMpLC4RAdQtfFikIkPBz23R7yg91hmCzkHgJJTCYWyJLQu0EvTKFCEkWimyPMM6R16kxEkDSovyBlukGJkT1xoUypDZHGdLIq1QpoYvLdpAw0d4q/HE1ERMmacoJVAqNICaiFgkFFh6roeUCqUspZBoYqRxCO2pxxEUjlhIYiWDhwVPLGOUdxSmABRagrSWhkiwQkIck5mCAo+TChVF5HlBklSJRElZCpT1OF0gI0naK4jihFIK7vjeif8TVDQeRRVEC2uTMDPXSUhxcX3uulQoJ3HCBqqd6JMdZZApqiiQJgUOFckgjPe+HwYSfvGEt6GAKxAqDr7qyKMUeFtgTQjW9c5jXYlyHkkIqsb3F6cK1i8+or9o8d6Fa2ufh+P8ukPK/Zxbgw5qH+/OKYDO6eT7zwXCyMmWDi0lzlh84VAkpEs9VA0OLyoufv9NvOXpr+UJT3oGZx9Z4OSuQb7wznfz6lf+Keff9lFevqtKMvcQN37+ZoZHt7G2vEK0cSPz7Yw4vphjJ2dZXo158fMu5ivfupHdkWbPpRfy8JmCh2dXac/lzCy2MC5iWDeZHBqkPpgzVlUsrR5AJREXvXwC1atwUb4Jb5bpiIgYR77W4/RKMJmdXpoBH9F56Bj3R5KhrMBkDWy1IPIyBA0Lz5op+Ob9P2Thjgf4/pUv5t7P/z1nuwV/8Efv49CDHY6cXuIFL7uC79/3Yz7+0Y9RtA1bd28nTZeZK0ZwUrByqokerrJz0jHV2siW3fsZ1SWVNcF49gC/8rTtfOQHJxgUJegRysKjE01HLDOz2mR1aY6V9iy+gEa1gfAdyjwlK3PSzSM081XM/H1s2nOCj713G5/7yhne98aMbZfv5JEb/pqPvGMHTRw/OhhBr8OOPcO87gM/ZOXkXRz49ufYtWsSWTsPkc6R+ZKTnWE0M8iNG7n86h0sHzpLdMFFrLa7tHt1ikgQmzYjg0PUx89j8egdNO/u0Ktq1uY63Hjrg6xFhk0jG4iyCrJMsdaQGUtpHan1pBKGTYO2NszPNDmhPNUzhm4rZ3h0mvP2nc/g4FZMWTIzM0vz9Ao1WdCOZcBg25jUGKZX1yinI5aXVnnmMz22tESVYYQXONeiXo0pXYjTLI1Ba93H+7pzmA1TGpy2OCspSkulWifPsqB+UxaZGipxTE9YCm+RCJyzZNaC6kunY0WRe7Aa40FqhRQW2xUIa6l6jSlAaIUWGUpZClujaSESObHX9ITBC0kl0hgsxhagZMCbFJp6UserAjKBjzSFdyAKpNQUOXg0XeuJEon1gkgotEiwZY8kjrBEOJ9jo4j5XpdIeKLCk3hFriRlXgSpKR6R53RxCF+QKU8iqjgs0ZAnz7qU/x/y4P/ffXiADt5FCFEGbXg0CD7Hu05YQAqF8GXQm3p7bonpnA0IXh/yS/ESTISXeRiz9E/pMOKO+uMbh3clQvWLqACtFdI6MC7MwYVGSPojoP773p9jzvv1oi/XTVHrIeCB5x469bD4DUHiQQ0k+mAyRBhLiXMYAwi6+AgtLDgRCr2okmeG+MK3c/SHNzJyzZv46v/9WT7z7c/w0otfzQX/+Xf48Ns+SKsUNM/ew8HOCm9+8U6OPXIsmJuaPebqk5RmmMXVRWJt8WWHtNdjQ32MDSZhz5X7eOAkzK9pzpxdYyHLGdYFo86wdTJm20hCt+No4xkXBWsxFO0KzV5CM95OnDoqqkGvWIQ4Yu/2YZbXVpDVjKxV0ik6JGWEkxHezNPrVWiomG7qiWVJaqHTaFEkDQ7cdAvXvfrldE516FRiWtWjHPvB7dx26LucufNBkkqLiy97CgvTs8QDw/Q6c0gsPW3ZsCZZbY+Q5iUHfnoT+3Y9mYnNkoY6RmQTXLSbsz3FthFHVmb0MovOPYnPoSgwpaeqBZnPkUqgdEJiClY6huV5y/wvbUEOneatHz7Lbbe+mz/61b9i/O5JfuPJ8MHPn2TfzgZZniN8g4n9e7n3Kx+nOHE3ExOD5FmXbRMa2Ei3fZzVfJV6Y5jNlYR2KZk/u8SQr2JyTZa1GPARK4OajammSB/lVDrLw0d+xl//zWEaYwmLeUE9illeXQrsFy2wEnRUoUgtVRmMRz1fUNoGpkxpTFbJTUJMQW8148ypGSZ3VmitKpqp5VCng2/ECAqqrkKHAmscPtdBeiwa/OjWH3Lts5+LlQXdvGRkYJBOliFFifeEsAw8cRxjixLXzw6OZAR4rC9RkSLN2gD95alDRRFZWaJ9/8arIkpnzgXCOywmM1QTTzczeCf63hGPp0RIgzEOXU1odzt0JUSJQGLQJUid0DOKmB6JBlf0ULKBFlVQHnJHVFF4CqI8JtIFCPAioyShqhrQrz+FLQhYFAHSkRc5SoGSijw3qKpGdh0TukLHrCGTSpBJ4tGVGCklUS1BOEktN6xlikac0C161OIGaf/1JGT2mEvr42NE88nrP/BfXjvShxc5rLM46cEbsCUB0l7i+ulD3ldROsEZEwIxtEBGYFUVJ0GoKPzQtA1aW1ntS3UCDAgnECJCyjiAgpTEZCVKClzeQykRWC/eBByvkDhREMI2+lp42c9B9S4U95+L3WFdFe8CC9j3l03eBKiSs+shHxCMWf/rweGURBU5qppgi5R4z1tYvPcOZsdfhxg7xuLpKg/NdHjmf34PX/zuzXRNyd9+6AN8+ZvXc9OnP0JSrmF8xKmBrWRqD2fZwcGpBZY7PRZW15g7s8ozL93K1mFLdfsvsVw0ePD4SWZWl7HpIjsTA0XOnp2DLKwpNtQi6sJSN56O8ow1ElY7QaJGq0bp2sysNKlUJlhMY5a7GblNqCvFSC0m1hZbhGxMW+Zg6zifU7OaSEaMb6pzYnkFrTwve9NL+Z3f+S986Vvf44Gf3sTtN95ML8rxecro9nG61jK9OAeVBlnepFAZa9kaQ1EdqhXOzC4ysWmQrtdMjg7yil99I48c38vDyzEzzcOszi1z2RMv4+/+4jOcWppHkdBbXMZYS6fXCYY1B0KXKCHwBZSlZdvuEb59x2Eu3jTBR98xxk9vuZnFZo2yfZr3f9mTx5qpNUOzrLIoJQdvOciOLRGViQ1MjA/RNi0efegBNu19EnYtZnjfRUzES7SXjmKbVWrDW+m4NrVKA5lspDawhU5nlYqqkYpBnvbUa3nqJRdw/s4qO7bs46pnXsstP76TIgr5BD3jcdbhskBWrUhNXBRk1iEiSeIiZuZWKNOSZZPS2DzB5TufT7zjadRrLTZvnGSkch7f+/HtWKdpRgZtFLFKMM4yUk+IRMHcXI9TR0+x//zd1OoRpS8RMukH6yikLZFAxwl0pJAO8A4vHM6EJsgKf04goYRAqBLrDc5LUGGX5ozHeot1MqSSOYFwCcYGHbkWLpgUZRBXCBu+Bk9JmZZYp0hQ4DOQDmMcsfIIV6CTBtrE5K6gkBmRj5GRxItAqdVCUagEhQaXEMcFpiwoSYkSTdKnZEZS40zZx6SA9AKpgiHTuJBYlUSDFJnDxVVK1yWR0Mt7gf6KB6cRBPiYiSKsdUSuQAtH5CQ//M7R//8vWYOMsMD7LMygnQnzZ1PSx6H3HaUDOF8Ny09h8Crklwqh8CKBaA2hLEQpTlSwLgZRDeQ64UPUnusXVu9xvkQqTWk8WlewTqJ0QIhK3e+whQiJUl4GHEJfCnkOEfwLb7/4fNZlm4I+D74f7FuWFimjvgM2LIPxui/3DMHZ3pc46uTdDLUWwdFHWBh8Fo0tPSL9TNq1cY5Mn+Bv/+bveNULns1DD55gdeVnfPmtb+XXnhfTLCJONA0y2klWmWAplaTNFYZFBZNlXLhjM/v37+VYuZGHl05xdvoBMrPCROKZrGi01qwsGzYMVcnmUhbbniiKSCqSqBICiL0tUTahGArPM8Gw1lsNC2Nj6Cw3KbuCtdWcWNaYSDzYHjhBLlPqY3XsYMFKzXG2mTM2NEhNSd7/X/+c57303ehIcnymzrGD9wHQKTuszs4yIKpsm5hEZWv4soldaWPTFGskRVFQKMf8fMzQRJ27HzyEtoYrr7ucQw89Ss9UmT3T5FOf/jIHTx2n2V5gpbVILhRxtUFNDlARUFWS2EdooVGRIncCKy0jVc0Trx7l3783zXNfezGf+eyr2RPDFgkmKnDVElsB4XKKaoV0qsnCoSk6q4Kh1gAb40nas1PMzB/h7MO3Uax1UZObcXGK1RbpY7zSDG4YQQ2WyLyF3JDQWjnOfQ/cwU9uvZd/+9bdHD18iu/dfTuVoZBtGmWeGE/DC6QGpyWpLWlqT1r1ZFlG1ztaAxVWVjo8/eoXc/zoHOOXbWKyMs7JuTXyjkZt3818ZhjEo40n0yVOWoR1pKWj1BE2T5g61qNai8i6EvIqEQbpPMYbjHRY4ajFEdKEVHcvPCqO0EmCjiM8AuOCCs0YE+iKRlI4S2ZyCpfhKIi8IBGC2DlEXhB7iySiqmMMAoPC5xHGFuTZKlo7jDE0hhtUquHmHlNBo4hlSGhKqoN02z2aRUaUDJLYgYA0Lj2mDCE8RgusbyF8hitbSOcRoqRWURR5G2vDgWR9QRwPoXSDSA/gXIS1IIRmQA0SyQZhyKuIM8uQGgIfEekaVVWhzC1KOFQEWnkGjaXenwT4c273x/Z4fIxo6JshrEMSBcepWZ9rS4TSeA3StPuW/wjnJGpdduiKUJTjOl6X4Af6UsUYqcDbnMBil3ilgyKmH9fnrAsBITiUjjAuRwmFdQU4i5IS74IOHdaZN4E9gwwmpXW9/Lnz0ttzy1Tnw4wf75EolOova50JHzuQ05AEZKn3lrhs4JwnsTHfPXUDR5dPcsuPP8MrX/R0zpz8Bl+9a5rhbdfQqBnmT57CLk9xydAG0uwQdx+PuXOtyqaNo6jxrXTKQZYP30zkq6S+ydMvv4yBoQ0czR3TM6dpLp1ho8v47ZcMcOq+U/zbvGK5HVGmEZuGFS26HJ3VVKqSRuIYRCLjCplZI3MRMrc0pSMZHWem1yVuLeFjcNWEVdNAiAqthR4X7d+JW5iimRWMlzHZ4jL7t13MybllrG4zFiecOHOYfVc9g9Wpo7zo6deA6rHtkv2sra6RpWtYIcgyz1pnlW5nlcHRBpnpkMgxNm+f5F2/8of89rt+k9oOS5422LBliLe+4728+3d+ja9+9evsv/gSWvEK/37Tl4m9Z2TzBLGOkFFG3NAM1xLO5h4feXxUYDODFAlW5Ag/yMjwKF//+hR//bHfwBz+Evp5F3L7zyJq2xNazQ4VAZWyi05inripwobzL+Dk0Qe4dLzO8SNn2TWYIOUYy3PTbN97HvFYg6im8fOShbUuuWkjTUxj0ySnHjzGrisu4Dtf+wbV2hALK6vML8xwdnGOQyfO0IzC7bUBGCPwUURbhgJXLfu3RwmVUqGcpWj1aC3BO//sT/j7Gz7MyMYh3vXyN7Kw92n8/uvfzP/4+3+mO7iBiy66gGOHD1OreAZ9kAeWStFdTRmoVdEVRVxJOHj4COdfsAsiiXERkFFRVZwJuI7COuqiEubISlKsu8OtI0FS0wndvBsY6d7197Di3A4KJ4KhSnhEZDEYctdFlDHWC5SPgtwxgV6eIqIKRSGJfIKwAlcopApok24vR8e18PrqBaetEGD8EiYqifwwECFFhEPTyyxV1cDqGJIqpe+hVR2MJBIixAH6IOOW5Dgb9g5Clri87PNwTKhpZZB2OqUotMcVkopU9ITBRUGVFxw3AXWSS0fmLN7ZcEN5jI/HRQcPfdKiFNj1fCMlzy0h7TrUi4AzEJiAAfZhgeOswdteX84k8aKD1CkqLvFk6ChIrnwfw+uC4BHvCpQvMaaHVBZDia6AlwZBCOBYH71IKfuimL7q5ReSns4ZmM7xa0R/P8A5vK9zDmstxhR9jo3A+nXUcEh7WidHGt3GDHRYaRvumvoyL33TVWy+YJKfnJyCbc/ht/7wL3jr732M5770TWQDFWqdE1w0WGOxlzLTUWytbWLD4H5SvZEsdaiVHo1ayVUXXsnGTXs4u9hh7WyTqHOaS4Yd2wcNC0spM2vQMJaGiNC+5PB0i00SOt2MNIuI8oQ8T5GtFoqcqoqIS89AMkaROmy3xJQ5PnMMyJgqhtiV6Fjy6KkZolIxUtPEwlFVdVq+Q2PLJioDkyx15/CZoGpyZk7dQ6NaYXH2FHGs0b4IIREY0uYcVQ1XPPN5DDfG8AZS3+P4Qw/wzdv/hsaIpzo8yeyJk2w3Q7zqydfwg/84yBMmd1GJoLdSgl2i22uxOj+PzTM8FuldsLtrsM5hTcA6KyGoSh2e/1DMlU/ZwIGfPczXvtXGD25m93kly6sFymq2xXV2jQ3ywksv4+LNo0QjmtnDDyH2/ToPPzrFvI/IJzRzrVWmV5YpslE6MzFd62kMVtGVCmeWjpJW19h3xU6y9ioX7r2URhyzqQFmsY01ko3jEcOJRHpHT4CtSBppQbWwxC7cCrtCYoSk0AI5NMzg8CBP/KUJHrjrH9m9fRvXXDnJI8BEe4Zvf/kG1PwC45WYnZsmuXD3edRUTCpSXCXEQkYuobWa4SgxpuDYoaMIk1LXBdqFfASPCcEZSqJdTCEEZZ8ZoxyI0iCFwEpFF4PVfb286xueZGDOKCFJoghhLVrI8LtlNMLXMd5iXIlzGWXZpjBNfGKQRoMTZHTp5R20IkiMS0+lUgvcHFOQSYGvaax32DJGyyEKJF5KlBRQ5lSUR8gCK9qUrJK7FE+J8walBKXpIaRB6aC9R1jKIgXhqVaSYPCSApEoirig1CledPFZCxvFpEQ0siqmVscTIV0EViJUQtXGDKgBhqJBGsngY66sj4sOXkiFVzFKgC0CX0K4MrAuRFhyutKey0uVIgrfQCnxDoS1iBJ83sarKsQCiELRxOG8gXWXKxYBKCeg0CAMygUXqZA6KGG0Q7oQ8uFNGsYmBAt7AIL9gkP13KMf3u0dDotUYaSjdAzOIpXHFuJcYrxUCmt8/3AA8P0xjURRQS/DaNXyvPbN/PHzbmZte4WNQ+NMd+/GViY5cvwLTH/rCyR1yyte+gxs6xh33bORXjyBqYxQecIF9I7NUawc49qLt2IaY8z2FPc/+ijxcheyk+zd5Wm4Nir1PHoyo+dqDDcKSHooXePAwwu85cqEf79V0sscZ2zO5lFHK07RKFbdInOdYd7ylBrfuvlRMjlEVFMYU0dXYnqlp5fCwFCVTr7GaV+hnitsYtk2MklU1Dj90ANc8MQ9TM9nvODy3+S3/uAPueqif6OdrVEpJyAX1DdO0DzS5PwNG7nx2BH+/EMf5tjxU5w5U6eZGiZlyo/vfh/PvOSP2Xzpk3nGVVfwLw8fpXnRKN85dIIh1vjpiUe4YngfSzalO9ehqCWk7SadrMWG+mYa9SpLPgtKKiPQFY3UMapXpUkLZQsGKgN874cn2PiabcyUG/net+7mJ2ckwxsVc52IxSTi4p0TLM9Nk7fqRFuWeMM7b6B98Ds04hVEZ5rbv3iIsU37iaXD1eqMj20gFWsMM8CojygurDLqFKfPPkJsBjh8aoF/+tSXSDbA9HSOqw0wn+ZUS0gKg48ViXCsRAoNSOEplGNTEjE5PILKA7+8U1jyGc3Nx6fREzv41tce4dqnvZgk6/LCHVsoaoIvfv9Gdo5v4NjiaXJrqTJIpCSm5rCRDKZA20CILocfmkLIu7j2+c9ibFwinMKboBX2rkRpgygtWkryLEVVK1glKK1FWqgiKUzYVZlKwOxqF263XmqK0iAjFUxEgPAizMiVQxuDFZISTd0OQSHJxTwohShrGCzKBTCg9Q5XQIQkoQ5W0mv2iIYaGAG+sMTKY0yJEjEAZSGRiSO2NSQCS06pJZjQZCghwSm0jPFxDNYhtMViKU2BRlGPBVlZkHgFKIwXaD+E9G2cLmgNCMa6EaWGwvWQSuLLAivD90ixztR6bI/HRQfvAZIBXFxFJBHGFEGfLg1eWgQmCBIVaC2RVAGBI8ULfw4vLEqLMjnKRUif40UZIGJOIVHg+oF4zmNNiicHk+NtHrp5SeC/RDKYnsQ6j1mCCM5SCC7rcNAELrwgQgpN6UJAAV7ibJikOVOGTt4FrbBQISfblS7ILp2gdFHg2luDs57cd/C6wOiUp75jnLe8TnPetjr33zPNB373uaw+cDOnfno7L3/P+4kEVLI5/id3bx5s2XmW9/6+aU17OPM5PU9qdWuyBkuyJGQbG9tYBoOhmGwmB3Aok5ALCZMNJpiAIYGYpMDEIQFjGwRhMiAwyPIgW5ZtNLWkVktqtXqe+8xnT2v6hvvH2poKSHxx1b25+apOdZ11ztm9z67ud737fZ/n95w6/jiiu4NSaro7b6BXbMNeOElHZvSTrSxWkuGFC4zOn2eze4bt04K2HzKSEVU7QrnAlrgFiWJaGgQVfZsQz8fM7Lac7FmKOmG4ESjrBFOWxHZEO97g1KkTmKkOaZxR6Yg6kmwUOS5YWt6jfUQcdTC1olApCxMZBTWnBifYd+teTD5Aec/M9pp3/8A7aNc9XAgM/Qbb919Fv1dw+00387o7Xs9Hf/u3eOaz9/PwFx5GRJJXvOqNXLATfPC/3MdaOsnmzpWcfuwETz5zgc/9zYP8xM/8CA89eoTrL9tLXgnivKR2Jen6iNhF5EVB0Oss944zERxRrLA6x4easi4h7qESSU9WDH0g6gpWzl4gzwseuesca6lnpuVZiIZMDRwnn11iacNB2mX9/DqPPfBJgh1y3TW34+MeM2mPdjrCRJrVpTV6/RFz01uxWZtOnOB76wjp2Zx4FrbXzLQ13/gNN1AuJky1O1xY7zMsHJWSDBKBkDFDoZgQjlbSNB6pUky3DP1imTU74lTe49ig5HGfc6Iv8UXBdTe8nKNHz/Id3/ptHB0N2LRvLy5PKDdgl5yg6yKsLhjIQBEc0mh0EOS9gv6gRolJTp9cwdp1li+cwVUKEdp4L3HOUHuFw+CEQKcxEkFkGwWbc/UYJCgg0kjhwNaNYicoSpejEwE1ZDLG6Agnm1EOtUCZrJFaZhEjRniGVFGMFQblSxIsUKBNs/OKooihLSg19GXdsNoHJcY6Ijw6NKNRGwqCrNGpR5uACDXeO2LdwfgIEgEm4LxAK0dtR0jvMcqjRA14pE6wQTOoikYp5wM+BKwqIeqhPYhKY0pDKSsq18MrgVOK2mhyV6N1RJAKnWVfcW3930JF89//6y+8953f3SbYsdpEaAhuHJYhcU6OlTSB4BWBAfgSKdogYwhlc0OwjcMVLAIJKiII2UCOxvS650xJY9h8c3MRzSqE51xreFzZ/CltMyfDhzH6V445BG68RJV431imtR4nQSFfQpBs8kJeMDU9x40PIYAUKCzSN7sGqx2xVbgiQbzmXYTHPsH8jvC3VzsAACAASURBVCv47K+e4qf//D3cc7Bm8ro3cR0P8oWP/inf9X3fxVOP382JfAulnWE9itix50Z8fZH5VoRKWxxfvMTJC0cYrp1i70SfV15ZMS369JFYJZA4XK6Z2WJYW6nQkYSyQkee7ROBmxYU9xwcUImE+cyDMnj6rPaBuMuB04aVnkZUCaMqUA6GaBWhfUHWSaiCIxMKqYdM5+v4pEIwYrMx6GKZLFNkWeD46UM8ee4opiOZSDVb5ttkUztxdQUbA7pTbT72kY9xKodRuU65sk7/4mm0W+Xhpxcp11Z47au3ce/n/w5JyR237OeRv3ucw0eOgBE4V9Hrr3Jh+QJlJKhtH2c9flhDnbBerCCdRQjXkD6toJICLRzzE5rLJ7Yx7Fnesj/h6htnuPvepynqBCrNmmoURkE07G+xtIybcgglmU1naO3S6N5BhkPLTLyZTrfN7KbLmJjsIsIUrW4XGWsmzAy1L0hv+AbOnM7ZsWcvO6YVb3nd1dzy6q1s27SfauMI+TDGo4ltc7OYjaaJVcR0BGkEJy6MuPkNX8dDjz1NZ3IOmaSEjSEL7VnEdMSzB49x/RXX8uwzp3nD9fNM2xHv/9RnePDZS3zv99zEx750ivbMJOVgQLfVZa1YpROlBGeRIUOwhhA1PonYunuBVjINqkAI0NIgVUkQFqVjnFO4yqKURTrZKG50s7dSQSC8wRjDoMxRClJp8EWNM74hxLoAtQMbcOO4S0KFxqOsQzuBsgEZoJIBrwTCV4RxIpsPBVJ6qrLCYKgjTdACUzuEh1J6dKwQwVCXAaUDZRVhogJlHM5bhKrROBQRIjgkMYEaZLNYroNEeokUDq9r2nFGWVusDHinMdAYGWVo4g11wNqKIC2p6BJy8DJgMkOU50R4VAh84i8O/59gdGJsJmheNB1PEIoRIjigBFU1RLegGp18UM0MTeZIsvHi1eNdQRASbc14qQpISRBN4RXO8RwWAdmMU1DjdwRBE3zZzGMZ86DHD/F8ty5EM7sPorFjj+8UUjbZruE5RvXYth2Ce5HKZowVfkm8XyOPUtKBbYIQojwwrEGNJInbQX3N15LJzXzntx/l3n93J186f4J0YYFv/NavYc+1y/TLZU6O+izqXSyHtAlCqdZQaxcoS8nFtXWK5TN0uUSyscKrdydMZREro4rYxgQElbXIqCAlo0NJLyickpT9QOkzRBbY2c5Z8wXLQ8u0ipA6IUlSBn5IHMf08hrrhmyamWbYc1SqRBGzUVu6ccqkanGxLohbs5xZWmHH7CRCbxApj1UO4WPSKEKGHq0kZTrNOLnS57Las9Cd4tzGOdY2VtlwNe0g8LOzhGLIYG0AchpbD9k9a/j277yMP/vDvyNPtnD35x7jpiteTjzZoZ+PmNceLQu8y3GhR6Fi4rqgShJGYhkhHKWXBJc0IdA44kTSKjR64MDXvOrmXdz+L27h0MOf54nTMDvn6JUVuWicx1NWcu11N7Jy5iDl+gbnRou0RmeYehaMqZmamUZPekbVCeayfUSTLVrJJpYunWbLri3QvQoqR++Lf8z2y67m0qnj9CtHXA9ZXQk88sBnOX/BETKJ7+fNwrCqqKNVVBTTKz1D59hzxX7uv/dL7N62h9LnlMNVpIeiyukdWeaON7+Z84ef5uj6IrepGzi55zJ+7Sd+iJNrmoc+/SBXLHQ4fG4ZGUFSBVoipbaSQdFHpYGZrEXsNUcefpL9m+Ywu1ImJtsEJxrVGqBEjLdNgYxiQ1166khifNMwJVJTliVONu9oW0nGqCyotcPGjrhqQr2HvibEkkoIRFkSxoW+X5ZNfF6kqepx5uuoRAlJXQXiRCAkVFWT5Ron4OsKrMQrg9UCDCQYXOlJY4FMFSIoQmQQwVBUYIjQRoOtqb1AxBEuNJ19VfcJsslh1kpha4e3NUUEWgqEiwjaUVUCrdp4FLgaQUUsPcrMUw4uoVJPZQP1SEM7QtlAMvJ/v07+Pzz/W3Twv/Vff+m9P/DWLbhQIILGh1W8GpsiQoaQHSQVzo8QQTe6Wt9Y/xtt+9hI5ATCAXbYUBfFc9hg0YSBjDXqMngQZcOi9s+lMHm8aEiNIUCoqyYZyvoGOuYh4CFIhBSNw9U3Bdw7np+jN6iFF2Tx8NKO/QVQWVPonVRUwqL6NVQOIVpEI8eXDlY89oef4MCjjvv+4GF+85kul73pdUxEEXOzhjfelqLCJs4feZCjvb246X289VWXcfmE5dwgYljA8cMPQfUUc+oSgzMDfuQdt1CsX2KnqZnMBGdWapwAL2skBqctS2uOVpLQEzVCCSZSh60rtmaeZL3mXKGpbd0AlEIGYcTSyJFEktIGbn3tKzl6+Enm25MUVQ8vLdpbai3Jsg5reGbTKbqdrEkCkqBzSyk9qe4zt6WD8kNCmCSe6nL86GHOrz9JXAaWTp6mF0asl6vMdncQG0cVSmQoiJRkTmTcc/8JlirBdGed1+3ejFAZzz79efxgA6E0l85faGabIdDWCUUteN+7r2Xx0LMsVzWy8mglKceeCRkc7SRhdqHL1Jzna27rsOvql7FyasTNWy/w54/m5FHAOIWWAtKEsHyS+Thh19Yu2RbB1//AO1haPM/U1d+KZ4Op6Vla6SZa8SaqekSQEcPRkFjB6tmHodxg+uYfRLkVquUn6U5txRYaIytmr7uW11x5JZ/+1GMo37ylN2mDxq2EwiuJMIq1lSVUFLjh8glmunMcO3meIGFmZp7hoGBpdZE50+f111xOtzOBnDP86K98mKXFRR5ZPIGQmh2btjCZpGwUGyjVqD1KXyFEjA4FeTFCtWc4fPwot95yJXVRY6IWxmjwETpI5BjUV9kKGWW0rKOsLCaJGdQVJovxrqJ2NTUS6wOxCOhgcVY244q6Qroa5aCsBXFk0N5jvMJI08DKdDNH72CQQmC0oSg82jQy6TTqYkfgkmIMS7MNhoAIZytMJKgqT0DjlUWLCEnSOOOjEucsQcRo45Djm4m1lko141+Fpcbjg0bLmOAkAY2QjRgkEnUTb+g9JgbjLVqkODvEq5hgMuqqanAItUZYiZARd//5wf//6+CFAKkcJooaSJjqQGKQUdqk+eoB3kQN0VEGxoJ2CBrpzQth2PIFhEEoa0RVIeq6YdeMu2kldKOD9w3ETEdDpHEgNUoLhKwJoonSe+H5jdGmqJcU6Ocdqc8tYcXzZJmXLGCb5/dCxX8OgialJCoESSEJaYbsdOnnHR49tYvWwpt5preLn//4eX7uyZz1S8v87m9/hDoMeeiJmmTqDYTiAuf6Q6qZ/fikzaGj6xxcqnjmwjlOnjqCrxe5fmtG/3TB8QvQ7Z8lq3NMKxDHjvlYoGRFZFLiEBj2IuZmWmTC0tWGqSTGu5R2J2JyW0ZnKwRTI3RK7mp6gx4rw5JWtgmlFLGJePz+B5DtlDNrK+ShYiqNCM4SyhyVZpTDiqluTEs6rE2wzKDmJsiEJq8c8UrFvJ5kuNbHDYekGtpVzTD3jHyzKotCSrceMKVnmKVLLFNCgPO15OzSGonNSds7eGo95b7lx1n1QyyKssqxdQ61IIo0ruURvYJbr5tk6At6weCUwgN1kJReoYVnRI7PK6qiy4knDB/60J/y5JMn2HP1NBM+YqZMKGUgwdOyAdqaKvZMT85Qrxke+uNP0Em2seeqqxFFSkcu0GrNoWdn8WoC6iHtKCJIwfTEDmxxgdUv/Qo25By+5Nhyw+vZdvPtrA8y7vrQh/jRf/d7lBpq7dBKUTvJ0HlGwxxTQTEYEQVBhmFtZZ37nniE9mSHzLQ5ubKC0hVnL11gbuft7Nr5cib3TXPPkWO84tW38szyElvbc8QdzTe+4Zt47au+Dms9Zek4O1gjkFBXFlvVxDpGWIPx0yyf61HmFVpH1FUzmhIqwlrbJLEh8LVlpD0+UlTWkwaJXxsQ65RIRWjp6GYJIsgG/ysClRqPLmuPwhCnjsIOsK6JB6z8kMRDlFuiOlC4mpGt6NucEEt69QgXKUauplISXWdonyBcTB0a0YOMW4zqRiMfhAZpyIs1cCM8FbiGfSOosc5B1tBrk1TREoIYj/SSOGgSqTBKYRJ4LnpTYJEqQZvxXqC0Te3QHqM03nuKoiDrtCmdRWoDWtO3/4csWcdidpxtXkQRHLrqgpzCKd2MT7RBRBoUOFUijUfQZDJCE87k3YggCmRoFjJiMMRtrBNGI6RrqJDODQm+HBuParyXhGDBuPHf0UbGKcEovAGvAigIpgnPDaJRKiB8w453gBfNEve5kA9e6NqfI04CBD2WksmAk75x3CUS1wqcuS/m0U8rPv4pydNLHX75znu469kTzMxPsattkHv2ErXn6Z/N6NgjPP3kEZ69+AR26hrSyLErM4RRyupZy/rxJ4jWHiWcH/CX963wZJGw6D1PPrNOZwo2pZ5tk5KrdwpKq6DyaCPJVc6ea3Zyqe+YCRDFhoNHNkjiLn4DLt8J18xnJGEdN/Jo45mKEnSZk+YpMom4NFzBVQ2SucM0mcgw2SQqm+Xs0Wcp1y+S2UWixDEz3SKLPXVf0wsOOT3DWQQXa5jeOk8qDUlXEk9tZ2LHNpLJzcx3mzzXPhVVR9PeP8v+Vpcjz3w/V4kVFlQL2VX0Vs5z8swJ5grNzul9zE/N0x9Y8DFaS7QAsarwC13e/uOf50xvilkydGiohx0MiY5IiDGyw8TuPRw8eIave/s3MR+G/Ml/O8qnPnGWgavwUUB5WHYSZRVzs5vYNTcFomT/9p1s3b+TQb/i/IEvctnOfRQThkhX2OVj+NEyydwUMnIYM4vJWvhRn2xmmgtP3cPM4DR/9J6v4wd/5MfZ8CeRw4xWt0k68lZQeUcoK1pBoGXM5s37WGhtRmvLht3gyXPrLPgW7dwik4pJVZKrGlNF7L/tZt75gQ/zsUeGPPjoMSa94b4738+51SEXz1bc+We/wYGDnyTpxCReMZWmBD/EuhpvE1xIKas+fn2Nj/zeXTxz7Chlv08UpVD3sfU6SsV4Z9C62asZkRErjaSGSOBTQ107sIJgFdY6nHOUtkYVI/xwnYGtsVrRtyWqMsRVTXABR4wrZUOqRCCNxopGiePqEdLXtL1nwgqkK/CyTx1yPDkyVJhIUvoKX60TCUtdD3FuiC0HSJFhrCLyhlwrgk7xNIx7XZWQF9RV1eBKjMBFjkoUuKhipAaUdd2AzFSNiSLAU4eIyAg6rZjgNGUlcCIQRZqpqI2oG2e38gXKF2TJ/wsFXgjxISHEohDi0IuuvVcIcU4I8dj44+te9LV3CyGOCiGeEUK88ct+JuOiWbtxyEcYAQMkJU2wliMQgYyRMsOLhCATMC2CMCAN0qSgMoRoDBe+tviyxhcFoSgIdRgrYXRjMHIKfAzBNJr6EBrVjrQoFSNpPT++USF5PnD7JYEhY0erC/4lodsvlVA2s7ScGulj6mBxVcC0JhBLBZyeQ8/GvOy6l3PT5VehVM6amaaMJqhcQhVlrNkhM1snOXTkIX7rz/4tx44/xWA9o3QpHR3T65c8ffE0F049TWwKFp3gjKtRSAa55WVzKcnGgAndwrUkZtqzZ8pztavpyJJCJGTS8NU3vYbCeoJRjKo+05nhgUd7eEbEVUJ73jbqBQlF6Rj6Ee2sopYDBtU6k9EExgtKKYhiwaAeImODSmOWXECalE7S5lIv58LyOrn3iMyQiQi/MSJTLSLVoZcPGJUFlW3R3/DYekRRX2RiZorSC1phwLw9x1S+g5dfliHyGCkj6nyJrb2YJJ1nI6wSMo0yGTqZIpcVlarIokBvvebN17eYqtY4uVoRpR3i2NOKIU4ktawbfrmKKXoF83WPzsl1jjzwOGH5JOubAoN0G5VKqIuAUoaF4FnUJSsbmrn5SdJWRnu2RV5c5OpbbkVvivGbriSSNa1Nu5jauYdtV+9j+fyAyZ2Xo/JTrF88whrT1Iun6ffb9Oo+V97+Nha/+ATv++VP8NmzFet1Tp0oSlejpcBK2LAQq4LTJ59ife08OS3iqI0KCSsyZyMM6TpDUdZ0ky6zM4r3vfc9xMmITx76Ah94/2+RzHepZq7gdW+4lZd/9fX4iRaPHTlFlffJcayPSozUtJRAZCW98hKx1PgowZaSZ586S1lXhNCYvYxuk/shakKRO0ek4wY5YhuLvnNjl7d1yEiTIAlljVQKnCeZmCaojHakkVFNS+kGwavblBpcKJhQEXndQ+uYvCgJVA1XSk1Q+pLcVdReIZwnCjFCGaTVGCKqUR8dygaB7MfjGWfAtzDSsRoVrDEirSHYGjOOHqwAncZNQxdKqlGFLxoppvIR2rWQMkISqFzRpF3JBKUMVSUoyibgSOnAyI8QPqcsVpFhHM8pmgZJCvNll89/7Hw5HfyHgTv+gev/KYRw/fjjbwCEEFcBbwWuHv/MfxEvnnX8IycEj3AFSjw3BlGEUCC9h5ASgsK7EqcE3ggwGTLrQivB6agJ0o41JBHBKJzwSGOQtUMWFaI3hI0eYdQfO1ibzFaQz5uUhAwIrRGqg9ATBO0hqpB6bHgKxfNu1OdvBiE0evZgx5mToWFLMx7rCP/8h5CO2ENwAaMioqEHWzLKOjz7xEXSW9/Bp+tZfu6e09z5sMD2gKBpmcCUSphaWyM+cYJvf+N+ulM5ibOEya/FtLby6PnzHHjyIGr9aXZstWxP4cLZIalLGSnB1jTmdbsmueaGhM0Lgc2TNTOJZmqH5HW3wdxqYEYpNpZz7nn0JO0YqqhiviPpdWsOb+S8/vbN9IYFu63gZfOOhQRUFsGGR9SO+flZdsUdSkqiyYiJuRmCt0gbY8uCpaVTZK6Hk316kUUkGd25PVB1KW2LtXiaIt0GURe0puXbzEztwgeFnt/E6sBSDRWD9QGbI/BygnytYoZP8eM/t4BdvJ+f+uWX8fm/+g7+4q/286rdi2xptyk2CrKZaUK3Q6YjEpWymlfs2jrB/Qcu4dtzzBrPdHyBWPdptRyTmWd7O2a6K5A6p0hrDp44x7f/2C7++Hc/ykaxC7Ma+MRnLjJpSoaiorCeJQ3bNifMtwsuu3IXm/dcgQgJWeiydvE4W6Ri1/Z97LzlTXjfhtRgVy4y2TKcPvwZ3OVfT7Ttzezbcy3Lcj/nDn4GnU5yae0CN7zlNSBLFoshIJG1RxpJrSQgmMjgzFBz/9/+KstO0A6eBV+ThcCMmUR5Q+767IgVH7nzPfzbH/oO9kzBG3fvYdJJZlo9Dpy7yMHHTlDEbZ794jE2zg6YnZngX7/zPTgvmTGG1aKk7zx+tSTzWWNA6kPqu1w6YbmwdJJRbx1PjBeWVCeoypColCBzCI36zOi4mV0LjY8ibGGxOExs0NLjdE1dDamrdUZVjak1lXckKmkMdi4QvKMvPSpE9Os+tfTUQYGHVHsMGhFpRnJErmEQaIKuqcgpUDqjLg3Omaa41wVGB7SooQ50S09WS4qgGdmqAbeFJh+2rJomqJIRIlYkcUzwSXMzMbrBXMgWSnTxxDgqtIswGpIMVORRKGIziRIT6PYUAUtiNIgYISNeNNX9J5//ZYEPIdwHrH6Zj/cW4H+EEMoQwgngKPCK/9UPCdGE21ZVjooirG9i9EAQZApCj59saMYyY4WMDKFZtD6HHxCAEiglsIRxWAiEsm46g7oE6xrFi2hm5t4avFP4sUwSUeFE2ejnx5r2gHzJK/XiJWkIzcIljP/hvri7/3u/Zx0R0oTcVdDJyFe6lI9VbP7q17DqctYOrrPqCkaVZy2C2d3bGFKxuDZk3sOmay5j042bEOlOzi2OGJmK06NFxPqAtFhl5oYb6Jl9XFwekVnL0DS0xD0zhiltEToQdy0hziCp6SaW+S0ZV18LG6s9YinpXTrLRFuhZ1NKAduSDlNBsHR2xFX7O1wMgdI42mlFIgdUMwovCi70lrlY53RVDA7qpR6lEahOmzqvaSUdKh8YjATWGlptxVLeo1SKxHTRIqUrM6zJGMQdXHeWi7bGqAi5toEuenhZsmv/JGaTJkvb1NEsC21IFl7FR3/9ADs27WejKjh03zM8/Mh5RtJwyefETiNrgcLQ8gmRk8zNtRFeQV6gdYRotZmZ6NLWGoHFG0eqPVIY5kPApII7f+c43/fPp3nizCLb98YMKXHBMBlpNJ4Z0SUV88zNXMbJU+dIleH6a69lZut+2nNTpFuvhC0Jsh9BOyEUs/TWctrXvZVO+ybqkw8wuPQQzsHCtk1Ee1/Jx+66lwMHzvLwgeNctILNyTQxCuFAukBVB+oAIx2T1Jb3/uRHmRKBhII1VTFSJWv9HklkmEoCW7a3OPTxT3Dh1HESmfLomWO8+2f/DZ+75wAfe/+H+fjD93Hp6TN0t80zk7YQwfGBD3wAoSQhlrx883a6UYxPUpzzVKFi1a9jdCNRfvrQYYJ1TTxfEFSlx4dqPI+OQekGzhWKBtTnPDKAaaWoAHVZ0StztJA4YfBSEZyl0g5vR/SHfdb7Q4I1UINxFUYLRG1QQaOCA2upRg4/sqihJipbyCAxpoY6ohUSUq8RDrSJcb5EqwZoVucOJTIKArkSjebUe4wSGCFRQjeEWedRAazPx8E9AeQIIQdIUaK1JFBjtMcgMbRRzmGCwBUVxitaOiP4HlIWWDcEPGVZY5XFqkZS+ZWer2QG/8NCiIPjEc7U+NpW4MyLvufs+Nr/9IQARqdolYwxuk13HfygMTspjRQCiWkUMN4R6gJnK2RQPLd3FU4hgwITI2NNTRPMrdDI0hM2CkTfEgYV0tpG265yhCpQ4oXRi/CiSZ0hRkQStMV63YR/j9eoL164QrP89aEiePvCEjaMbxC+SZyyukKGGnNREzZ/J1p6Dg5fzS/8/KM88ecn+ZuHHyMMLHGqGJ0+x465bYh8ldHGGmmnxb5kEwvRjXz4Pz7IaRfx+UNPcf7gE7QnS7ZefRW7F76KiydOc+zECnUU8E6Qe8HOlmM6HrJ3V81st0KnI7IoRiWS7mzBLbcYsmHMvpkdjI49RWciYmrgmYlTYu0RM4b/8NcFKxuW7bqgUyZ0lGRiIWG+9nQnNS1nWfAZIy+xAeJYoD24ep2606bdmuLal72GLz5yAkpNuVaS5WtsnslYGV7EUFOFmq6ZZMpP0K4Fm3SL5V5FKQVrw5rzvZJqtaRYWyIUno3eIofOTPIvvuc/c8/5m/iXH1jnZ/5TxHs/tpeVyR1MhoLpDU0dBXK7Qe3Wuah63H55l3//b/exritmp2P2bmqzKdZ0Uk+aWWZaloU40E4tnbJiwsd8wyvhl3/nW/jUX67SWzNcONpHq5i4Y9kQnkkUWtXMpYZnTh1i5/QWVvIBRdqlNRExaTWjM+coD57Bjc7T2fcKKn2RqhghLj3O7LQjHVmi7mb+8vd+kRPLi9z6mn/Or/31Gd7/x5/hiWdXKfLAouhTJQIbqUYMMBKkpku3hslNXe4/8wSyG9O3Kctrgd6oJml7jCkpTcyhC+v8zh98nrsOHEROdKlb2/mxf/VTtBc289pv+BoO3vVxbnjVm0jTnSxZz+JyhXJ93nTrrbRdweOnTtPvjfBKs+Ecri+I8ozBYIDzBc8cWOfezz5IqIYganQUI1WHyjbsJjl+hy6EIaiAlw2TXTlHpYA4ItUZlRcIW5DJGK01MrcoNYuRiiQKeProGNatxwqD0ZZQDtFBUlNRZQU2qxGphKTG03xNJhH9AF4qcAMqvw5jMKELAisceeiRSo+0oGuPFjVKSHytsLXHRQKT6caQ6QxUFb4qMTIlCoY0BLzLUWKcGy08QeaYBAIRRk3iQkQRSiI9izYpKRHSZMh2SqQNWio00VdQnpvzTy3wHwQuA64HLgDvH1//h9rXf/A2JIT4QSHEw0KIh5dXPcJoghRUzo4xol2E1+NlpgZnx6aFGiEdUnkQdpzdWiBE2ZgMbI2tqwYDrDVKqSapyUGoC6iKppP3FSLUzSw+6BeNXUp8GCJkBXJIGLPndWyf1z6+wJrxTVCvDHjrUUKO05xe9HvC80vWRCSUOkfEhtEXfpvh8laeWbvE+z57P9XOIX29k6SlqVZ7/PxHf5+/vfev0bLDnlbK+WMXueH/ehdrlSaa1uSLq0ybdTZtk0xEc1SF5sGjj5PqCD+REssYWSqMV/TKku2dZracpRlZpJGmBgsjPFlHsH+HpJf2KSZiQpJRCg9RhBSWLYlFRiV3P5qzdR9E2jHdVWzVlk4nRVc5Kq5pxQNaiWjYJKlqXps0wVSe9mTMdBT4iX/2NkxbUCUp9ciyeOYcAUuwjiqOWR5cYL08Ty5LbF0ijCNtpzjpsS7n1JGTRMOaoV1GioJzg5ojFz2DlSXqvMuTlUTqOb7nVVeyf+slbLyC24DIRfTLCNvXHOnn/O3HB3Q7mpaJ0XqdZFqQRIJMe6SukakkkoFh5JmJC264eRvbJ9rc/aRm29ZZRjEoU+FXIQuhkZsKx9r5s9y4eZ4LbhuduMPq6XUm0lm6m1vQhg27TrHUoz78Gbqd65jbcxXDjTOw6zaquUm609uJo0kO/9WdPLp6iduun0M7B+0RVkHmJGVd4kqLUorf++idiJGlUuBLx2CjZE50qVzBZNxiKlVo29zs+xsVoTTM7rgaV80yu30vw5Ulds/M8K9+7Kf5wG/8Kre+/iaWjz/JD73je/hn3/stvPMdb6M7u4Xb7ngzE1t3s9CNkW2QrmDCBTaqAUtiiA8NhKybzXPm1AqhsjhnMNSockg37owT1ALOCrxTuKAIWtIyMZ5ALBtHtaPRyCMMQXioPEonGJ+jIk8d19RxQuUkXVERKBBekcQZXuZIFchGhqyMccUQVwNlGyoY5evkac2q71F6hQstgiwoXR8VKzwBJySl8BSRJhcCWUOwGq8VQUikD5T5EIdDRBCkIcgMG6D2AkS7aThlwFFQU+FVROUdjgInhtR+iBUOI4ZUdgSqUW7VdYmUGkWjq/9Kzz+pwIcQLoUQwgqJiAAAIABJREFUXGhSLv47L4xhzgLbX/St24Dz/8hj/LcQwk0hhJvmZpqsRulKItXMwXEDfKhQVY2o8kZ9giSQNCMTZAMckyO81wSRgNXIIJEmoMNzuM3QqFeURFmD3chhUOBHJb6sCV7gRbOtFjY8jyjGA0HhlEDEMbVvtuhSOJyvcU2advNWzAuUNk2AvFDjkc0LNEnn6iZKcGIXejCN0m1G8i0c9TcyYivfe8sraWfXoVYOc/jYKtmmzfzS9303P/+udzFdjfjhn/xe3vEj7+CayQkOHr2XS48/wBtf1+K2K6cJ/RilpmilbdYPHaQUl5iIAjLTiNSzfTDg+u1zzC0Eko6hjj0+DoQ0YLsJSSJw3ZibduW89vIWol7gmivfyKxytOoKbxKGQjC3VbOOQDDDzVdp2rrpVudmK2o8qa8oVRtRO0LYQNt1OllON99gS8cSqFjZuERiRgyX+iTLS+R1D9EyJJWgDAMUimGwuHiKoV2jciOmOzFrq0u0hWNPR/Mt3/8zZN2rqYoBgxCx0itY9B3SfgHl4yzky9z++jlufdNX01ucIot34OSQkOe0i0USvYLKc37/nkO88837+JrbMhamNtN2mlyMSFue3d2EWVtR9iyTnYgLiWH1wCI/9YMfYUlY7nrwGOkEaB8IsebSsiJNLCxViLJk9VyPvbv6zM5OwqY9uNYCXm1hLtvC/MICQQwwooXbOILKJnH9DYoH/oxk0+UsPvUI3b0LvOJb38EPfvUtnDu+QVARfgQdZah9yTfdto9AQhQ53vb2t6KrnEwU4ArWiwRbWd7y2q/lpl07yZRniKc/gBtvfBm3X38L3UgRxZrjJ4+wfed2LrgeMqv4xV/8Be7+2wc4ceokDzz0EJ+85z5+5/f+hLP9i/zcj/8I3/LaNzHqeebbbWonGIqYVMakA4ntW3pLaxy/eIHVvuLE6hnKqk+QllIMsGVFIiICFd5bgnBEOiBdE6AhxXPQPYHyAiUad6urwXuLFAHnR9S1ROWKtBoQkVPQYuhgpGoqGQh1gvQtCjXAJhYrNEpH1Koi1xVZlpFUMR2dUuORZgS1woQMb0cY7bGloKoN2jZu0kJqrJaI0DSZ1rmxNNohbdQE/AhLpGOMSLA1SBPhpEEaTSuO0X4EIUKLFCMzkrSDkS2EmsYqjVPQlYK2yqitJyiNksk/pTy/5PyTCrwQYvOLPv1m4DmFzV3AW4UQsRBiN3A58OCX85iKGCHbWDvAVssNIhg55s2MEEQNGEwLhIoJISKIlCCyJnJLeLyox07Uhl3RBGzL8bikmb3rGsTII/sFYjBAWlAuEHyFkB4pImylGoOFitDJBEQpJgOpxw4moV4kgWyi+nywzxMk67oeI0mbEAQlNPWoQi4dQxSrrHdv5q7Pj/izB05QLx5mRaT8+m9+jGTyFuJuSnmph5rczh/81p/whjuuo3/kCVbOnePOX/tJrr1ac259g099ZpJtr38Xu2/+bma37IGqZMMXDC+M2J3Msb+WtPLAN75+M/uTs1z1NRqRKoSukbGESKMZ4qMIMbKsT25h/y038lWv/GYeWUvIncGGIRuDPon1+BXBd/7sx/n1v1rnkSOB+a2gVUniIzqyouMDo1ET6D2t5plimgnRJVhDNBggLh5F65ql9XVkW1FEnkxKxDAHUdINNfb8eYIdklZDhLPUvs9gMELajCqdZKNIuOsv/ogNO8KlO+kk80xOdslUyqIN2HKNxF3k8cMF9x+aInrF26iLNer+GivlJWwtmDIRCy3FB79L8lXfdysnTj1LrAVGer7njRHv/aGr2Znl/OhP7uY//NyrmAyOYrnizDrc8DLDFp2yJRWsnoeWyJjaqCj9e7n/T9/Js8dfxScO/jA/8DOXs3NqOxPbX0f/9AOEashwcBGzMMXq0S+STm9Hmk2cWVMcfvzTxK02ycvuYOWhP2fTDddQL7e491Of5PaXX83qao63Cp2kVM7TjiUf/+IRrtqRsDIEJ+Bt7/g2VpdgemqBb3nDTt786qs5dPhRnnjiKfZsmWYyzpjoWPpyK7PbE26/fJmzxw7zyi0TfN81iu1z01y2eyu3pAnbpia453Mfx43O8rVfdQd1kdPVHbpb5vmZ//ib7Ns+z5RIGTlJFSy+36O32qewo0Yeaz35asGDdx9gsHKMUZWgmcTLmiJUEFLiKGqWq3WMEhqhBM6Pw0BcINIx3kEtAsFoVNZhWHmUSRtZooqpgqG2gjg00sUEQzzOgC2DoypTSqtwkaPIN8gqzaSbIJQGVYAsFUql5NZTastyuUruLT4oWjIhsRVKlAQ3pC1B1yOCG4AvEF6gyBB1BFRY20MIQ16OmoQ3ZVGRwlU10kfY2uBEBxtZrM6pZYmrHcqPUGVOpkzjzFUen3giJNr7Rr79FZ4vRyb5h8CXgP1CiLNCiB8AfkUI8YQQ4iDwWuBfA4QQngT+GHgKuBv4lyGELytY0IURdSjRIsKIVqNQeS5cA4nQ6VjOaACJFxBCTSBHBolCIAJIoXHWjQsvTeCvHI9UxrF41AGfWyhr7KhopJShiQcD0DrC+Sbnw4UCL2psLbFj6mTT3nsC7u8bnxjPqZQayz3DOFoM2L0JJm7kgSNDtu8t+d63bOJpN0u3vQU1N8GGv8hUaehFgXYnx9l11LkNRF5QnHyIr/+2t7N/8qsYhWu5WBQcO7DEzI4Jlk4eZ8fW3dh+wZCYb9o74jXXwmXzHTqyxyteHRFPeIIskVGDLw5e4wyU3iKNZXZiM1mo+fpXnOCV3UNs3jLH9VdfzxVbr2Xf5VdhpyPWli6wdW/gwWc8myZnacWKrFUijOYXfv3HSMWIgoALPSQVhVvBZx4bB1rtmDzPcS5gNwqK0hLHgVZbE1GAKvGZpyPb2MhRDgXd1jQhBIypmRGOfa+9lThOCS6i00CjkU7Q0orFUQ/dniIfDZnYNs/bfvydHPqr+5kQGqVrRtUQbQJqSvHyG7az8xWX8/nf+V12TO+lCCPqruHwswN02OCN/+bbiSdX+KNf/zuOnKl50ys2cfzSGg88JfA6pwJ+8aev5ZpJxXe+PWbj6JeY3Bxx8dx1MPoCOs/50G+8j/07P8+rX/savPdMoBBDQRRNItOYfHoTk6qPXVrBXHkzK4e/AENPtbTG5OwGw6UjmERwxeYY43OqctjkDHjNbDrBiSXFTiV59/ffwV/f+aeEjmc43OB/fOIYnzlwiU4rZtv+Tayc79MbjCikRh25mw9++N9z94FVTHeGj3zwnTy1scKpw5e44/or+Mj9j7HqBnz/N7+N/NQ9XDl/hsQGIlGzvrbG5Xsn+dyhSzxx9jzbjaBdCoY6otCOAknuJc5aXF1y8uQyp47lSDEEOUQoiZIxdRg1bCdAiAFSVFhrUboJsdZaNqgDBboqSQWo4EjTiNKVBKMY2BJvFEQaryWuLHF5ia+r5r+7d0xkEbquULZApYKQePrVACeW0CpHhhHBDlFWIdcdC+k0mY5wtqDyDae+yC3OaTaKmqDTpu4EMZ4g+LHAIjTsovHo2IuKot7AusYdG7zFaNAyYEILI1vo0OwVTKzJpccpjY4MsTDoSqKEGuNSvvIlq3ipXvv/m3PT9Wl46K834dUkCo3zq4hQjQu6a0YgyuDs6PkuWWEQ1jfbVRnhKCDPG2iYjcAX+KoE51EuNEW9bBgYsgkGJGhQkxE+EtCeRqiI4Aq8c0hhEb6Zz4faIkcDQgXk1fOLBu/987LORg/fFE8pGr691M11RQxScuGBATM3v4dTq+f44EOWuHs9D917J63OLM88/nds3ruLbtzhyQNPILuWV+/bw3SrZqqI2H3NZXSufRPp7DQf+9y9rB9awmrBu3/2XbzvP/8SE8NFPv3wCSbzPgf+YAdHH7nIX/z+iMv2R3zt21dJZERQFUKYJprQQhACXEBlsLh+EyY/RVyl3POlExxefgsnOhvs5WHayTwUhsXkcsozn+RCEWjbijuuMKzFBY8+Juh2AueGcD439IppECWmoTYzoVssDWqcSJrXBw++xClHhWYuNghKKtdlKATGK6IosNavMKZxr27fNs3i8eMEM0GkMqp2QSxaqKFk7cI5iu0tfH+JPd0u2+e3c+bYkIW9s+zf0uPiyWf5woMrdLYlbJ6J/2/u3jNKz6u81792ecpbp2lGI2kkjbplS5YsF2G5YzBgTDElgUAgwYHECQktQCCQvwk1CZAAAduUEGyqwYCNscEVVzCusiRLsrpGZXp521P33ufDM+acz//wgXO0lpbW0tLSmvY+7973/ftdFwP1EmqghyDt0PY6qLhMkjYZDGqsOXWS6KTgP79jeM/7VjK1Y5ROGLN9V8qRRkatU2FM56wsO669/T2o5BH++tUP8vK3nsc1H3sYWwqotBIuvXAzopxw1y92s+O5W9l+/w2sWbuNcbuCZcPLce0jpHqQ8cnDVBrP0nvGm7j7++9naMFK7n3wfo4dlqzdvJp/uu6ntNopoqxxmUFLj1aactrqhdCJyIISlfoiBk2VXXsf4gUvuJDf/PYBVLUbkc+yadvrePy+XzLUP8jK1Ut4Ytc4uZvm5a96JTfeegdWWPJWi3Ryjk/91QV84Wd7WbzpVEqJ5skdz1KOBOvOGGDn/iM4Olz80j/m+IE9HDi8G+1J0laTzAlKyqO7NyDzNKoU0NXrqPTlfPi9V+NsiVwmaBuQ+zlZ2kIKjcQnyyJSZ4rXj3WFI3m+b9JIpvG1pp3EpGbeFBFRCH7IkULRaGVYZfGtIjdpcXvGK3LyWU5ZKpxx+EEPrSgGDFpawkARpwlB2Sc1IWSGsl+EPHIl6NYSqxzCQUmHGGnxAn9+QVw8AZQMQBb+WaEFzmqgyMr7ysO6YpRjTFHeEqqFcgF5bpA6pJ0YPAdWJ7g8R9gAI33ieLY44FrHX135zSecc2f9/322/mE0WR3zpLWU3GUIUSkokYCQMUJEuCylYKPJ36ntEAmQYEVSPGhVGesUVhXt1udHKM+jDKRUYAtZtnSgrMRGDpmBcAnOFJwKZTXOhMUbh8mwNsdaWXyMzsF8OgajCvqlc/N+1eLfPF9+Es9vXLWm02jTu+1c7Kk9fOfffsqvfvoDfvmj65gZP8KzTzzAyiULGXnuEMeOHUG4BkO2jq8X0R6Z5LQzuxhe2s3s7uO0xzwWGR8Vtuhe0M/Ctevwog5zjZzhYJaLB2exvqK3pDl1Q4vpIxmV1jJUdxUtQeUSjMEqg9QJNlfgAgb8MtlcD83eLjYtgUvOCumNOqxe202zOQE0OO9lV7J8zRJ0bhmJFHccjhleUmfRIsfgkEJqDxdZyjKjJHMCmSJszGw7w84zg+J2ilAxpbKk15boTgJmk4woyvBFiThqY4loRdNoM4VJFdUs54YvX0nKKK2wRFbVBC2fditlJJlksg/CzJAkJWa1Rzz1FKJrlM2Dq9i05AL2HI/JBip09yu8QFNeGJImCdZzhJ2QTAukC5gyHe56NOCL9yS87PyAZOQIwaI2J49FRHlGT1hlSmbEnQQjB3js1juoiSHOuvwC3vCBa9j0on5esWURm9ZWeHL/FAdGPLoGl3D7F69j7Xlv4GDjOF0qQlrJzMFnCUKJmpkm95dycvftPHTnbhYuWcv4Cc0nv3cvH//Pm+jWIc6BjQ0VFxKplMEwpOIUXfU+1m9+EbXKIE8d2sm0hO/f+xBCVfGEpX/JUuZO7GSgFNNpH2fl8tO59IozOH3dBj72r5/mBaf2s2lAcNrwEs7dtI4lKxbSGJvkBWecTXtmistetpWonvCP730X3T19OCWYHJlm+55dSJniuQKQZTOJDNS8ZUmiZRs36zEzK8njBlJZpNRImZGlLZT0cU6SmbhIsziJQBXeVD8ky4uyoy8rZLFDZoqyLEGi8JxEyRDrPNLUUgtKaKcQbh7d7RU6v7Bl6CmF5PPo7qgzhpYZWnaQwpBnGYqQPMkJEofMHbnNkcrhixirSkRojA6LxI+QJHGHPC2Kl3lWMKo8L8DkhSsiSWKciLAuJbENMtcmSts4WciMYlElp4wUIS7tUFIS57fR+DgbkhiHwhDqLkq6gv49oCD/IGBjX73+U9f8xdu60N5AgRyVeYEWkAJrEpTU8zTJbJ43U9AhndMgQ4SRBTTMWQrrtgObI+bxwNbMG5NQRdQdgXUUur15W5RVFMiBeYm3EBKcQM4jg6UDsgLr67AFYlgpcmOQxhR/5yukkEhncK4AaSkVIkZjcMsYSV/BT7/6TfLlg+wbyZBhyIrhVfRWF7L3+Ch93QsI5nJqK5dQik/wnndcSl88ipItNp++kqef/Ca94gQnd+2kJ5hiVb2bPff+lO5Om8d3buefzzdccbmimp4gXFKm7En27lvElo99ATG5g87kCTzpYYVGZWVimeB1DKkAHUs6cT+isQOvXkJ7e1h46BhywSvxpp9g8fqFnBiP6FMzPHdgknrFA6+EN9tg47YK/QHsfsZQqlhyrw9feIxRppNXsPEM/YOLmJ2bQ4QOz+WIQDEd50TOUAkEUmua8QRShAR+gDEWIT10oImyaW77yYMY+gllzuEjLZaf9xKOHXqcuqghjcWZiCB16CBkRgR02xTZJ9nfEgjZx8KKYagsqPbWyIOUNITSbI5VlrzdpEtpplsNfrO7icw8WifbdA/7PHlbxMJlilUbB7nonBJH98xQWryACRHzqW98jLu/9R1uv347zzzzELV8CXbpFuzAWvr7NYuXnsf+HQ9x/9GTPHfXPtZ0R/QtW0Q8+yS+m2X/iUOsXn8h93//X3hqzw4Gghof+/TXmZaKv/7bD1EfDImTNhHg6yoIQ1cSYrKY8aRNo9Pi5P4RJqYPMttukhufaujRsBntJCNvScZOzNG3+Vy6lwxxyupFbFpyOuNTB/jOD39G7Fd44MEnOH9LF/fdv48wXMbwphpdpkbep3nkF/ez5fRtfP67N6AjxVxuufLNLyc/9Dg9XpvjMxZf9+CVBVkaYVQBF+s0U2rdZbQnWbGqj/7+PkxWIaGDL2rktomVAiUqeNYgXQjWYFSbmBgPH5cYUpOiRA1DghS6QPuKjCSeo6R9nJUYWWTHhcgw1seJDC0dU1GTpCyotC1TKmGBLDFJ8XOCzpFS4TKN5wlyY1EyJxAeeZxS8z3SNKUkPEyW43SOcxatSkgVIGSZ0PNw5GTWYZxDC43vKTQB5AJBCemKm2iadEAWkm6ZOTrCYqVB+oI8SXC2gKqVvDLOgHY5VhgyEXPbd7f/3w8bwzkslsy0sNZgbELmOvPcZw+HwklRLFIxCKmxSISS89HFHCVyhMyK32RIrbFSg/CRIkC4sBANzF8Hi5N4UW5wOZDb+athoWqzDhzFiMWKrLg1aL+YwdtiTIQTaKdwoYfQAqIUk+WgQYoQL8swEwn5opdw+Nk5li7v4oWXLKE5CYt7l7Dcr9OcbvHksecQ1qBrHsHQAKkTLF03RCNewhmXXkFv/1n0DE6hpiJmjj1AGo2wsP8Udh3fw/4j4zy+9wDrvJQl2yTdFVHk76cA1Ud5aZn02Cim56WUBWQ2RWYRqZmlFAukKuGnlkZjjO6FCaaT48kMr9pF15acdGNOvsBwcMde1lZOsGP3CF0hKCkpe44js5LO/jYuzekLDbbkEdk5Ytug1k6oxzkqrDM200Eajc1A6QARO3ypCENJmkiiOCZJc9pJhyTK8VWOsRF+blCqj1kbIpVjppUzMNzNww9cS613MYOLAs5aV2NxrQ8VWowwlGZietEcHJ8GWSUvS5JyibRHkIgWYZyxICleZKErYbTDLPBZd+Y2KhMBb3vxEC9YDctrHiMtzabTQm75yRQXDtf4k784i/FDs7z+NeuRtsXeew7z3Z3v5qo/O5O5uaeorQ35wbe/x9ILtnLPjh9TqXvM4XHXsR3MiC4O7X6cqdFJDjz9LMdHd7Hvoc+w8Oxz2P7o49SXDrL+9GF+tXOKcn/I5GhcLHTKEpGljDUThNemUfWQnSL6O51O0xqLOGfL6dR9jUsjXGioih76hlbw9vd8kOG1yzl35WYuufQFHNz3ICad4vBTj/CRv3kPmxd18ejuJq+4/FX8yz9fwWu3ruTo7jvobpT5/Mevwsk53v1nb6QhctYvPYMbv/QlPvyxb3ByroJvLTadxBcGRfG9jaMmQVBIq9NOm8bsBI1Om1Rm1EUZIwpzkswUykbgclIRIZRF5T5+FmCsBN9HywpSFa8pY5o4EeM7R8XvoZFKUuWTpIYshEz5WFlIQqSVVPwQYsi1QPkekVD0pJKyX8JzZdJYYaUlyQQ6F5RkCaslolKmYy1O52iREABGSKKUYm9lE4xtkdsWlgytSsXODoMxGXnmQAQoVdBqTe7h6SoCr3hmCYnvBcUzxBhKpRJ+oDE2otGexrqMFEfmBE78z5V9fyAn+E9ec/Wbh1HOIV0GroxCFMJd5yOEh5AJCoNzFIJqawvJhzVg57V8pgzCokS5uBbNW6AwOcLlhdRbyiKbbm1hPDfzeF8N0vK7G0KxQLUIp5BWYtNWsby1IGwhFjZ5kdkVuUE4h/ALVGkqNS6LkK4P02nhiXEaTcW1d03w1IMNVLXE0yOjtEPH7p072bByDboW0Gi0WT+wiEWVXrat7OLUzQsRQ+uZm3qcyamdzB1uc9O9MY9N13li73YOHdXsPT6KnZ3iNRsWccY6g+oNSZoaHfYxNTWHLw4SHRvk4FM/Z+mKC3FjB7DSx0sEWeawaYE5PXYkpm+oxeRoQN6IiTopi3v7WHDiKRbWBf2DAVannL35AqZ27CBWKYNVn2MHEkphDydHLaKW4+UeLaFwiUTUKrRlQk1YkBFyHryUZ21sKcPGKfWqpt1K8P0KNoOusARZi06SkymPOI0xMsVTgEkRmYO8Qb20ENuc4iWXncKPbn0YUQ6RSNrZcQa7M9LAIttT1PIZAr+b3q5+aLYoUyItt7BW4PspeZBT9UNu+/E4Wy/ayrmrD8HUKAODgBdz+Jjjh0+E9PgJ/3bjOBcsaLFpRco73/t6ZkYeonFwP3/z+t9wvPJa1l3wFhK7iA0vvYifX/vfJM0GUy5nXXcJOerzuo/+G61jU6y/8mpGnpuj3+3m6Mg+5o7HjB9p8uUfPUGzupzpYyNYlSNFSG/N8K9/ex6XXXYm77jqIh64bw9T4xGnbBrg2OE237z2I+TGcdedO2jFEZ+95q1UXYPdO47T0ys5efIEv/zZHcjp7fzwP2/m1t/s5eE9R9G9Mbf85Lv80VveSTJ9iL6+Js+cCLnlv29g0+Y6x3ft4OcP7yJKHbfd9RCf++y/MTZ2iHimxR1330KUwHijg6xWika5L5Ba0G44cD6VoIewGlLq9lixYg1BIOkkDkmGFT5CFgsa60BaD2EUSoLBYFVM6iJM1kJKNY8BUdhckpIQSItNU5yztPMOVVUIgSwpaZaSGgikwpgYUkfVqmK8UvKJYo2viox6nil83yMTGZmwKEtxW0ejvJBM+eAVbxq+p1CeK8qYTuAIcFYQyhRPSPLUFNpP7UDkheVNKnKKhTFSIhJQFQk2RTtvvmBlisCIEPihhxEWjSgWtyLhZ99+8v+FE7zFKYsVipwYKaLffQENpjCnzP96nvEicUWe1pnCo2pNkZhRkiKOXkGqElIUs3R04XEt5uXz34znF8ymKFOQ5rgkgTRCmLywmlvD818mqYokvqOIXEoJ2lqkp4t4VFrMDj1h8STkWuFd+l6eeGSWsFli89ASMm8Knc9Q0SX8dsZLLngFk+2Yvu5BJqdnIZpl/aoKJTLMsUM0dv2aU9f3snf7cUbzkLGKJPcTlCxh/DZWp5SrBs8bR1RCjGuiyn143mlUskmGT61D535SfQpOn0or7iVKHVFkUKWQZmpRIkDkQBRQZoZpP+PEaJv2zBx5ucTcTMLwcJ1FKwOSzginrqsQzCmirEMmQIQwSowr1zC+RRpFuVtTnWvgtzokIidup2S5oN1o0+cq0BD0lhYgE40na1TrCisSMqXIhCFPDRUVooREuzKRFcTCR5YcfiBxeULbE9x0w3aWdC1BBh5xIKmoKnlYJrGaWQxTcoYF3Y7RyUkyVcHpMn7L0ptoyASi7bj7kQlmOhk7b/o+SxcbvIpD1xVhl2J02nHlhcvIy5ZlK0Le9MWvc8fIGWTLhujxcoav/B4br9zIygsu5/DJWZ48eIKbv3YLnu0gXIbyDXMJTMgx9n7931i+WDOy/RFUdY76inXcc88h9GA/2xsCf1mdZ5/ejy81ZJKpTsIzz17Py5cd4pl9DV72xms5NN7hqne8lV8/Mc6HPnAFV764xSmt5/jiP67iI++9iL/6yI08vncC0wUHRk4St0PW9nazeelmek9ZxOYLh/nZD77BUNcq3v2Wd/KBd23j0LGQXm+Y6/79B3zgb/+e2qzlwvPX8qkvvQsnp7jtth/Tv3It9/xqH2uHl3Le+Vcx3Zzh8ku3obwEVEBJ9xCoMn4YkNiYtm2QNFKmp2ZJzBx5bumqiWJP40xRUkThZCGXyRU4WaBIlBF4uUbLGsZkZJnBkSBkjs5LpHmA0yHGGUqhJokkcSqwGQReh9DL6ZgMIz18K+hYgyiVsGlGNcggTSBN8FSOSSOssCQ2Jc4znJKY0GOuM0uaJmRpjHYaaxSZSUnzBCs6hcJTQDNJSIxF+z5aajwniti1KGTcWpXA+ZjcYWshLklIXQaeIsfhSw/tApT1MTEo4+E8H6MC+D3k4P8gjE4OgSPFiQ5aLiE3E0jZBhRKzkO9XJncNgqOhU1RWkBmkEoj8hKOFCNjpOsCZsmEQ4oi0SK1KoS8DkwrKTRgQmCNQ6j55EtiCy0YOeQC6ZVxUuFkinUWpWrkpgGI/21mkrJACZtCCmylAOlg82Xkjz9EY6TKyG9/wOBr/5F9JxTH79vJqkveyLdvuJEs72ZwQZmRyUNsWbWUaSNZPbyG1UvKnFbPKenFbPhDCSJ5AAAgAElEQVSTj3D/l/6YR/c0aIs1PDN+kJbw8W2MpwfJsyaikzLtJNWBfsr1Nl5eRtQiZqd/Tu+KKrOtBkPDhu7JQzSyjbSTAbLpUYQO6Jm1zE0LAhOje0t0WuNUV1VZMmto7I9JVIe+vi7yOTB2MfnJCSanKix7ydVsXruL/onHqDcbvOjMkPse0ByaiFmzuMTYZAvfVagM+XR3PHZNVSh1aTqdDnRV6FQ80smIvBQzM96iXC0zOS0IwiozE+PU6n3IcBpnG0xZQ3dSwM86StNfBoVBBnXCxBD2ecS5ocevks/uZ6CqGKqlZKlgY5/gzBf1cv+OA4RBL3FlgEbWYhmaFMtzJyN2HIjor9W47MKYyzcJyNucd9kq7vv2EWaN4VUvHeBHd+8i0mXqpYQb//4qNg4v5nVbruGpp+YIF+3hPZ+5jvPXlxgZr3Lo5i/TV1dMxw4/bCOzkHER00TwrYMz1MYSkrv+Eukv4wt37aPVvZSnP3crl15wCZ+46T56uhM6qUfDWfq6NX7tHfhelZ7aQ6xc2sUb//LDvPYlF/PDb97IJz51K1+7VpB0HOsaPdz/5P0MD1ZIWjHlQNAvFvLGtVN8Z7rCrQf2EsVNzPEqb/rAX7K8vILpk8+y/5lzeOP5A6x/0SI2b/swT9z6fobWnMlL3/9Zvvu599M83uCRH9/Njx++E089x8PPjPB351zAlo3ncevPHmbVakUjjgsTU2ZQ0hD6PiqzaC+jM60haVEpLSA1OdYIrMwIvYA8aSKEwBMlkJq4neH7AS2XYAKPvNWhFJbJsjY2l0itcAiSuE0VgZMBY3mG0oIgSEkig+ssxtiYUBjStENbSWpejZm0hQ58ZtuOitSEoU+aGZQEmenixK8ds3GTmnP0Kh8lYqJc0VZtytInMTlhAMZIhE4RFkqqNv85ecX4RhfFrMCEOANK5+R5TOhJaEVYqQlyjRIW4XVIrcG6Mk7pecMc+G6+UPk/L7L+gZzgMZDGuFzjbBupFRKNwCKVQ1kNvkUHhWO1IK35RSbVWhApQoG0GkEH5wSeDFDSIlROrnKsNlgvAGwxZnHFbcDMu1VdLsGAyAUkDpe7wivyvGibCC09LK6QZhtwmUTIcvF/OY2zFtVdhUUlWNGPOOUy0k6DR777IE9/7TscGp3hhz+9ie4FK/HSJuMNw/6OYc/YcSaPjrAmjhgeGiJvd7H8ostpHz3I8OBa+lduZbjqUKOiWEDJkDyAPDFoJWlmGUF7DL9eIg8tjph6bwmjMsplH2cySv0BXf5BGuEknghBJuSiC4Rjaq6MjBOirtMIOwULO8t8FI5KOaDTWEFrajvJTMT0sX2sWhlz9pBl4fKAwa6UvvWn8MpXLaIeZFRrOWbBWgY3vZK8ZzOmOsDgohgpY3rDEoN+N3mUEPYNEdkQr1JDZ4Kg3ocqh9T6usi7BNJUSV0Nrw2mnrPABAwHVWwckiUhoTUo2nQFZbwkJc0ian0lwt6YzBX5fq8/xJkK8Zih5C8iaQmiWUPs1ZiKGzzzTETS0Gw9BbqaAWWds3x5F17Z51jUxakXn4pSVTasLrPl/HN569+8ij/9xh185Atns3hLL4vWLuTv/+P7XP3GK9iXSq778vXMtCc4PjGCEIJWWqYlDSOH53jX+97P6iXbWOiOMtlMaSvHYycMJw6Mc1R7fOFnv0D4CXlSwneSuhEEuWV4gc9gd0rczOgJF/HwLTdx6dlbeeixB9m2uY/OjOPiC4c4/txB3rRtgPX1Nt0lj7wl+OTnvs4IIQeyGabjFEkXGzevZ/XyFUyePMb4XJmbrv8Ptr4s42v/fhe3f/XzbH3BNlauW8GTN32dqH2Q4VUL+M2OnTzy8HZ8T/HZD/83L778MhrP7WbxMp+Dhx0VIVAYMpNCXsDmYgSJsTTy/cxmbRITkrkOSgt85ZFlWXFjz3My28HSwa9oOjYhU4A1eKp4CAs/RPkVTBoh0oSSHxJLTSs3VAlRKkXOKaqmStnP8LxsvrTooVJJaixSOHSUIF1KrCStzCDRuEzjrC76FiqgEvagPEFiNS1VJ5L6dzHoUJRxUQAiIs0MGZZYd0hkRGrboE0x5jEKT9jiBO8FaK+CsT6Zp8mEQJdLWM9ick2o+vDw8KVCaYeTjiYZiRS49H+u7PuDOMEL6SPCBGsDtCnm7k43cU5iqeC8OUTk42SOlAJrMywgvYLsZqUtwGHzxQChLMZECAwCh/KrYB2WHOcVV0RhRfHuNi8IKPgyEpNkCCWRXoqQz0v4FMKFBaNGKJzM5ufzGcjiTyO9YgF7OMVujLGBJjAzNGwZoTp0Lxjm4PED5JmmXKsxPbycbi+n88TTnLniPFrLU06vXYDWMyw6ZxM3fPPbvGTJJF0bz+bErx9ipNPmN5M5pcEaeZ6TJEULbsoJauUqxw81wK/giRzSDJvnaKsIS5IsyZGZZurA9+gX3bQ8i4kCRqfHsFkNf80G2PdrzFyK7fh4eobV6wR+rU47ajEtSvS0hpkJfQbtbmZ2/wY167Hy3LcSrBCU5Z3UXvgh3r/lW+y+7wgjBzu0MsXCoQv50Lc/yt+cP8Bwj2LnsSlsTVDyamjpgTeMq2R41TKLFi4nFRFTO/aT5QmultDdoxijjm/naPRME2QRdWWwaWH7qng9oFtUezXJ1BQbzsgwnTpRMkfFl5w4ljI59SR+9zA6aZE0FFNRzJAyxKbEuae2WeDK7NsR01uR6MULmWmnLOo+yMYLVvH0A/vYetXfcc1rP8svb9yErZ7k3q9czKc/4lj3qlfy0j85j6v+6HxGZ07y0+98gcFujyTP6C37TM81KZcMGR7XvusCGoee5k3r9vCBDx5hVmtEeoxICjpZznBQ5ooXb+Bbd+2kRMRsBO+8ciXX33aEC5es4uTsBGmPRNYj3vySNXzuus/gqzs5OZPx6CNvojNp2X5ek4k44uCOe1g1LhkNLF/9p9cx3OMIsjKTjWlO37KeY+Oa6oLVLF6dcefO3ZQa07zu1cu55PyFrFu8nsrCSc64rM7ojl7WLj6Dua7dfOXzP+P6b36SL370P1h7quKZQyepLF3I8elpLnjBxcxFY0wcO0hVl5ClkCiLKOWCtsupNEocfHaKxZedQNsq5EkhsRYS5xSBrpCkbTIcOI1nQqSJyDEkQiNzjcqaCGVIS5owKwpzzhZWtyyDFEmiU5RNibJiXBKmOalLSXyoWgmJpKUFWmmEdTjbxgoJRiOFhxGaxLRJsgzfq+FLEJ2UwAuxRuCqiiQuKJEyr+JJiXQWYUOkLPhFyhZgMU87OlYhpcDl2fwp2hJIWYRGEgj8EkpkZLKDMYqSDpBYjLUEyscC6e/h6fwHcYJ31mKzAbTzsOEyqKybT8+khWtV1gtDuwpAaJQK0aqMkIX82QkPhMKJ561KRQtM/p9YX8F8tLJonFr4XXP1dx+HKb4gwoHJwOUpzhS59pxoPntfjHeep0SaPAMpUHTQqkJeBvIzyH5eJWmOc8rGPg48FzE9dS/XvP9yelptju9/hKce+S3nnn8Bi888h7nugAO3/hp98iYueeNr6N+0jVcN70GK7Ty7yzEzsYtHdx6l7AcEaU5PrRsXNVDSUq0F1NMmdeEwZhorwfgd8AV+SSJ9D6kFnp9QKfsEXoomR3sGFwSkSDq6SktK2q2cqJ2SW+j1ylQHlyL8PowbZGBLlUDUUOVu+soJC6tH0C5lmXkaOztNag4gaquonz2EWhLw7HMHeM0/XM0/feZzeMZnYqKBVjFBcwSiE4jWOHE2iRExPV1DvOoVV3DF1iu49KV/xGve9BYWbzqFih7CKwlaska100s7K6FMF6lXo51IMmbmG38h1R5BJynjapBrg5IeeRaTxz6ZcTRmp3CpYOmyFQwuXEH3AFSF46JXegx0O1pNQX31Jg4fnuLGLyfcfNOzHD62gNtufJirzoOHHvwRT117L9/5cYUn5ixT7R62btqE6jhuuukhpsZG6UwdQ8WWVqNJqCFJwfcE9WqZkpvkP28Y48nUMRcrKqcsJsIR2ZyTEwn/dfOzBGGNVkmwvF7h0X1V/Mhw97H9TGcNSj3QGJvlxm/cwfv+5BWku+/hr99Q5Z8/+HPu+PUhvnfTbYzs20M6AQPdiqEFdYQX0al309PvmJ3ZQxKEvPrqP8crD/LcuEcyd5SF69cgVlzImS86j/pAhVLV48ivnmLwjCq9Z5xCb30DW89cyV++7aNc9Xfv5Nav3s76/oXMjBxmy8qtPL3jV+w4fBinPSpBiEsjPFnwmKynqOSDjB+egkgiXBVPlvH8Ek4qpBYYmyBVAEicynGeIQxKBLKEshmegEDqYmSfemRJG6WLHoz2FNIaRCioOo3nPMpO44wlyQ3KCMq+T9N0cGUwfk4uMqwypEYRERAH8ypQF0MEZVem5nJqVuNbjTQ5JW3J5pIiLikkQrewurgZaF1wcqwBR1HKdDbEUwo3X3AyGKSnsbZ4Ynu+IMHS9jR5XkWHCuXl801ejWyn6Cij8nugSf5BpGi+9tVPXnPVn5XRhLh0CsEozqYosQJBo0jCyDLOpfMPWLCiyLMLpXAmRcyDipxwCCmR+NgsQVoHxuBMhkpyZJYhjJuXZxfibKWe39SbImGDQHhFtMYpAIczXvFDkBfqP5DzLVaDsD7YkMymaA1qn4bSUe57osLD92lqpw7Tf/FV/OuX72HSa7L5ha/n9N7l3P7zu3nnJz7JE/f8kr/6/66netYllPZ8l2dv/ABR71IWD1/MROc4+3Y1uPuEj1/x6Bs8jeW9sP3ACVxQpuo7XqoFb3l3N0E/JH5R11aej8OSO9CBj/At1tPIMKFsAqSXI8mIG12w6kzC8YPMTc1RVj7RdEpLCrrKCYFSLOjbQ0V2UesXeLqfkpvAnnE22dFHkBsuQHd2YNMTqFKZrlOWc/+/3kdcyXjgGz/hkafuAi/j0MkGZ571x7T2N2nXNWEoSNqCxIVs3/k0D//4Fh6652a6Vq7jnIvO5O7r72H1lhVcdlEfm4eHGGsNUqnXGB2bpSU1xnZQKmGRX8eZWVyYs67fpxp2IzOFlopW7tPOPUQiKSmoL91AICQjeyZYH0xw1Qc2MHVyhF/9wDK0IeW3dx7hVz+3nHrOUvqNpDE9wwu2jPD3n/s0Rx/7LbMdxwUveTdPt9v847e+xf7nJnnnn76esUPbOTn2HCptoV3KkN9FWwl6aglrfMOxY0e59c4R+rZtY+fOIywbCJieSqn5EUQ1fJHywjMrHB6NGQx7aKQpa5ctYWL0OB/6u7dxdOQxXnz2VkKruORPz2H3A4e45a5Zztr8Zp78zX3UdYro6dBpp+wcVWw/ljAWG440DOlYh2/+8Gv8/RuuZvXG01nQ00t1wUoGwioTUxNMuhZ33z1OdcGZLHQnSCPDwoWXMLJzF8mj32Td6TUuPrvCU3cdJ8HhL+8hSvr58T3f59RFwwyudJw4ME3oFeGDvJSBH4Cdo6QUQU3S6Uxz1pZz6Q2haTKMy3DOklsHSpGbTnGwQpE5iE2EkDnGWXIcsRMY4WMyBRKcEaQmJRcWqw0mzcAppJHkGubiDjrwaOUJuRBo4eFlGpEKtInRgNauGNtYgUGRGY0XGKRK8UKPpsnQ2mKzFIMkcoXRqaiAl8hTixf6uNwHAVLLIoon8qLHg0SKedSKcBhrEb5H5hRR5tAKdB6TuRzfC0nyhCzPCrmRdlgFTkh+9K1H/u9P0Tgh8cjJxTy7BB8rEiwHsCbC2haoBkhXoGWlAZGRuxjn0nnj0v/mrlsnsNIWZdN5ZoRzDqcMVhSLDChO8Ir5pqstNFm/AzdkBmcThFVI6+PJBOZ5N8yzKArxd4VYZViXIGoCd6IOZ76evVOXIY5MUllqOPLrB/j25z6OqPo0RzTtw23u2v44Sd3wD+/4KzZuOp9ffuXtrM8epbHvLtafuphThrppj+xi5GSZ6bmQHnJ6XI0li7p4+tAJSkEZqWOgxcZlEV2nB+Q6I9Q+uLAodmmF8sFIA15vwb+QFWRgqAQBfq1K7+A4S2YfZq4zR5g40sgiFy3HrybYdgdUg64F/cRuFluXBIsy8pOTWDShk9igBb6Hj49IxkizJxkIHd2p5enGfpozs+wcj3lyQnL3A7+gf+tWGuMOr1PDVwm9vZLe3m4Wr1vLRZe/mSceu5cPvPsaVlzyAo7sneBNV65kamwPh4/sZW6qzRnrh1jVI6grQ1cloEWDuTghmsloRpLJ2QbWNFFKoPwaYVDGlapEtJkdm2CuU2Fpb86LX76efHqapQMZ7/noYi45xWdiSnNIBzzy5EnGnpnlo5++gk2bX83U0af49g0j3PxEi1e8+yMMrtnE47f/lK/f8A2m8znG48PMxi20F2OVYCJMqZY6LEKwQJSYmW6zrMdx712PMVhVTLcVOXM0WyV0LeHyF2+hM1vHb3tQKnPamnWIhYYNrz6fe79zI+Y5ePnb/4Eb79vF3BEPGSf0l+aYTPeycWMXJ6KM5skuToxYGnMKjzLHT0g69HDtPbfyJ6/6Wx44eJxpW2Mi9RmbmYKKj6j2Mndoik/8y39y4sAoi7YMEvQvZzqcYvWqkOWv+yL1hS9i7IhmvFtz9MBuRqMa2/fu4KVnriFPJ/nav3+BNAbr+yRCYvMSKnNU/W5aWHLnM930mGuNMClCfKGwAjJZ7MA0gsAr5NzOFagCkBgKrIgSOb5MUKKF1ONIEyHJkNoviomRIxSKJM/JnMUYR6g1cTMmwEclOSZNmHOzdGSDtOwznXVwUhbqQKlwLkJ7hizxsaaLuZYE6ZAuQFNFOZ9ABxgd4mSA1j6q4hGnTRAJ1hWOZ5vPM57y+Ya7sEhZNOqlEGBTAhlR1hEua+DrAF945KZAsGhdxfc9VLmMLJVwvwdc8B/GCf7aT1/z52/pR5bWYfIWfiKRxptvu+VIJzGqq8jIC4HJNVKGKDkPHBMW6wxazcufDLg0Rnmi2JE6jdYOkRTLFgzY1KGEwBoQsthgC1G0W5WUYEEYV8QibXFzwAisNQhTkO+EVIg8wuUeqqRQDJEtX4157EcsWGsYH51l1fB6fh31MzE+zsDwRdz/6L309imCuTb9S0KqlT6efXwnmy77S/omvsfCZWuRfSuZPpowpgc4cmg7h6Im40bTTCMaMw3yqMOsF7O23EO/c7zrHSUqiwVKxIhczWsJLVCMolAKpxIkDqldkSxSjiogpSBtTyK1wKqcsVnNqnXL6RwfZWQmJ2lGEDo8mxFEDUpEZDKiXI5wPTXE1CGMaqJKMWmYwVyZww+cwCtbdk+FVCXcdt9jrN+4gYmnfsUCDH99/VfIrY/sZJi545QWV/mz9/+IrZedxviDM5y1posvfb6fN7yiyfuuvonnnjvKnb/4JL+9506ufkM3crSHS9cJjh5q06aXru5VNKNRTjYjTuldwKw3i4015dDSSiPGpifYs6/B8sW9dBMTDC9j4ZqzePqXP2HqOByO5vjjj72dmz/3Gz74vm2896Mvon99wqwTfPyTd5NUz+PBw0fo6unnHz77ZSZGR7n/1pvJp3ZizCQCQVfQouw0qlJGZxmTExHvueYTeEdu5y/e92oeu3MPQwOS1EjOWa14Zl/OygFB06SMzzkmoynOP7NGV+MEb/7rt/PtT93C5a95E0HQ5C+u/x5fvvr1LF7kkUcNBocCwlJM3DnCnimPkweaPHM8I6MHzCwiKbN4scNXTR744U08+NTN6PH7+Nndv+XsDefy0D33seGsLVQWDJBMTqL2foXPfPvzmKOTPPHYKGv6E6byIUZ+8SGIn+b0Fy3jpi8fwi4YYuyZ+7n7sXtYWQ946rmjfPmrP8JfUCYoKCKFdlNaUkD5br5z4li+us7w8Gm4eI7ACwr5hskLXovTKO2RxFGRhc9SPN8vdA9ZDlaB8sm1JrcZaI1JDL6SRC4hcsVYMQld0V6PwWQKg0JpQ96OEZmHVhKXWmTmU0k8EqXQcRMZaBIbFY1uXyGZITECTylil2JdRCAB4+GpwsPqWQi0Rypy8lzie0VxCQSoDCU1uUkwTqCVj3UZymkcGic14GEyhe9R8JYsSGFIsTgRIZ3B5PDj/xdO8AiNl/vQGcGXCaYLsiApZu6eR64yhDEY10TIHC+MkV7xrocMUEqjtcbgsMYUiRpdUCeF9MBTWDyckoXUm2IsU8QcC1Xf8xIPKedJd9LNc2Xyomk7v9D+P6Xa1lrSIER7GaSSaO0G1MQYLOrBHks4bob5/j2jJFnOD35+C9Mnfsvb/uGjrO3u4ZRtG3DZAt7zyc+zaSDlhQsOEh0/Qk9ZMrfnMUajY5zYc5zjrZy0VUbkCUGoCZIY3xoWWE1J5qzv9/FqQLWEy+cbt/PBU5xBCjcvQlG/U5cL3yG9HCcsvhb4nqAUaMqBRpmYPJ+lGfuQBnhxhdkTGe1GhunEKGdQpRqmPYVoTpOZ43hCYBtNvJlxfDXNlm0eA0OGpcMVbKa5/e6d7NpxlLGm4mhT8pk/vxpvbpRN521k2fp11Eenuf5jr+XFmzaSTn6JnXddx63/9TAvvui/OPPKl1DJcm69+yYGW0c5cCJg3+hRVp0ziAwUixesQWlLubtGva/GmMmpNruhkpFbj8hGzB2TbAjKbFvZQgWO0oIzeObxBrFXZ+XW07lg3VmcfGAHXXVBNrmXb37h+8h4FZXVr2F6zue6r92AbQiGTt/KVz79L+x9dhd5chjVScn9gG7pCG1OUBcMlhp0+y1eedpCnvnch1m7WHDbf91N2A9h03Da2pDbn+zwmosXMm0cQ/U6frNNT9dq+k79Y75wy1185cOfYcHakNuu/zg79xzi6re9kUwvJY8F0/E41s7hhT1MNjQnj83QkpKXbe7m0ZHDTI4rwp6UrriNto6zB0O6+l5NddlaPn7VEm7/yXV84rMfZWT2AGHmWLx8ORPDl/CRP/0wr3nth7DTjzNdOR13eJxZbwu2+kIsGT+85xMc3L6f2azM1uFefntijtkMSnVDlaJUqLVHHMdF+sxYfFfCkwbPVpg6OQW2BapOYvLidehKaF3Bphkmy9FhcWothxVc5jCA9EOMUDgn8XMPhE9u0yIQkWs8EaCdQnqScmLJjSHSBuWlBDLBZDmZrzGBT2wlaeTQvmJaRwhjiegiN4JQVFAOkiQCylSET8NEaBxVr04MWDVFbls445OYiIy48M1KjbBegT72LZmzpEmxNFYI0qyD0haXGRIS0DmByhCVmNSCSR3WSozQlHQJaT1MrvBV8D9+tP5BPOCdcBDWEHKQJKsgbA8qXEOGIKeKDJYghETpQaAXY+o4uooCk4iwaBwKJTVCKaxyWGRhYPFDZLWGrNYRZQ/lz0u6ZfGgF4J5bnsh5/jdOMeKAmNgCoUf+bwHVsrfQcyEAz/VCFVGTMaUfn0X0k5w9MnnaJz/LvJ9v+ZTd9zOgDFc+sq3s2vPbp668zb2BzUu2LiUP3vdq5i99Wv8+YtWUz1wP+desAEmt7N6wxpUK2BOjXHmpl6q9TH8rEYWSWaSlMk0QyqftRWfVy6v0LOoDGPjGCw2auCiNiKJkCZHmAyXJWjZRngJMgQjKZZcyuBr6K2X6QotfV2wdqVkYs9BXCeh1bC0ozbtExmNiYxkzpG05lByFhLLTNIimJgin53EtZvEzVls/xks//OtXPanC/nwe66j0c64678+yBP33EBbtNl5dBfd4Qn6Jn6KePIHnNI+zLp+RdeJR/jxNW9h/SkXcM7lb+WFr76KlcsG+MJHvkN9+Qq+9MGHWbrh5Xzp0/dyaGaED/37b/G6yvSWY2ZGR3n9RQOsKhuCdJSp6VHEBHjtGSpjikvWe4yfbNKOA5rHRhkb2cvjzxwkHFzLzddv55pP7OChRzRX/8c7uOPRCktOu4aHxv8Xd28eZcdR5/l+IiIjM+9au0pVUmkpSdZqW5YleV8xNhgvYMBshh4wDXRDm2HYaZqGbgboYemGGdbGBpvdgNkNNuB9l/Fu2dq3kkqlkmq7a2bGMn/kNTD93nn9ZphzhvfinDp1S7p17608mZERv9/3+/ku44oX/EfunMx497d+yGX/6b8wftDSqO2kteN+mG0SxUdZ5Q1DXQYRK7oqllaqmJwQvPXqSznl7JA7tkhmZ0JUGvCRf1zBsqzNx67S/NUljneeDG/bUKQQFamYSdwzP+Oii5/P5otfwmh5Hp//3NVsGFnO2uwYw2sqDA9aFhZSvKnwyBMN7nqmCa1umm3BwXiYdz1/I88/WTEvbhH2aRZVJN/c0uTZH76UL3/y11x7wyyHD7Y5NHOQv/vgO2hnbQqJJtobcWRmjvWrPZdcfSlu/Kc8vff7TB78HUsG4LavWq77zEc49OwljKQTnHbKJg4ebLL+3NVkzc6CKCpwdKZGGEZorSnHRYRV1OswNTPNoYnDZKmn5dpkZOhY4EQT65u0pEEXImindBeqtIzDhxE+tkidoZ5zr9qUIh7fsoTColUTTANNSDG1EGi8CIkyR5wqTJaSWon2IYJmjvYtpngsWhTzZCnhaaUJadpG+whsTJo6shSKNqBlLHNJG5c5IlfFZ4qmz1C6RNqUZNRI/Ey+WrcWaTXalwj0H/wyWmqskdiCo4hHJo62jVFWE0tBJg1Oe7QzKGeQCOJAIcWfLpP8s5jghZcYqZDRMuLkMD6bwrfnULpIIAeQcgBZ6sUHYEgQYYoPLBkCwkIHst/5kjnOM4/SAqF93ohVKdYZiCKII5z0z2Vy551un4d5K6nzTrhXeJfnvTqTx/KJzqo9V+a43JE3vJha2oSF/ZhDGVvua7Lw7Mtpf+d1jA+dx5tPGGZ+4Pn65z/JV6+9nXVL1/CLm3/CTY8dIVp1HGeeOR9fuxviSSZbB6C6gLFdMywcyXjD2zeTbH+S8fEqaaypdndRc20EhlIIS4aXs/i8U0hlSmIS5JzACYvs1C9dluAzkydU2bq5LrAAACAASURBVARMgs/aKJkbvAhzIrNVhqioKBYlxSigvzugpxeG+xzTsyEtAc2mZGKmjfMCWQNnm8TZNEiJMy28zigcEzA3iGgdZuqu/dz6+bdw+pI2C4sZm+ZDS3WRVRSHts/xxleM8JpL53Phxu28+/I5/vqEgC17J6inAc0993HNJa/m1IXH8aKlJarjY7x6k+fCNXdwqKXYe8wxb6CLqdljjLoHeOFGzTO/3kq7XWTuUMCUqnBUClaVS/zTuwbo8XOcuyFicuc2Tlq/EGMc8dAAW3f1Me77eO/fX0Nl+fP43dYBDhf7+PTnPs313/sJUf8wH/j0PzN7tMhX/undHDx0Dy1fRBTKmGoPYbGP573wImyxyeA8oC44u2L5zKuWccfPr2XbLlh5guId71nEoqWGd3zgGc5e7Tj/ko3IrMHmc/uYzg7zztcuo9At2fLgPkary7j3iTs57UXn8KF/uJa+JQEM9XHOQsdwn+Cy17yeJx+ZYtrWyTLL0ekZ/vElJ/DSzcOctHGQY0dTTpq3ANtSNNqa1Us1tQNTvPBkz/X3/AM77jqTpKGYeAyOW7KEwqbV3PrYLWytS7bLtaxe/2Hk4OWYRV9g8ZmXcN+WB9g2vp9LNi6m9lSN239zOe//T8/nBf0Bd/xyKxlFmqZI1mwT6ZA0NZjMYoUnac3SmrU0G7OM77Yo0aAQCkIRI0xAKCUSSaQl3mWoQkDdtDomxgSRKqwTuDCkmeWJbikSF4DR0HCCmlQ5E94H+LZDtDOclMyqhLaE1DnqWZukY2AVrgttHamZRkkLyTQq8NSsIRMJWmZ4G9L0dTwZoRc50NAXqYuUVLQp2BSSFkolxFIjbQHrU5xISLylZWexBrK0Q5aVNo8ezCxJSu7f8UmuzpOK2IboVJGKgLaHgAhnVc7a+hPHn8UEj/BIUYDkaRK9AuUDhGrgrMDaBNIJcianRuky3vUgVYQuRHh68YHEBUHuBnvuoMhO4IfzeJvlXHitc2iZFMhI/0HyyB9KL8+NHPH5RzLLP/ou/uj3sqkdVGIF0zXGmlW6rcL95iZ2HQuoPnEH56zq5vDaM/nbf/gcT235OZvPO4H5GSxdOJ+hvn4KT32T45aV6emxjM4bZmzvMYZGF9PXE/DQZ2/hUC0i9W2MqzNVn0ALiRAxaTNg4chyZtMJgiglNBFSxYjsD9LP/DN2ykz4XGnkOyG5QkIocRpsQCe0wJJJg4odTkJUsFRjSZcApQTSh7lBRVlU6omQWBegRBdBoklPPAG5cAT/6BgL1naxQB5CDEqamSCaX2GwOc5osZ9eXSRp7IHmAyxYHBMNNlgwIAhqNQpTO1kctzlpWYly8jgnL5xh01LHsuPbTBxoMO+4eSxdtpyZZw2bywNccLnmP97wRXaOOWabU0wkdWZmEu65u8ZVf1tErOxl9UjM5uNh/Nk6W+7ciQHCvgpBPAJ9mqfGPNPjni996QZEVmXp8uPYu+cgL7vyKnbccS+//fl36C5pMtNiXmBoizo91TYrTxzmzsduJO7W1GYs9VpGUvI8sXUCGZXoiyxd0uFtzOrhCh/40CDn/s0app8ap6AXMXFwnE2nLWbRQA+nlOAbv3gHb/urlWw+sZuH77qT1hzs2d+kJz3MQXGY11+5lP/2wes4FghGuwZ43TmLuGKd5o6HtpA8UqN7wQCXv/VlHJ0+QG/sWFQQvGjNINVTjuftX7qPn7z3Mzizg9pT93Dz/T/iaFdGd2Ue3V0nYXa1ePzhx3jTNe/n9m9fT8RtXPuRn7Nq3bmc9fK1HDwSUFm7lg+88qdUV53BSRdv5mc3/pBK7HBuFuEswhoiHXayijOkikhNhtZdSG2o1xKU0iiZ4XyGFzHWOZxUuaNTSlomxQOlcgUdhvkqmAwdCqSE1LaJoghMShRYIh3jpacp84amD6HtMyIfgo1yOaUwGJPSpoVKLUlaQNo+ROCQhRSbhYSiSOAFSdrASUNYqCKaIVLGZHYO55qINCAKqhhhadsmUhWxIkBqCULjpCazDkuU9x58vts3nbQqL0sUg15MplCFiKIqIa3HCzDeIZ/LkBYWhMX+EaLlf3X8WTRZv/KlD3/4ja+bj4tjwtJroLELlwpccAT8IoQCk5rOitvh/TRYBc4ihAVZB2JkJBAiIpMpUgqUKgEJIstyaaNxCCXwUiKtQniH8zJPkzF5g0Qqj5AORIQXCoFCqTxtBucIZIi1Ht8BjjmhcS2P8l0UixnRqoXMLrqWY82Q5a+8ht3NxXz64x+lPz7Coae389TOQ7TiiHde9WLGtvyCAbuPocVdzE2ltLcdYOUb30X7id8SF1PiSoGv3+eYtBayiEBEdBfKNJM2K+Z1cfYp3awvjiP7JpGWvPNP3ojOm8adUpLweK8RUiAA5ywuEPhqCVGqosoFAm1R0lEylkh6IiUItaJc9ZRKgqjkiAour7YqkTe0pUOFCiEdDoeemsBtuw1fkAinGOpKaI4ZRkeW4ycaNHUPU8Uifae8mK3X380FVx2HqdbQ9ZCfPtLHtnuf5WOf6mWN3oXyGXWfsnBUkEQB7/ue4clkM4E0BHNjmGKBSy40nP6e13PT3/w1V77tOH523WFKvsBn/vYkPvqeGD8TUUxbVPobHKq16OuJWX9ylTDsxzLAvp0HcEmZ6Ud/TZJNU+0qs/fgIcaPHubMV7+WKT3I3gd/x+5tt6ELlqLM9bkjw4qVixfRspOUgwJhu8zKSsTZ64rMq1Y4JhV9wtJTbKMyT1+hztFdKTM7Un5x/RibztToExTZeJ37tswyO9HmyYOHOeXKTyJ7ynzpQ99nZBQuu2IV8WwXZ29aQHP8CFu3zjKy9gqevy6ifHgXF582w8gCwap5izj9gikaD95P83e7ec3Vazl24BCXXbiAmj3Efb+4h0986uts2dbkW98a45JXvYqR+UPE0wlnnbWYjetW8fjRgxxX3MHo2UW23fswl162hupIlaWjK/nUtdtZu3ArPjuBZ33MVz/2n/nyzfv51s+/S1QoUZQeZz0yjABPRs4/p57iRB4e3dNTZtX6ZfT0zqOVGGQY5UhuNMoZfJJLJ2OhKFhBlqRkvkEYKHAKYxJwAT5TCASpsKRO4OsWRBuR5k1Jm0mUENRFk2baxmWWICygUkXBS1KXkogMJUE4i1WSWCgadg7hoCsoM2tbBGmGCxKsaecTuDeoQkaS1Kl5iwrLtDJDIS7TbiVo3SYQMcY3kUpihcd6h3WWMCpghYBkDi9yYUOAyHOdPbRxqFjjTJqraZwgkDn/5sav3fUnNVn/LJysuQIWErcH0bwJISaQVY1LNUodwfgmSnuk6Aefi6i8B0vebFFqI+g5fOsQIoCQECc7klURIHWOHZAuw5sSKIcPZvM6dNrq3OVzlLzzqhPF1SKP6JK5xlaGeGHx1v0+DlAIgWqmSFWhZSdp1TZw4JFZnn3ySs741BO8dPNq5q1bzJnLe5lp9yNWLqfQOMymAjz4jR9z5pUnUd+taLhZhpdNEa0d5ugey1RjjN6uQZ745X6mfEivVwQ4RFxg3/QUYUnTmp1hVfcwIngEJ0OkbOFlkKOP/82u47kmMi43akkp8VLiA9VREHkohEghsCl441B4fIecKZwnUAHOG7TuKJIEeJeXrCwGKWXHW2DxicJIQ1cJztpc4FeHirzmw2/moTe/n2K8lKVdglNedyqmNEUYDUIyzdNP7UHPhOisRbCsl03zW2zb1aIgQpZvGKH4q8Ms7IYgqHHLmGLjWS/nezd9npNO+jE/uknz0QvWURzczlGriXiSHU8IVmweJJvR9K5dTXDbw4hjLUTXUUoiQrePUuqPuGDJEi544xup79rD2jM/xObzzmHmwFEGB5aw86GteD1OGBraokRCxtqlZRYvK9JutakmkrQn5PCzM6w/+wxSM011yNP9xEOEhR76tEJHlnZ7hv5BCCT09Ammj8yw474jFKq9HBqfYuF583joR3u57mMfpDz/BJ6canOZX8J1H3uW0852bPlByktfs5Gp2x/jrE2PYnvrjAxYhjYfT9feMWyUo3d7ly/kSDDJjp1bufjcUX74q708/8VFzn/lS9j94e9w9z7NgkBSO3g9//zNOnc8/jQvfte7MPfPUOkb5oaf/QypA75fr1DTJ3PgiW9z3R07uXyNZ3TkHOrmCfp6TkIeX2fBlscpjfQwPTdNloU0gzwbVUpJ6AXlTJCUNcp4rK9Tr0Wg2ijXpBgFtNspEocOFcYF6IIgswYhJRkWicSlBabrDVTo8FpifJOgoHCZQSQG6R1WtcksWB3kk2OakqZNQFARisRBo92gEEmyJA/arHpHmhoapRiX5eDActCDDKBuM2JbJiwIpAaTtUAKnHW005jUtanICspIIqlImnOEQqNsgWbaoFLqhrYn7ApptjOkjmk3DaW4hAvysBCpFNZm6E5FIfAxGEFBKhCOORKUAFn402fWP4sV/Je//NEP/+XVvQTKgGshtcClhiCoYq3NbcGyFytBqgArPE4UcDLCyRJS1vEYXFhFujI+DPEqQgQRMoxwOoaogIx6oJAh4ghRHEJUyvhSgC9GWGFzXKTLDQqOTmB34HHC56sNC94ZlMilht4LhCthukLcY03CzRto7XiIwbWv41fXf433/OQWJhsP8MyWI4SL+tj7+COcvHgBb/3Cj/jhj/+Z0vY7OW9Tm9JgNzLuRmiN3fkgQ5d+lh9/9W6OVPrZtqfJ/rYm7O6i3mgiFJSs5gXHLeb0c+YRB4ex4SGUzpU2qOD3jWOkzG9EUiI7aVQei5AyrwmWY1S1C+IQG0pkOULYNiJwaOGJNMTKE0WCuCSJIghCEIFHhQEEOSJWSfDWIkXuIlYyj0oMq56+rpCjj+1kZs9uTlu3mPbD9/P04Qk2rDzGcqZBWaSER76VcPZGWPGCPlJ9iGBkAQN9Je5+fI5//dIxXvjyl5KMbWdrax7v++g/M373pxleJ/jZrbMEPQG/vPlpxmY877tmiA0jPZSicQpdw0h3iCPb+wmK46xe38/ksTLK99Ls0cxkvQz3l+ntP8j1H/lnaq6fdRtPY8HoKu66+ZtMTT7FVGMSoRVLezXzejRDxw0wM90kqB1gyfGncP+3t/P5d66joRLmdTUYkS0GKgmjskV1yNNXkpQrAfMWKIxyrDq5wsxkmzUnFFmyfJju8gxjj49x7nmr2fq7J3jqvnt46Xs+wLM7tnPG3/wl7/2r1yJGXsS69i8pz2vQFx9kpF9SqASMPXYAExq0P5nJgwcYXlNl4Ugv040jjO2f5qT1ntnE8Pj9zzC/fxmXnSypDM3nsz88wMj6czh5w2lURTfnnzLA+Pg4C+ZXcWnKa9/+aeSUZWDxWZxwwYUs6t3Grbc9yqVnOArqCLc9mbBq81pu/u12CnE3lD1kgiBfFiO9R6WOonOoyFEoSKKiZuFxgr6eIYwLcjyJysFcRvq8NGEtXgkSMlyWu8iVkL/HhpNBZiyt1KDjIiZzBE7SDhS21cImLaxSZECSThOEMda0KShJe9aiZIBUgmM2IdAFikYhvMKHisSlqCzACYlUTawQaKkIpKaVGFInKYiQkgryG5JMET5Fym6U9lgvIVC0m02KZUcr8URBgCZBa0dmG+AEKsgJkWGoMCZDq6hzTVqMdGR4CipAkqtuvn/dPf/fl0kKH+RGhzQgQ+CtR5Fh3RSBHsIxiA8XIOQAnmGQC5F6PiocQetRfLAOgpMI1Ag2XIigG6liUBKvQrwKIYjw2uGCHryu4ksC31VEzF+Inzcferuhp4SoSETJ43WKFykKgXISafLNjhLy97F/OI9lFl1vEa8dZut3f0FX3yjP1mfZOTdOPZplyca3sGLxakaK3axcvIgHt+3m0G2/ZK1osL63jY9bJEmC9AYmD1AcXsrBB3/AoiUlZprbiNs1BlxCJa4RptMok4JIUJGgrD1GHyEQpfzzyE4osM/rer83Y9FZzf++K+/wwuGtAywykKg4hDDAVyW+EuBLAlcUmKLCFhVOZMhI4AMD2kPgQAqsy+3Yee3QgRckFoJA4k0BQsW6ExV9us3i1eu54vVXEM4d4pYfTKOrAowlrWquutJTUJZENgi8gmNNOK2b278qSTet5Ws/uJWjssHoghI3f/NLFIcF9QM13vXxd7HnQJPyUIFLQrj8pAZxzzgz4yVs1Mb1anqTaVZcuJq9BxMWr0pRPTOYeouBSh/704RffvPn3LcvYHDTWp58+h5EPIWJqnghiaISMpEsHaiydKSHxuw4vfoYoRTsfmCSl5/b4Jb793NcJeb4hb3Ioub4RQW6FxfRBRB9FlnxxCXNYI+grAwjCz1aNTH2IMNDEfMWeC68SHHBasOrzoXKg59C7dxG5fYvMn7XXzLxyPXsOrSXmcMt6sNr6H/p+zBHG8wbGiBoZHQNVhFmimyuQX1qN4M6ZnSVwnUv5uFfOp4+MMgcRa67ZYqvfftxNi0N2fXQb7jxpmv5xa0/YdvjjyAjx+c/+Sk2nPpqyoUid9xxF92FjJ5aHTOvxPLTX0F9/mu5+dYdvOddz0ON3U93GZxoIuugHWAcLrMkxpAKR915mqnB2IjAt9FpLvuzroUXTZxPMD7NXZvGdCa1lMjl/SJnHIFUYCQ2S/LrUEIxDrDtOt4mmEBhmk1CrfBC0M5SpAiIZQGMwhqJIMTLHMsxnbSpqghMRt3lpdtmq04mBUTQtk0SI/AupNZOOZY0SEgQOsNisFZjSbE2wroYF83QNIfBJyTNhLDQRzMJUcWQWtKiZQRtE4CuEuoyzgrwimbbooMiszbB4MmcJcBTFAFaKCIZEv7/p8nqCJ3GyhSvBhHBKDY6H1W4BKfORBQvwfgNeL8RxBI8PeAWIOhHiBhp447Cux/pYohUzo3xLYRNUWRI38Y5meNBlc+5z4FAFEBVQ4IFg4hFw/glQ5gl8wiGhlD9fbhKiC0rfJhgVdaZOCUIBVqBDmH1VSTTKb1DXcyM7+VnP7yXd37ux9z0tr/j5+/5Bx488DAF3QM9g5z7hndz6P4beMNFMHJCD4WeLlypye4nxpicqDK3f5zhF5yLaW7ng9/7r9gGzF/ax+FxT0N347v6GV60hkuedyHu4G/R3QmoBghH5jKcNDmyGJfTOIVH+nxCl0rnJRmZp9wob7B2FkcNtMMHHhlrVClElANECWTRImOHjQVGG2RRIAsCFzjQOQtfBXljDelBO3QQY5xAiDbIFkMrShy/cjf3/ehz7DzUxfKLLuaoTRA9PbTEFGEzZWST49y3KyQJqlClNSr47KqnOXjWC5jYF9ATVrnkrCt46v6HSNwjyGaTJcuX8tZXfpTTNy7iDfPrfOju8+HIJAysZ9HKNn4uRS35S4Ku3SBewspTu5HuEsYfux0x1+BofYoDOxwPP7mcRraW8b2H6Yq62bRoIWb/o1TChO6wwakXdBMPSIJWDXOkwbGpgA3FEht6n+Hy559LfOQIo8Wn6StsI9r7JLpQpHcwojvWVGVAtWSIi5bqQBXKEbIS0NWj6O0LKZXgzDO7CWZ3cMqLujn3jB42roJPv7/IaGmCbeNdnLF6D339gnKxQJkxxn76d4h5m/CL+ulb1Ed7/0109ULPcIViT0z3sKZYKjHa7zn/IsWBAxO0uk+ip28xi5b0s/WIQkxPsnl4Dapd4/FDU0zs2cfmK97CxRdchkhhf1ny7KFZukcL3Pihx0nu+ya33/hZzrnildz8yx188wlHpbuCzRw1a5FKgZXYVBKoIkk7I/ECrSu0WpaZBA40UmIVE6sCoYxRRAgTojJPai22swPMMg9CY4MEJ9oImQdXe6XxLiCxEuMDJBFBApWgQCNpI8KAMPO4ZpOW07SSjJAi7XpGOc4Q7RpdYYmaa9PManiTMOsTKoFGt5pMN1MsMYUwwjBHmCUMBkWqJkY0JcYnTNsZpMpIfAMnLe2WJfB9eC+I4wBrppHe0643CAKJUxkiSHFZPVfqKUhcCuTlzS4VEllJWRawjhy/gMELUFT+5Kn1z2OCJwD68cEJhOEaEBtRahnWLcNFRZzVBKIXpVSO55UWryZxYgzrj+DkETJ/DCH3IfQEaXMcaxK873TrbdJJYk/B18A1kDYl8AneNvGkOOnz7ZmOUWEZV46gWoFqH6KrH1fpRUdhHigic1i/dQ4VeNo3/yvxijU44Zj/vA286R2XY2pzFLMJZHUva1dsYqo2zcSjOzl39VKazSZzVtN/0jnYbk88ltLdUyBeWiYYWYb0c2y66jRmjz1LvbvEgcNTIOsEKqOcOeSRIxzdt4d2rPGygXRFMgsaieB/NEcID/hO0+uPlEJ5dd7lDVKfYGwbT0aiBEZLfCSxkYAQROhQOkDpABkEOchNKZ6TijprkTLf4bh2hs8MGk0WSKBIGDQRXUUGBqGpAiqHf86SJYtxzcMU+npxtoEY7SbtnYcWKXjBw2+bYGLlEpZXp1lfOMKH3vtSHn3oXhYt6aG7JAh1iZtvO8D5K7o4fm4/Z7+jh/Th27CDvaTycbKuGOaVyW77OKy/gnTiiwSLNzA4OMtjewMemy3SUywjCjDlZlh86ghJ27Jj++MsWz+CzRSukFEKBdXiIK2kzZyfIxBF+pMmQ4OObhUS9DXRA+W8IdhVYP7xAb4xhtBzFIoRha4SQpfJvEUECVnSoKpU7qJMGvR3t9HJDNI7+mpHmbegwYYTwGtBaajA+cuLDBQEI5tP5rjzllLqG+SY66Y1/iDF3btpHqvTUxpEhwFyepauyKCtYnAwZP+e/Rz1r6e65hr2H4b61CFGyt2IsUOcftHzOFi7mR994zr27jyIDgXfve7bSClpNQ0nVkYZXLCOW266l1Pe82bM/HP4189l3HfrDxnxD5JMWMLIE5RCCqHHSqhEBWxmccbmMZbek5oMEXgCV6SQSVLa+NBjMHlZNLBEKgIlaVuLFQIjBCrQBCIiDGKEyev21s1gZQYYYkB7i9ApEklJRvjU4SOJUFD2KaFzWNfGu4x64mhpx0Sthm9AQVURUlIwjlYrQ/mQorZIN0czbeAyQRoGjLVmSQKIykViX6TiIlKTzwFNa4l8mcwImq5F5nTuqQkEoQpQVlCgSFYHKJO5DOstSkmUljhv8zAS7bAqwaoUFNggIPM+xy//b5hZ/88P2QOVVxOKUo4diFIkBqUs3khc4BCE4Crgj4A5AhzuIH8d1iskIVZkSAzSJvnE5lSeviR8Hs/nJZ7i7+P48OQMG+8Q1qJwnecDYYgLgSDIaXNZA29jxGwG01MEGdiWxJkIXXIwbdi/q4bqgm9/8hu8/NLvoCZaDBzM2D46jy++/3L2McIn3nw17197lKWbBGbfD9BNTba0SNdklpdYNr2ch7/8Iw4/+1ue3nMH733jcbzzE9sYWjhAzWTMuqMs7RpEyjniyjGEL+GlRdo8kFx6h7AdQKlweVtBOKQTeZKVkHltXoCTEoVECp+jkV2Ckh6JyFHMAFJivUd5gzc+jyuzAgQINN5neYO1M0SgEOSvpfF4UccZj9RNzjtXI7Lv8C/7K9y1e5L2VETQ6wmCGEROCMxEiOgxfGx3QFSa4sK+CqsvWsXXvv0ZKgO99OgEHwlOmzvGB7/YRWNnRLiqxaE7a/SetAq5dCGhcGS772T28aOEs4upNrYx/fgU8//+UtyhT1JLuug+doBwaJT+apXpxWvZt/tpKm1LT/dSPv3hj7LxrEWMrqrSOFLHTO4lbAqk62KwfYyNq7vYOaY46wJBsRhy0aaMcKnCto5SLVZIZApSoeIUJ/I6sgg01lmCUCNCQcE7nHYoL6guKJKljqCiMSs3U939ABSgMphh0imkyJh9YpyVG05n/5a7GFq1gphj2EIR1Zom7svor87n2I5ZepYsYP+xXZRPNpxwmeALlw5x+pqTWX3eJdwXT7J3zzOMrlzK1rsf5O1v+yeSJ7ewZtMyHvn17aQ0+eFj97K4v4+ds/vYteMIp64+i59c/wVOna+55k0jnHyxYWoy5Qv/up6/e9PdpIPdRFiMSWm6LOeae4EyAh8rwkCTpQJZNDRTm6etGYtAIjpKr9S2KQQhJkmQwhJJh80MQsY00wwjJda3cEGRrFEjFoKZLIeBtRKQgcC3c9Neo9FGSkm97ggjSWoNUheYnm3gVUBXqMmkpdluU9IR0kKqwQqDNYJCEFIkImu2CWJJT6HEtHD0SGhlTWLlSNIA4ZqEhR6aNAhVTGgLhMrhjKBpUlykCGQAzqK0IswaEBXxrp1DyJIQHSsym5GKXJkXRkWyzKDxWGOItP6Tp9Y/jwleaFCDeGlyYqOP8F4ibBNnFV5ZvG+BqSH8LMJNgzdIK/LoLFdAyEZuPMIjTc5lxglELgXBI8nRBflbOgdCeKQM8qBt8VwIiMxlhSLXk3vZSZwKijglEdU5FL24tsHLGibLCPt7savfQP/ufTSibt5yTZUV77gR8ZWPs+3Gp6j96vt86sCvmGx3c3yhxfzRmLQQUSh53MIyQe0oct010HseB7/3CXqL02z8yBu5ZO4gr37xb+iuKry3dBUDJp0gEB4XJMhiK/9bfIpC4mVOukN1/hbyWDGBx1uBUCJ/jEeIvOHk80suB7WRb5HzGn2+6s9fR+CygNwf3GnQCvJj+0fDO5HfPMlvIHlNXuJUCW9rBIUQbxznr51h+6NQHO6hvbcJC/tAJxhmodBNoSl4vlJUNxWJW44H7txKxQ/harPIQkz1nkle+OgSsEOEB+6HJS8g6tnO9h+Ns7JnIb/7/hasWciNPzrM7sSQTY0R9AtetvutFJcuYGTpMmaPzVEPoadYRM7UqSaOWTFLK9KI2TnaRtHVCmlPT6KCkOndc2xadQo79u3HrejltNNCJrcfZPG6hZipBmFxJaa5BykCVGRw3iCFysMiOn4EOhgMAgFZfvPF5RxFXQghBbbfgY27SOoJRaVJSImCiMH5Btd8nO6VESPfqgAAIABJREFUi6msX8r0/eOEvoEu9BIkgqSdoUoZB/fu4OiRi5jZsZUHP3EDDVtny72/4+PX3stb/uad9E7u52d7H+PQRMBxw/fwuR9MccfNT3LCkiGe2b6HFSMjzCXHiFSJ3r7DbH75lfz6Ex/gkQUVznv/Gex88m6O3zjA6y4Z5b995EGMychMRiA1xluMlLTbTQo6wElPEER4EkyzSHeYY4Sty5EgUohcJpg5rG0SEKGUptmYoqiLJH6OQCu8CsnaDrwi1JpWmuKrEbNJm7ILmW3MkTmNkW1kJMmMQ/VopmfrOJFnABdKEiEgSRJClwsPWjIjCwQlFeGyDC1DklbCrBb4UKOFwaSGflWkNTVHHJZwLkVGkjiOMGkCgSDzHh+GGJsfh2JUBGMIdIAVYHDoQCHSlKCQ584K5UhMkvOvnCNQKpd0SoUUAq113tf6E8e/O8ELIUaAG4D5+VnKV7z3nxVC9ALfA5YAe4ErvffTInfXfBa4GGgC/8F7/8i/9zG8LANNvDB4FJYErSKkzy8WGAcexftjuMR39NwhQqZ58IZzCC8RwuKsQSiD9zkGIccKOLxT+erc59IoD7kMEtkJ0gXhcwmm79hcpTG/V55I53FxGRsDTiLMEHGW4YVCNj/PkvUL2POzO+jbfC6tb7+TW747x9s//BE++eZ3ctz6V3H5+HdpGUn/6DziY4dJuw3hxBisey2t+z6P7voWunSY4aFLoHwAlhWolh0i6yJ1GaGJWBx30V+dT7k5g48kXrRQAqBjcJLNDpzJ463LQ0iczL97m6trVJCz8zvHBiHwzufHsCOLxOVSyOfs1k5mKCXwBiQSawRCJEihcdZ3Tg2FyIW+0JFZAmhn8DLE1D0iSli7qcgXjnc8e+8Mo1etw4yNERAiDURAaj2v/HA3/+XdTd70lQsRdz1Ja+wp/EMZH3nwdFTbMHPzFo7t3sf4Dpj+zW/p6R3iSf063nPedYRxzIHaOFlXitADIBoMhsv4+tH5mB3jvPnFp3PT169ldGw3M33D9PT0kEQpc9sPUIlaJG3PxtWjzOyYoBiXyeYiXnPqEo6M/Y4TNguWL+tjoE9R5giVoEa8MEbMTOQlMm3wKTjjEdLmpjokzuXnLA4wHqnydDC86dxoPU61CYrdeKcoRinex8jQ0DM9hx2pkM41aE8m2DsO0HSWYt88elxC1mrQHkvYNfUKttpzeWbsOyz5zotoTReI04zj1w2yYaPm4MTD3HnfE+z89d9z4y9v58oLP8PCU1/ABeeuILABz45tZcfeaXq6ilTDOXYeGOLL/3QnF/zFKVz6xn5++o3tNPcoTnlZkfahh7jroQ0sWPkAuhRinMB6ie7ovLNA0C1jsmaGjjQySEEKarUGYRSjtc6lttYhihbbhCD0ZNQJCwGtxhS4FAKJ9CG+DYES4MI8sa3ZJHYlTNaiKMo0dB2T5pJmkeWCjVJQoZm1AIdrOZSKCLQnaSbIUhE/06BcLhM5R72VUNQxifNIGSIzT9puUiwVmM3qVHSMEA2M0SjVxiXdOBICHyOMJ8ASBLmCzZg0bw77jMAbMg8tV6AQehJrcnm1MGgdkWaeIBAIB1IIFILE5Fz47N/kVfyvjP83NXgDvNN7vxo4FXirEGIN8D7gt977FcBvOz8DvBBY0fl6E/DFf/8tPEIkeBfiXYQkQxDgHPg8rwWXHcXbKbydRYhZIEfZCiTOP1erauZ0OiEQFqSXSJc/Fi5XeeSB239wowrv8sm/gyGgs9L3HcwwTiCswlsFBODaeYnCB6A8tium7iXEjsKyWVZdXcFN3c4Pfv0s923bxlw9YG3Dc9JCS6GaUs6O0qO2QVcPYeJwFz1FI/rPREvn4VAMvfA17NuzHfQyGnceYMekpaCnMMZwpFknbrUZ6K3gTA1BhHca73InnSDKKZjS58f0OXDa781OHktHC6+e2638jw5ewXP6+T/o/fP/CHFed5DMEuHF7+WYolNeyXn8/9dTzKvca6AKLZSWZK4JJc2q8/tg/ChBuUC92CCKJTar4wuOf/rIDNNDczz+sy08cNejrNgd8u7PHY//6a3MffdOvvHhBnfcVuS2e7p4/Il+rv3JMe7csZ+xWsaYbFAerlCNF7OmT/D2v3sfzx7cwebBiOXF+YyIGl4ZTLuNT48S9ZYJSxFL+vsxmcCZGo89tpUDx6Y5NtWmP9zHyPEpfScrVi7vozp/jihKiLs02k1RK8bYKM3PIUyOu/C5WTg/LnTonvmxNdbgbK448l5gTX5OurgLbyTe1LCZQrgmQqakfRqVxoRBhcS2KFY1QVhA64yGFQjjuG3uR5z47u/S2P4VTp/neOZAxq5dxzjp/PORe6dobYdN8+f45tev5vrrr2fyybvYcMlS9u/7Hc88upPfPXIL7VrKa1//13QrxdFpx3Rlkge2/ZDL/+VT3PLxX/H9H+/FpiXovwCphggzS6kZoqM416DLgCzLwy2klMykCaqoSVyapywVFEEQgBQYZzHO5ueijQhUKW+kGg9eERUj4lI/MijTymrEVYNXday3WG/z1LUsz2+Q3hCHEtNso8jQwtPK5sh8C2cTfJYibQBtRbvpkbpMVncU4i7aKUwnbXS5TKPRoqgjklaTVHmKhQJpO6ESFGgGAis8MsqIfDeZb4KXpE6SOUXbSzKpEToiNT7PcnYmz2UVYHyDLCXfYYsSTsQ0jSRDIqIAIxwEkgyDUBKh5P/NtfQ/P/7dCd57P/7cCtx7XwOeARYAlwPXd552PfDizuPLgRt8Ph4AuoUQQ//P72KBGk4ew4lpvEzzYGXmEOYI3uwA+xgiPYxIM5RrIVwEJsQbi/AyXwk40RGQmNyA00lU8d53ygkO6QOkDxBWI5zCmTRHGYjOhNZhUqtMIGxOr3Odic46gTQaZQzYFpYElbQpViym16AqRfS8AYbOWs1fvKnBm86t8pV/vI5t4y3O2OhZdxmc86ou/PPei5jeiQtHSG5ZT2n7KdQOz+H6zuex629g5K/P5u0nf4GvfeoQdv48dLCYsFCmHGn6e0dYs2YJvSvXgZoGJ3HSYn0Kso3wBbxTeNQfDFmqM+mrDihNgsNhveC5mpWU+W3NkDPzjfB5P0QJrBD4wCI1eOkQgcjLDULklE3h8lAWTB54IPLwcu9tXuJxGgixHeOVVhWsboKv42QdsinKoo82RUS5j6jWy2tfaXjf5WUWrujnXe97EYfjBj3Vo3znhphpu47p40a5ZXebm8dm+RWafUqyrFrjN/f8LWHfyYwMns7bvvNfCdefwFe//UmWjJb5wc138Oj+R/j7a6/l+9tniVp3ESno7iqwIOxjrtFmRXeJq553Gutp8bIzYz509ULWnnoiE3u2ctrZp7NqWYmpqUlUOMFAV4skFvTrJr4FKgKvDUQpMgAnTafkle8MlbAIn6ICjwwlCN8JmxE4YxCtJt41UTICmeBlSIDBV4tgm0imGO6ughb0VgRGhbTrA/i+a7hAXMnUl0cZiPrwzSIcPsybPvlpvnPDLxk+4zwmGeOUF83n4NO38tDDh/nENxV7H95Lq54SVoucctZxXHDe6Vz/+X9BpBnHLT6R9s4Jztt0Ik9+9FKGVvRw1jrH1defS7blt4RzszB1kGeSq4nac5BkBCisDkilR1hDKD3aOypBBekTloyuQClBlmVkzpJmGW2bETiJEAkCSWAl0mYYE9JOm/kOJy0R2l6cgUBZVJBjNdIgwWlFXQqaTkNJY1JFC023rhCbiNCWkSIi0ymtsE7JO1zaJhSW6WwGFViKYSGPtUTTbuaCDOsyGkkbgpA0DKklCe1EYZICqT2ClXHHjZpiRB4lmqWWNE1IbROTetppgiPDZhkVW8iVfllCiCMQBqFBB5IsrYNPEA6UU53+oe/ImP+08T+lohFCLAFOAh4EBr3345DfBIB5nactAA780a+Ndf7t377Wm4QQDwshHp6cnMP5nJMSCIHwTYSYwZkGwu5H2gNgauBUJ24vxmMQdFbuPq80WSzWS7xo5heOcATCIbB5I9IHWNHEyhwk5J1AyRi8RmQaYRTCKZTSObTLZ3mogO9wrYXAS4HHgLVoI3EuQxlF0A4gS/KyRI9HdA2x8dU9vOq0R3jde/4DrfGHKfdV8DLB3/sPmMICZMPgouUw/2J8eR3GSk58741MPFLg8lccj1wokLJJSzSZSJvoYpmVIxWWLU7p6z8CWXdey3R53c47cORb0ue0+iI/2Pnf6jvO1A4mWah8SsdLvM1hTs/V/RQqx752SjWB0LmzVeerDKkyvNOdfoXAYXHSIkUumZSd2K087jADZ5ABZNoiIosLQ1AeJSRpVMUkbeKpKeweSaOm6elewaL1Z8G2I1z3sV/w2ONdvO2qcX7xmOXLvzjCLfcd4NGdimveezG9E/vZl1nuuPU+XnLx33PZX7yCF73jLbRufpiXn3E8O55uoWowNNxNV8kTH1V86RVnc8aJpzPalRG0CiS1BrNW8RevvoAHfvoA8zes4CXnLyVK6xy/ICSpecrpMwSmwMCiBRR8Eacr6IYlkwGhMHgDwkYIF6ICgXAhwnYWEJDfCAOBQOGzvKbsUHjrkIBSWa7/thlSaISzGEKilsALjXMVRDnAuxYEIbpuyI61SIKb2TW3mkdmylR66px+xcuZwvGTz72OJcvLhHM7OHAUqq0eAlVi7aoT+ODHz2TX2By9ssXyie30VjazbGAn7/jv3L15tGVnXef9eYY9nOneuvfWPKUqVUkqCRnInAAyBiSACiK2IAgoQ6O2jQpo20tfW0UBxbGbUW1EZDK0UQQBEyBAIJCEDJWxKqlUUvN0pzPsvZ/h9/7x7FtB37db18u7XCz3WmfdW6fuPfecffZ5nt/v+/sOP3Ex286e4hd+8RI2rXO8+se7fPThXez1T+P8iwPNocfJmuPUvRLn4f/6kb+ktJqyzNGZx2hPaBSVVgydo3HgM081EbqDSJZ3yAtDZsBqSybpGqmip84jtVE4JdiiQWlhMpmguoHFcBKtSiTmxMpTuRT0M6prfOnRI09nnOG1RxeG+Toy0RZvBK0VPnq8eEY0DItIHFV0sjncMEeWHSNXo4wnouhrnbD2vENVQTMeMYgFfZVhJQI9GE9SqIdECmOpR0uITIgohBJrOhjdIZJT2D610ki0oC3j8TipX32DoMliiQ1F210bTBPRXlLy3Hd5/KuHrEqpPnA98J9FZOmft/bf+aP/L/f9P56piLwfeD/AZZfuEPHLGCUQTyJxESX7ISwQw15UbMBFZCUCSwQttENV0BJbCmCGwhCkHbJCO0JceRINKlgUOUo04FuTflAt5i4S0kImEMUnPj0pcxGzBL5EkyPKE2UeFfIUGqJU2jREIDaY9avZNrXMGVd/H8tffB/THcvYZHRXW3wskCXNcpjnpJ9hfu/n6fSvRIrV3Pobr2Pdhds4se8eztxU0PGB9XnO5n7yw7n4oufRnV5D5r6F6B6oNHdIIcRg2qHxaYhFBAmhVbWqZKDWQlGJ/NLu8S1coESwrXpQtEKTHoM2ekyhUFoQL2jjT9spK5Ww0aAiCk2IgmmZODE3xEkDpovKG1SuyPQAsnmy7nomN+5j//R1mHIbH3nnX1BsOIe4cIJv7r2FU5ll1NlEvzdir5tmehY6ix3szAY2r8v4q1syLv3t6wn//RfZff+j/Oj37+Lu//nzfHavZ+1lZ3Lfrfv4h7/+GX7+p/+EqfXb6ZmT/M6fvYkpO+TbX/wgT1t3NR/+xiP43iz/9Uc2s8ke4/d+eT2PTYRszXUouYUzV8PSvgHdq57OwUf+kl2rzyT6ZWzmkKgpO5Jeu0i6hlUkunYGJAojJqmjQ0zNqo0oa5CYUoyM0q3KOPkFRR9SLkGM2FbOrpggyoN4cj8gjmtkajVNdy2LixcRul/l0L1HOetZP01nleeNv/w+/vDX38Czn30l9x64hzPnpvjVP/w0mzady7R7kP1fGrLn2ISJXsf7b3gVN925l4v1ldzbex5fesMb+MAHrucVr/0wev4AF2Y3s33VPURlaOZPYWfW4Q48Ar7H4f0nGA8VRddAHRjZjJ7OqZRnYHK61hPE0Z8tkGIWVIFWEIJDW5s8oFSN9oJ2kGmh0QYXFFUzJhv00ZOaQd5nsR6hdEPWNbjKUFcNxhjKeoLLakQC3SVFrBxNnBCDYeI841iTe08fxYKzDHzKiMAvUmceXU6RVUKVK4JbRuIA7TwhNmhjqLUnhBofSyQWZKqhZ3pMaoc3kUkj9DpdDAqZjOmWJU3TJGsTJQTnKMouE2pKSfDVuGlwBgoTUKSOWCGIpA0qSCTa7x6j+VdV8EqpjLS4f0REPtXefXQFemm/HmvvPwBs+Y5f3wwc+j//AY+SZSQuEsPjBNmLhL1IOIryye7QYE57q2h5AkuHZG+UUm/VP8PYW1xYWh8WaXFqSfEzKj4R3vHPbyma74l9SSmFitNolQKCEYOWKTSdxBZRqR3XKkfZAZWPUK4hyjEGl59NWL8V7QLBTwjLK2/cRrCe3OZ0uo78xF1c8OynMLfjbC557i6ufetPcsXGjCvOLzlrs6ErJ3nSeXMU3XmKfgEmPw3FpGHoCjvon+Pq5vRiDyC6HSrLivVx+/MrKliR9jyG0/h6e6qIK5tqO3z9ziNtvLF9PrbF/AO+6aBzjS49VhtUXuDywNJNHh53fOwfHDf99SH2PdBB5bs4fGSR+44MqQc1sSusNw0j3yClYnrasuWaaXq+ZjAYINURfuiKsxgPlzhr206qPQ/w27/1PH7mtTuZnTzC2p7hzS96D8+6cBdXbziHFz/5qZz4ykeJR2+mHwd0NgQu2DzLNPMYVzKIDzN18TYueuFWprccIfMTivVnsGHbNszJg5x57mpizDGFba+9GlGWGB2g2/Ok0do+Mef5DsdS1c540rwnef6kc6daypZGSEPxpD4OhFgBOcROS02tsJuexL69DZvWbGf/fTfxwIEN6P5FmFO38vsf+l98/l0/xwsuW8NL/sMmxg+f5M57D/L9F17AJn2cr+85RnNU+PCvXsTznl/xs2+5hd94+xc4Wh7hB645yvve91b++q8+xjv/06u442ufYXDuZewfGjafdwZlbhif2k9/5rkw3MvVz53m0HLgpK45rCqsKCzClFaY0jKpGiSmQB6anFg1qJA+lx5hgkNipCgzYqaoIHm/+IaizBAfqJ1naTjGh0ilGsbeE2KONSXSVNhgKSVBMVXXM7Q1UljqXMh1yiQO0mVZBuAjY2lQ1tJUGdmkTwxQa4MdC2UjjLE01mC7mqgndJSiJxZ8wMQWzw8OwghtDRNXU4XIKAR0mTMJdWv2Z/Auvd8+NJiQoKvMWGKMdJSBSYOOklwlnUNJRMeUNvfPP1//X45/cYFvWTF/CtwvIu/+jv/6W+An2u9/ArjhO+5/lUrHVcDiCpTzvz3EI/4OorsJab6Brh8kjhaIkxNEv4yESJBkE6BiW8G3oRtEQWPSS5EIVOg2Z3VlYHpash8lbQYqyf3FBLRK4dNK0k3j0fhEH1SeSFoQdaZRmcXrimjHSOYQY4naECXJloUu2ByhwijNKBthg8PlmrhFka1ZhVk1YGncsOyXkWKBTblhkA9oTj4EHUOjCo5/6ZPsfPYPsXTv7XzrUc9nbl/k0X01z7tgI6a4h3DiFkb9WYhrUGoNxE4Ll4A24YnNr11cRJL18UpVn7qfFpNXpmXSyOmhsoj6jk1D2uGhnOa7G2NSJ6CK0xizSEC0awe8EUzEGQ+ZIp9xuE4glHPQKfCLiuyBCQ/Md/nUX2hOLJ9J5wd+jNF538e35h9kXzPihD3BNZeeD8dr/tM738TGtVtYV85weJxx+z8cppzrICcPcOWmwBc++BaufdXrefoLn8KaKy7nfR/qgr2YXWecybs/+hG+JF8gm+zl+T+4lcPHPs7Op27njvu/xhW/cA2T279MT27nqh3THBt1mD2rT8gWYONlVHGRdduuxLiaLU/fgj/1CHrVVrCn8JLsGZTPUFZQJscqC6HNFsWmc2ZbKqrEBIMpQWuDFn06UCbG1uo2+tRdiSA+ECNEl6F8B8hQURFciZk6k+bxAygci+O17Nz6ZBiPueCHX8CX7juD4e4T3PRgxfs/cZx3vHsfl118IZc/9woeWr2GRyfzfPBLf83Lf+08emt2cOiRjRya1Dz/h3+Yd/79Vl78mi9y9MBt/OMX/phPfPAFPCKruOVb3+aqc3dQLxkkX40ZbOfUPR+ieNJOnvuCObausuRVQ0dZ6qxhlDUpFNs3KN0laNiyc5YhR2lChfMV3jVoFcmMphaDCwHjPLlkqdu0AUvEBkfUimCT4V+cWApdYFRSbavCMFLQKAfO0R2V9KRHRNNp0qd51VhhM6GgwRjw9ZBJvQxuQi4NYVzRjZpGayZBYUcVZaWISx7jYaEesyBLRF+3hnqeSdPgVU7XKwYUdJQhC4KvHN5HJPp2sQ7gBEIO3uCcYeyEsU/utMYYGueICoIIToOJChsUWfi3qeCfArwSeJZS6s72dh3wO8C1Sqk9wLXtvwE+AzwC7AU+ALzpX/wLUqFlPyocQrsFpD6FjgEdSJBKyNLw9LS3im7hhJbxotL9K5WnUilI4J9UTxIQNV5JKU1eLCqAgJwO7Nanq/vTj6dSRdVOC1GUKOmjpNs+9wyjPGZluBiFYDMy5ehWNTFrKKxBmwI3W8K6gtwqst40OutjVEnQC2S9tfTOuZJi/9dZu/MpPPb1G7j187eyYS6nu2bAsVHDiRAoT+7m4N4vUI8g+IzoM5ACJIkiUkEeWw56WwG0cNU/qSQ17Wv8jkpBSStAeeKtUS2XO1X3gDyxacToT59j0QpldDq7mUJsROcZqsjwocAUXbARGQphtI4DD08ow3NYc+5lFOu28vgdd3Hj+/+IPl0qHKdYxRdvPk61OOLa517EkUceZFAOuerMHtvXjrGHTvHiay6knn+AtScP8aRdz8bdcStHvngv524U7v/m5/i+CzUPf/jlfOpXP8mq9dup7v0Ylz//ZcT9n2br+nX4k0eY3lIz13Ws7lTE+QiLy5Bb6iO7yfsD1FpFrJdRsoDJZlBmkszsdEuDlBR9mIJgwmmmVjp3/5sKLK6c53ROTRvmblTqTrW075GAkgpjh8l2Qzt0fgo/v4f9R5YppUvd3MmB++/i2pdeTvWljyHV/SyZJc6aLXn/597JJ6//Kr/1F3fjNj+NBwYv4I7F8zn17b9lz9emuemzu7nr4W+zodzInhMDmge/RrN0kG0XClOPz/Pb7/gEn3jfJ3jT27+OHQXW9GcIo2nypcNMfd+vECPMbv1B1q0qyGyBFk0ZFIqYwuyVB6AsDNP9VZRqmkZ5Amm4bEVjPMRMUUdPEPDRUTU1AXC1Icac0hqmugatR2RFsuBQjSH3GR1tMSJ4HfG5oi6FiQ1MrGOiPdSBqg7UtWK5EqK1ZJ0usXE0WhhaQUTTDCuqcYUjwbuuHKMyBRGm9BTdaCELBBqcF/K8xFjFycmYYXQsu5pGCRUCRUlUsWVI0XayI5o4oWOhUIHCKJybIJLYOjoKHZNB5VkKNWMVWG7P33dz/GtYNF8VESUiF4rIxe3tMyJyUkSeLSJntV9PtT8vIvLTIrJDRC4Qkdv+xWcRa9T4EIxOIs0EEzUxVCBpQKpUnZ5LUEjUpytJkdByiG1bzTeITziWhDExxpbnbghiUgyfTpVp8gbWoELi0qsJUCGqQumGKBplSpQVMCQr02BBZ0kUpVLCk1JZgnQAnIKgsU0A39I0mwBRQS1k5RzMbGT68rPoqiFm+RDzo2UK36Mb5/H3fYJizTqKEDm1qJk7ey2LpyIcP8Kls1OsyubIjOdJO84khjHLwyGTicJXU6hYoHUfVNZuSOaJr8aiQ0mwNnHcVURQBDVBSUiViarR0SA+0UIDBVFpMIMU8GHBNxNM3kUmDdHmrXAnEub6qMEAvX6OsK4Pc3M06CTq6BToooNetIQ9Z3Hwjs38wSe38iv/LfLe244yOTnDt+8+wPTmS9i9+5v0Vi+yrTPPJcrxPz7+DK780Y2842d/mZdecxZrVmvs0Yd57S88j1e/biPWPMKO3HOy6XHDTz+Hn7hW81O/fR4XzH6Fpz01ouu9PPvZaxjMv4cn6QNs3Wg5Z/4LzOy8kPzAPPbQg2Sz65llmXXmKIfJqHxBPnQUdgl1+FEMjyKdBuM0ZAbta7QxaALBp66mqQLWqPY6yDAma102CyT41O0YjTI2MZhi0moQQ1u4pCO0lhJa21S8tDTXemIhKERVDBc8zdJOlh8D5Q0P7J1w09cdf/Whv2N688XUB+9H7juAd/CFj/w6H/vEb/DFyrPzSVfw3h+P/P5bNvPYoyWbr9vJmlXCzjVPZvrMJ1PsuYvj+0+xaXyAc6+6jrWXbOKPfu/ZfPmR3+BDf/QW+rOGfMd2iu7dSL6K+qHfxVSWufJOrrhoBxNRdI3gy0gZOlDM0FHTlGWDneRos4zNDZ1ygNZpYB+8QyvoNoGuSu6kDocRsF6RlQLKIRgmEyE6RRZtymDVGV4i3nUpdJEgWAwiCu+foBp285LYUYQwodTCKDqKYCnjFCpm2MqhghAGkc5UjtNDbEczbjKcTwlmi24BnxUos5oYNF55GmlYHjaID5ho6EZN32d4V1F7RwwGHR2YimgCvapHTpeJh1ETMcaicstER6rxmCCO4FMgSm4LNIbe/w861O8NJatEvPdJ2iuS5O9GI/EJPDiR+FLlKS2NSAOEZPMb4xNVqoSAzpKsPjFIdKtazQitnzstLEFIVgWtr27Cm6MkOiCGGFoXSVMQTZvwrgwxpjZcxzFa95AYUdYTVUy8eYnJ/4X0WCkEPBAtqHU5dOfImsDsqSFqMg841GQBzFr0zAaetLHg4D2Wjp7QqTrMbolsX1/TOXsN1cJJVjcLLKgc7yMjyej9DDJyAAAgAElEQVRmPTLdqgpPy3Vb1SqAGoPLiTZDe4e1QOwQy0iIXcxkCRgSteAnDbmz4CrqLZZCbyLM9FH5FLJnN+E5P44dDnHDB5D+epAK8k1MvKfjFMEsUNRj6GxH7b6R8fJGxvu7fOxTx7nXdZjb0GfHr3yA//KqHjd+8STPmlvPgT1/zhtf9xzuuvs+Nu7cya5Nm/nb93+bC/NtLHTX0LivcM0lz+CcqRG3/8UneeZFO3EnH+aiKyy1uZNNC13mBsfZ98lFnvMLF/GPf/lNLvqlp3Pr797Gc17/gxz6wrfo9I4wdeYZuAP3s/Hii4gH7iZf5Vm7vuTE6Cj93pk8cKDPWYzprTqBeI8spNlOjAk3N3iIDpQQRTAqI88s0Q1BFYnl1cY66hhZ+YgJoWUtJSZTen8SC0NWuiBSfKRtr8FgDEoG5HaIhA6Vj5jxBDYappbneGRpK2dtf5T129dz+1drPn39DTzr8jU8Mhu4d/MlvOeGG3nQH0X/54zX/Oy1vOnqr/Nbv/kU/uTtf89rfmwXW57yIl685j4ev+tPmdo2x1zH8PhCzfMufitFt+GqLQ2Xbr2ZN79kPXu/9hDnb9uJWywI5SJVk2N7l2BnLG962yk+8rIKCRnKRWq1zLQDZ4QmjsnznJjlmGgR6tYqV/A+5Qgo00bbRUWWlfigaJyDMMHjcTJJrK9Oh3o4nwo360AMQS3jQ414TYZCB0ceYMYZlhvHQhiDqzFNIJoMG4XlbEyUSNcPGAVL7DRkI41zy2iVMZQaK54gs0i0dLvA2BPthNpU9EJGqGNKfipX4WJgKI6ubSh8SVZFRrljWcGMK1HWsGBrOkAMBbaweO8woYfVNS4UCBqTKTSBWjxGEsHhuz2+NxZ4YhviYRAxQFJdpuHfExa37UqFBDnNW4dWDadoh6spgUl8liT7hMQ+UBFIMMbK78XWxkCRMOqVoWzyD8lBBzKVogIjChUt0XOaO251B1RFcD6BOjESCBhVtLL+tMekjcYQaLBRgc5Qc3kS9Q+moBrBiRNE0yNjBMePcNPnD3PrHsXWjSWDrV0GSw5TgJ/fR1afwOVzdCXhd0JO9Almym0JyqdTqELC9pTGqgIbBK8NPotgLFamEgTjJkg2nTqX0pJv2UFt+uTlauLyfZCtQsk8YXIYu+NqZBzxMoPtn0ucHEJlATc6TocjOD9GGGCcJUz2EHtnU4wn7H7gm+w/443MbdvEgU+8G3PPbfzJY2fz8je/i0/+6R9w5zdPcPHTX8/RG+/m4g170GWHrdkB+lf9HvtOPcB1B26gN7OHcPgwl872MLKHH/yvz+S2v7yZS370qRy994sMY83c1kjsWs7VBcwL51w2hY8l/bMNMupg7BLe92CjQT2WoZYduR3TdYusHzhOHnNsWxWISwvoTiddh06AsvXr8emq9AHEIdoRA1iVp3lNaKtHleTwhCfYWCuFihYBNKHdCIy1KK0x7RwkXc8WhcVmFdQNYh1hWKNtRre/i0fml9FxB3NbC+yGpzNb72bhbsND9z3G33z9JI+vHvDaV72O63/nnXzordNs/skf5eO/9Fle/Mgs3/fcLnpDzt2f/DiHQ+QP//b1LD+0m1e+7C5UPeaa9ZbLtm3hyms9g9kO7/n9R3jyM+HYqb1snV7F8OQBBvZC1OJfEYpXs+X8MeW8wq9yFMuWqgiM3ZBSlZjYIyfSy9a2/lItE04rsjynqVKecJkXFJnC+QblHXl0NE7oZBmZ0iwtDTFRY2w3GQkmRR9BQI1Kis4i4yoQMgtFQ30MVG7RvkHbjKZrWHITukNhEgJlXiAskKmAry0D6TAZQ69XpkF3ZgnhFGXeYUyOs2DQUDUUFnLbo9IRrZbpZFNQeaLtkhWBJgg0ilVTUxgXcaGhW7YWIlnVdsYdhAndqMnKAuOE5VglEzKXEqEy/a9B0P/Px/fIAq/QyhJiQLXpQBKSr8zKEC+IoFWWMHOENKsnZa3alUFqgkuClxRw4VNVFFUKvlUq+ycME6UUYcWDBUk4vFJopRMgGhWixoCGUCKA1Xmq6FAk5cU6lDpEDA1GFWgVCbGlS2rVZsNKUtjG5Lon4lFOyE0GPQ3dKcIgks17qgfOZuncP2Hx+OfYe/sv8vZfu4QDe4+h9u/l7C0nkRIcEzp2Cfqz0CjcRBGVImhBKQe2VehG0kQ+Cs46tF2HtYvE2MX7gKwtCGVONr2epnMexA62fhQm+9B5Bc3DqP4cwWwmNjvIy2lcfIxMG7xRKDfCdBXiu+QGxJyFmRogpw6Aq5AHVjE5eQtv/uURh+ucF77xcdQdH+OcM45z2Q+8mge++SCmcz+7/Gd5yss3cPyW/87P/ciQyaPLNHfs5vtfcDGf+6uX8banGSZXdbEzxzBnTBMPnKR71gbGjxxk5xk91N03s+u6AbNrcw7fc4LFoyWrn96weGgPenYb2Z6PoxanGFzaIeyeJ65ZhXz5VurNJZnJMWpENsgpl+Y5/LjnYG9Cb3UOIeCIFNYQ1LjtDsfE2GAzA0ElrYFJKUOCxmiNSCCKgiAYnUzulNYE71EqVemKgG6v4RgChNCGxxuUUQTvQSlCsOjcoHSGUZHhzFu45aPvYv32c2mqb/G5D8+wfu1x/vxP7uEXf2kz/+UjYy59wQu55ZOf5ra/eZAla9j4st/kWTtfzYkjb2bPrY+yfs15fPWWo2ze2LBp3Rt5xuWf5947buG8M3v8+OteyckHvs31j55i7+ePoI9ELr7Gsm52juaRA/DCZ1Ev7KE7t4QcX4cMHiLvG668qMut9zfEviJvMoL2VEbRdRqVRTadMUctgTyoxKjxniCCzQxG99NQOTM4iTRRUWQ9VByDyphMPFneBT+hGS+nYmoCIhUdX1CXE04tBXKT05WM0bLC2DE6QN3EpH4NgnIKbfqYzhgvgkxKKq0Z0OWYLNKxXZxUNEVGGQyL4z4hFvTyVphkJmRZj7zOWNJLrCr6CNMsLJ5iupujmWfYlNheTj9oJvWYbtSYIsOHBtGGLHaTlUoYkZtpGhXwbpk6OgZ5F6mETp48ek57fXwXx/fIAi+INO1wNG/L3lYcIjo5PJJ47NAWxDo1MKKEcNoLpR1OWY0KIFKDimgyVCwQnZCZhOZEtAJR7aAwhrYqWOkakqBJ+RxQaF2nlCc8sR2eiQEJyyRXG03Eg4AiQ5sUwKHaR0IA0wZfx5isEIgo5RNXOu/TzEbKiwbk8R/ZVt3NVRedx8c//ghbsobnPGcbthSsrsmKKVwxAPHYrkYHIXqf/Ke1QdcODJAZEvvOokcNvjqE6YNev5E868Gmp6U8zNDDVg+jZRGvBsSZS4DzCNkcdvRJjAKxC0g8iQ+9tJGZLSBCyAc0OlDGY8SoMNWEujON+AK7rubYXs3q2dWwNOShv72ZZ1+es2bTLOtm9rPmgiOM//4nmZke0VdDqmbIpp1XMOwe59DDlqldW7nwyXdid27CPvwY02tnGC6ewnqFf+wELj9KryOojRuYbpZZPjxiw/kdyi2e4mCP5RNQTw7CzBR1HNFd9OiypsOIaq2lNOBDBdkAkcAqmefY8Ch5rIm+IFQNMYOoi7ZrTIuyEgtUEA0xJspolDqBYTrBfWmQbQCPanHFlRB0Y0w6dz5R6TQKdDK4izGiyEAs1uQ4W2NGOcdOLDI8UbD7zrdz9Wtfx6lbP4TZfiX66CmGseS9H3kRJx69l3Ov2MZPvuAcmn13sufEIi990i6y9WvZvg7uuuVR7r/lm9x88xE6eeDdN/4kX7/RMP6LfZxx1g7u3rufP/3jD7PzrPO4cE540St+nRP3/xmTY4eYP3qSrVs3M3z4fub6m6EuaLqa0FPw8DF2ru1z821HMVMZEgxWh8QaKkqihrwPhekhOFzdkOc5odUBRJUjJp5WoJdEyugIKqeZeCyGGDzRW0QnwzLJx0QilSwmcaKNCaqVZUQmONGEWJFZhZ8YxDXkNuDcEgvjmjyDvukSg2UyOUy/16GKjkVnyU9ohnlNf6ZDHcfEOIWgmKJLE2py68gokagJcoA8U2RhmgxDtJ4YHI0pyLShVhZlDAVN2nACdLoFgYg3kUIZGpJYMYi0cX6pk8vsv5sKnsRgUS27RZ4wqtJ2xQxMUmUdIyJNEpEkGydQGomgdSsyiT5F18nKMDbR08CkrA5RSVXWKsdW2DMSI8oERFRi8FAQtaCkRmIH0RNCSNms2iRTLlFDRFyq3Fb8bowD3XpuKYOmaCv+ZK0bgkYZSWq5mFLURZbJjIXpPQS5iye/bIaznz9Pd/mnOPztOxmsfhDfy1GjYVK85SMyM0VEo+KJNOTDoXQBuoFg8Nph16yjmV6PzK6mGPwc4kcE9w2Mfxy/8C0kW0OmzkeyCyDOoVSDZo4m1Bj3MNY8E7EOGx6h8ScpcSg/Q/A30eh5TNhJpzmGZAZjpxFdUwy7xOoQh+5y/K9Pz7K/OsjEOf7sg5vg7Kczv/vLzGzbBAcfplkasCU7i5MP3smOl/wAy3d9hiLXXPmSnZy46UbW7FpFb+0wva7RcXobNlHtmSc3mt6utcTFEeQNma2YmbXMnL2auP9rcPa5rM6PJuWszylWawhjWNtBxhPKQQEholVG8GN6KlB2DtOZOoTtDRjX0wwnHYqmw7BvmOmcwkoGRSC6CaIDxmTJKTE0LaSniK2uQEIKW0FiayrWsrpWuj+lCJLmTlFpjLFJ2KTBh9QVoDxST8FUydxwzNoLV5NnByn767nDXUJ26wlGi49x9qUzvOtt97LrwrO5cHOH937+UR7ef4plU/PVg3t59zuv5+d/6lKa5ds4tLSZX7r+bey98TCzxQforza84kcu5bINnpPd5/C59/05p1hi73AXn3jPHzO0mmvOigyP9Fhz5hLTVYbMnEs1/Dpm+wXo6ihqzSZu+8bt7Jzr8rDkRCrWznTZMAsP7KuYWtdhNp+lsBlNDGhj8dHjtEG8w4rDZOCjpyDDh8CSq5MeoAioGPHjGm8cOQOqqsY7Q6Zz8tpQh4aYW2JsEJUTckG5Ch0sznusieRR8HVGXQW6akARLZUa0m8qhqZLMB2mJRBsQ1YGuq5ETtbYLMN0BB8iwSi06rAsEzIzYthUlEVJVVVMDxpqNFXwDOgl76s6pnjL2oPNQCnKLMNPHFlREqsJ2nYR20CW41xFafpMQoKyvP83Ejr9Wx1KErYpIaJNROtIjI4oNUrH1LWoleElp10PaQ2dkulYTC2g8qxU9SLmtJ2ByBM8b6VUckMUUu6qjqAt2pbE6JN4pa0yIKKiaWmYKpmhKUgwkrQwUCuwMrodzKX2WySkEbEyoBLLAlWjbU2MobUEyIgBQpiQyWridJ/O6ilU/Wes23GEel0GZgFfgGQZNkIII9BjTOZQfoRqRsTREhSKZu0s/tIr4IIXk2+/miJ/Bm54H14to935xBBQ5nJQV4PZhqMHdgklDTGOsPlRrFlD0Pto3CFE9VEMiNog2Tgl1eidmGCZdC6iaXJcfRQ39rjqMHJS8ZGPK8arL6aYL/mBC0DOOYfYUcys2g6DyIQxIRyHbk13qk9cegyvkiNfXDiE9BumS01wkQ6azEZyt0y5BUbBIzHDrjXQb2jI8ZKnrqU3RTSdFBI+vR2dWYqiOC2AixGCb/nmIqcpo7mcZMvmgm4vY7kqWGoKDlNwrPacHGYMqxQe47ME24UQ0kBfp+9d3ZzWZ2itkzFzK2A6DQ220nylFHmegwSCa/AuBVmgDMYq0IEYCvSMhfXTmFf8Lg8+fpBxs4O7bvoHxvfdzfpBxlVPPZ8zn3IdazdX7DjvMXrVIi9/7ohLrh6g7RTHguGOz96ATI952+sf56zyKO94xTv46jcP8EfvfT3vfecbmPPHuH//mHe95T289Od+nIXHD/Aff7hi667VPGNnw9HjlnN2KOTomMI01PkE6y0m75F1L8PE7fRXdznpIraOrHaOLes8Z9mGvFPTn1MMtWUcJy3MFbHKoCeOLga0J0ikaiJjN0IVGbro4TNNLY7aVdjCYk2JDw3ECVNZCUHIeyluMlc9tJQEpTEmY6jGKCNkOsMTEetReUPZsXSYYExNYzTjbodeV1M1Q5pOTrTpOlrOKxpr0bEgTBwdq8lthfZDOq4gD6uY7U4T6oqy02fSRGJoyG1GUwe0C2glRBWooqMxoFVB7RqUgdo5tCkZZQ06BsqoyFtDP61TB2j+/UA06RARNLG1CcgQZVhxPBSBVHSrNryCVC1HgZZ1o2IKp4jBJF64jigJKJV8LZROgbdRIkqtwDDJl8C0gdshJn68ae9XKmshHUFLDqpGqTw9YeW/Y9GI7aA2Yfpo2r8BEFPXQDto0619b7RobVLqVEypSBINTTxIPplFZAm/swTtKFwHp3tkh4/A4jFqrSgWj9NMdcnPvRTprUJ1HMZneKWwKCRfT93MoYoZsngvlBeSNWNiWaPLa4h1B5iAWkth1hOrg4ieR/RjaBmAHSNhjjzrovRZ2LiPYJcIypNl1+CbCb7zEJ16MeGFR47R3HsP9sKX8bXlLovjw5w6/g3e8evLbHjqVVSP3k4WT+CzGeJ9NZ01a6HboT5xkM66CimWYFyxZk1O7GX09Ai9poMansB3dPKzqRexmwr6qwpk6RiiCxhXlDNdXD9C1sGceQa+dti5M3AmoHOFaMjKnOjSbES3tYH4ACagNKyf1azesAovwkgbjk9K9vkZss6AXfEAaxmTFfNksUCyJoXNBJWuSSR1jBKRVtVK1GhRqYJHp3pDkt2yb1yKj9QpgzMGwUUh7xhcHdAqY/HUcbr1HOMDnuOffyt7vgm7XvQTcPtHmR1cQEct8tnr7+YZy8e44Dzhg39witm1J1i9reajf7fAGdsDz1y1ije9aiPHDjzIS35ojntGY171ll/ivm/s57d/9Q/ZdOZGBrOreMl1V/CXH3IsTu7kh1/+a3zgKzfy9POv4+Mf/V1+5LpZvnn/MjvO6rBtZgfF3BYYP8bCg19j1fbzqGzDz76lwy+88iTLmxsWOyWP7VeczDJCZ8y5F6xlTV7QxVBJxOYlVV1jypzaN2Q2p64naBNbMVAqikoyQi1kNieG9PkWGnSnz8JoHlNqwlKdCiwXcdGjco8eOXqxgzMapwLWdBgCsWlYMEtM65JxPaFTFlTU9MYdBqbEDScUpUargIklTayxVohG41wk9wW27DHRY3qm4USlyGwXXCSoilh0iN7TUZFoNd47NCnkBRdx2lGUNr3mvEPVRLQyaG1wAjkQ/BIqTwVq+HeTydrSgfSK2lQkiSSUR2tBaZ/e2NqjXTg9wAzWEvLEb4+B5OEuFq1scuPDkbhoOZHin1gRxNg+TgxpxJpoyRjVphdFlVSFUbU+863Yqf2UioRk1NU+c9UqQpMrwop4SJ1+bSlFSoHOE6yCTV2FTgwbdPpdEYW1fYJZRjKDDj2Ut4RejcpOwtoIW/pQTAiXPov8ipciU2uJFvzSCCYjjNJoXaCdIvPHyKsxys6RxaNU5UNocxCZLCQIQS1RxW/j/ONEvYw2HVTcglL9hAtLB5EeUd0O9jHQdVIexwdR5UOY8QH86AEmRY+oh9xxV4nPNFk1xMWTTB0/woZrN+HiUcr+VtS4iyUkE66lI8QYULOCLzKi1XRWgd86RawnZIUlVMcAgaZB5TrNF6Y6uE6DjZ6gku8N1RhjZ0A5olLoegmmVqVQGGtAm0SzXRF7aZ04zCYNwIJ3qLIi6ApvCppJSZH3UKViLJZTwTAiI9Q5Wg1OWw3olimzYukg3qNNslXWbXLWSkWvlGLFA1brpOfw3ieiAKmir0cjjDEYpZidnWZybIn9Dxwmd1vY+dSrOPzIl3j40Ijujk00s/C6//ZqLn7pK9lw9ixv/8i1XP7UbTzvZy7lXf/javY96hnNezrb1nPw+C5e9WMb+LEXnE0c7WZoTnHVuetZOPA4133/K7j59qM8fnCWe3Zv4dShW3nzr72NO771eV71+qdz055l9HJEtMeMHiDYijoPzJgC1BEyc4RnvPAp9LYqvLLk4wnLwTNfCTPdnE1zU+gsRUMWGkLdkGtDrCsybZg0gspKtE6mZcSkJnfRkXW6NN7QBKFySwSJ0HhKDNoLTilUpahjQx1r/GSChKRtiW6Ij44gGhMLsnKKQZhOmhajkbFCjXMaA8vNiAhM0LjGEJsAEhnagPeeLIJkDic1Vg8II0PXDMisJTgPKJpEp0K3/u5W6+R065N3lo/LNPUoedaEQKeM5FKjLXgCTVQolRFcxLuIc9+9H/z3RgUvERWbtl1OzBejO6e9VVISk0mDQ1bYkobk7KESR1a1fHYdCDgUJYgnkjBxpTyJJtkyblRSxAZlUFGwWlCEdnptUboGUthvElEF0A5NAeKIBJQvUCZCbIeZeEQ0Rqe/swLnnH6ZotCtV0xaEJKTYKq38/bnA1qaRNOKYE3aVCIRpTrETgfVn0ZPnY3EU/jRBKsVNpum7q1Hun3seABZHxUiPgRiLlg1IKrNlG6Max5CDaawVYFSnowcURPQk2SIFY8lgzc1h1Z3JzvbyV6UlBhOQuxBaAgxYO46gvS/n+PHL+XQF/fwj3ft4M8/9TnWrl3NrtWP8prfLMEKDI5DlaFXd4m+xm7q4fwYE5cxEcLG7eQnDmJ3bMTqAkqHP3MKqybETEPjMOOGuEYTh4v0TIHrlyi1jO0rWL0WlUW8nqCaMabQ1MFj8yFalQRfYWjfRq0Jvml1DoIKSa+AMjS1JhTJIKuOkTpEdOaZb9Yx0z+CRhGaEbps6Y/olKMpAZtlaXP3Kvmam6T21dbgxSdTqRaXB/ABbFam56IEIWJtjkGQGDj6+DKYKc7esobJ1vPZ9ze3MTnn51h9+YMcu+1PydduIFPzEKb54hdOgfoKx0+NeevzHyUOMq78oYv45edOc+/DA1ZvhYe2vQa1++/4/A0P0+mdyQ++9qcY/93N/MOn3sPID9i7/Toue+nV1IeO8OHf/BQxLuFOCK967npOHDnK1kGG7mxA3/8NbL6eOt9HtuoK5PhumNvJUy7tcfyuMbW1SQ1qlti5aTv5woj62ALlpmmqOCHvzVBXDVlh0ganBK0KYnCIpLCL4A0m1q0tBjTeMxDNQmySYlxlhCaQiafRmkIZlIJagzURnwXKoNGZZVJHXL2EM546CKJLcjQxSyZikyVD2cmThXbTMKCLtzkujlHBMTDTqEpjTM64GZJPO2RQMh/mmTUleV7QyfrEUBG8o1EZSkaEWDDoTxFCYBIrym6HgGoNAcE3EVSGn4CxQtCBoGqs6qJQZCtIwXdxfI9U8LCSyRlb2XzC00O6KY/SLlXHLa4pEVSIKB+TRF4EF1JFraxKQ1AlKSBa2ycqbAmoBHoDEZt1wFiiakO6IyRFXBv/12Lsp/n4yoLW6LbVRhyxxdhXpOpAa3NAu/BDXDHwokFpB5KDdFlJWhKaVsJeECgJvouhC9RENY8aCcp5vASoHkepeaw2SRKdzSB2PVZ1oHaorJOo236RzIyxXoh+gvKRJuuTFTuQusH7BYIfJasFiZigMGYROEAMe4Ea4UHcaB/aNcR6ghorwmiIqsbkdQNbp1m84yGWsxex7bI3kh0VdpzTp+zs5zWvXEtYM6AONVm9CplaojGR2LPgDHnso0kdkhmepM4dGY5QHyB0M+hHyFMVEyNIr0RnJYhNi3cBYnMoSiSMUWYaMsEYIO9SZCB6DmXtaXsFk2VJt6DVaWhN62STnOvkWZI6wTlGwwIjfSahwERDHBuGfhGjp1ofd30aolMafHxiQ09q6eRXI+oJd88VHrjWGq2S6ZQo2gxcg+CIXqjHDes2ztDvlMy7/Zy46y5MX3jShdvpuCXO3vpMsjXncef1n+b4oyMW68CtdwtffGwD+7Ie198e+fqndtPIHA987UbOPWfM8Rv+mD1fuQF3csxTLt/Cp7/8GAujU6xZvZGFRx5n55bL+caNX+ZrH/oIT758J8++9hf568/s5sjyPLEOFDaJ6vxsliyL6cKxm9Grt+LkYV70+idTmgE0Hl0p+lmPvHToHpjmCJPFI/QD6IkjF8CHVJwEm6pybciUbedqqYtumooJFVm/wFtDLCy1j2gVwHhi0UW0ozAqwW02w9PQr6aYxJymadLj9nv0TI+1KkPLGBObVLxFTdk3iNSYPLl+LnuDqUd0sy49+ngM9DIacXRWzeJjiV4Wen7AsHYYJVSLi4zHDTooiuDxGVQSqKoKrUiMLUlrx8p14h1EcTTekbb/HFwijqT1498LBq8SXxwRdBREQEWb2CVaEjUtNqigEy3NKBCf3BSDJcgEZQRjk3pVvBCMQKaT+96Kc1+sUSa5MCoJafmOFUZFRDSQePjISlBDQNmIYFMaqbVEcShlEWNQ2iCqTIZCJD/1xAYKCYZf2VBUsn+NktBYFSwiDiF1Ce1JSBtEazillUPFthtRhqg8KmpMDZEOuJqQaUysCGEGLRZUB3yDzxt0nCZmGyCsQmcVKkyIcj/UI9BnY9QCqhhigiWEY4jcC3EZG3qoOCRUYyTuRocMYQFUDnGIiE09kDYoH3B5ZO55I1aX/5G/f8PXOf/Jc+y+dYmXv8HhLtqEkQNkUROygHKeTEWUU4hqiJJD1MSgUGFMLoboR4gzKNNgxBIxoPXp6ljbiOrkGO+IymO1RiyEsoOxNbbYSDQWrS3RaXQ2JrhlFA5FB99azAYPSkfM/83dm0fbdtV1vp/fnHM1uznNbXNvcpPc9IQEQiBAIBjpQhAEBUHssNQShNKnWHbPsqrU4skDLbW0HpSlRkuHYocNFkprIU1oJQmkISHtTS65SW53ztndWms2v/fHXPvcm3pjqG/kH4Z7jD3uufuc3a299m/+5vf3bYr8egRlpkNoBWRMYxMbo4Iu5iHvvISmKgjBMLObDIPtXR97mqupe8ivNw4zoNGhKdvBWizEROqLezKCxGHOInMAACAASURBVNhbMieSyT87WxGih3EFzZyT85oThx2LtSNc80OvRW1LimfxpY9+kssuPsg9h6/lkbs3+T9/62cY7D+b553/I3zlocThxQf5+Nvewvf/57/kpS99Dv/uHbey/8zLMSc79r7kdXz6yEnKyZcpHryLa66/iO/4+ufy0PR3eP1LruI37riDE3fdzXSyC9MVrB/4Rir7Z8w7y3h+NGeIuhGVjXTrl1HqCI5+nGdfdTE7mylrdsgjKwXDYWJoNtFHLceKO1lPF/CIV2TsGQ935e8wBikWuZgHjzEQUos1Bu+Utg0MeiuCGS3lFIbVCotmkypGTGyxqWKzMMynE2pX0kYlMGVgYKaO6D1FmOOT0kRLm0oWZaCeDii7yHzkcMESWiWox5kFbRxThJK27airSDOfsHNYUrcdW7bL4qQYSOLYmjYMhyUhzhlKTZOq/B22FtM3pYuUqNki+BFlYSEp3i0gWmpbYjyk0lMVQ0LKLBrf/Usp8GSv8cwLz/J6Y+m7XAN0RE2IzYb6MOyl4D5z1VPMHboW5E6pJTDIoRP9iEsgC5dM7qStIcM6Zil0yjmi9DFrOaQh4+OKZM57ykEZLJ3/jM8UTYReNpvvu43K5Fe7tB9emk+JaP/FX5Z3BY3bAeH5tSraW++q9IHZCkYjGCWKIGmCREeyD4F9DOxq5v3rk0m6wMp+ojySt7uygjCllL206Q4qziRyEuIMayYQYjZG6jYQabE+YYwDU27bReSON7vpqWZ3P6lKfNPh7ElW2pLxaM7oas95L9tPNIdR5kgsMdJBGuQiqBF0jGgOPIC8y1o6MVprCepRHzJeWQAqxBhyNm5hcpPXf35iTQ7ScAZctmzOFKeIcTljV0j97i3bU+Dbnk0FQbONReENXkYU1HijWIGqHFHisItjWVizNqYsFNjMn6Uu6bKnnc19nkB+4t6/SBUjDYjvKbo9VFe4vOlzubtraaikxKQZqdjDfHPOlvE85eoXcdt7Nwizd+DG+zljx4gHbr6fP/rtRzjvyefyFPsgLzzYMVoxVKnkI39yA7/4rjv5nte+lM2HP8P62fvZs3GEo3su4BJ/P9OVjiPHH+aKr3s+9x33/NGN9/Pca1/KzZ+8haf+m1/jwrW9fP5d382ecys++7cf5rrnFWhZYlOL011QNbShwq7txT9yP0V1AH/iXn7yF67gbW+6icFoTCxgRVYxnaCTCvXCuJ0he84gpoYSD6lgFvNn2YnFqoKp845Np1T1kK5JdIsFrk69ZUNEy0Q00LaZdRWnM9aHA6bTOaPRCoukzNqGWgqa2NBIiY+BVBoGTXZr9CERrKDzTVw9pOkilXHEoiSJh27GwBX4ZsHq+hrTqLStZ+xglhpsMaQIAScZsy/qipAiKTaMYolg0CqLM6kNiy4yGpe9kyaMyhFKYGAczaKjKyztvGO9rrKXlfxL8aIRg7iV/IWIXTYNE8mDVVNDygpRLcucsuRzOpOaDjEBQi7s0Ya+QyowMTNT8tbcggjB1IjtqVo+c5i3jcpsxvjzMNSgpoAYEPqQXHUkKxhrwShieyGUrSGm3mhq6eK4/Bfy4pHvk3yXt+Mi+feiedHQ3nRK8lBWnIPgs9Q95+31SVLZMlklgS1QKUkyxcYBBEXdUayMSPI5lI4Yzs82x5oDfAuzixQjljUoRmi4H0KDj1OKNEd8778TatAWUkliljUC0KdkBUQcUedoKikWCVwFG8e56FWBPZeC27WDMKiwTUuMq0jZkb34Ty1qSrOtdXDG9rRNB6lXLavFpOyfrTFhjKUUycd2m45Khj+sQ4oSb0uMOKzEHi4zELMfTIoekQXWRGJs83ocJQdrIBDBuZpUrrC12KCx6zg/pi22iFpSViWdHTH3wtDlAS2yFC0JmrJ+QsifZ1LFiJBIIF2mZBpBtUBKgS71O7CILwQTAqGLVMWYmCKLk2s89NWTHNi/A+vg5t//GHuuejW7ho9y9tMv5dHj59DWKzz72M1c+cxVnnndN/D2H34Pz376bv7mb47wvg/BT3z3Pr4wvZH7Hx3ylVvvZv+L16jtbj6x0fKO//BbfPJT13DojtvRlX2Md6/g7zzMxjzRvPfDVFcOOLcuecoLruSmz5xgz6UJ3fLomZfg44MUbUVVezj0WfwZFwOXU+zvuP51e7nqGdfxAz/936jLfaydu0qXoPNzjj1yhJ3N2exe7+gsdHWFEYeJC6xVNAYyT8LRtdkihNjhNFLammY+oE3HcTYQ5hUuDLCxIfot1tyQE5MTYCw+tdln3wu+8hQoq03BiaIjLhpCBaUZUKjHuJoNX1MXFWX/Pa5nni4p7Bwy8Z7V9VW6kKjalsFwyESU3bZGg6epB2jsqARcLBESpQS6RWJ1vWCqDYWHHW6Nrl1Q1D1yQGLRRoJ2tE6gNAxioENpt/MZOp7o5WujwCdIPoD6XlVaoO0ITTbnhmpLig0mzBH1qN/ovWfIlMVx0fPXC4xRVFqMqXLnbQQnEURJseltDnIXabCkpaWw6SfW2ROXXHGzzHybeSMtCFgpQAtSkm374eWQLD9GTljK3bnLnbtqzi1dtnoiedfSe+ggkuP0BCCgJp8IYmyf9pJhI91OYdIsaNIaENRs9WKwrJo1MqDTWzKXOu7FmhJNxyGeB8UmREHCFEvMPjZxnrtkk4jGY2IvmEoZLtFQ9CrL1HuWRyQFQuFxydFVhv1XDkmVoV3bSTXLKkZnchEW05FtjXulcCoRaR5nTZzXwxw/KDFCikTNkYkZm8uUw207XRHUZnUjgCtMLqh9whbqSX6eZ+qSIZIYfWZNRZA+NyCRHUI38Gz4Ga7cy8IrXdkR2wrrCkya5i+cizStZ2WQd24xZqZXZk3Z7fOyx+gw0qEmkwByNq1kQkFpsF6hdBgfobIUtiQkz/RIw3weOOfAiHtunrFIwuVv+E3S4Q8ynZ/DodtvpDnzSezf/yxecNEXuWdyBu/+f26kvKDlg3864fnX7OC1F76HZ//Qa/if1/8R//CVGetnHuDT9ywIx47wvT/3Mv7k1/+Qi3bs5Oa/u42LnzHih77pPA5VZ1AfUp727HMp4xd4yY/9JFIcBfv7fPXwUfYdPI/CXcDiq5/D7DgXM2uR/U9HZsfR+CH8+CziiYZdlz6Dh2/f4sWvOo9d6ztZLFqsc9AKs6phvcr4ssY5rhiTXJnnaNZkeGIZWuPBFAOs8XShw4jHuRG+a1BpiXiQSEjKrI201pJ8pqCGUFE4w6wXTAXfUJkVVKfQepIzbKYAXYMplUVoWGfIVtMQBxXFKAuoqrqk2dxkbTDCjyDZFiOOWWpQl5A5jAbD3JsSmXSe8XjMqB6x0cwZp5LSlcw1UZUFbdviTMGgLhAxOEpSjxJgwatSqsvN1JJV8gQuXxsFPhbIiQNADm9OKaEuTyjz1jbh4hR8thCOCzCpJMYOXIOdTIjk4Zg1FYol0Wb1o1S5G7fQRUOxNkOtYqoKbEcmzkTULtOLemWr2OwxktnOWDemiCnXVktOgwpLyl1CpQ9rINsEL4dqAkRJKGG7WxT6jlwjriyBkthNEauE2FKkEWoyVdRtm1QlNFmSK/OALnmIg6x4NPTdfX4sxJB0QWl2QwiobvRMIwG9CdtakLuxoSEupmjyiArGRkyosayQxGO0AutJocweO8HnQVDqMGZAkohtEkikLISwZwWjHandJA06aEpMkRdpUYcARnp/S5cQn5koxLwdTdJDHpKQ2Fs5lAb1vQgu5vxXRDKGbcAUJRQDorF94fZg82eWd3qRGBvwCQkmX6NkhouHGJXCWdq2ozSClDUnFx2zYhcT8RR1TQwtyUaSUZSS0tWEMEUKg7Gu901aLirZBxwRovTZtqFCcWjKgj2cw7QeLbOEXosC68mCPrWsngHDSceXb2tY3ec4Y/Uyytv/kBOrFxG2Cm66XZi+7xZuv+/jHAqJ0B3jqU87l6LZ4Hu/62WkwRFe97a/54qfew8/9wc/wGNv/x9sPXSY7/jOy/nELXfwqb95A8+98DoeO/4ga086j3vjTn7i9+7haZefwao9zmjnU3nPu+/n9Zd/PX/19l9jfPJRDnzTBdhmA3/8jzEjg6wI7Y6LcY/chjvvaXTtPor5QWz5MfzkTj78d2/gj3/mfaSzzmH3quGSnXdzxStfRReAI/dy130bPLbvGppUQOFpfcju3SLY5JHkCa4lAV4UbwKpEHyzwIvHFpa2a0hqqG1FWwdcu4W2loAlyTGClFRdjvpsBgWpnTBPHWo9RXKsVkKzaFlJY+a+ZVK3WKC0MS/OSSkjqCmYzBeMu4JiJMRBYGRW6HxAxmPaNGdQCqFVRlWF7WZs4amLIUlKJjGwK3TMLayUQ7pOEO3wqqyYEmmUelCx5QOrxZBoPCF12bblCV6+Nlg0wqki1icN5W6nz/TsuzfIgyubDKozrFkg6rODr3FYyjyQjIbkXJ/doRgzRBgyMGA2LRxX0vEZuhHRLkIXYMujmwEWHXRz6Bp8mKHWgHGkbkLUCXF6HCbH0e4kMghI2aKSFbQphWw0JQaxA4KC2og1WTW55E6r9Whc5KGbn5K6TURakIhzQ9QEKEALC/UQigqpV5FyBesD2mxm3FFmhLRJjJOeFGQheSTMIbaoP4GmLYQOdJGfy08hbJG6OTrPCTXWVdiyQNwgF0aT/fdjCKjvYZOUdwWGmLfTGnOrWlYkkz3vjQimrKkwSFdjypwnat1y9pBzS9EGoQGTh495JtFlbnwflKAC6gyacj6sT3kBJiWSJCgDFHnAmtwC4zKmbl2/w/CChKbfjJl+RxB6EV3aPq/KIh9vrfJOwHhHaQ0VhkrHtGmY4YQ0wLgBRgZEGtACmyx+0UGvqUiplzTZArBYTaRkskUFLWLzTCfGLn+2GKIYIBBNQLXC4gnzms3jDmeHjKsRKwdei+64lkfu/3ve/4H7eNVbf5nRgYoffvv3c/yhyLHplM9+9m52XXo1d9xzmK0NZde+IbcPPIfvOQZd4JpnXMDNd32Fq595IbZ9EqvP/EbueORizrj4bIxruPIFL+YfPv1RXnx9ze233kZV19zwtvdx562PYnc5XDxKs4g4v4dCR5jpSerpITA7SMfeT1nuxxcdoV6l8J7Rnof47hteyfe9ZsJrX3GAy7/j+XT2EezoQcpLLuCy657DJekBds8eQq1hFCOr6vDeY1IWClkG4AtsrKntCNMGVuKA8dzg+oCVEsPMOoouMWYdVw6oVDBVgdiKUHo8c0KIzE2gkkEOCJHA1BcYt4O5F6AipoLgDUUwLBYNRoXJYoExhrJ0hLHJ7qEtbLQLoquYzicEW7IINaoVM4GTtkSTo7CWFBtcinQhUMcKFkIhELwyjI7QGlIFnUSqVNDGhEuGQu3STvEJXb42OniWNERLrz3CmGW2Z5UlrAGSbIIErLNociSfMemkS5xbtod0aNl7yyiRFpLFaA2YjCM3Hcw75OQMiFh6S4KeEmlMpLSArVCTgDliR+CEZLOlq5U5KbXZ2gClGIwyBGSyC15hhBRbUjJYM8QTEEm42JJC5v5G225T+FRDb8egFLKCSQFoSRqwwSEsiO0ExiukzuNs2Rfa7F6ZWAAgMoKUbZABku0wMRdkgW2rB+3teDCG5DtMz99fmrFZa/OihWLVIOJPwU6mywZcLHcG2d45hpil1kp+DUZIvatihsVixvBjzF16JKuF6XNwNfX/ZqO2pBFL5poqihowxiI2e2hjqzxcNWOUok9JSlnkllpSaiG2SPRoiEhYQmpQlI7UdSQjpORoYyI4mDQFk6KjLQwxdTgzwDAl2QqPp22FeqzE5CnrOr93W+RMTXGklJkyMSWsLbd3dsvP2UhJjB7rAjkmoEAVTGryQNxa6hVDHTt27vLc8he/zujVz2Lnhdfzg888k8+/9yOM9Ap+5d/+AZdfdRYPfnXOyXbCH//BbbzjDYYHq4JzbMeJwvDO33s/v/CWC/mLv4OXP30fDx86xFOu+34+8j9+ndc9/0J+8+YjTI8dI84Pc9YzLuHOnd/Jsy6BT376C1xZfYbvecu1/PV7P8MmynjHEB1dgNgHaN0GThusbpDKNxE2PoDb+WR0vgr+QRiMKeaBOB4Rx4KbbWBiSxs2sfEoWg7Zf57ngEyZbxTcOF0nyQnEWeZiKdpAh1BUFu0ydOuM5bgu8GOD+Gy9PNeAdJ5QgG8WmGAyCjBTqtTRlgXz2FHbEa4tOWlmYHZQRMXLHDWB1FVUSUlNgy+FWYQ6VZAMRWGRnutsomXmHSu1waun0w1ibelSotKaUVmj2pJ8i9oh3jckF0kRxm6ENidYDEY4CkpTs5BEJYkiOYqs4SeGhO+H8Mb8S4Foekx1eTl9W5GLLtnT3QyyIlHmPZ88K8gyf7gXDIlDxZK0yzakqUBMiUiBkJkxqj0LA+lZHNDSYSQXGYOQ4rDHi8s8Dy0y95429AZo0puOGTR6DBY/mWVlZFVhi54LH2qQlpQ2KNiFAj5t4MyIKDHT5yRjx5oEQXBSgAu0XUvpC2xU0vwkPnnqUU1aLGCQB4dAHlL3HjsAKpmCmU3WDCrDXORsDlRZ5tqKSF9YldjTNK3NA8/MEuqLbRKyu6eiWvRD1wREUjTYfk6RtPf1722XlzxzEUtKsae3QgrZ7lm068MuXJ9JmpBIHkjqqdeovXdLvmY43po8bNYioGaAisPKki+V+oxdRZPd9iwyUfIxFkhzA0WEQggxYswQH4ZMYqQrSzprUVdSmQGL+QmcK4l+SiwqWjvLHbwdoKHNw96YPUb0NN8k67LXTFZBy/bQH0u/01wqXMmzgVjhU0OczhjsXeXR2+eMxy+hu/QsvvKXn2fnvjt4rFrlpn/YRFYMsqNkY2M3J/Q43aTjtuP38/pfg59481WMD5zBOZfuxmx+mbj6Ei4Y/zblOS9ix0VrPPjgSS6+/hv5wAc/yGuePGT/i5/B39x4D1dd8o385Vt/jC+M4dyza172LZfwqTunXHftkJ17d1NOjpHGI4yJSHouVHfRmTHV/EOw79k5oGO0ReP3U8s+GBa07YPU04eguhr8YZyxaFeT/CbGAdbiDjzAdSPlg3c+mV2jdWbdSZpkcvBJ8D35IkcZjo0QI3iveFeQiNQoczFEDPOeTaflCvOFp10IQzWonxAlYV3C+ESUQKVCnDS4gWNuFOkiq65iYiTvcIuC0pbk4POEth3RKE27gq1q2kbZn2omKeBXhsy9UpUVWpTglRBrKpvFly0GrYYEO4A2sl9zQxrGhipYQoI0tCydbZP0BIInePnagGjgNGJ/4nFZlss3Kb1/zHZOoct0My16cVHadqLcvpsEEJ/vx6nfbf+BMVgUq1CqxSpYTRiVfHs2AM1D3s4Qe3fIGCI+RAKgqcz2sSLZMzotwM8wsUVS7hxNALxFtQEmWEriUrylfYBJJHe8SbPHSzelsIYUhOTBEKhXx/lEr02GVyTP3JM2RJmDW+SrJsQM0GIH2CHiZ0RrUZOxa3oGCoDGPk1HFGski610ecj11DHTomccaT6m/RBal2quuDyZXC8sK1DstsAsDwpOP2Gzadyyq10uPCJks7aeSSBLZtHy7/rblhbSeQex9AA6dS7lH0yOIUwpF/vT7i824+8kSwxK03WIsSQZE1JFChVGKqbzOUXpCAjB5vdR2xyAnXxYPmF/wEJe/FKF2IoYOlLIFsBRU54TqRJDnyksaZtHjxbEoZIWY0b1Kg/dKTz10hVu+dQHiIdu4Hmv/A5u/ETLcRyfu/kh6vWdnHdwN9XQs8cqTzo44r2/9K28+kDBG77zhWzec4THTmyyMlrlfX/6Ae56LPCOn/0IP/4df8jZ5ixedsn1vOJlL0CGV/JT7/wg43MvpjxvxLe/6ft42Qtfwb03HuVjH/4UD/z1F3hsqyLoCep9e3DpUcxkTjm4CcoaM1gFKUAP4bsj4AtM6Qn2emL8Bob+hZji2wkakXILTZlVVZh17FRheoxC9sDaTs4fnmAuxxnIAC08NnWURvOOvhzQmZJWMrWxqCuMExhVLEqlUKFKK5ikFGWi9i3OBGo3J4UZi2KI2hUKFdRFYt0hKpSjdaI4mk4ZjNaYo5TJUtc1IsK0aVmERCdCaWp2VLuxS8qwaziBw7ohMj1BYMJWu0mKgbIo0HYCNhJE0abDdZ46BGoj+FJpywRN3lnGGImhI3S+1+5YwhN3KvgaKfDS51VKnm+qkW1nxtxPFwgFRgaI1iBFP1xUkglAhZg8kIhkrDbDCyVihn132WU6n1pELQaHodxOPBJKDAVRE0EjsRdZqdSoDrAyxHamZ19YnDHYLjM9SBF8IPqQ031CR5h2pEWNkVGmM1KBZvFNbAKuE3QBRNNHvGXfHCOOZCOuKzEbHcyOIWlBHIzwIaFdS7OYY+sBJkRMVJyWOEYQRuBXMAqaWkyc5gVGq757t0haxvlBQDG2yErOBPhIbFoImbYZY8wfSNbWocnlhZgIsezzTLQf8i69WSARST3DYRm8bEyGflLo7SU09tRJd6o+IkTiduQdnCrWOUv3tF2Bah7LqGEZ5JIVzjbrHWJOtteUqZ/L95EfLOsKnDNEDVnExYCFwFSURdEQCsek9dkPJxZYF0mdAA0ahRCmGPE5c3X5WpPJjB5riN5jTY7w0z4HWKxBncEWIUOQrofE6DBFIDUGaTeZdBVnntlx7yObPPfbX8PBa15F2T3Gs5+/woEDT+WpLz7IxTuFa644jy/d+WU+efdXeftPXMkN//U9fPOPvRn27mZCRbTHOXj2AS69ZM5rX/Esfv6dz+fSiwe040/xrj9/PX/2vo9ytHuUX/rF/0IVxpx/wWXc9IWvcuut7+Vt//WZfPJTibVnXcyaO8EGa/jjLYEE4yF+x8uRjfswx09Ce5IUhkhxMb4+CzsDSTPEHSfUTybxGVx6CerOwlSWZM5AyxF+tALlbmgfRVPB2aMj7ImbNCFStAUdoNYhhbIIE6SYUarDdZGZn9OlwOBkg62GbFWGtkrYlGmVPilOxiyI6IqjJFF3mwTXEhgR5tnKwruI6Rp2U+Al0pFQLCfpWEgCZ7FBYBaYRpg2E8KgInUtdagZSEPUGSs7VymislYMSG1g2rREWxOTzee4OGY2YQVsDHSNp+gsoS6IZQG1xaeIVbOdAGb/RfnBk8iWS5lKpqaX/5NT68XmoZ9xmod+XS7+qgFjut4KwGFchUjOfLQ2syrElhipUZ952CnGDNdIryqMkaAeQ8JK0W8aMiMiRg9G8RpymhMBgkclZppjyHF9GSPMWHJIueOWxRYSSkzlCOT0dsMAbMqsD5M9rFPej+FcHpKWncV3EwwG40o0eIgnMWJRaylxxG6ONYPc52pCU7fdIWufbZu0N7SSBhNLonbZoz5lWMtmjk8fP5cVlVKaPISM2nffCsQ+5CKhSg9z+AxzJQVdeqlArvY2G24Zj4Hsl45AT0JKSQAHyfeh6FmdjrEYn7afU1Uwaok2ISHk3UW/IwjeYq1DH/clSAgBQ6bFoqlfjHpmS8/S0piIjn6GAGJqJDjKWCPFLtroKa0FKqy2tCLM2o6hVeY+sjJUJLgsbHOREHMxp2swVVi+FChMVuvSU277hG3fR9dJ6ggSKGyd7YutpzhjB4PZFg8/DCc2dnLR/h9lR3o3H//4nzHYcZAbfvl9xMpx9g7Dt3/jJZz4T8o5UvDgfR/nhV9X8Q8f+zvOfvpTqCvDSut44KbbmLUr/MXi02wNdrK++wAf+4Thld/2S9z46Zv5luvO49///G9w8Mrn8rv/6d/jhitMusRbf/Mx9p8B6dDdpItWWDk+x158Hs4uSOsD3OwOWNkLoYLFiBAfxRlD2bTEAcjkf5HqXThzEcY/nbn7U5w/DzfbJI4KCm+JzoCeicxuxa1fiNsdeWpoudHXJElURgihy9BWTNR2zIaZEAtloDWL+ZxYJYyPVPMIWtJWQya+YXW0m82TG+ys1zl6coNUwCIKdqGIbjGsR4RZQ9cY1Akz8YwbwXmLOFA/ItJRlxYvnhKYdZvEcshab43SxBabHIUIR7e2sGVBFINWDjGBUM7R2QoBJY2VyhpCysHsaoTkOlwr+CKQOsN4MCaERPBZe2PivxSzMUCl2DYRywk5FhXdlvsnuhxppsrSX9vaEtRmCXsmR6P0QRp2FcVhpEDVEonZ9Kl3+0MjMYXMvCHQYyRATnZKFEhKfUaIzfUiKKeUq5nwLEJvP7yEFKDY1qcqsWshdLmwSZdpmCZi3NJxUPv+OBBTLg62rLAm9HODXoi1nBn0GDJJCBJ6T5NeIi9Z1q+p7a0Lmn7AOQZmGdtXQ9QW0YShyotUsrDUBMTMoJHe7javGWkbKUN6uCNlz5BsDpfDWuKSb24hEXqQJWeQKmQ6oeQ4wfwcWf0bY8T0kEUiB05vK4slD1m1X2xEXM/3j0jK2gFc7taX9hIpNJn6GSOEPE8wkh8jHx+DlbwtNsbRBENnDBMCnTuJpP2IFpSlZ7FY4OoxtilQPKYcsWhPsFJbkvGkLlAUhtRlL3iW8wgEEyJKzEE1shThGVyskKR5eG4tIXqMcxTTEVz8bD7x7g8Qjif2XDjkA2+9jvT8b6Y7ktjRbJCGjofmA64+syOtn8Nv/PzTKHe/mvkj7+TaqwM//a7b2f/Oh3njy5/M2//75/kPN7yALb2Xz30ocOZlZ3P8UWEaO+64+bOsnPxDPnXTt/Azv/CfWZ0nbnretRz53J9y5b4HcXsn3P7RVVZMRHQ3+y4aYnF0ayXFrhehj34AM+9g35nEicU9+EnYFYhru3D12Wh3d46o1C9gXGLgLsPHw5jVAzQ8QqHnYn0Hqw22i7TTW7GrlzDWB5jP9+Zd2HJRx+FsDu8otUSsMFtMcUXF3DekCEVZ03Vb1EX2+j9qjuJNi+1KytJhQ0kVLVuuYEunTOMEax2SIkNXUPhAExOmKEndhLKozo56UwAAIABJREFUGLganXS0Y6GRyGoaYqPArMVIzaA2BC9ICYMoyCLv0pwkbDAUPUHDlYbGd9Ra07gFg6qkjAZllUI9scl06RBbIkJhHUvH2yd6+dqAaOgLw2mWu0YsyzxWNX3Rl1xMVNg2eopRt+Gc5ZBR5FQwR0R78Yw9JURaPqdmt8htw6geEsq2wXLqC5uWw9geMkqyfdV4WnDEEvrthUv097KqmJQw2kGcInEOfkFqZ8RuTuwaCIJJBU5qYhezOjdCCFOCX2QkKGMcPd1PcQZEY96RSLZhiCkPOZM2kAaQVtDYkKIDSShthpQwfVzh0g89zyuQ0M8sQj/D6Mi+96eg5uV7lB6DX0IUrjA9FXT5h5lfKUty0uM+7yVzqBcunbIOzaKgbfuIU/dVkyEcA5jsNte/3sx8UgL0nyc95LXN8unvLyJZ3etyNqhIvzIZQ2GGRO8RE5kHxQtoOWIeEtiIs5mFMyxXcC5DgqYPRja2h2ZC2P480vL9mFNWBoLdfu9GHBoUZ/NQ7vDRkxz/7JfYt2vA+IySo4fmPOe113L1k/bzmhse4LznvYq9UnP48IO88Z0P8dYf/y1++1238Ze/9x+59qfezvtvPM6X7hUOPvdZ/PnnD/Giq+HNP/W/OHJTyY2f2OJX3nUnv/oHN/PcF7+Og6bliud9PZc940l85EPv5Uh3gv27CgZP+0EqaVgzJXvW1pAVwfij+DCEg8/BrL4AOXIXafhCollk18jyBLHYgdE5EoQYGpLugnqAyhWkcD3CY+hgBHoOw3adpphQlLuxRx8lFXupaDB2CsNV9g2mefhvtDeDy6yrlBIpRKIPFDbb9DrnoPTMZxuUDPGLhtIVVI2lrlYZMGBYDGkkoSESzJyqNYz8GCkFU8CmzpjQQuFp4kkAmgaOpQnNDqHThkFZ07hEa5RghEWpTNRT277GFBX16hBQylSwMbAcN5GFNUxU2T1MGKvUdWbyVcMd+FjiKVFboaZm1nZ0wW8HwjhX/POK5z9y+ScLvIicLSIfFZEvi8jtIvIj/e0/JyJfFZFb+uvLTrvPT4vIPSJyl4hc/089R9aO9gW9L9Zx6cposrWuCplHbASkJEnuezO2GzjllSJoMoiJKG3PziAX/xSxktWey4GedZr/RpfpOxZjClJI2VdaBLEu7waW9D05bS7QQwxmWbCiottUQ7bzW3NcW2Sb+ZQyP94kRUO+orlAmJQQbUihxaQKq2Ue/sZMz9SY8W5iyP4tvZxft7neFqsrmWkkEwxZHJU97m1mFZmSJCXJFAjZtpjQD33VZGEOBUZqMOVps+7embMfsqJCtkT3/c6JU4tlJHvq9HTIZZpWSnmwm19/XgC2eeR955L63FqAlE4VcUn9Lku1H/xWLBXH4iOSGkzMzpniFW3bPDdZxjdKyjOC0hBNopNcLESEttgitXvwbcugmpPaSCGwUg0Y2poCRzkQgk7zgDUGTJGN6FQCPgSsdVh3ytrBSs4qQE32yMERxBOkF0dhoYFu4rF15s77xzzHvuK58JoD3Hr/Gh/60Al+4Zoz+eSNn+f2Lz7IDX//+5zbKT/7+/8HX3wg8ed3wTP2fS//83MV55895Od/8cO8+jVXcN7VB6l37+Ddf/8wxy180zVP4uD5+3l0+knef8unOR5exPzEHr7hzW+h2rvGf/ypH+eOP3sDW5MaSVPG56xx36EtdGWL1dUR3P1OOi2gTqi/CVtVmMcOUYc5susg7dRQhIhLD1OUJyE9RBV2QnU56g9SLXZBvJfW30dVnUl0+/HlWVgOEs3ZmMmM5Kc8pVywwzZIUqqqImpABawtkMKSiL2aWjFL8XgpTNkkFiUxVgyKmtIImzInGqVtFxwvO1yoUBsIskWZEqkEbw0MCpIGRqaitsKwaBlLwVqbmVTTbkGliSIoRVSGac5QIl0EDRGTEt28JaqyFaashZq9i4K9NrFqDJPFiE5btFEkOo4vptiiRU3CpIiLkZVySJ1NrzL5ITxxHvw/p4MPwI+p6qXA1cAPisiT+9/9qqo+rb/+LUD/u28DLgNeCrxLlq34P3HJVsAxr4gqWdkogsqya++LBzZ7WORWbjtMQVVxridha5m/PKc9uvS7gqSy3c2l1NPurMVYS9KwbWql/csWXXI6lo90iuVByotF7q4zHS93Zyb/HE6JYBKOFA1oiWoObM6xgZJnCcTMUFGPyABraoyE7cclSfbgUZuLZYiZ1ilyWuEHpSOZac82KbPlgHR9p22ysRcpw0Dageau9VQCVT41Mo1Stk+TvFbJ9lGgmJNsQ1D6hWglh5drdgSN/eB5+/7S00uT9KD7coh6GkW2D8LgtIUxd799EZd8XiQxqMm7D5XeghffLyhA6jJ8E91pGLwh02gL1DusHSCuN4giITpkuLJFVSqS1nHG0iRhczFDdYGTjJnWadgneMnjPl9rLdp38Gokx/Cd9j6XLBqjBoPd3jkiiZAiI1OS5seYWs913/psTj5whGueeZCzVg9z5uo6K/d/mqe+YAff8vRv503/90vpuuP84Fuexac+/JOE0vLgfR3XXrPO119oOXPHhaSjicFAuWDvKj/6pms5euQwxfQxfvYNv8mVB87j2EM38YG/eA+3/e6f8Ps/+av86q/+NRed+ww+9+AJdO/TmBy5haecn9g3OIe2vRO9+HcZpk26nQeRkEj1WSzG59P5kzh/Ai0j2ESXIkEGuHZBN56T5GHEJjR+mWAaqvFVyFbAdg9RDE7SDY5gCyXJDMNJxuvHOcNk108fO6zNTZ73HmeypUMMmlXrYtBOGNiCsRVqG4lpwaRTCrVUJGKMrJcV+0c1vmiBTJyYIcQuMEwGbQO2GOBtQVXsYJEsrotMdE6QwHpZ0ZQVrYEg2Riss/3kUBOxz1u2MTEeDoihY1EWPNzOORkiOhgSWoORCmcGDHRAahx0FQVDJBXb2RPJGJKV3i7liV3+yQKvqkdU9ab+5wnwZeCsf+Qu3wT8saq2qno/cA/wrH/sOYTcnYuzOYBYwDjbF1l60UvZe2fnv4mqZBMpiCF7K0NvcyAGtM3FPIExQk+IzJt2Y+jDAUFd3hH4DHVYU/Xb7mVR6sM7Uo59E+351HEZGuKQZDER8qNq5rGGDGs449CY48GNZhqkpMxSsTjUFhjrejhIMTFDSbGbkXyDdn0EnBb5ZI4BkzwSA3ghdjmUWEwuMNZYYl8vcwj3KT/1XDwhphZNHalLpGB6SiNgCigdUriMo0tEdOnWmb1zkLhNq0ypxPRzAqxDjc+LT8zOmNaUIA7pB9hGNb9WAjG1iObXk9Rnz08jPWTWn5ya0Jij8FLKZl1Cf9hJWeOQSugFV0CeY8SARMk6AW0znOY9lgzrmV4VG0UprcMOIs51DH2CuVCmCVVoUZlDahgUBqRkXKygLpFsztvMGgYFSdiUrRjyPCF3dUDeKfXukg6y0thHbOFQ57IdRgnlvjHazZiuwtlPfwoTs4vNqeXzd4342Hv+jutes4ur/tUraWXO7/3Cy/nhf/cP/Ogr3sXnbryHZ1/2SxRmxMr5FQ8ejnz6Ufi1t76Lj77/CD/8PRdyz1ePceNnD/Ohr57gnGu+ge/62R9g9fLnsXLZ9TzzX72ex4pVDnzdLvShn+Y1/3rCfbdOuONvP8HCH6TRPZycb1ClNcyxd5Gm9+L0YrAbaHUeg3YLM6jBHcAMKwhzxA8wbSKU67jmGFa+BHoJWu7FNR2RDm8fwLsnE7oK15wFdgoFxKImFcpl52wRwgaD5ABHpTWDIp83nbSYYUUscmiKEIjOMYmGNkRcIRS2ZV402bzMRFrmnPQnSEYYJthdr7BWlIgxdNGxVqxhEyTtOLl4jEEptCsWFx07pSJ0kVIU0YBNBTbUtAvDsG+AjB0QrDCqatIiMiiVtDjGigxZLyyu28SOLaayLNo5FIq6RCee4IRocj0Kxva7w9QTTp7Y5f8XBi8iB4Ergc/2N/2QiHxJRH5HRHb0t50FPHTa3Q7zjy8IANvbc2OySdPy/0uOdO44yzxoMxFnCsRkKIY+3i9T+PKgUlN2mMwFJGU/dTlVCDKvXnp4JmCtyR18P+hEIoonpkUPPaR+kQmIyb4yKebCv1wAUlRSVLK38CmMeQlCPQ4P1pgVl7Ejpuw2GEPKYqpU9Qh0piRKElIPw0AfA5YEJed6pmiJwZCiJ/gWK5IhpFTmjjYlRJe0wbC9I8kZsKdCL5AcHBxT1xdg/7hjm/noORJx2RGryjZf/nSe+XInAxDVbu968jF3OHE5wKXfhprepnm5Gzs1tD71c39DHrCbnge9vcPo+fzK9kyElBcaXQ5vYdvLRvG9QnaJpUVcoawMBdMqTjqsVowwrKSCykwJ8TgaDJWWWS0NGaJSwObPxLnMW3DOIVhiaikK26dI5Z2E2ERs54hEWg0YM6CcCdVgjfVFxcbhR7njix9jomdxVvNunnv5FWyGktn0Rv7q/TNe/bN/y4nZCb76SMeP/19v5pI9jieNBzxzteL8tZI7P+t5/Xdeya+8+7uI+y7hW195FaZ5lOccvIDLz1jl+K03sTKIlIeOcMsv/THTz/wVr3hax+o48pHf+QJPveZinnzNBZy18wGecfVO9u7exYKAGw1J60+j3fowuvPb0GYL1q7D+JIuHqEcrtKlBygteByuewTYpNUVVO6C7hjNeEBqzsbICBdKyuF1pPoOkl8BtVi3g8gCTD7nTKG0oUWMz4EZQPJKDB2x7TJZwBiSsSRXUZSrxM4yLixFp4gbklCcqykZUQVDM0xMwibWJ9YxlIUnhuya6qJhzdWkNMRPDXZo2IqZmzWIysiWhKA4AutjaKqGwdDhFSotmUTfU7gdsj6mKx0dIMZh5onYKsZVGFOinWAL8DHSdoGuDdkXa2Boe/jwiV7+2SwaERkDfw68RVW3ROS/AW8lV663Ar8MfB+PRzOWl//PXkNE3gi8EeCcMwdY57YZFJAtCE4hIf0AVSwptWRbXiX1gxezHaCdC6dAr3zNw84Ylh3yDKLNLMGU3RpTjFlwlKeYGBuJEQyuZ5LEbGImFSHMcKZGkydpk31PKPoFqB8SSoKwHHpmu9jt+pSUnqfYb81BYuhNuAQlK1lTYhte0p7yZ/tZQsKRxJDdLy2aFCk9VgUNZbbe7VrE2KzQdAFNtld5hr7QgiJEnWN7ame2DHC4pYoUtofceVnJx3ibiqmCaJG54CYfc5Y5tkBMCbQvzur6Bc+TXSgFjVmJKkYQtXl2oXmnswzOUHoBVK9gzU+cB8QhRUw/eDaiSMi7HTRj/qJp24I5xUAKWbiWbZfzvEI1h6HXxoEz+DhBWsNqtcIkLSgpmA4HaNtRxQ4plVIDibZ3tVxSMfv5gaQ8XBchpIjFYB19tJ/FFvlciakmmWyoVhql25og0fLAoYaJg9FiD7uf9OPEr3wRd+33MD15A7/9ntvYODTl6m99IUfvO8Gbf/Q5NGaVc/VhqEq+/4eu5N67vsKjG0Ne+pqzefsffJHu9w7x3Atadq8uGCxGfPzur/LIxowrX/QqQiq4++hf8/wffg53fOrT3Pqlz/Gil1/JZz5yiPMvHnNXcZSihqNH7+aCiy/FrV9KiF9GZhUDt0XyC8LaW6H+EGZ+IWWIUF2EdHvp/AOI1kRK7Og8Bs3DdHFGOfg6XJijaYbwcqK8GzFPIob1DOF0ufg5V5BmDQddxQMaKZOHqqTpMqxojULoWKkdi9Qxw2CC4JsG44RgF/hgqKjwnWeHjjgS5vigjGIFtiFJIhVDxFmKtsXajo1uxoqpmdshrj2BG1jatIOi22LgBgSfkMLSmUyL9fNEWY/ZTJ6BneFdwkSHd47UenbIAKJBhw4wlIMKiR0SlFSAK4cUnWBMpBqWtCnPceo4pLR2u/d4Ipd/VgcvGaf4c+APVfUv8hdcH1XVqKoJ+C1OwTCHgbNPu/sB4OH//TFV9TdV9SpVvWrPzjJv0eFx3Zqq9oyQ/LuUuh6zLkmasU9rKrINbZljvLTo4YEhwdfEYHFlRGzO4DQmd+3bB8CY7ak19FvsZRTbEiOFPmWpzguI8b2kv3icQjLDEPRMjtRL5PPgN8aek61xu/tdduNLap2QQDOlchmyYXquucYOXTJoUl/49H/roJeDmSRomiMskFj0dDXpi/upqzVFDx/0RV9PiYyWg89tW16WZnCndlZK13vV9NYInOo4VLfJpHmwTeoHs6kv0qe69lzPtWfq6OOuy89kiXNvs55M7IfcChp6AVo+5vlp8s6FfjZhkubdS8j/d6aPy+sHwFYzQbYuHCaAk/xautCSVKiLVUaygqsEX3Z9I7G0UDhdSZv6mMg+ZzOQh+cxEn2bfel7VpH0yj5bVIhz7KlqDu5aYVosmHzpv7PjDOG/vOmNxHsPs3vNcryOfOXGmyirXfz0G3+bd3zPO/jet/wZZz9lTLV1iFc+a42Pfv4uHjzq6QphbWWdn7vhB/i3P/ZSdjx9yOuuuJBLLohc/4Yf4Ytxzplhzi3v/jh33PQQdv+1fOxPbuYNP/kypJpy7OENzlkdMlypsbog7Olg/TmYlRfR6jcjaUQ5fxvqbwLuJ565Tpz8KYUbYkP3/3L35lG2XXd95+e3h3POnarqjXqDZOlJsoZIliV5tsATMpjBYV6NY0yc0IFOhyFJM2UgdLqX09CBdIfuhECHFRZgGrAdh8kBgwyeZGPLsiXLkiVZ1iy9J72hpjucc/bev/5j73urZOgO4Pyh1XetWq+qXtW555576rd/+/v7DrjGZ1vt+T10zmGYMKs3sP3F6GSG6e+C6nLMbBcnQ6w1OHMR3eKZ3GzZTa7ayAEgtTqmi5bKCSJ1qQEFshSPdJY4jzRuQJ+Eyo9IdWLGHFtFUpwzcLDeWFI9w2lLtC0xnSd2WzhZIH3HepMYuB5N29R+g7E7QDd9hqoeI96xO+rZsS1VqhjhqYzF0jMRi6lHECzJC9lAQQmGHEaTlEkw7Mw6jB0wHG2QoqXvQKxFrWF7Ps3W5pVHa0dHWu0Gv5zHX4RFI8AvAvep6r/a9/3j+37sm4F7yue/DXyHiNQicgp4IfCJ/8KTrP5ArHEoebAKUvzai72uuOJFAyv1q10qQXPR2IvijhhpMdJnyCTVwJBYGBhKvUfvcyM0GVI0GONzcS7KzD1YJVMMM65fEUMWCeV9fy6YS8qgWQ4E1WCiYmIxuYomfyTNEvY+rH4XKGItIaqsXs9yMCo4jPpSFDRj8zEP8VKfbXCtBoi7Gb9LFJOk0l3jygZDSzeeiFHKwNVkKMjseb4saX25CMYSDl5mEOTjWFOuTzIQZY8ymRQrBWpJSopd/gjLxY0CB0U05hmJNYIDRAPLHFxTztwWaCdj7+WW1QSpy8cIOWdX+y4Hxqjmwh40m7rFAqP0+WdMSoQ2YCSi0uWdRYqMAuhilql/cZeazGc2UmEtLMKMafKE5DHIPuaPYFRw6orMPLO9Erovg9WVaw2GLlMurUGqmvluh4kNemTA5jORyfwMw+PXc/ajH+ClVyunjnY8+/jTzJ6ZcezKGzn78Cd52999Iy+/aYP1AxWnnz7Hj/yb+3jT932aI4eOsnh2xhUH13jVlev87A9+jH/9f36atHOAD9x9N09/MXD6P7+T4R33c+i6r0GvfDmXv+zFvO89H+Qn3/kQ43qN17zqBo6ciOz4xIGLLsVvjGnOrWPs43Sppq5/i274OAxeS+q2SC4gZ+9DJ18LbWDhHGb3Aj1zSJtEu0Maj6int8PadWhvoPooEp9CzGmsWlQr8C1OG9ApWjWsD89xVXWOvnY44zAJnBnQB6FPSky5MahNZDRQTN3TxwWCMopjvIxZLKBzHpkpLtQspIIwwvUTrIyxpmEndnTNgC424MesNQ3aRObdBSaDi7iQpsi4YbRbM4pDFs2ULXOe1reMmgHb/YzaJNQpwzQixgmVqYhRIAmV8fQiTOoa9ZHO9piB0gwsNgWkjwxpGGvDKHk0ZOi1/6/Ag/+LLBG3AG8DPisinynf+8fAW0TkRjL88gjwvQCq+jkR+U3gXjID5+/pslr9vz32d8r7cVyKMdQqKadkmaaw4kinlGXf+fd8YV+Y3MnbnBIUywJgGWaMS7MlABLQqKSQsGIRSaRI9uzWTL/TZHOHay1Wa5LOMDZmzDU22eYW0EJDZClK0ozvJkl7MEPyqMnioLTiaWcHTTWO7ABpwORFSVMpvJIhDUxBg8qAT/FFPJRNu1I0UNKjNNmSKKSr6C+joLJk3oCxKZNzUsI5XxaXMiRUXUE0GSXLtgVL8UXe5QRU3Z4oiWXnLqvnW+6+8mNZnEt3X2p9CCFboxb1Kezjvu/XKJQPkeKBIzEv5mpJYUEqpm/7uxaTNzTPwQ1FFOfzgDT0WQFsrGFHW0ZpDY0to/oAZ902lTmB6RakXrO4Je4w6Tymztc6r1dl1ySQ+oBUNt9jzpbox73Zi4hFo2Qr5D5hdnYZr60RUqReRFzsudC8DHvmdq5+xRWYx8/TDDrm3RY/9INfz4c//EF+7s4P8upTL+HEQcsNlzpuOHqYL/QzHhGDLI5w/MAG3ljMxoBnHvg8f/0bb6LdeZpve92LMHaHz3/iPWwcfCHPnL2a++76IKfP3Ee93fJD33KYu+//Y6bnztAmz9fedC1Pbl3gSJsI65/Hua/ENQ+iZ05i7QH6eDfeVujaDWg6Bu1TBOMZ4cB1mGBBWzScwcfXQPo9Yng/tRrm4RQDswV9T7BTcEdAE04ciQFqPckGLlkXPnWBTIEVR9dvMRx45u2C0CmVM8zigJQ6vCSGzYC4SMS+BVkwrhI7bUvdCK3OmHSOs67Pc6BO0DDnyPAg87AgmURghmlNFjoOx4idc1G9wfzZs1THDtEuetZ0zDCAWsdmammaIV2Xd4H9qMebkBOaxdJ5xy6RsXiceCQYYh+RgaN1inhP0mxfHLTNTUCq6bvIcPjl+8H/Fwu8qn6EPx9Xf9//x++8A3jHX/w0tMADuXBkoWbJI01ZHi9aYUyHRodIXZgWBpLNgh1sVmSKIZmANS3IGqrZI94YyRN+fKEDLlDyH7+RmC0Rlhmv5EFv7iGX3XtYMT1SMhiElLLpVqa+ZQxWNC8Qq/APY0sOrJLoMRhSzHi2sTWpMFQkRRRFrMVEKf4uma0BMbsuAhZDTJoxXO0K9l268wKBLBe/3C0LUXskZHwzpoQaLXNfXcELscuQgy0LZIZNMjVJRDHJ59dKKMVdkHIdRECTyQHDmjJzJukKxtAle6UsHMgezJMXm8JoylcFyPSz/FpiWSRz6hLWFFWxgHiMZLvfrH4OGf5pBQ09EltCiNiQqbFSaGcpRjK8n3CS+c8xCfVsxMIrvUJHwOkGEpRohcYnrPQEaZk3u1RLSInMiImxxXqf30OTMIUqiVZ5t5LSitFkpPDkhyNUOsJsgasadhc9M9Pz4hc9Q3zGc/+Tn+P3f33Oa77rh3nF29/C5oFTPHvmfr7tltezsVZDUK684lLObH6Bq668hN1nnuDbv/o499//KH5suO+euzixcTG/+9uf5Klnt3ndGy7n2MW3cvHNr2dy8Yh/+qPv4EDa4PAkIk3Nr95+mje96DC2hrVhxSc+dj/XX3cEe+Ik3eI4fnweu7MN9UlwG0i4n2zt+GlMnBCal1EtfoXkjqFhBOMFcW6oFjUyegBN65j5XVB/BS52xHQGO9jALTYhCsklOhcRNvDM0WFk4jrqhwWGDTFGPBNCCNm11czpk6CVZCXxXPDJE+M8D+BV2JSIGwzoiMzmiabxDBeaoTIPGrO9cGwV00ZGa2vMYiTogtRHApbFzODXD9FPL+CaCQ5PN1DCNPPqkxMGgxFd1+Gos0mh9rS9sj6yMG3xkyG16QoxpMYwRLuOYTXCiZCcI/VKiDNkLFirkL58u+DniZJVEe2xSwHLaj1RxISMn9sZSVqUDqUjls+RbsU13o+H7ldY5kcJ3xazUhNC3IcDa2F4WEDyzuA5zJBsGrQ3MMyc9ec+Ry5g2VdlD5M1ygpLTynto7fu4dT7zbVijHnculLn7n+bInYpAgu6UviJxiwsKoKi1TUphRXI/H7VYge8FBUV6ma5fpqyk53g8s2qDpN8mYfwnOuaos87KlVUFuW1eUQDsMh8/oKJ53N4LrtG9huAlWu1vAYisuKKGzWr49iCr0MiaUfS/DzoAhMGmFgsBUQRGkQr/rzb3KCFbZQpr9nFb0ZPS6wSXh0aDD7laEYbE8MAruuoQljNAqyv0KKjiF2Ps3kHYY0ldpFkchhMjm10KBXRthjTEdttggk4W5N2I6PYcOqSCWHa08qTcOYUj64f5+d/4z/wR7/0P/IzP/Q99GuHWGwHPv7Hb+eGK5S+9lx3xc28/a3Xsr7V88j8EGcffZRL1o5z8wsv5cjJc1xx4jL+1c9+F9V0yhc++QccOHGS0cwzUMOBnQfpNw3PPDnnR956DTfedB1HmyHeTJmMHZVESA3V8esJuzfSrV0M24+hF1pcI/T9UUL9RsKhSzFyjjC8mRSFFpC+JukEGTiSfQyCgSoQ+2dhvACTu/VFOpB3Yz5QITj10AIVGJ8YDs+RNIdkp35Gij1tH1FbsVChJ1CHCuebbNPhDdMU6UzE2Yoahy+U2K5f0MWW2HUYZ6F2tIsFximj9XU2py3awHC0QTOYYH2NqyHEKXWzgY+RWnrstsHWQ2o3w9PRLXqMVBjbg5sTXKQZeapF5Pj4EK6LRDfO7aMRXOwZG888tWyFHbbjlFYc1egIsa9IsSYt/sxt+5d+PE8KvJRBoxRvlfzdTFKwuXuLFdb4DBuYCmt95mCbqlDryh+95CJewkvLo0A42Ww8ey2r5HmfLn1Olqtl7piXhcgag7GeEELxoe8Lb7zGyCCf/aobLVzsPi3niJnamDL0IGXYV/h6+ZxiKhhyoT/GPFyJIZBCXAmmViOBJKV7ztfKYrOcylQ2AAAgAElEQVS3SxFDiZrsWhtB+7gaIpIKdq6KRVZF1yTN51Qwayg3RYqkmI2e0LTCTPJ1Kh08PaIu4/OFkrla9DQrPHP3Xa5nTKvXkPZ5oebrEPa44yKrIaRR8xyDiaQ5+9YhWa0aUw5/6AImTtHQol0i9QlSi9CtFoxygDLHyOcQQ/H/sT6ncBlD7B1KjZhIlCkDF1AWkDq8OHrrsE6wbmlUl0gY7PJ9S3m+kHM1B+RcXl9YNh1GhEUIiBN8Ei6c3uTss9tM1ivqSw6yOHeBx7a/k9sfWeOlt/4NbvqKl3PoolezfqDmsdN3cfXV69x048/xJw8mLpmscejkGv/sx95H0IoLd3+Ex59ckA6eYaN6jG/4ltdxx50P8Pf/4XvYGl/Nwdd/K0+d2eWe0wtec+MpPrwF3YUzXHtwxkf/9DHuvucu1i5yvOjll3Dz17ye0aWXg3wBu3kOV9+Bm9V0w7MwPgHTGj8Eu/sE8uwXcd0XMdOvxXnLkAUSFpjJiM54THcBTETiFvgHMZtPYbGkdoqtFxDGyPxKQn05ad4T6xdkJk46y0sOTtBwASNCGFT0NmHUUqvlgI+s09OJImlOawQxnsZZhnbEwDsWixm7IdJVKVtUNI5qVOchphq8jBDbsNV32LWGZANRZpgYGA/W6KJh2AzxRhFv2JkY5j5ywDUk2yC+wtBT2UhlLANTMTFDmuiZG9iiRwYVlhZvqqyU956FUUZVg6smrJkJTjqUBdYozijG/hny4V/68bwxG2OJ45YhYlYV5c5HCWC7bD5ESevRiBSu+p+HH60w2y/5OumSyVGwenLRkyXTsVAe9x9Ti5GXiEVSyFRI9Eu690x/BINz+beX8Es+iCDFiz0XNGU5Rlz6oSRhJXjaz94B9j1XxvgxefFYJiElybEnIpnWKalgz2n5+stiYwqksqQw8tzxiNCCSLZU2LezIfnShSaWtH4Rl+0QErAs9upykacv1ySbg+XBNdlATjMNdInPWzLDJ79FUp6z7OSsybsNyT+3fLHKgjzOztCLplS8+SOYVIawJitZl++jLmmXmZlji+Ctj4GQIjMSLg2J6unEEEKDawwNykgMvgLpA24JPWl2p7RiMQKhD7hBla+pFdou4Is3RW4wyuLl1pEwxySHiuIGhuFkgE62sfOO+x+6nt49Qd/ez9Onj3DsyEkeue/3cPWQ+czz5GNbvPFlhsn1L+c9H/gox+pj1M2Er3rDKwmP3M3NbzxMn3oOXXcdD955L0cvXWO6C7Jd87F3/T53n/t9vuZ64Rnd4Ae+8Qjf/bZbaY4abvvpX+PxM5tcc/G1tN6y+8gdyPGD+PYornkBMR0DjVT+lfT2XsIQTLgWg4H1Fl3s0ro/pRq8ENLDefedjtD0LciUWK1j2lNIehgdjYn9HAYzfBjRVnPS9KM0g2PoYA2JF1j0gaYacHh8nmM7a5yeKqZtscnQayQaIcQBvje4aod2XpSgoYOuxyRHkxRxDh8sKXXMrKElMEhCN+uyUaDvGLiKYeqJfYfYIZU1RC/szHYY+gFd10FlGdaOtu1ZsxPOz3fxFsTnhDGxnk5Dpka7hplm+wE1lpmHOlhql311KmvRCNs7ih0FtqVl4ocQd3GuQZMQ5Mvvv58nBV6J0hfhTSImxbksH0dSwUslc74LQyVqtizIYpJc+I2XUoRtjqMrtG1ZiWzqjONr8Z/BICm7Jy6Hs5pM6eY8MSimWHxiwBWVqaZlklLI/POCMavucayRhJjlILjME2IqDJ9S6GPK6tAS8WYLLhyNrAqrFtHQkuGC2myLoBFj7Yo3nq1hYqaPppqlBH7FUBKHqJBiym6NYlGz79yWw0zqvNClPZ/zPP/MkXzZF0YwMsg0yZiyc2UZrkqK+bovh9NkzHu1g4oJFV/emD1YavlIoahM055+wFI8hjShxQrBVsXnpU8ZgxePFLiIEmLiUspDZbGQlos8xcWz3DdLdbSxNH7IprHEvmFBTWo8fZoy9o7JvEXqjiSRUQvUmZYZi6pXRTHOETUv/gJY74pYzMJyOU8CYUHtJSdmGcvokgMkYzl/zznuu7/n4qufoa/G3PiWH+f7v+Md9Dd/Jdu7j/Jjf+ct/NJvfJg47TjQHOKD//Gj3PiqSzGLQxw49Dne9B1fSX/vgJG/n3/zvz3DS2/cYlod4zWvOsZoHDgfnuXYK1/MqSfnzM7t8PgzDY9/8E958O5Pc8Phno1TL+GN3/dK7nnPv+Wm1xwkrt1AfPQJBi/9GiR8BMNxVI/Q2UNY7ehmf8iguYV+cJi+OkwzVQbNEO3P0tse1RrbPQbNKaKeoJ6dQ+szaLeT/VqriwnTB0E8bhSQjU+iT36aOH4HnrM0w5PEboqvW64de862EW88MS4K+6rBm8TcWTzrTLmATeCsYWENxgS2ZoEWwboOnSuVE+bzlmgdMrHMNqfUdkgiMvQelxQnShcS3iiLNKP1BuccvhFSSNTNGA0d42FFsENiUqTKwj2ViqCO2vSsmzrvokWopMHagMViMfhkca6iNT1IRdUB1tNbi4tdVnWb/XvXv9rjeVLgQVIO2cjBDpY+FaaLacsfZCTazE5xtsIppHYXksOYRCrJOCmB8xFNmZUhkju6TOXzJF1k7F5Ded4BmgJR5wiKsVmYYpPH2iYPw5wBtWBS4V1nQywpGHEWIyWWRJCsNC0MFJEyXMsde46nkQy9LHcNsldoxRhsLO6DuqfMFc2ipqUBVx4quzKEbfPNFcv3XDkXbM6cxGMl5kEoeYBdZqy4wobJi5HkxYkEYrDi8vNrJGkOzTCmIvYdmK7API4lHhVijxQeS4bJcpebCmsmbxbM6vjLhWXp45MHpblzz9RDzf7txZog0w6r8julYNs828hsI4PGiNUG+rY0BkVYpZKZTQgpKbbYEHcxojJEg2XqOubthPmoAg1UMsabDmvOUnkYdQ4/iPRpB6dtZma5vAhL5SBoxoBNNqcTSXnRs8U0bel2aisSJSPX9ZjOsfXEFhsv/CauXruP0xu/wKPnvkj9we/jX/7k6/nDu0/z0svfxs78YehbrrzmGnaHwv/0v9zC+sWv4cwTt/OR33mK3/n5/4vN7YarThlefmuL7ys2z0358APbuLWL+IbveBtpG37/oQ+x+eQZfL/DsbWL2d16hM932/zAV50jfurnOHCJp5vVuPOfZuPaKzBrp2Axx45OEc/dhTv8U8iFH8G4b2dr+6dYP/hz+PTfE/UjhO5OXHwMO3ot8cL7cEeuJZ59mHp0Aqp1pD9DPzxMmj+NeqFqbmTBgoaDhPD9cGgXEzZo8fjUIa7GcJ5Doxl6usmsrV5xwx5Cm0ljtme23WGjpzdZsT0S2O0DyUcGxrFYeDoT6GbKhvXU1CzmMxisM64qtO057xOWPNeKRPxgzNhXjN0629MFxje4GpIuSANPq4pLHc1wTCLR9y0DCw6DdhZX5cB4aw3VLBGaxEANOE9oDGERqUY1NjlcZSBGfFCMTFCB2n75HfzzBoNfytdFipDIW0zVYaTKBTLV2ctlKbYhsvSOV5Us6NHEUlZvbINIdj5ElCgBbFuGrDXZ5yGB9ohEnEm58LP0eC+DxJSHi6ns+jUth4yZ07IMoMgfxfLYhMKZL3bEyZLi3lY9w0F5MVLVFeNEtaRDLbv2Mg6QVPJgtexQpFAr6TJtUYvYi4SxAZLL8ARKhkg6Ug+ashjJqM/+6ymRUr8qmPnf/HkWCmWeeoogmkO8NYYsWtI+0/32BW+bIj6OMaCaCCEQQiww1JcOo5emZLrve2UgW/Zm+Q3Y210Y+XO6/mIVYYxBosVoppbmBT9g9zVB+0V02YdI8VaImkjGE3VCkJ4YlJCENvSk1tAwzrMKG2lCwJrnxkqKaAmWyfYPYgyqMXfsJoucEoq4HJvYp93sBYSHWBNmc9rUsvPwnQxm17O7+wQfve02pvYgcdzynf/odfz4v3gvv/kr9zE4cgVPPPZZNmzHe399wWRwB34u3HjzCaannyaOT5Emp9gNwuBgNtTi6ElOf+4+rn75GxlfdR1f/MBvMdt5mDd/3dWwfo7R1hZvfJFyJg244A8xcUrSc7gjVyOTV7HLYyz8a4mLy5C1mjT9t6i7l7q5jPWD30jY/QXS+Q+j5j1YLN3gemz7EaqD1xAW29jJGslPiGGLVlq0bantOpWZMNWzNO5S5vEITt+K00sRhVpbUprl5s0p3gWSCYiJNAODSTXGQNcmfKuMBp5B43HRo2nCvHZUrqYajonJMEoVfTJIDLhxxYU4I9aCqw09gbkJjFWYqGFWCX7o6eYLvLUYH5BhC1XH7iKgIpjC6qtsQOMc1zhs3WC1YVJtUBuHDhq6oaf3hqqqsFoThw3BLKj6jkEt+FgapVTwiKYiVD3R9wTb/tVLank8Tzr4vH2PqqX4Vqj6TGdLxe+dCOLYs4rNHjRK9jRX6REzIMfwKaRZptQVXqkRS0rzvO2XLI7JBVNJ2csWKzYPBxM5n5MStYYl2cxWcUbz7qDwvkMImTdP2FOEpqrkmyacBU0dxpqcZCRK7BOU3FGxFkjZeCzlYycJmUOtudNPAks4LunSZlbzriJp3qEgpLT00+/yz6QKTR3WDohFQEQoQ09JWC+EPpTXmNDSfxc5LqQyI8CAyYZfRM22EikzZZAsdFry3zO2nd8fZ7MnTwgBWWoZyFCbMTl8JZ/okkuUYZiy2rAUdUFZCKEMMDN7haWSFMmCpqUmIWWuv/RuNYhQzV75khQnnmXCU15Pa6KpaaUlusO0EbSpUQHvK4JM8SxQN6PrDR01g9SDlUJKNYjVzLySbC63vAYxKbZy2RxPyPS9tBSXeUKYkxZw0bFL2O2+h531mwh//D8w/+Iu7zUXc7OfcOK113LrrR2b5iCzJ7b5wpOO7//p/5UnP/JPeOgOuO+upzmwseCqKy9BRgv685H6sovZOXeBY8nywPlnOfV138o3f+vfZmPg+aff/RJu/JpD/Iuf+BCDrcRPvu9vMphYGBm0G/HsH/97Dg4vwtY9YX4/w/ot0FvwPcy/DueeAn4c2gW4iAs3QFXT+/8ODT+Nca+D2Zw4eCsi76Z1c6wq1oyo176TuPnL9HaEj8ooHQataeafh+ZBkMeJ1YjkTiA7T6HSkUYW219gzY25EA4w7XPnvmjBN4qKEKKylVrCGKRfMIw156oZYbZARFkMEqPtFu8HbPVzqsYT5y0TDLu6yOLA2hGd5cBuoB5vsKln6X3HrFvQ+HEO8mhaJIypGgO6gTWCqBJmCyaTdRKwGc4zqtboZy0HhiM0JHoNVGZC0F3wghePdR6iJ2hPawLSWLwIrs321TH9/waiETC2bMkTKn3mj6cacXM01ShbhRlTIIoSEZfZHIXdUTp7Y+oytNujKqqCsYpGAXUFDlFSKNtmk422pMBEqg5NASs1Wnj6ltx5Lt0URTI2l2ETV9gyZfArWWFrUrYItpKHj/nJUsky3dsN6FLgsxzCLs3KVsHUz/3/1YBUQZMr6t5EShFrM2dcU8wpTykgBFLMAeXG9Pl6hZKWFClZqhTfFMowOK0StVQ8VkwOFYnZKTOHD8NqWC0GQzZcW56zpoyhK7q6JtkoTXDZPD7j+PlNyDsz/ZIuPd8hK6om5B1FKlh33glAonTNKZFZWZmXL+W5M4pmIAmhjysYKWJZ9LEwGyLqByTNPjImzaiDQcUQ4hRjOqzMsp4hBYz3xcyt3JvGIGUoDJk+GUKHWYaLoJBGIAFfz8H20EyYnY8EvZ0TL/sWfLqGf/1Df4NPvvcX+ef/87uYzXa583bP8Zs8p+1Zrr7yxTx61x3E+c380R33cHRwjnp4kGM6Zzo5w8GLr0e3zvNsGxlef4oX7dbc8fFP8LqbT3DRyeNw0ef48B+1TIYtf/sHj/Cff+njhGcf4Oobj7BxKHF0cJTZpUeYDE5ij1yEdE8hG47wjGDqU1A/TZ/OQv8swU0YVELiaupFRfJXEcMTaHMcuB3hEPVMaJtPYNzr0M0/gtRRbbyUtnsYb85jzNWk5hW0+ocM+y1is0ETDTE6sB3iPeoSB92AhROSqehapZUZfWeYdT2DHo6kit2dBT0VW+0Wa7bmgrT0NhKmc6JT5g7cXJiYmq2BspsUGwzjqmbLQd+1HBqvsdtvMlpfR7vEoUHDPPX0Knh7EMycXmEwjBAD1jVsTA7ThZaQhKqeYKwlGMl6FcANa+rYsZM8DWNSNyeNBhB6vLfYxhA1YTulK3Ui/VfAV54fBV4yuyJpNqgyVsinFvIfUY6HzowNiVmwkjLUkjCFpjhHig9I1IBInTHXVHBcLXBD8UKR0tUZa4mhJ4WESAVaEfuIq3PVy91mBSGhtoNYbAMMhLTAJkeImmcDpRiK9dnoyzhCStkaNnuDZQyf0okvz29pD1Dwfec8GvIikLILGGpzJ61Jc2GTPNC1YiHFXFCMWTFEMII12d+elLucvSDz4k8vWThlyMNNY4oBGCWWUKUMiyOSLDF1iGb2CerLjmTfXah7oR2GonaVuHLojLF8XyxRIzaWAG0xeT/Uh1VhXHbwmolJq0XDlEVRU8JYyQ6Z4hExxCQZuuoyjRKbsC5BMCsOv1GDBsWLpYs5xzeRPeKTjFhITauRKCBBGcsGB/UpHIGxDCBuUqvQMcN7Rwp9tsuIKd9TcRljmO/Z7GYpxS8+L9DGXSBEwYYjxDini+cZVgNq93me/vUb+cXfuYwjl/0MN3zlNte/9BW87z2GV3zF9Vx6w9Xcet0u2499lvs+8y7Wr7yUq90daEzsPDPj8lu+giuveTn+yV8m3fpTtO95B49/4aNceuwkconlse2Wxzbv565ffYgf+5kX8O3f+6P0sweZPv1e9IU3E565m9qNGZy6Ard2nG7jK6nbPqeBnTuHPfzVdPM/oY7HEfNFXLOB6T2dnMcNtlnoF6nsq9HpQwT3Cly4HQmJOPz7uOnfwRy4hq67FD96EXE+x08PkNZvQfhP2BaGgwGhugzsUQhbmDTPc6hujB0H/tqxwEOf3yG5EdouqMSSkqcaesysz9YfA49FCbFDU5fvASnpXSnRLwJRILULXJWtUKJaNmOgLoyXqSqTwZjOdOgwEk1iGJSZt4idMrAb1K5BZxFtHPVoxPZil6YZM0jgJSDG4p2jNdmuYhCFTg0b0WJ0zrwecqhRFnREPyR2Sl019D4hwezNpr7Mx/MDg1cF9SADFFe65Bkqs1WnatRlqb8uC7gQ6TN9UhJqMwNHYsjWnOJyMXepYOD5eTIxroRoqECft/VSfEKSLDBNRGNDClWJCgyoLSwW04MU06toENPjvOLqRDIRtQZCX3zfbXYwDMXSmBJytByg9nudanYjzGyMvJPY+z9VLaHTPWLA78tPiUTEkD30kxZ+vMtQi2r+F9C054GTvaYDrtgHZ7vgLPAqaRrZA15D9lKPHvospDIm57r2qSu7C3LHnLKAytii/xUh9SHji0u/m2J4FjQreXNqU+bZa4pFAZx3EFJ0Csv1Q0y5Zin/bEp9no1El/+wSdjC7XfeEIuqNUcqFuhMJLsISiCZHPGOE7re45o1dqgYeIut1vEiJNOh8gySeqxJaNrFaEsUqGyV8fV8mmVInVW2qYR5L71nnDHFkyaSVEipwdMQksOmOWlwEEjsPvsCbP0PufV73sif/P69PLn5OppwntPnPsWlJ6/kSNvx/vc/zSMXxkzHB3jkgnDqlr/HDd/071jsLJgf/zoe+OBt7Fz4Ak9+/HYac4Zv+iffyy1vej1PPf0IL7/uBm556St4+UsasBfziff+JJ/5wG9z1S3X8+JXj2B0EWH9CJw8iWsHuHQv+GdI69+Aeksbd/CykQfu6RLQS0lmFz+qQU7i5xuk2XmkEtzuBm39SqK5FxbvwtqjpJ1P4eQOdJFZWqnZwJkREs7S1Tv0sSHIlEYPkXRG8hmONTV01rK25piMhBg28YMhdthjHZAW9GmHTlsWcYdO5ogVrKuwMsaEigpDMgriEQupiWiXGDQCyTCpBjBbUKUWI3knOIgTqjQG9exUnmY4QGoYR0ewLaZSjBtAC+NoaFLZjduKziZi7NmwI0iGZKAmItbSa2Jo5ky3W2odZ9MyVbqU09kG3mE15fCiL/PxvOngk7a5yBhHShZTTi0PPnOHqlpyLA3FVtigahBp8yIhrmDuFnEJU0Km1XSZmphstgdWweALHJGVnJpAEIz6TEU0e54qWSWayHdTDyy77syuiX0xx5JcsHTlzVLYE0XcI64M30RJomSEYJnqs6wUCmJWrpbGLm0HbKZYJsmDVVGMDjK+TMjYPnlrZ7XJC8PS7gFyclMp+illm4FkYsmqXXYLeUeD6qpIUcQWWoacy07aiSPFJUiy761Mlqg93jiilLCOvscUrDxTMDXPECRf8wy7SEnpycfNGH2f0480U0qXvvViJHv3sCycrKikoe/zwEqW70Fm2MQUs39PBBGPaEfQAW1sWFQV271l7iq2LGxqxCTDej1mPS4YyC5GLuCxOCMYmaGmRtDMQlpqFjSHomRCUvZAMibvmDLbKs9OrFqolBjOoO4I9WwAk7fy7o9Zjr7+Zm7/xZ/g/77zW/ijX/44n3sk8Gu/+RN86zf+Y2698gjDtcPMT0euOTblyZlw/qnPkObnuOHGS7j83O3Mj32S4XDCYPyrjK4aUZ17hn/+65Gv/9F389I3fD3dQ+8mvlkYVKc4+ZpbOfORX+OJB7d59Lee4JpXtZy46lq0uomwRmFpvQp3/j30zYtw7YOYepfWHyL1v4OrbgYuZt4u6MKCoTyMTDxy5hH6I5/B9uexfAsMryeYx3G9ktI6fQ2ilyP9bcT4YWyY4OwhVK/HOgG2SQzBDBA1hCSoqdBJ5NUXNXziycS5NCPOJ3TdBdbNhO0qodZSpY4uRZyvkN4wSJZtTbTiWYTAZAizTcH4IYdGNXMTMX6HKYkDB5V2GsBMmBOpXbad9mKoo8e2hto0bFebjOs1Wo2MVGl8Q+8ibVpwoD5AkgofI6PBkD4khqZi3vaYZpIbIO+ZdonBxLGjLeNUU5G9p+YxswlNQQ++3MfzpIMv/+zDmXPYcw9pyXghCyeMgtYsAz2y3cDey8hK1uKDUqLcoODdNodIi0mlo87b+/QllyHz5M1qUQFwxhaBVVWe3yIaCGmK8xmOiBFMsvnYkl0Gl4KifE6y0hWtOvNlrGD5Wpd+PCsKYSTGwFLpu2SsiAiJKUnblXWykRojey6ZS18XI3vK1ee+0L1jSSnVK499kwNQll3/Mq5vv5WvqHzJAWXFhCkA+L55wXMxdV39v+wxocQWjr5Z/czy54hgYlbwZinynnvnyqEylTjAuE/dRWY+Pee1pohoQ0iKWEevPeIHYAeEkBiIMnBgwoJK51Ta40UhldwADDH2xBBW57hUxIrkucnqc92Lel8+fzQGFpbaH0BlhmsO8NiDgfrEIcbVQZov3sMHfn2L2z9yJ4d5mB/59u/jV3/v3axd/jSv/aY3kXYfwjYbzKZbdL7h2CXXMFkXNs9/jEODAdYPqCctRuY8cN+d/MQP/gDX3XiQ//DD38WfvPMH2Hz8cb7w4G088d7/g3Z3k0teOOHm1wnHLztMOKJ5N7KoMOkiND0DzSX4akqbLFvTG6l2ptj0Jnb7bUycMAy3sCabVJNXYRbfgIyuxM0W2Ok2wW2j4X7i1JDiLsY8Sm2uwPTvwtkxkfMox1h0u1i9m9h/HOw1wIlVwI83HmcHiPccOWl40dHckFQasW6Nroo4VUy3IKRApMf2kdROac15qmaBrXvWKku36EEDrluw2bWYhVCPa/zAQldR2zHqWw4Zx0gsFQ5cwchHFfNmgK/H2N7i2kiykWf7CzSmYsOsYzxoOo+rK7rQY42hJWKaCm8UE2ukEzbGA2yyNNExTS2hTnQaqakzPVy/VI3/V3s8Tzp4xVolRrIjoPOodlCKZVSTC7vk4Z6m4jljAAypc7nW+zy0EwIqzaq3VM2ddtb9ZMw6pZDxUesQXUbSZSaHFNoghZeuib1BmWYmB2Iw4omhcLqJWMmdtGiNSMoLlJbpJfl5lvU8W+AujcdKMZBc3FMf8nMXyX7uc4vzY1BWSlrpM6FQTaaMKiQx2QsmE+mJ5OtkNUMwGWZaEvDzYHJVhMv8Q1WJ/dJrxWbWTIwQcweuaLaTUMlDzCXbJeztRlKIxeNGMw++4Ol5YE2OUiuwWSyBGKolqqxAH97kwbV2hVlD6QVUEFdnUZlGkhiMKikKFo9omyGfMrzOFtDss0fQrISUQ8zCkGTXaGOFmhqtDxG0xqULGGY0saeOEUtHQ1YLx9IA5ASwvEOwlSDGEslxjdY7kmZWzX6bhCQJtYZoAyZYxAx59PZPcU+6BXF38x9/+Oe55mWvZKgf45/9p/fzT1/57Xzzj7+YD/7LX+GzH9th4+BvMDnQ8aZ/9F5uvf/fUR33nPv0p9G1Fu8a6gMV87XI+AV/Cz1zPw9/KvHY8KOEB96Ddxdz9PCQjVMb1E/fjnnhpdhDl2PrUxwY/y569NW46lZ6s43ay/Czd4F7A311Bc4eYySfYTZ0zNqLaMQz3pmi9SbBP4Dr7yUuDhBtTSUX0dVTfPXqbKLXPUNt3sBO83tMzhfKcHwRKZykkorgbqOW47T+KRw30fYfxElAzAZogNhjaBG5irD2JCevOMkVW0/zEAlNC3TumKqgrqYJiSYp27bF+pqowjg4pmGb2hgWbOHdkEVtGC7mQILgcWJJGvE1ECsWNVgX89+bMRif68eAiDFKIiDNgCoKk8lhpAepItMAw+HRFXQpxrARXakJHrUphwZJhbWJ0HcMtUZ7wapBqnzv4y2eL22g/vKP50cHjwI5Fg/yQBTGxbArgZln5guF3iZ9TmFSQZNj5YioeZucZawGZKm6zI6SkiwSbV44li+9DA5XvOzCq7bWli45F0Xnir2w7HyW/6MAACAASURBVILdAZmTNKtvV4ZfzhZqZi4kMVhizIuRMak4T+5ZA6T43GCLpSGZsdmPR8hWxilkdoyqrhwRMywxyLsJLfi69qRQbtqC4VsyXJFIJKOlKP45W7/SBcd9HXwONSn5skUIIAUO2T8Aek5nTvGbWdEeC4ije8ZqGWNfvu/5YUhIwfLzMNUUAdTeTb4MIVntIPYJpUwxUEayaCvbK2TTtOxMmUmgqGDEEdOINtX0oswQWmtY2BHTNCfaKZUHowZr5/kaxhrLrBx/wNLp1FiL83k2hORglBxb8GfzNJevy/YN1iV6XeCnz3Lsr/+3HGm/wPEXvBRTbfKWrwpce+NXcdvPfh/92g5/+r4Zr33NYd78Hd/JpJtx5VXrfPadP4A9fIz73/+/s/GCC5jxJmYYoU6s6YD28Y8Rnn0Ac/RS3v+J+3nnbYYHPvcE17/i7Qziwxwc12w99Rh3vOs30Sf/Pf0LvgE9/FPE+hjeX4qvGsLwa+kHI9R2dHoX/dophlWHdV9AjaUbG1L7hfy3pCcIzVVUcQvsNnb3ArAg9h/GpKfoRj2TrcfpRteAvR3cA8RwAZo5fboI8TXVrMUuHqP2V4JYkulItidRojSlxbVHMKOWGw+MOOFmTPoBXqCuPcaWnW0KTMyIReiZ989y1j6DlwW2b6krR7TCOBo6L/QKYwGzmCI4+oWyxpy59PSpo24yLKOpIbia1hl6BlSjAzRmDUYVqZ9jbHaTHTiLiBLEZ9dIO6CLhmlJWMMYgihtl/MRqBzJWRp8hjFLs9n3PV348nnwz5MCH0myIJodjF3CAaHQGT02jUF7Ym5UMa5kfIpdKU9VAoZYQhzyUVU8igNpMVbppc9iE8lvhpYh3IrBkgSJIL2QouTAjpTVoYopi0yVP9KeK2UuuoYYNAeB6DKzdEmnLIrXZDE6yOeeJCM5oiU1XkqTarK8ORT2SdmpGEsJ8pACueTwaE1hiUQUy17PqvaKIlYIMe2DDfZlnBpB+2UQeCmahSuenyt/TRFj5HSkUOYEgoaITZINz+LesSUlcqaslAFqtoBI+5woY0wQNVMXVTPDxniMMdnYLRQoZBWUUuygMw8nWykkwUiNNVUe1gJGbcnGXa1Z+fv7PovBoC4SqgFzaqb2APPqGLvmMCITBjpkLXo2NOL7HmN6nFngEJLxmBDQUC6ySSRi1m6k3ExkyDDHupnKr1TKqQyWxc5gZqgk0U2OceEj9/ChLcttt32MYbXF5/7k00xGx3j4rs/ztm97CX/re69mN/wpD3/uNi55QeC6l13MoROf4IF3/k2uvMhw7p4PsjGoqA/OsU1FWh8gzcMYv8kX73wPVx+1fONrXsErbphw56cfoZOL6d1hjv+1w7zmzdegx16LfeJBuvmHMN0TRKmQ7fsgLEjdIYLtMZzEzz7E9Nwf44YNMv9DTHU5UnW4nYdI6QJ1t0kwpwiLy0jDy0h9h8THaNc+j589TFw7jLUtJvW4uI4ZnyPETQbu5aQU6asrcpTW4kGSnaNmSLAjjF6EoULNBZK9FBOOMHjhLm+84SA0SlcrYTonxRmzOjBD6WSOtcJFo+OkuWPgPQvT5sB0E1AfSdIyXneoh+FwCK6jGdgc/KGG2tSQhESPd3AkwMjUDG2NTT02Laj8iMo0uHqM1BPWBwewndIo2GGDdYoZCLUV8BZJwtB4fFK8WqrO4qOhE6FLSuoS0gZqU6Hmy7cLfn5ANIAxCwxD1HSlaxuSMfJZCamwWUEYc3ExzpeQi4TIIFMfC0eG5dY9RqyzpYC02To2BfJwNiESQGKez5JtZcXEYg9QfFS0zXx5zbsIJBZpPBkq2YeT7fmNQzYdy4ygmAJJA840oHlYaIXCg3f0fY+1grWGGDus8ysqJJDFTqvz1tzpkgtqdjkp17DAI8uGemlqJoVauZxVLOGYJd/diNnbRZgmUw8J5TiZFrp/PiIKUVM5r1QgraUcP2ZPnT6w8n7Rpc/N3u5hyfWXAq8oy53MntYhn4/Jxmyr38viK+cMMQaiskpNeo4bpXF7v1PSqPLuxaDGErWBIFQ2M676JCSdodKgLiEyozJKY4Q6lbSpPC3PpnHWkkiYlMr7mJ/DGMmigmJjEfsc76hloJwFLA5T9cQ4QEYT0vhmvvtH/gFu8w+ZXHIZv/sLH+KR+97NN739zRw+9AI03MvJo3czGiwY+iOk8w+xceglbNefJ5nAgclRTLNLZeZQT1jMDebKt2Kbz3LohYd5w5v/LpvpAv7OFpmeoTv9AI0Z4puedMn1WDmBvfgm/Ok/wK7/A2K4B/RaHKdxnGV3e4x1hk6vZqSP0C9O0LmTDGdr9O4YUV6IMQdRuwnts4g/hzU35nSu4bXEnW36+m78rkUmLyG4C1jdYSGPUpt1knbQ7mKbKYRTJP8sDoeKw5iW1MdsNeJaIucwpXky9YAJT9IPD5B0AvMp1dzijedM3CJ4z/mdbfy4Zn5hznC4zpn5nMHA4rrImo5Iiw5kQnAO13YENfjDB9nYmdKLZZEC47UJLBYsmgFqIo5demnwtcf2AVc3bMY5Qz9ke3ubejzEOcGkwCL1DKoBQ+tIpmNBYBoDpmrwxqIWNCW60OKMpZaKbRuzduXLD3R6fhR41Yj2iyxqknWwMXOvtSGFQd4Wu0jsA7YUcU35D824iph2iq1shWiNUuci6or6klwojOuI7QIjuVuUKGRQvEfMomy7HSIVKj0qtnTECWMqioNAOV7Gn4EyLMycclUllaHm0tvd2px/GkMoMny7YqloMnjbAB2h77BFOr9U0GYGSBngQe6gi43NSi2/5Nab5ZfL/8gF1pjMdZd9HbRq7uqN7A1ORSEyz3AWDZAyBkrILKPlTkcE5ywxlCJOjtxLKeGz1ClziGJx0yy0SKPsvR8r9eq+RaOoapeUylXCFft2WeURu4g4ixWKXXIWnEkquwHNDJ4sLUirRTCpEE3DIhqmVUPnRuzaAZsRZoOKSI8n0ahShylVmlH5hI8zVFuMyWwfbBFPLaMl6bLtdBJUDcakwoDK5yCmcP1FsNaD6zFhTt/vsnHiLD/79lv41u//b3jsD36JV112kO3ZLu3pe5lPTrP44ic58NVv59smn2N636cZHb0GM/sYl1+xhhwewuIM8wE0wytgeIimu5/u/+Hu3WIly877vt/3rbX23nU5l773TA/nwhnO8DLiUNRIJERLtkzZkRwrZhQpkRVYRuxESJAXP+clCOCHOC8GBCSBnRCGISS2TDuKE1lCIDmWxIgUGVIiRWk4vHPu0zM93X36nFNVe6/Ll4e1qs4ZkgBlMyAIF3DQfarqVO2q2vWtb/2//2X6JPQnfOhvXUbtN7hw9BpHN57BwjFz55DlgjTcJS1/GL/4fWK6jF77z4n6dWSMZH2Gk3gfy/lTdF1BX/9N8uENbOkR9yVCVFa6wa+U7vD9xHsfbjYF62qWdfpJ7KDH3834+Z8h+RvI+BGK3cTbc9A/Qb/5AK7bw8Z/hu8uwPoK0/xJ/PQJkhzh5a14SRC+RI730NGgTySUEO4DOebPPXyRT3z1mGc2AS+ARqZyzEAgkYm9EFOiD0JvxiUZGAF8JRaUwbG/zqRcKPN6zuV0j3udZ+E6Ftmj04YohZkzJhFGt2DPzetn7B2qcHW+h89KOOw4KZGJDofjYr8g58wk1WOms0wIoYaDW4E1FC9IVnI/MA0DfvUGqCOHf0sw+NpLTtXD3I4odreKFCw2PrhhuNoRa41GE/WNdVEQBsoOLimoy3hf465UFdcNddC2lYhbDYxOrM/gCilnHaRow7crjlq52s2zZRuo0TjjZ4yVsxR0JdY5QYNnSnakqVL0soHimhpfcDqR47qapOkMSr/Dn3M1kKnYLWc2wtuh5u6523W7YJFvuM8Wp96KkHZ/8w2ffu2yB2pc4BorYx1OM9RP6Ty2XuBN3XgL5Ng+h21tBKR6bMg5Ro2yHR43zH27aNrWvvnNePvuOewclNQ8eUoppNSUys3AzJ1fOKSeL3k7my5CKgFxczYpcBIdqzRA2GPhBrqUWZowxMSeEwat/iW0GQgJcLrbDZX2wBViU7JVN9JYMqnE3cJ0nk1EPMKYk8XTu1Nefu4VPvTX/jM+9pnPcHTsufXqK1y5ENlf7GG3PsWFBx5l9dn/heVVWN7o8Y9cwF19B8OioJs7OOnx7hJglINHYfluyvwy+fBBtLvMV3/rI3zsV/8Baf4IFx76Lzg5PaKsb+EfezcavoBP34/fvB+3uk05HvDddaLdz8HeE5S0wd38DOPFRwmnKyTMMR4h+MeYuzuEwx8ipc8Shj1s7xozLmDdj1D2PoUfX4S9/wBbfxQ5/iOmfkPWz2P5EdKd24h8nVg+h+r9RNsHfxvNnwI9RvViZYiVe4BSzNW0M101keEpa7vHlWuBq8uB/fkeQQNdqP7sog7vetTvE3SPYbaHieD0FMpIscjefo+Yctc7ThczinSIBZJ6DrziXUZ7R9KA+ks4jLl3dF3CuzVeYSaekhJTyUQKzoRBPaETvDMSG6KtKGEDFJyrViCnp6e11MyUrMrQ7yGpWh7MWdCXns6+C5F935VLW8nMrSF1qBvB3wTpUXcVI0CxCmVQmuugVCaLC5RUuem10HgszdDeIT4jqqRpwokny4S4UpkrRaH0oBtECjlpLRpaB2XqE6hQEq37NkQn1HHGReeMClcLUQ27toa5SxssVvVlhRdMCttoQrSGQouvTo0xRXwXYALnq0maSYUyKs2vDk/NtmySs0Hj+ctZDqpV18p6oA0ysR3cQt6RU84KUdl28M21kVS9dEwoO9sEgyaOqjj7GSVyy7LZPmdNedIdbHT+/apFr/LyyfVztWx12CSVsSMkqmHaWRZsEUN85fe70LVQk2Y1YVSaqlXXyKqqrVz1ApjO2HDAGmFyA6mbM+IYBZDC4D09a0TeQDQT8gaTDc4yWcFZqRZA7X1znYLWXZd3nkImpQ2i1W8nW8L58KaFNbuAy6eYBKLc4MLRy/zWx36Jv/7hn+XlX/5dLvyZJ/CPP4Z77mvY7GlII/29R+D4VRYXL5LfeAVLG9Kwj7cI8wP04Drlwrtw62cp8Q4crbnz8Ze4FW/wxPe/n6vzeyyXr2IHz7F491+CdJN09G6kmzGWqxBu0feH9NNLRN7J4G/AFPDpWexA6fNDsEhVBaKvkGfvwU6ewtInmORxNvH3mPu/CjngVr+P5+cpw9vR+EtY/xQlDPj476FlxjS/S6cbsjvFxzeYfEHy/eAP8KsNefEAOd7GaQc2glN8r5BA8xuIzskx0wWlLCNPvyPw6X/5AtZf4DRvKDKAHhNMSJropHAcVyyHGSMdXgrBHHdXI/18QZzWzGPNWR4H5VCFySY6XSA5s98v2KQNawcLHD4NRC90s451geA75hbw6jGnxJzZKzM6HKMqulwQsqJuZDRjGPZQeixJhQFNKSWCZJwK96jZwv3/D6X1e6KDrz18RnL1z7YyQYSSRoq9gXEXMyMEjzgln2NfVB61bwKYVIuwTFWuT8EsVfGEGiILVAac+hqbFaRx7IUtL15wYF0VGVgVz9RowLEyJIT2uFu7gOb9vrVdpDTV7dYu+JxTJGdMmuqlU7n6KVaBlO+btQA1JclK9W6vuwZjZ098vpM//y7K+c717PpdV21vZr+cuWCePZbQIaKNXz8iuB23frsL2D1X3mbPyu55to/jXFPQwk4IVG/Ts+GnNL1B8/fZdvO7bp0tq+rNiyhU2upu5oG1gnrOf0jOzykEZzWXNYswiVJsQQ6XWctFrL8IfsYmGX0oSDxlj45uykxxA+TdEFecNI/5ZpuA7URpkHChDsS9rzvM7W5MdtCTkb2vhccOCKVwcnvkR/7Ck9z9quP+n/o5Ztfuw6URu7KAQUn7S2wZ2HhIQ4eb6kLjB6P4GdlepSwOcNMXyeqZrj7O8MoFLr/l3Vzb+1lOXn0Wd+0pNC1In/0XxPUJXJlT/EO42Qfo80DPBUiHHE3XCNMdjsqvsZHbJP8KZfYYG/8SiVN0+jhsbuLu/Ca2fBJffgiZdWj/k2h5ALVPYUNm8scIf4zxCJlHcAWK3wDH6NGzkO/h4got0KU9vB4Q03WY34/FRKcdKhcwPDltwAJFX0M0UnImOMVZM4SYDSwXh5TsWPR79DGgfs7pOjOSiZKJs4HXx4l1cQSvFBWcDugmslwu8Z1nsozkwuQFZUKDgi9sbET2RmZ+hgYlzUdUQUqm9z2DeZw4pA+EIhxKR5pO8L0y7wJpPTKVQnRzBl0y3tnQF0cQZSyekuv5cjDsMSs9vc1xNjC5+betnN/u8r1R4AXM+/rlydTsxpSxMWKbe5DugtwhhYkiEQnVbkB6Q3wCbViwhcrYkIJKHdZarjYFJRnCxNYyoDahqaa/SMbpDCueXDYgEwU7J7qpP06WlQsuEXW5hjs0QUIN5NgWtfqYqGDnBpigWFKUyut2qXmUNCpnnkoLEBmrj4n2jeIpIIWUJ9Q8ar6yOKxK96sSt3axsoVpUmWokI0KIEkzxTqzHq7/2u5xrNEjq4ioDgopCcpUF0SpXi6uKBbjmRAqR7ZynlIq1l9SezyR1u3XxW7rlw9NFJYLmiuVswYmbBeJUncdpWvQUd3lZKsMKodrn6/VYVRuGbONJivFt+cWSIqkwqbrWUkVzp2UwIlfsNKBrD04z4FzHE7HXAmZGUcE2bDwmblNlEanzRS8KIjUHV6jrlrFiCpv2pS8VQJTQALgq5BYhQ6Pd4eU8jLrQ+Whf/8BbPUM/+y/+wjTydcgvoRaIk+KSIa4QmKkk64unLMVcbaHhQuUwznx8FHgLqV/C3bpg4Sba37v4w/zkf/5r/DMauSlI+ONO89iS8Etb+IfvIK7fYvOblM2/y0cPgn0lHCToXsGZMagj+Psq4T+EUr+l2hIjHId83+D2B9yOjxNSIbMM335AF5OyPY5JpYEeQDnXibJiHCN4l4nxVfw4onlZdQ9BPYEMe6BfydRbpDjEZ5HQJ4A2we5HxGP8/cqZbkURC/X76W7S7ablUHdZbI/4V2XhEO5TUp3yP0RnK7ZHwacOcx1LDLMvWfRz5jpjAFlP8CCGrRTUqZ3DqfGUmDPX0WcIZ1DvYe4QEshJWOue9Ap1ilIIQ6ejYdsGekcMnh8P7AqIyurzrAzMnOfiLahW3ZspAAdLo+MXRXHlVLImlGX6V1mltN3XFq/Nwq8OSz3mPVgATFXWSsyQV5jcQMl4k2ruZYbyG4gR08Zt1QiQcTXuDqRhh1vO3tD/FnhqKjFdstsDXMG5zvUKkVKcM3LPdRjKw6j4mhkD7lDrWt4vINSMX4r3a7Qb3NSoQ6SpUnot5144cy3hZLwzu3UkXWlO+Pn12OVXRe9NQ0zq/gz+UzZucO0W2ftGoyyVbIK2+79fA5t0xLk2q06BC/aOuxzuL2VHQPmjK5zhslvmTLnsXPZzhzaJien6pGzfT1bT/3zqtSy1TY1GGeLcxsZt03fajAYtPe5yG7xoMFZokamxp+FaU6IgXvSYcsZk4PUtfQvLwxU58gQwSVHkOrZIyRki7uLq5+ayNmMBd92Y2/G/uvx+jZH2n5ZlewjKR2jh4fMTo/I773E9bd2/Nm/9B9x99XPgrvOsRR0BmYOzQKWqIHzK5IEwhDAe9StCcERuvsx7Sknn8ItD5AXPs5TT/09Xvr0/8tDV5Zcfe9fptx6Bllcx27/AVz9GRIvo/YLEB9gLNfRuxmf30/KN0Au4fobrMcNGp6GaaCXFd4+i19FZDwiptvVTfHoF2F4DHfvJbr+z1LWr5L1LxLcW0nzd1DSHbruPXD8aZw8TvEbJv0abjYDdwNlgR8GcBOZU7xbEPNdyF0Nds9LYKjVKl1Gy2WEg2YLXX2C7rvgmekGx4DkGf2BZ6UjWgJjNkaNDC6haUW2RHZCVE/qOkadkCAUVZIpSQdWuqne/SHQO0+aOfpuSfA9Oa8ZnEM2Rl88ocGgzjlGy4wOUGMYlnQm7M064lB3cp33OFE6pxTLlKCYU6ZpxVhOMfEQDSWcncvfweV7A4NXKH7aiXIq5a2yVhxVWpxpIQk+oK7H9gKyNmTekdcZkQiyAfEU7WrHakYtb13tSuW0WuiSdsO+JOBctQLIJWOqOO2heIKfY7ZGfcGigPXN5mDd5O8ep64GRrfQbyOR85ks/k0DNitnHi9bGmajP6qvrpZOa2hIoXnWS+Vo1NCPKqwBmofNNr5PUd8cDMvWNx9oCUnnoQs1SFbNGbZeO7uPoc2Y7ZyQqboa107Zdotig0mmWIteEyN5DXWXsdsFNGGWVIOwis/XRWIbeFLzVetOyUlLatrm8e6QuMrMUUfbCQjWcO1tAlhJsZrINYqRWWVbldxMxYrndbfPGBwSD7hne6wYwC1AHGEcGSjsuRO68ZilHuNtjSsrCqn6+lsGfDV1aIuZSEBQXGekmColsqrUKBjmMkKPUZ0vxRkqHbbnKVdukLyjm27x0vPwOx/7FX7s332Ey++7zMwf4la/A2mBjVOlnrYwEWRBdBO5nxiW7yfNhFIipVe6Vz9LurPgofdMfP6l9/OB9xizpwemzVeQBx5FrjyES0vKyRrPY9jgYfVhOj2hdE/guo6NfaUmVo1K7zqUfSZ3TLz3Gn4/IuE6w/pBirsD7grr+V9jlu4nL2bE00/iFuDTVxg10Z+8AovCdPwlbP8pQr6JH+5jHE8xf4FcbqLmSHkflQ2UexiXCLoC+RLkjLkOkYzZAbh5pfICxk0036HkiauHSx69dsBLz4e66zn1zNSxsSOGIXCyHvDm6ftMmgJYxjNhFnHFYV2HHzPzYYaXzEaUjjl7oqCZy2VG5JTgA4QFsQj9ogftEIylD0hxDMFj44ToPmPMKCdImdPlQ9SXqtkoGacJ9ZCjx08ZQk8xITLCQlmlNd/Egvg3K63fCxdF/YC5OkAzl3BdqFJwS+RpRKcVNp0gJ3cpnEJeIVf2KPMOnFbbYHwtunhy9k1YEkHWICd1eLotdltMHMNipVM63yPSUyiIpupoSa63OUfhlBzBcuXMmx5jNp3RPuSs0wV2YqGaKdpeqWoVKom0uL1G+7S6tacdmrRutK4bzeNly65pQSjVIbNx93OEXCp0Qe3Jd7bC5xeZLeNGFUpszJtKM7GSKTlRSj4TKnGG4dexvzXvG9lBIBUaSTXVSJofTClQqmDp/POrnjFQtqff9tdqJ/zNrJOtX82W6dRGHbC9b66ewiVX7UPOlSpJYyoVAicaKN6Ry5yjxcBa5+B7XPB0ZWLBmgt2zL7dY9+d0uV7hHSKSwlnpTl0Ss1pMUMC1RZCCuriDkc9LyhT5yr011xP1VulUpJQPyD6Bl15A/Iet0j89H/6A4z5WXTzGnn6Q2S6jG1WEDd1YC0ecz10AvM9uuVbscXDFK6Q+45S7mKHP4HceJhXnoVnPvUq8/7LjF//E7pXnoVLT5LlIay8iC0HcNeR8XVwc1KZg3wGs68w2FW8vx/Hl7B0k8ScyT6OLq8Q5fsw5kj6ZVz5Yya+wCw9S5Zfx/XG0L2V4H8CGV9Duw5LHifvogtP0funWZuDmOjDHLX7ce69FHcMeYPYgLOr5PJVCFeq2NB5kL7+X6v4rRSHyR64JTCr8F2XuX4ww/uJMCijjyQ3EbqBnA3pPCuXmbIj+jXMEjGOJBzz2ZJOHNOgrGaOI52YuSWikVRGVuqZVLgXjGnm8ZOy5zokT2RXh8AmHpUmRGRB8dCJw3EBdEnxJ2CJlKYK27qObI7oCyJ7pNKDViZcGANDGvD5m9XQ//qV9dtcRGQQkU+KyGdF5E9E5L9p1z8iIp8QkS+JyK+ISNeu79vvX263P/xtj0IEaZ051kOZ1SGfTqgTnNf6JY65KlljrAyUmGDocc7VqDQ6ZNu9ypamV2GTHLtzEEe9i5lVP3RXU4isjIhEtmZXxVLrmDOF4xbakZDWRWJLpimRU1VrCl0bSJ4pFLbFqoZe5NYmF7bDQhXfCjdnAzlxDduv1glbQ7ItNLPd5aij4b81oLpa/rpdEd0WVBFpY8IKC235+ca2+MquWxaRM+HQjgL6DR+XncE/5ws35xYUoD13/Rvnwtnrgl2nvn3NZ7ucdiDlzRTP8zuiUsrOdEwaLbTaEveo679p+IweILaHyzMoS04ZmHpH8comRbqU2XOOIZ8QMvic8Di6phPA6uJRE6hsN2jfQkbFKn9/N4htC2l1+qwZrSoduQl0WPRsOkG8J9mCON7hxV+7y5dfeJivf96YeqVnH9JxtdYw35IJDfFQuoTz+4g/RmxGyIrb3KLPp2Cfx55/lWfdj3L1yfezWr+IyRIuXcHSQ+jpvwJ7GOKPkvrnyPkFNtMRYfZWTB5hlYQiCTa/RdR9MjMsfpnl5gWCXxMmxW2ep/SPk/p3EqY3qj+7HsA4EGd7pPVH2RxcJ9iPkw7vQ+JbWes9bP01FroAEYzDusOcVpQJfH8ZSRuKfRHxM8gTOY9Us0GH6VGDP4WadbuPcEjhLuIcxsRyJsS4ruefFLwqJU90TtExEkJgIYkej051xzmfD+R0yuAd13XB4UZYyoxUTgkMqAwsHfhSuC9cwq+EEgrZZfquw2etDYoa3vUkVfKgBA1ImGCZWDMx6JLBAstuTu8Ckidcmei1R7pT1K9ITrDeE+NUv8/uu8ODH4E/b2ZPAe8BfkJE3g/8HeDvmtnbgDvA32z3/5vAHTN7DPi77X7f5iJkmZPdDAsCnSKuR7SjiCc33BMp5HFFnu4i4x1EJ8RNsLdEZzMyhaYwAqq3chXLRNTFGtaRE0IkSya1LM4iA8X1mA4UFUwTSB2uWCu4VjzJSvV1afi5lUTf9627zBSp3vTbt3XbxW3/LzhKsgodbDtxSdBwRKGqQ43Ydg4NSrFMjbJ2WDrHP08t3NtCZRnkmh4VYyQ1UVUcK3XQm0OzIangCrhSnOW//QAAIABJREFU+fiCYVaDNqwNYq1s6ZiNzbI7z+qQVlryUtmyiYqwDWKSNjSuRul1PmGWyDm+iQ1zxp6pENDWLgGofjzim1Cp+vhsk6Ccc4gXio44n+oilUeQRMyp+ndsYZus5KKciHGv6zhyPfcKpH5JNo/5mtp06E5YlpvMZMWMY+Z2D2dH5HxK1rjrHLVvSVnSdjVS6s5RO0rR2p27SpuUFv9Yzeb6ughYRNyGfHjIcLCE/AB+6NEv3ELuu8Rv/KPf4ckbwPFtcIGS7iAkcgKdz9DZJcqlh5Dl9yHDFVI3EOVrSP9ltNxmQ0ZeHLj38kWyPsbrf/QrHF57C/2Fu6wu/CAlvgazn0f6HyPnU9Q9RWJBmB8ypRdwBGam2PRlCE8QMnRuhOl/JXU3oPTE+Pegv4B1CTkeGaf3om6GL57iVoR8gHQLhuk2U/wYIR2Cf4NZeYDY3arjKn+JxMVqgUFE5++AdKUqiOkpZSAXRbvLFPbqZyzXUe6v5mVxAu0ochXcPqih0bGHp7MNZqWmhWWj75aoznFdh8+FPDbxnk3E2cCY6gxDfOK2HjMNkLXDdcamgnP4sIfNlHVaE3yF4bQPTYjombueIEKySHFGJ4XglZB7FtOSIQ0YnmkQNkNhzUSQnuCHWtDpCLrAj4ZPQvCKiuHsuzBktXo5ab+G9mPAnwf+abv+HwIfav//K+132u0flG9qqb7hIoYPFXt2fokREL9EZFbTetQ3I7KKy0ou5JgoeazOid0IC4MQqyd7dm+yEK7doWKcKS93SUo5QYoQDckFSYKkc2rK3dAu77rWs2Ga1bxRkQoP5YZH74aONDz4DKIBdvS/7aDzfGe67dq/8bI1AQN2hmiisTGDYvsZURfx3hNCVeB673eU0fMUxzcPQc+68vNiqPP3PRvInnXtTlyFroQmQvFYqXmtplXopeZQm1c2DA6V3BhG9eTV7TwZGl4vrWjXFaMukLornDXMui4CNZhEsKyQPQGtC1mpJ7bbCqqSo8sDa+tJB1dYx56s+7iWYh/MCDnS50gnI8EyWjJBGjSjHpqwzjiDy2ALLxXMItvB79mb5zA6TGsAe3FzrJvhygGlLEjxRSzvY88f8dIX7nDp8jtYPvmDpMUh+fQOGg8pMhFCIOpI9gmGx7HZEunmZLmMlK+Q4im5+3n6/l3Y5QN+9aOv8sb6izzyzsji4IBSrhM6QzsjnryOZiXzG9jq0+TZHm66g3RLcngMK0oOJ1g5Iptg+SbkH0dtHx0/Teiu4/OXKOU62vV0epNxOqFwAdxV8nQHFwdE9yE8hzEj52MSf4jTnsIFLO7hdEma3iDyLGpz0BXiNojMwRvqUoWCtENsQO0AdA66Rv1IKUJJPSqXSeqhj0xi7MWePQlcDDPEB9YBkk1IziSnyIWuUhpnc+a9J/oZ2RacnAomA3MdcNkgLJDBIUOla3fFsegGfDfUuhCNmCc0+BoiUxLWBwJtt5agqKcEQ7uE5lM0G34V6U2JUjMV0mqDqMcZyBDaOeZIButv0Lf8m1z+VBi8iDgR+QzwGvCbwFeAu2a7JeZF4Eb7/w3gBYB2+xFw6ds8A+iiYmrdgIUe01JdBLSmAGG+YdKKRoduQMcEm5GMkn1Al57kImjBSoeZYyslryyLsTFWDCeKF8ghY14wX8kxeEeuMTG1A5O2sDC0LXfZVSRtcEZliJTGlqhRcmwVuFYHoyJnLI+cm/GYdK2gni/02+5Pz3yhGy3QcbYYSBHiZJSsTbRVfyyf+ZNv049qtF1bPLawx1YYtm1JqV2p37oglrMojy3Ms4W4tji45K1mtF23PaVc/clWFybBKnZdEpSaSE/qa+dPFZiZtmWvsW5wjX1u6c1FU6uts6qvr0etBllLrINpKdWsLGciHVn2WAXltq+iqBILOhSKbNiPifv8ERe5y/54ypDvEOItAiucZoyJQqbQAt6loM5wnatK6nOfB64OuqXxVE2rNbAGrepr16EXBtzV+7FOiZ/4OnH8RfLFn+Tv/PeR+3/wh/EnX2VegJRRKcBIHi6SLjyI7j+AXnySYp8h+a+Q7fPoeAuJizrjmd2P3P4oH/nbAy78JD/9vnv8xV/4abh4jNu/jNp1XHqAcPFpUr9AeQr19xHoyP2HCOkAdMT8y7juEok7lNkciW/BLX+KYkeU6ZPYXSF2D5DjXdRdpoRMPztEx89T/HtwHBOHt0GGLr+P4h+EvKYMA45HQQec20Pj84Ruifdvp5hieUEq82oPYhchXcXJVcQeIjOjaF0ojUi2E1Rew+ltzK7i5QraBYY56N6M2+PIScl4c/hk5FAYBs++BNhERt8zbgQZC8vNxCycsDfAZbeADBf6AT0tzBEGMZwXuhBYOcVCh0rAnNLPF8ytGg92szlBlD3niaQaG+GFFA0nAwwHWKdo7yhEusGxySPzWQ8WWRMhZlys5oGK/OmK87e5/KlYNFZB5feIyCHwq8A7vtXd2r/fqlv/pqVIRH4R+EWABx9oW1jvSbngwx55AnEZLYnCVP3bc0WRdcvdTtQCrIDUHYB2Shlr7JvK1n2wJSC1DrUOB3Nlb1jlidfw7soUqeyS0uCTam8gUgdd1dCM+vdl6yR5/oVWhgi5IK6xXEqqnjQSasEXzy5FSc5w+q0DZU0BOvc2SivQ3/BGuoY3W3sNWGYXlC3NB8eHnSfM2dCyUQ/tzbucLf5d063e3OmXFpNYk2akJh5i1fvcuVbIK71Vc0/tZmsMXx63A94zymMphqtjm9oRA8lAnLT3rB5XZeRUMVUR2mlfU36cc+227UC2gNWU2Ur0CZTSMbl9Jhas0hzdv1QDtd1ET0/ZbLC4ofeZPk91FpNGso07im2m5saKUoeolneznqrz0jNGk2gNQnHSGElK0ROEHnH7RHGE6XXCT3yQT/+X/xXP3LxI+f6H+fUP/y4/9OcOuPuFU66858eQtGKE+lrEIXsPYutXcfpzmNzFLV7FzZ8n5ZfBllj6MuW1x7lz95jLVy7z4vMv8MDVl4lvfzeea0zrjzKb/Q3inY/j+2s49xYKEMYfYgxfw9kCic+i2lPWGzQHzGawfxlNL0KpLoxc/gFcVNQdY/eeIx88ia6fQxcHTMcfhXmC9U1ieJQsp+h6g5ut8elBcG/D4lcZ3Rzv7gM6it1F/Rvkcg8n9yG8QdJjRGY4W1djt3SpiQirUMrJSEmnFPO4zhPzgA8Z0zUUo/MDUTMFYyieNBVWaaJQ2AzgUqDMeqZyyuQG/BSwmFh0a7ohMKWJsPSUzrEeM/POk0umw+MQlosFU1wRSv0+ateTcsZrYFUiKRi+wYveg1MjlzV9UUbNSHAIhU0PkiOzomSnhGKMnRKKo2CE9F12kzSzuyLy28D7gUMR8a1LfwB4ud3tReAtwItSvwUHwO1v8Vh/H/j7AE8/tTRItbh6qdmJYVbzQi3WSToF0lTd+3LGklWOMb6FWRhZPdYJztbkWL+UJXetA7WdUVixXBkkzrcQ7QBNeVql8O06M4pVD3CyA22Li1C7bMmQq01CLTIOocbBZaTRnOr1TqTenyrNV2o6lW1jBQu1ADr9Jh656pmD43aAV+8DxUrzrqexfbQtPm3xaAtDfhNcA87pLn7vfHGvtr1bRoq1CDU7977ROtcaio5Vv3qVKihT9WhKlNLUrAK2VZyWatW85TzV17cdSLvtOlxZPduAcNMKf0j9DE1qZJ/328WzfcuiUXI9H6zU7fPkIWvAomMYOk5kj1VR5qbM8gUWfJ2lH7kQT/F2jJYVab1i1udqWVE8melsYTMqa8l5EptqJ4HiVMmASo+4s0GxqUIxfOcQHWog9GxRLQ/uvcxTN+DKg0se/sv3w3/8LLm7jLv/afCZlCd6D2vnGBaFHF/FwgFrvUNIr/L1f/zbIA/z8hcDlx8+IKz+Ry4f3s9DDzzF0a0Xeexp8CmT3AVs/GNmew9hZY2bvw0tEP1Icu9G3RcJJ0dYNyfqil4ukO1xNBnIA9jtT1P2fwCXPkoa/h04uYVfvAVOnyX6S/TT/8bp8jJdNPr+acq9XyfMH4W0Ivpr9PpJbCqof5wsvwbhPvrck+02Wu7hZNFwOl930LZP736aIq+g6bdb3vK9ek7IMSZHlORQO21Mt6ugDgn7yLLjbVefI9/Z8NrpjGwZE0ff75FZsS6b6gxqGVlPdEtPUI+Q6VxgrRW7v+0T93UzxlXi0t5+Mz0sBA0okVgiDB34nq6AjWPlyKfAplcOrFBshsXELITqVdOBiWdoBA7JQp+V5IUxCHPpGOOEM8cUMwIM7jtnsf9pWDRXWueOiMyAHwc+D/wr4Gfa3f468M/b//+P9jvt9v/bvtEs5VtcsihGC7nOTWDjPBICqG+ddMLUVSsDCmoJyZGcanyaZsMDRTyQcX6N8xnbRv4JlQVTyd3kMmFWKDZimipMIwmRqlhUIk5cFTQ1xkmNImqY+XaQ6ATztYCb1mFiLaAJKxErNahbrKBacKqIBLK1YGpKY2IAxUhWdyBYzfe0Jm6q2zaHFkGKR5hwbdBsrBFfyKRdEdcGkQBoDpRpglRhkpwqdGQtBbzkhmtPCabGrW989mlsE9SsO0bPWQxeNf0yM5w0M7DGMik5NwO5NjzVbZyfIs6BVlZUvRS8rwddNVQN/Mmx4rB+m0YlJEvVr1IVNJBSDQ532dBc83m9m5GiUKwjoayTstbEZCMHZMLsTu3YxxHhBC2ZuUW6bgRSDWrRjEpXd4fUFKtCxnz1JzIa/KKCODBbkyVivvnvi6MEjzCw0QVl3HD6wm10GSl8he6//p84vvg641sS6cp98OjjpHgKUwfJKFaHeRMJK4GcIzM7wdmK67/wdh7+hUN++K96SvoCDz7+Y7x8/W9x3wd+inl3yrWLC5jtI3IPCR/EuEChR+XTjPkOWu5jtv4YnVyqEBeKk8fI+RI6jfiDt6F5RZodwt3/k42+FZ8N7waiXmTqr6LDXYpfMj96nuDfheiXCMuHiPlVrHuEQS8ifsueSkgMwD7oiOOE3B0QvaPYQNQ5Lq4Ru0fht9D83I7mG1yH8wc1ijFPiKwxVxlpKos2cJ1DKLzzynWW00TnOqZ5T9DCuLnH6DcsWaLZI6FjNnjGTSSNG0hCyY5ZV+vEZd8jlghzz5jWkE6ZaaJIrTuSE8ECbjJCXCLdgHSwwVDXMTqP94lh5hglkpzgrbLlrK/nYZaMR6qwzoQxjXShkjWG3uM75cS+88CPP80ScR/wD0V2eMc/MbNfE5FngH8sIn8b+EPgw+3+HwZ+WUS+TO3cf+7bP4WibkCcUkyRXsgpNlbJPiYRTRHbzNqgsAOqUZeWgsZqB2shYYBzgoUCZVEHfdqk5hW4ofJFPNZqROWfdxV+MdeYLDTM3VqhbRi3bOo8QDPmY0VxWpRd5f5VeKWUCgmhoTEoaodtxe2EVkJoyVTWYI7qWRLOcdctNwycLfXOEOpCoju+fS14ztUCCIHqmpjrjCClxrVXnEi1AwBKPJv8Oqn+NdJJMyuzNohWere1LqjWvUAVVeUaRlJKqk6QUsippl+JWuPAC+qm+h6bqwuD1GAV2/Lpd/TQs4FudYnMu5SvXWhGAd917fcqahJ1SBByjtUmwZRsHrp9NnnGFBbkfskgSs+E9xvcZs2BnRBY08kxyoZc1qhYPR0EkmQcNS/AjDqoty2ltb7XKtV7XlQqX9tVrYCKgUskBpjNGMzYeEf3xDVW//QmuvcTxPkv8fYPPYJ0Ebl4AZFT4gXDb17BrTNlPsLyYbr++8jpd7HsKeUPkO6d9Edzkr6Mf+BB3v4fDtjFOU+88g/I0z26F7+EuHeTb1zFdU+TpmeY3CX649+H7glC37GW/53Z7ENI/gxJL1J0wrkbuAniwbsZj56hj5ns94jzNxi6D2InLxNF6NILRL9H3nw/zn2Bzf676cufkG3Cy2XC8ANEW6KbmxCuk6XDlXtoeSvTOJHla5hcwMeI5WNwN+uOWWaoXYB8FeR5NAyU0UjlBVxYIrKsOzpZIbaHyICwxOcDjDXIMRevep5+4jIvfDFxY+q5GzL+wBE2hSNbkaTUdCdXmElHwpHnkRMRDnAMpkymbFSZJyGEwKggEukIuOBxztGLZ5UjzO4y80vMKX42MkuQ3Axa8AziK/TsHKMaF5LQZ8cmOLrgmPIaw4EKY54wqi06KM6+8w7+2z6Cmf0R8P3f4vqvAj/0La7fAD/7r3sYYpepRWWN6khSqZYFkneDPGTCLCAyIeIpZYNtMXLnzxSDSUFPKRzXopGuNah2ff44ge0QsDbmZ7c1PLc9r1Gn5EqBvA/uNSBAnmN2WgtWC6LYuiWev9TiV7vxUnGT+vc4TBvV0NpwjtyGw1JhFVdzZrGGS2/lpgBsMXJtStfKaFGXGn++7EB7dYVSmsgJBVGsZGxbVKmWB9beG7/1htmygWQL1WwJ8xUi2mL19fiqP/rOH2c732i5syVXqmgNTmf3nuVcF5JSzgLUVVw1KSsR07yDpbawTV2QGoe+DYSlVGVrTobrOqbiGg222j0nSag6hk0iuMQ43aP3iYUkprjBax06W65JWJWptG0JCia1eRAnreAL2y1IEShMqIbGnqm7I+2FOAy44BmCp9y9Sf+zC/z6Uwzzy8jxCZQ7pAJiBzh5lZL3SLqhSyDjKcKv49P78PkONryGpd+H2dtJ+YScC94v0Nf/hPHFL3By63EeurZgWm6gfyfkCe/maHmY7E5wwyFyekK/fJw0fhnjrfQ6Yv5VTJ6G+BIhPk/SFXHf8PkFXJ6TZA8/7KPd46T0BTyHSPcsTHeYn07Y7A3I18hmlQ2jzzGFrzGUq1iYI+lLpC4TzCHMmRhx2aFksD2EOVip4kEcUlZIuYRwgrM5YvMKycpQGw+tMfGlHGL6Rs3UCgo+cmmRCSmz8gnbGBsye8UxDAOnaWJj1ZV27TwlOPoN7OEo3hPxiHoCATpHVMc8dPgChD2mPNFLJsnEMJ8hKaIhcDoW5p1npUrxnn4cWS6XoJ7VuCFLoZOqZ5imiUJgtEyRanZY2MKhjq7UhmycvgsF/rtykYrA1ASXOSJ99RuxiKSjnSKSPOBIWJnVYWIrIs5BThGzqUbyAVqkCqaYED1mZwhmlb8uWqgj025HUay2vgbkc4W6Li++6o7qIIfKmjFbVziiNIWdlMbltlaTt3a1dUho5s7NTlPdIVj9260CdMtYSW14aVaNwGqqKq2ICmaBLVddXUBap+vMyFtBVEsacq4WPZWazGQSq61xaZxtKslIEZzUgO3cXru4qq6ti4W23U7lpFddUmMB7QbVpZnea018yhmkbzj2dsi6pW3qjm5pVo+vbHn3Vtows3L1RZWC4UNPTNUrR1TR0g4+1ahHywWTnlh6JhHoHFkjFkYOzeMRBt0wxDUHLmHxmFOOGDpB1rG9VsGKtVCXOhcS56tFsa/ipSK+bdaao5Aoqgu0qbGVRCHjS4HgsTCj9Gu0dOjKmPo9uuPb1fpYFG9CkVNCApM1nSskW+L72+AeI/kVPmbMXUOnY9ZmCIFejomSkGGP/l0/Q59vMa03uLCPWx4y5gXOX8HWz6Hu+2DzUSb/k3THnyDODnF6G06uIAc/SoqvwLAkR8UWFwjjP4LZ08jqSXy5Bes1Pn+BsjxExjXwAkhHCSdo+k8Q93uE4CDPcZrw+QqmP4imr5P5Pvy0Bn9EKZfo3B65fBUXlCnt4fQi5l7H0kXMvYBaIUdBKdS0NXDqyekGYkdgq/q9IaByvZr/cQz961y4Gni4v8fzp4mMZ196JmcUr0jKKHWesgyBkxjpuyUxRxyO7EHmsEwwucwgjqEISZWJmwxuYKnXkeApecTcgnVJLDQgKUBwzC2is54xJ4iR3nscUGJinDvGwSEx14XEIiaKL9WozsQYLdSGIWy+49L6PWNVUOwK2H2YXiNzacd/L5qq1DsXhJNaH90JxggyITKRS0QpeHFoyVXMJLlmX+oakSoE2pp8bS9b5orhUXVteLj92crq6/WlJLJUz2Ytc0SqGVdpJHIjNYGLvOnxt9xumvhKZarbUXwbrvodrl2L38TWzGrHSRcB53fHVR87g0xtVzNVvPtNRmRt/th80cVtMGLD+gPCrAZktE60+rpvfakLTiqzRcvWCuAckb/BNVtmjWvQV4Vb2vta2kaFqkLOOZNSrHxhFKPRHHdWxtVYbTsQ3wap12H2Wfe+3R1UbZU1bnyzMjYH4jHtKK6nqGOyTEjC3Hm0nDB3Gw5kRWdrNuUE0YlFzriUwNWdQhuG1EVebUfNZPfev9lmeXus1fT53IBcYK2QTiKp3KUUITsoUujiMaRTclkj40XMekRugxOKM0bn8FqIkyenjryaSAp650VyfpCZVrvk0Q7weQ/VwqY8D+vPISdvYHv3czp9jtJnOPlNSriPLP8PJTzKpP8XxZ+iHbhxYpN/mfRH/wM592g2VD5O2HyEwgGsPNm9hB078oUDSu/Re2sSNyD+CCyvkjmk6OdAHiKlDxDdBsk32fgjRP8Atedw9gbRv1FxdJ2BnFaqLmtUX8fcC2Adqp6SPZQ2KyuLOh8rHZS9mvbmKhZOU6zmUi0oUp4hbiCbslgYw5DwGijqMItMm7GKwbSDfkZKiWCGL8qQYGOCasdsY5xKYBOVaI5oShqNLsyYinAvnnAST5BZhxMlSNVpBBPmVuFFZ3Xn6oeOojBaIcwG8maiU4eo4tVhsqxCS+1IJiAdiVjtQOK/LWZjpqg/gHIJ+ktYBK+fo9jLOBvA7mFRyAw187MsqH5OU8PIcx1mxgj4JqiYEJ8qpmrHqFtWtaUIKnUHUDv5GmBdC42rND0q3lrDp6uDZIXXByQlTDKFNsi10BaaSo2sEAfULl4bVFLZMzhfu02rxlsiinPaYBkaA6ej2AYp9aPZDVm38IRrxQcB6xBpA1itvY7oOWzbWd3JZDA3R4h1VJAVY0SKp9iEy7VAOuew3Lj/3leSUG7JVCatYwHX8HRnAUkV/xcCxELRqlWwssXx2R2f19qhk1uhVKviqGzteJWSGu3RgWjNXd1F/ZmrFq5e2+cIFiu1MVoV8qQ0YwpLUunJCB5P5w03bsheSSd3ME5x+ZglJ6hOWJ+wKaFWPx8rheILLtQPwFrwOUibZWSknRNOB7JQfVe0wzfefg4e1YHeIkUSbuWR9Rpza6yfQxcwd18lBfgT0BmiCyxvwL0dXd4ibb5cNRT8EX14tJICDhZkfw3dKH03UOwOUgI532GIEboLpP0Zs/GAkp9HV448G/DxBTIB4Z3Mxs9i+4/i1nskt0EP3offe5Ai/xzcZymb+9HFe0BepUwrXIjY/jUkXwAR4vxFjCNyB2xuEYpj1AXeH+DSZ5Fyh+zex8BLGAtEIzHfRd17WW0+zjxcBW7hdI6kZYXDcm3Gsp3idANcROwE0+fBlqgMRDvCs1c5WPIigpHtDuoGXHGI7KP961A63vvAIe75FV886ng1nyCzgW6d4f9r711jJMmuO7/fOffeiMysqu6e7nlxNCOSI1JcUg9SJG1JpCGt5Qck7cIwZPmxMGB9ELBf/GENGDBWMGDAH20D3oVhw7ABPwDDsA3Da+xCwEqitQ8v1lpK4kp8SJT40FIz5Lxn+llVGRH3nuMP52ZWzYg2sJohu9WTB8iuzMis7IhbETfO/Z//+f9FmVQQr+QSnsjzXFEKQzJarjgjx1TQTBEoQ0WGxGq5QiuK6sDGE21xcp4QVmF+s87UqpS8YrvdMqxWeGuYO16OmLdhBG7mZDmm1aB2qifc6b7MQimZWiuNt19kfSAyeHcneQs8uxZECrZcRfxxqi+0mhFpJDeKEEz0VmluF+YR+wIqNK89G+1cbvoqwHv7uEyRSQbmElKwKe0bVZJ0mV5aN2oIIamdBG1ojli0TbtiTWiVTq286GbcZaattX0n6r4bte/3ZQOOnXyBdj2Zy6uBCx77RTeq+Rzsl1gA9v8zhXwBhN46PeNsM2KtyxUIWgvOROqYukgBKaFvE5xLkNBx187n1lRQS9Qakr8ieV8vMJ/RciFVvNtHtBckd01f7BJhR7zAFMwDlQKzE0qVEgVtF1LuWjmXVi87raDLgmRJlDrHCqXVM7CFMRviE0JlmbdkaYzJUd+SJKQhlmWitoaWoa+KQoJATGhLYPLSC9Ntx7fXOG53D7hLQBjCa5VdQTZuUpqjcUs9R6GnJFTWVFFM3qCmuygpGsOmmaqn1LVSbt/G5SRWJ3oP2jNUycjyMwx+Qh22LLaGdMxdvUurLwF3mPwO6/UxZ9PvkzefIq++DuOHkTFh6WuY/yo+vA/aSejH8AK5NZheYph+G+49Qzr+OWTa4O0jaPkUSz3Fzl+guaJnX6Usr5MYSXaDlB7BN+8l8zTelOozjRXwObCGyItUf4XtecbrDYr9Odxfx6ogDLiek/RxRI+xNuI2I2TQiVhXrkJBlNz9kTeIX+1F2a4gaonoUu8yIXnhxqPHXL16xJTuciTC6bJwO59i84JNjusRiwurIeFlQdYzJpmUjkhD4rY7SylMomwXpTZlWwzP3VuYxiiCtUyrIbO9nc/x5DQx8vExLUVPRm5K2lbwLpomCWQhlRk6vBfXUGNezvEt5JZZ6+qfZhr9tvFAZPACLNuC5nPEXo1JL98GOUX9GsgdzLTf6bq7jjpUxUxI3r8l5V7gqlgVcvbOTAmjjGZKyhNt6Y48MrDrUxW6eUYvJkbCJoGri2LLDLl1Q4pun2cSxdeOz7cWd1zZpfESuL715pxg9LzZYi/wdOvyA12jJVo62blBiRSiFf5ScQ+LCaXfULx1iAfiwpAdDz/w8UYLU3CJRp2AUUoc246iqYYtFzcnIcG++NniZiJCzgreMAsmz1ItsPul7YXTgD3ElNy66fROwz1YPKIFioDPccdLOw2fMF13S9Q6knTpt6EYm2BK5ajLLEarjeQJlRXVSoi+PNMPAAAgAElEQVRM+UJulXHIuBqNiXW9h/hCWW6FMxATwxjyA3VpUcvZnQNJSTlH8UVB1DAF79X4lKKQagSM12QOLxlp6LCCE8UosAyInGI6IeUqljcBG/hIciGV1FcmiXztX4CzU+obn6HqTbJfw8k0PgLLliIfAP9HOD9Em77CqoVe+Yl+FE83YHTK+YjpiuF6Qc/vYva9pAQT5yT7KAzvR+x3aYsi/hJ5egE7+ihn+gLKLzCuLawcNz+MirGc/hFpaKTjEdoXmDePkPOA2OMs+gJJn6RygvFPGHgGLU8yz7cZUqPqGqu3yLLh5GTA/DfwdennyasYG0SuhLyFv4+Uvgh1QXzTqYWO2wiSMNnilkh5AQy3K0GasBeRdE60jl6hMiC2pg03+aHrzj/41pZ5/Qgn92bWsma7zqymSp7h3lLYoqwYKTIwjjAv51hdc5JjdV202y3mRG4ja1WkLJyJQhlhUXJeRw1gCZmQeZ5Yr0bmuWKaIRP1J1pYh/rOnCYkzM1z72wvNMugM84FxfntxAORwSMCuSJaUXkFkRdxOQdLCHPY31mjeaFRMB2QFO42eecCtYMiWniA7owqVAOGCANkBQak8+1F6561YUjg/mR2ZtuwY9vscGK5eHQ1w10XJxACWfIncfjdQy6bdO9wfkLiN2tnhrylRkDfu93rCwxeAmrxnbVdV49MLRhBrrh1uYLeUbr7LumPvQvSjqve2kUh1Xeyyro/ZrjQ6LFO33T3mLAu0Rwv38D28r+XFCN3x9Bao3UOu7vv1RpjfAug5Hz5mGOFoM2xKfSDOn0lvs8clURrM80CusFnxLacrAbYnpPrOWOuSA6Jg6VVrDVyEtBEKgMy6L4nAo8VW7A7+woCcA/OsqalF+wz4jOegEFhHKklY3nbaa4D6BjyE+qo3KXJKSxONcdrxc+/Sh2+QbPXEHsRP9qwHX6AXG/A1e/F0sicnmI+/To5PQK2wmuB+Q7L+COwPIMdrSF/BE4zDE+iPE4bjsinN9GjH4Hlc0jJFL/NMt5ge/3jaLvFZvNTDKvfZN48j8ot6vwCuv01hvEutv05XB+nlscZvCD1KeArZDMqj1LYkLmCScW5hXML9AZZ1gz5RmSs9igqkPIa4zTOjbQCTahuQiPfheYV09Yz2v45vYbqMUjIMsf5skb8uPNPXo/VIJUkj6PjbYou5GHDVU+stzOUE6quOBdju0qcjpXVtWOkjGQVlnoPq46ZYrJlYYV5YTsb2QfSLGQ1qjUqiZIKS2uUzcDSz4/NZsN0Pkcn7blRZI0wkvKKqcW1V2vd21IKGbETWs3U6jhbNJ8xuqJzI7dLy+E/ZTwQGTwewlB23tB0TsoDzU5QuY0vhi9nqF1lsTudVTFidUGmObpaNXjrYqCaOs7s5E3BnTDsqAuiKRqLUsNqsGcS0VgTE652E6aQEcZa55oLUoZg6fQMf0fdXJZG0txpgbH8cqmdURjV+rTDcD2+682TYEAXu5D+WW9d6phK50fGpzstEt/1gnYZho5vW5+kYy7a6agL0qZ9pyg5aIqaarwnEi313S3pckETjGjOHDqVMjL5pCX44KRg/OQEvvSCNRdyA+bokHGTblISMgdmbc+gaQAS1Mxdt7H0FYNVD8kKD1EnmaNRi53EsTlJEsvc0DRSWyVlpWhlkYWk0gv0MHCXLDUupFawQUhlRD1BW4KymsElxcpCw3owFjI5MH0P6Mo8OFhRbJ5xW2iDkK7coI6NvF5T2pZWV7FCEY/Vn7coJmbI/iwmW8xfAv9esr5Evvkt8ngCVz4N7YwxnyL37sAa2rIhDc/j6yfI0zeYVx9BGpzfvUPSr+Cra1j7AdK4JutPMtvz5PWMnX4Wv/I06ewraHoKWxKVY8Y2RBLQtkztW5T6LKNnFvt81KzkB6H9MYxfZ8tTrOuGlkfuzpljfhhb/oi8egOvlVxWmN3G2pohf5wmr6J1wniDpE8wpa8z8DTWjDwsSHuG6rfIeoXm5zivkHyDDlcRG5HlDJe7uOdgQbmQ9QhvivgW4zam52jr15C+DPYIxgR2nTrcZRzu8ujVNa/cVoZ5xmUg+cRxGzhvjbnew91YyoDmgqSFYjCkK+g0sxqiZudVKOPArDNIYWVr1ouEVpZXhlIoMqKubDalowwDLqHsWlJmnddY2yIoqkJJlbN5G6J4u050hMTAmVS8pDgv32Y8GBl8nZHbL6N2DzjD6k1SfQ4/X6AaasdRyNsqqTZSa3jdYku3yPMBsYSmRHWjWUF1iFqmBJ6r7jTf4nKX5veQPAfvfCfutdcvUVrH8qoXqAlZAJvBps67rniLyb6ooyxIqWjuOGDHclU8skhsP1nuxb5khtTdnDrOvRfAdGevENl/R3a0QBISMx9I7SyZvlqRRuqZZjCAognJqscyMAmWQnEHlRBFM49mHVdSp4yKeHRs0usKloDzwEcdkmos46VS2xwU0trwUE7Z/1m1/x9eg+evdC13F5Kl7sMa3a/Sufa7xY+10LYPbZ0Ny9Lw2ZFzwe81pptblrsLdbvQtjOenLlOpBy1Gmsz2aDYloEtaXmDYudom2LBmOJiUoaY2FMv8o4ZO1khxyt0lZHVCk9DX6kIpgmTEhLCUoMGnwtaBuT4BD3W0DERj2LzusH6BBlG8IHGmpoWJD1Fs5eg3mUoP06WR+H0FebVRFs/jS23cUtM5yvs2jNwx7HxDjY7i92kjp9kyNcp+hzr428y1Iy0GW+niDVse4bkn0HufA5t3w/8BZbNM6B/Adoz5PoGnn+Uswna8S+Ra0LLCq+/RfI75MVwdc7tBmVIrJaXqO2PkfoFrujXyPwDknwpZClSAQbEX+qOVt9A6y3Q19F0gquQ7anAndNV8BM8vdFhLkPqKULFxPpK8Q4ip5CfII3Pgj2Gemdt6QTcDfxd1jQS5tvO9LoL8ijCk4ylYtcf42l5jae4Qxrje0/GDWfSGMbQeB/LilHXJBM2umGdj7gznSPjyL1JsHmIawXhJI8ceyZ7pQ0OGazG6rUy4zmw+WqNoTTUZ0oGScai90AGPC9Myxm1JZooKRVSg0GF8/mcybYxLUyVXN9+Bv9gTPDekPkcP59o9yaYJmw2rN2lySs0bqFyhZQdqwttnhBCJRHdYdVBzUtaSClHhqU7TJdok2cCXRCWWLrrjBUDwlh6b3grjnOGynlAG91mLeehc9N3/pv6JmpcqxXzC5riRQG4HyYLO957YIY7sa0Lqt8+Q+9QyMXz2rPz6PTcwSNvNheRS/g3XBQnU2fvXKwC3B1sBO+qiH3Coh9RNJbF8525uOpAGJsLkmro68SB7aGYXUF5D01ZsGX2MJWwP9bLEshRyGRfiA5+fchD12Xbs5ygm9YlLBTr7OFF28BNyEnxFgXVIWVK6pRLN9zO4wZP1DL29ErfmYloPE8hgSzaVTG1U13ZNULEwzsyZBpKpFKUPAie1yFmVk9ZHIrewOuF7k9OA8I1fL6GyhpNE1ZfY7Hn8asLafwooiva8iTenmVz8vPU0/dTV6+z9oV89JOMy4j497MsC+dTog6J+dpC4wZjfg92/irn5/+Qsv0W26MP4etvUs7+Nqs3tnh6jpS/jFz9Rbb1eTj5EHL3PyPLGcv0+8jq0YD42hlef4/16jbzkqhymzxYaETJHWIy3URW3E5geSl02ttj1HkOKWu/2i/vDcgEXmjWhdf8GHgUuAblDikbYfbjQYKQ2mUtGiqvA2fgcd2GYOAatwHNAyrr/v9t4lxKMy1n1I2yuY7kEwY2eDrh1TZxVAqCMyXFFmdplZoSt7dnVDOOx5EpnaEj2MYpeWQGWhVMGktq1CaIKSkL4zhShhX3zk5ZrFFW4/68VsnRfS0D4WqRyHlAEUYTEo06KqfLxI1yzFEbo5hfEmd1fjuzKvCgQDRWaW/8AZRCWq0xU1oZkLyg+QjxNbQFWxKtCUOvRrsXQqU9RLpMLCAaX6NDpVpFS+CbrW6R5TQ6+bHA+D2heoznhOcNUlu4RCVQ2wQDhnPUncqMVhBamAl4ZMFYwDut1d51mmi1dZzW2WMO9GxYfM/H33WCsvss8V07+OIy37r39rOTCwjGR3+fBA47/RYR701WHTfeo0GBUe8AIe+YjaN7togRkg47u0AL+yqcjFkj5YCn2hIQVtaMLbXXO2IZEll6nwujcSHm1BTtWt4MyRe5xeXCs1WjjAOtxc1GU0gBGKGSWedKnZ3TU3BvrNdGKQolmqxclM0qg0/Y3IvC2hCWKIYTN7rWFrR01tXS5SBKDlptq/ubs8sSBeLqAZuVXhMYxoBzNqvQS1optQykOkVGUY5Jc7C1SIZYpnkl0RC5gchzkB7DKWh7Hs8jLn8Ra/8ELe+jlIbpEW36m8h6wZarNL+J2XX06BPIvf8UH66T80+S/Y+x9hPo+WcxeRWVH2KzfgTTX2GU92LbH8LTpxD5L6l8GpF/Dbv3KzR/iqMB7gyfoPg/ZNT3MW+FxAeR9CtoeYx5XlD9FlmcbXWUT5E9YeX/xjBEVkBF/ASGMEkZhgHqguWXUHs2Vsrd9CTJFZo8B/YMyR6lyasYW7xeRfUmcIrV8D5uskVkQZYG3AM7grxG/EWaTf2GAAhIh09FErSRlEZ80/i+px7hldM3ePHuXdY6UKY1L8k9yjiwnkrU3rSQU4qSuWZqFlZtTbfRYh6VsgR1eOirubnNzGXDUc4scyRZm/EKu/4VBVLOiCjJhTZVFu/NVNA1qeCuKXmqjOWIm/UmqgO5xfVc5e1n8A/EBO84Nt9CpTCf3yaXFVpv9GJgpzn6Ql2mwIOtD6ImVBzzCe2NQ0ZCc+vshuikzJKCBcKO4dLxVW+o3gEdQs9EMmZdRIre1ehBYUtVQquEusfM3S2yDHrRNek+w76Y2C8XTS+wdCBw+z1N8c3smoh+9hKZa+jB+Ju+MygeOxbO7nsqookd3o9btHbvvr9PXl2iHuViu/g6DLLjtwOGILqAVRyz4NgLOSCXLkWsGjr3cGGWvbs5BXvGQpIgxfGYXThFXT7uoKN2SYUUzVDW96enlphFfTWlUJZs7lCNNEaRO/j8xtC1eVqduwZOyDSH4Fzayyl4L+xiDZ9AhktMojjh4qaJI5qDG58UKQrDADn07C9uyt34RMLNSiGW+Ro338obZD3H/VrUFfwG1h5D/Hk0b0gcc5Zep/AKyY+R+YzEjPqE5DfQu1+D1cfiVn/0DNSEyd/H04tw8l68fovz+ocM/kGoH0Pl1+DoCrx2TDl+mptSuX78kxTWeP0Nrpb3Yv4ENd1D62uobmn+Cap/lSw/gOeRZfodVvmTnHNKThPUa6g2VAttO7BNL7LyH4T8Km4V/HtQzuKkshn0RZI/E6tEO+rJREJ8Q9aT7lU7YjZhvu1J1HE/B0+JBowFZIN1a06XIbpSWUDO4rz1oSt65DB3H+7w+LVGuaMkHZjXyvH0CDSlyUQ+KjAJ2pySBqZpRoc1TqaW0KN5rBZmcfy8YblScyaVIQw7TrcMJTJ4ny0WyR7NcsuyoBqr2TIU2mRkU4o45tE8t9JGysK8LMi45nwyjnZd9TwkLBrHUXN0gcHWaC3YdAu2p8ic0GWDNsgaBdLZtpg683IPa+dYLTiNNFRSNswLmgJeEFdYEl57Q1MNeVuaIa1hbcGnM9r0Go3X4egUXW+xfEaYdc80OwO/3fncsVRzS1RfAvP3sIhr3bdz1x0KUQPYT3S+gz4u2CSX/bp3RUNFLnxPrbOEPKOpwI7YKSnYQr0i3wsIfULPuO34+B1quQTl0DHvfRetdn69N1QhSY6sNQW8oeJgLQqeNsT7vrD3kSUMRC67aMUSdTeBtv3kFsYo0QWsmtB+I2puUWdRwZaoPdCLrM4SFLLzgHQEZaiKVu8CblETCPXmxjzPuHjQWm2L4iQh7maLI1WRWbBtSGAkB1rtUseGbBd0sSi0e9ggIlFsrUOirkf8yhq/coQfb/D1Ogy8LWNlzSIjtAlJ1yjjeyBtonuzJFydklY4z4R2fr0G/l6KbhAG5noTpiPW6dPBna9PIAO0YWFefQJvvw3jLZo44jeQe/8PSIZa4Oj9tPkYSddY80GSPgvz30BXH8fPJ9hcw8/vcUWNs+038dP/gVl/AkuPYPNNnEpO/wbSzhB5BONfZ+sT83IP9SeZ6hfRNKJcx7kJWjB7Dsm/hUrD2UK9hzSl5S9gvEE09FXUVzivY/JHJLkK+houLwFnsEw4b9DkDGQi57OA1gyoBdPzPkvcwf08zltfx99dToAnQRIuK1yPkHQN8UexobC+sua9V5XiNzHOqOf3OFZh5ZBV4d42aL8qiCbGMnKia0aH0RzfKK8NjZuy4OOId3/WNEcT1Lg6obrEaib381iFpEpO0eOdUmJ2Z8gjpdepqk+06pQqTAZ5zOjdLUc24Sos3jB9C8b7p4gHYoIP+qJhdcZawpcNUq9APcLnKK7RSlCYDJBCzkNceE0DP/ZEnTPNhFTmC+Eq6xTCDovsss2djoguYfQtFLRPzK1o93v1jjuPKEd9X0MtUVX3TUGRjUAaCpLTpePacW0IESbPvaBrl4qo/XNvolZepiX2fWqRLe9xen3z4iu+Y4fhBxYfvPWL/Qlq6EUDlbPthdq2pyaGBn9l52TFrmjsQ8/Gp+hu1WFfB9hn8B49ApcdoHZm3BA3C/r6qO0azfb7FpO97xqsJJpHoufgArffUzrnSjFoi4XxeROmswmrzjiOMe7uNN2NScN9Dm/YBdpcwx1MLvT3W90pV8Y/3tpeEE01HJtSyeShwJDxEpl7tFUrNa3IUmOVpwNp2WJ+j1RA84BQMBdoG5bW7QzzDPl18Dsk7jFIA38Vtl9A2KDjPWr+CDI/yXD2j2iTUtvvo/lDJJvRdcLTPZiEVK8i9evYcg2jsAxKXv0ctG/h4w22q08BHyHnp8nLt0ibf5PEXXQeqP4ao72fxT9H82uIbxF/iU36OEP5VyE9TpF7Uc+oV2lyh+RXUbvOUmHQJyEtVK00XibbB3FbYX4b5zTgSUDaEVhGiN4JlYr7WVAe/VGE64hdxSuI38G5i7Zrcc62AZENOY04U/QmuIKXgG9kQSWhHNHSe2A6ZilXWR8/xtVBmIZGGzO3dMvWJ04o5Fw4twkwfGmUlNnOM0sbqMOa6bSRDY5yiW5bMWQcu6DfzFINKZnJKs0bMiizXzQ2hsyJgpaY2JNw2rZMGg5o23FG6kzbbpH1hlYG2rzQS39vOx4IiAacyhSCQzwSWOpyB5unmIQA2xloW8EWYzvHhGreSBoWbykrZk6bneQJK12wSxKSCjIHXoz6nrfdmAMXaxZYc72L6oAej7R5Qqct2YWmObIySbjMuHUDb1GaLfH/1BYTw2WIYkc5VAXC4XC3jMcFSb2Z6bL9W/ceDdelTtPc30y6Ln6bAl7p1nVxGwlRLtoS2Lh0k2wXkBJ4cI+YnHOsENy7KGLcKDsTFPcSjkZu3UIuGqcShttMaM9kMMNq2CDGROyh4eNgXY7Se+NV7zslaYqO411TlnV6KBAcftlj8yprrNWQVNAohLbsMUluAxpKtbG5MiDmNJtDtt9bXwl1YlOFIl1vp/WJ/MywMZybVLuhSJJQUxBIWgN5SyPkhF49gVWJ8SolVogOponkFfeR1OsQNm4xX1A/jow7F9RPcJ/J6UqH594P7Yt4OY6VWjpmzrfwVhn8+2D5Jln+ANu8l3b6OEXeQ61rfLmFcINEY9q+SlptqPI8ZfxnqO3L0WG6/QLV1yR5gtw+h+Sfxdav0pbP04Yn2TYlyx/jeWZMH8eXW5jcJJeXsfY0I58kFL//Nkl+ii2Pk9oM+TPI8mHQp/H8FZSZpi9gnlFfcL1NkzWJFaadqOAnuN8BogBr7QZJzkP6Nz+DyTHSXkY8YzqgPnYD+xnRY1oTNN3E7W5AanI9rg9CzwY/QrxgtkU5QZqh5RlMPk++UfiJDxzx5ecaX06ZuhiaC2e6oXllZTNjyiw6cO5bjvIK8Yn1BDltyBYQY15fQRwWM8Z8FDWmwSloQLmSUHfWCK1Eg9R624ImvJ2QVJh9oqQVabEwBTor2E4Su0bj4JyNxeoFNPk24oHJ4HPZILJG/SrChtQzbmsztU24VaqFrdsOErBLHPZYosVSy1ul2hIaEO60Ze6Tb2dy7KR9L3Gp1TscUhdoE/gS7IiVYAOBa+9B42gA2hldqL6FHihvXlp1gmTPsi/wXbsMm1xqapK3/P6bvwliUgyIpluABsRj3tUs32L+vPst3/1Xl7juu+d9HN7aqBR/n7dw9/erAt3XGv7kPocyHsTYypu+2/o4XG78+pP/X+yrQW37CbkuwjI1ZgsM1zzTLLGdff833u9TH5tdfSJuPuDVo7GsN8Or7CQIonPZYW/YgUQHbiuKjzl+pjD6uDjSYDpdHtvd30pkE41NykUWm84xfQPlCE+voHJEYwY5QqUgdsw4XAU9ZS5HpPwUcu+z1PVzyHKDMn4oJHLlOWp9naLXkXmFyLN4+zpbeQPmQi4/CjqypK8xL99Dc6PZlyj5e1iP12l1S5ZnaJwimlj0FiJH0D6MpMxU/y9q+ixp+VHI3yCnMyS/BPaDlPy9wBVgpshVkn2AZE+RUgltGB5hb65j4/783WPvDL2IKoRMxkTjZUxvEnjbhLPq4xuruyjohu6R+AZh7EyreC4iQSFmi6Yj3DMyXMFH54kba77vRo3u5rSJa7ZNpDm6z6sbyY2smeyCDgVyIq1HmjhpU7rao2JJ3sQUW6xF1q47RpYwnDtHjOTVmtM2czYKMzWgT4GmRpVdL0gkWkkSrRrJIPtOxO/txQOSwYOmGzTJLPZCQAR2B0zYnmc0QdbTaEd3QVIBHMkdj+7sDe84b9IGOtK6fK+4xQRt0ifFXaPSDiMmZhgTklZsmmF9BVLF7BRdr0BPQxqw7dgq3c2oy+26tf2kuZ9gtFPqtN8TltohkoE9E2Y3GQuEgBjsG5vcA1gmTLsDqujvdUs+egFT2MExuj8usSgkixQkObSO66fIMHW/WvC9Y5R0SiMO2gW+MO/rRe0TdaeAqWJdpjdCgEvHRC9Udk2g+IrA+3dyxKn71u6KStYzdxBaM3auZe6xAmhmNBLTaUOzsUyNRIIV2LGhJWwJcy59BRFJgFqH5cxQizWj19ShqH7DSgRrJyd0NeIEL54E6XiAISND/C1co3GJlBHC+MOdkFjwLnInvYPVKsgA/gQiJ1BfogzXMTlF00BLL5PlE5i+RGOF6F2sbkHei5z+Y9rJk7Thg+TlBr76IpJ/jHZ2g2H1CbbT/0QpazQb0zwz6DOsJSGbJ9mZqYgfYZs3YHqepNdoXkj+IcrqN2jcRu39tPo7DOUDkQ23jHKDnE5J/ueY8zdRG8j6GNiA6ZbaGp4+T2kAd0BfjS5sv46wAf1KNPv4nfBO5UpcLz3TFXXMV6G97xMir5NcUb6fJiPmL+B+pydj9wKGIWQlNGmHDWMlaYQ/hPgtRG5jfobr04hcJ8mj1HzCeO0lPqDH/MjNl/nN2yNnViGNQLikJRRLUGpmEWVlkFBkdoqO5BnmIVN63a6tjikoxVPQiDs0Kx7qqbk42+7ru0ojPlski1OjLpU5bcmyimPAuzdDzCM7YQ7Xt59/y7fL9L7bISKvAqfAa/d7X+5zPMphDA5jcBiDXRzGAT7k7id/2l9+IDJ4d39MRH7b3T95v/flfsZhDA5jAIcx2MVhHGIM3s7vPxAY/CEOcYhDHOKdj8MEf4hDHOIQD2k8SBP8f3u/d+ABiMMYHMYADmOwi8M4vM0xeCCKrIc4xCEOcYh3Ph6kDP4QhzjEIQ7xDsZ9n+BF5KdF5A9F5Gsi8lfv9/58J0NE/nsReUVEvnRp23UR+YyIfLX/fKRvFxH5L/q4fEFEPn7/9vydCRF5RkT+roh8WUR+T0T+St/+rhkDABFZichvisjn+zj8x337+0Xks30c/jeJhglEZOyvv9bff9/93P93MkQkicjviMgv99fvqjEQkW+IyBdF5Hd3jJl38nq4rxO8RGfOfwX8DPAR4C+JyEfu5z59h+N/BH76Ldv+KvDr7v5B4Nf7a4gx+WB//GXgv/4u7eN3Mirw77v7h4EfA/7d/vd+N40BwAT8lLt/FPgY8NMi8mPAfwL8tT4ON4Ff7J//ReCmu38A+Gv9cw9L/BXgy5devxvH4J93949dooS+c9fDZc/Q7/YD+HHgVy+9/iXgl+7nPn0Xjvl9wJcuvf5D4D39+XuAP+zP/xvgL327zz0sD+BvAv/Su3wMNsA/Bn6UaOrJffv+2gB+Ffjx/jz3z8n93vd34Nif7hPYTwG/TLRCv9vG4BvAo2/Z9o5dD/cbovke4PlLr7/Zt72b4gl3fxGg/3y8b3+ox6YvsX8E+CzvwjHo0MTvAq8AnwG+Dtxy950i3OVj3Y9Df/82cOO7u8ffkfjrwH/AhcjSDd59Y+DAr4nI50TkL/dt79j1cL87Wb+dms6B1hPx0I6NiBwD/wfw77n7nf9vcbWHdww8hIg+JiLXgP8T+PC3+1j/+dCNg4j8ReAVd/+ciPz53eZv89GHdgx6fNrdXxCRx4HPiMgf/P989p96DO53Bv9N4JlLr58GXrhP+3K/4mUReQ9A//lK3/5Qjo2IFGJy/5/d/W/0ze+qMbgc7n4L+HtETeKahB40vPlY9+PQ379K6Pj+WY5PA/+KiHwD+F8JmOav8+4aA9z9hf7zFeJG/8/yDl4P93uC/y3gg71yPgD/FvC37vM+fbfjbwG/0J//AoFL77b/O71y/mPA7d2y7c9qSKTq/x3wZXf/zy+99a4ZAwAReaxn7ojIGvgXiULj3wV+vn/sreOwG5+fB/6OdxD2z2q4+y+5+9Pu/j7iuv877v5v8y4aAxE5EpGT3XPgXwa+xDt5PTwARYafBb5CYJD/4f3en+/wsf4vwIvAQtyNf5HAEX8d+Gr/ee/29mIAAADBSURBVL1/VgiG0deBLwKfvN/7/w4c/z9HLCm/APxuf/zsu2kM+nH9MPA7fRy+BPxHffuzwG8CXwP+d2Ds21f99df6+8/e72N4h8fjzwO//G4bg36sn++P39vNf+/k9XDoZD3EIQ5xiIc07jdEc4hDHOIQh/gOxWGCP8QhDnGIhzQOE/whDnGIQzykcZjgD3GIQxziIY3DBH+IQxziEA9pHCb4QxziEId4SOMwwR/iEIc4xEMahwn+EIc4xCEe0vh/AUnYQqIjTPMgAAAAAElFTkSuQmCC\n",
            "text/plain": "<Figure size 432x288 with 1 Axes>"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "import torch\nfrom torchvision import transforms\n    \ndef preprocess(image_file):\n    \"\"\"Preprocess the input image.\"\"\"\n    data_transforms = transforms.Compose([\n        transforms.Resize(256),\n        transforms.CenterCrop(224),\n        transforms.ToTensor(),\n        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])\n    ])\n\n    image = Image.open(image_file)\n    image = data_transforms(image).float()\n    image = torch.tensor(image)\n    image = image.unsqueeze(0)\n    return image.numpy()",
      "execution_count": 30,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "input_data = preprocess('test_img.jpg')\nresult = service.run(input_data=json.dumps({'data': input_data.tolist()}))\nprint(result)",
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "stream",
          "text": "{'label': 'bees', 'probability': '0.7750759'}\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "## Clean up\nOnce you no longer need the web service, you can delete it with a simple API call."
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "service.delete()",
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "authors": [
      {
        "name": "minxia"
      }
    ],
    "kernelspec": {
      "name": "python36",
      "display_name": "Python 3.6",
      "language": "python"
    },
    "language_info": {
      "mimetype": "text/x-python",
      "nbconvert_exporter": "python",
      "name": "python",
      "pygments_lexer": "ipython3",
      "version": "3.6.6",
      "file_extension": ".py",
      "codemirror_mode": {
        "version": 3,
        "name": "ipython"
      }
    },
    "msauthor": "minxia"
  },
  "nbformat": 4,
  "nbformat_minor": 2
}