{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Training: Quantum case"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "raw_mimetype": "text/restructuredtext"
   },
   "source": [
    "In this tutorial we will train a ``lambeq`` :term:`model` to solve the relative pronoun classification task presented in [Lea2021]_. The goal is to predict whether a noun phrase contains a subject-based or an object-based relative clause. The entries of this dataset are extracted from the `RelPron` dataset [Rea2016]_.\n",
    "\n",
    "We will use an :py:class:`.IQPAnsatz` to convert :term:`string diagrams <string diagram>` into :term:`quantum circuits <quantum circuit>`. The pipeline uses :term:`tket` as a backend.\n",
    "\n",
    "If you have already gone through the `classical training tutorial <./trainer-classical.ipynb>`_, you will see that there are only minor differences for the quantum case.\n",
    "\n",
    ":download:`Download code <../_code/trainer-quantum.ipynb>`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preparation\n",
    "\n",
    "We start with importing NumPy and specifying some training hyperparameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import warnings\n",
    "\n",
    "warnings.filterwarnings('ignore')\n",
    "os.environ['TOKENIZERS_PARALLELISM'] = 'true'"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "raw_mimetype": "text/restructuredtext"
   },
   "source": [
    ".. note::\n",
    "\n",
    "   We disable warnings to filter out issues with the `tqdm` package used in jupyter notebooks. Furthermore, we have to specify whether we want to use parallel computation for the tokenizer used by the :py:class:`.BobcatParser`. None of the above impairs the performance of the code."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "BATCH_SIZE = 10\n",
    "EPOCHS = 100\n",
    "SEED = 2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Input data\n",
    "\n",
    "Let's read the data and print some example sentences."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def read_data(filename):\n",
    "    labels, sentences = [], []\n",
    "    with open(filename) as f:\n",
    "        for line in f:\n",
    "            t = int(line[0])\n",
    "            labels.append([t, 1-t])\n",
    "            sentences.append(line[1:].strip())\n",
    "    return labels, sentences\n",
    "\n",
    "\n",
    "train_labels, train_data = read_data('../examples/datasets/rp_train_data.txt')\n",
    "val_labels, val_data = read_data('../examples/datasets/rp_test_data.txt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "nbsphinx": "hidden"
   },
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "TESTING = int(os.environ.get('TEST_NOTEBOOKS', '0'))\n",
    "\n",
    "if TESTING:\n",
    "    train_labels, train_data = train_labels[:2], train_data[:2]\n",
    "    val_labels, val_data = val_labels[:2], val_data[:2]\n",
    "    EPOCHS = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['organization that church establish .',\n",
       " 'organization that team join .',\n",
       " 'organization that company sell .',\n",
       " 'organization that soldier serve .',\n",
       " 'organization that sailor join .']"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_data[:5]"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "raw_mimetype": "text/restructuredtext"
   },
   "source": [
    "Targets are represented as 2-dimensional arrays:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[[1, 0], [1, 0], [1, 0], [1, 0], [1, 0]]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_labels[:5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Creating and parameterising diagrams"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "raw_mimetype": "text/restructuredtext"
   },
   "source": [
    "The first step is to convert sentences into :term:`string diagrams <string diagram>`.\n",
    "\n",
    ".. note::\n",
    "\n",
    "   We know that the specific dataset only consists of noun phrases, hence, we reduce potential parsing errors by restricting the parser to only return parse trees with the root categories ``N`` (noun) and ``NP`` (noun phrase)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Tagging sentences.\n",
      "Parsing tagged sentences.\n",
      "Turning parse trees to diagrams.\n",
      "Tagging sentences.\n",
      "Parsing tagged sentences.\n",
      "Turning parse trees to diagrams.\n"
     ]
    }
   ],
   "source": [
    "from lambeq import BobcatParser\n",
    "\n",
    "parser = BobcatParser(root_cats=('NP', 'N'), verbose='text')\n",
    "\n",
    "raw_train_diagrams = parser.sentences2diagrams(train_data, suppress_exceptions=True)\n",
    "raw_val_diagrams = parser.sentences2diagrams(val_data, suppress_exceptions=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Filter and simplify diagrams"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "raw_mimetype": "text/restructuredtext"
   },
   "source": [
    "We simplify the diagrams by calling :py:meth:`~lambeq.backend.grammar.Diagram.normal_form` and filter out any diagrams that could not be parsed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_diagrams = [\n",
    "    diagram.normal_form()\n",
    "    for diagram in raw_train_diagrams if diagram is not None\n",
    "]\n",
    "val_diagrams = [\n",
    "    diagram.normal_form()\n",
    "    for diagram in raw_val_diagrams if diagram is not None\n",
    "]\n",
    "\n",
    "train_labels = [\n",
    "    label for (diagram, label)\n",
    "    in zip(raw_train_diagrams, train_labels)\n",
    "    if diagram is not None]\n",
    "val_labels = [\n",
    "    label for (diagram, label)\n",
    "    in zip(raw_val_diagrams, val_labels)\n",
    "    if diagram is not None\n",
    "]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's see the form of the diagram for a relative clause on the subject of a sentence:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 900x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_diagrams[0].draw(figsize=(9, 5), fontsize=12)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In object-based relative clauses the noun that follows the relative pronoun is the object of the sentence:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 900x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "train_diagrams[-1].draw(figsize=(9, 5), fontsize=12)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create circuits"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "raw_mimetype": "text/restructuredtext"
   },
   "source": [
    "In order to run the experiments on a quantum computer, we need to apply to string diagrams a quantum :term:`ansatz <ansatz (plural: ansätze)>`. For this experiment, we will use an :py:class:`.IQPAnsatz`, where noun wires (``n``) are represented by a one-qubit system, and sentence wires (``s``) are discarded (since we deal with noun phrases)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 900x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from lambeq import AtomicType, IQPAnsatz, RemoveCupsRewriter\n",
    "\n",
    "ansatz = IQPAnsatz({AtomicType.NOUN: 1, AtomicType.SENTENCE: 0},\n",
    "                   n_layers=1, n_single_qubit_params=3)\n",
    "remove_cups = RemoveCupsRewriter()\n",
    "\n",
    "train_circuits = [ansatz(remove_cups(diagram)) for diagram in train_diagrams]\n",
    "val_circuits =  [ansatz(remove_cups(diagram))  for diagram in val_diagrams]\n",
    "\n",
    "train_circuits[0].draw(figsize=(9, 10))"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "raw_mimetype": "text/restructuredtext"
   },
   "source": [
    "Note that we remove the :term:`cups <cup>` before parameterising the diagrams. By doing so, we reduce the number of :term:`post-selections <post-selection>`, which makes the model computationally more efficient. The effect of cups removal on a :term:`string diagram` is demonstrated below:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 3000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from lambeq.backend import draw_equation\n",
    "\n",
    "original_diagram = train_diagrams[0]\n",
    "removed_cups_diagram = remove_cups(original_diagram)\n",
    "\n",
    "draw_equation(original_diagram, removed_cups_diagram, symbol='-->', figsize=(30, 6), asymmetry=0.3, fontsize=14)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Training\n",
    "### Instantiate the model"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "raw_mimetype": "text/restructuredtext"
   },
   "source": [
    "We will use a :py:class:`.TketModel`, which we initialise by passing all diagrams to the class method\n",
    ":py:meth:`.TketModel.from_diagrams`. The :py:class:`.TketModel` needs a backend configuration dictionary passed as a keyword argument to the initialisation method. This dictionary must contain entries for ``backend``, ``compilation`` and ``shots``. The backend is provided by `pytket-extensions <https://github.com/CQCL/pytket-extensions>`_. In this example, we use :term:`Qiskit`\\ 's AerBackend with 8192 :term:`shots`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pytket.extensions.qiskit import AerBackend\n",
    "from lambeq import TketModel\n",
    "\n",
    "all_circuits = train_circuits + val_circuits\n",
    "\n",
    "backend = AerBackend()\n",
    "backend_config = {\n",
    "    'backend': backend,\n",
    "    'compilation': backend.default_compilation_pass(2),\n",
    "    'shots': 8192\n",
    "}\n",
    "\n",
    "model = TketModel.from_diagrams(all_circuits, backend_config=backend_config)"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "raw_mimetype": "text/restructuredtext"
   },
   "source": [
    ".. note::\n",
    "\n",
    "   The model can also be instantiated by calling :py:meth:`.TketModel.from_checkpoint`, in case a pre-trained checkpoint is available."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define loss and evaluation metric"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "raw_mimetype": "text/restructuredtext"
   },
   "source": [
    "We use standard binary cross-entropy as the loss. Optionally, we can provide a dictionary of callable evaluation metrics with the signature ``metric(y_hat, y)``."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from lambeq import BinaryCrossEntropyLoss\n",
    "\n",
    "# Using the builtin binary cross-entropy error from lambeq\n",
    "bce = BinaryCrossEntropyLoss()\n",
    "\n",
    "acc = lambda y_hat, y: np.sum(np.round(y_hat) == y) / len(y) / 2  # half due to double-counting\n",
    "eval_metrics = {\"acc\": acc}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Initialise trainer"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "raw_mimetype": "text/restructuredtext"
   },
   "source": [
    "In ``lambeq``, quantum pipelines are based on the :py:class:`.QuantumTrainer` class. Furthermore, we will use the standard ``lambeq`` SPSA optimizer, implemented in the :py:class:`.SPSAOptimizer` class. This needs three hyperameters:\n",
    "\n",
    "- ``a``: The initial learning rate (decays over time),\n",
    "- ``c``: The initial parameter shift scaling factor (decays over time),\n",
    "- ``A``: A stability constant, best choice is approx. 0.01 * number of training steps."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "from lambeq import QuantumTrainer, SPSAOptimizer\n",
    "\n",
    "trainer = QuantumTrainer(\n",
    "    model,\n",
    "    loss_function=bce,\n",
    "    epochs=EPOCHS,\n",
    "    optimizer=SPSAOptimizer,\n",
    "    optim_hyperparams={'a': 0.05, 'c': 0.06, 'A':0.001*EPOCHS},\n",
    "    evaluate_functions=eval_metrics,\n",
    "    evaluate_on_train=True,\n",
    "    verbose='text',\n",
    "    log_dir='RelPron/logs',\n",
    "    seed=0\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create datasets"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "raw_mimetype": "text/restructuredtext"
   },
   "source": [
    "To facilitate data shuffling and batching, ``lambeq`` provides a native :py:class:`.Dataset` class. Shuffling is enabled by default, and if not specified, the batch size is set to the length of the dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "from lambeq import Dataset\n",
    "\n",
    "train_dataset = Dataset(\n",
    "            train_circuits,\n",
    "            train_labels,\n",
    "            batch_size=BATCH_SIZE)\n",
    "\n",
    "val_dataset = Dataset(val_circuits, val_labels, shuffle=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "raw_mimetype": "text/restructuredtext"
   },
   "source": [
    "We can now pass the datasets to the :py:meth:`~lambeq.Trainer.fit` method of the trainer to start the training. Here, we perform early stopping if the validation accuracy doesn't improve within the specified `early_stopping_interval` epochs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Epoch 1:    train/loss: 0.9327   valid/loss: 3.2976   train/time: 13.22s   valid/time: 2.94s   train/acc: 0.5143   valid/acc: 0.5645\n",
      "Epoch 2:    train/loss: 2.1623   valid/loss: 1.4883   train/time: 14.20s   valid/time: 2.83s   train/acc: 0.5929   valid/acc: 0.6290\n",
      "Epoch 3:    train/loss: 0.2750   valid/loss: 1.7657   train/time: 13.35s   valid/time: 3.04s   train/acc: 0.6429   valid/acc: 0.7742\n",
      "Epoch 4:    train/loss: 1.0235   valid/loss: 2.0944   train/time: 12.91s   valid/time: 2.75s   train/acc: 0.5786   valid/acc: 0.5161\n",
      "Epoch 5:    train/loss: 0.7471   valid/loss: 1.1852   train/time: 13.02s   valid/time: 2.87s   train/acc: 0.6571   valid/acc: 0.7581\n",
      "Epoch 6:    train/loss: 0.9657   valid/loss: 2.0165   train/time: 12.81s   valid/time: 2.91s   train/acc: 0.5571   valid/acc: 0.4839\n",
      "Epoch 7:    train/loss: 0.8952   valid/loss: 1.7803   train/time: 12.73s   valid/time: 3.00s   train/acc: 0.5714   valid/acc: 0.7258\n",
      "Epoch 8:    train/loss: 3.7057   valid/loss: 2.4827   train/time: 12.86s   valid/time: 2.78s   train/acc: 0.4571   valid/acc: 0.6613\n",
      "Early stopping!\n",
      "Best model (epoch=3, step=21) saved to\n",
      "RelPron/logs/best_model.lt\n",
      "\n",
      "Training completed!\n",
      "train/time: 1m45s   train/time_per_epoch: 13.14s   train/time_per_step: 1.88s   valid/time: 23.12s   valid/time_per_eval: 2.89s\n"
     ]
    }
   ],
   "source": [
    "trainer.fit(train_dataset, val_dataset,\n",
    "            early_stopping_criterion='acc',\n",
    "            early_stopping_interval=5,\n",
    "            minimize_criterion=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Results\n",
    "\n",
    "Finally, we visualise the results and evaluate the model on the test data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Validation accuracy: 0.7419354838709677\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1cAAAIjCAYAAADvBuGTAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAADXZ0lEQVR4nOzdd3QU1f/G8fem94QEQgIEQq+h9450FSmioCJFsIKKiCiKgBV7QxT9Wig2FBAVpPfee28hlAQIkN6z8/sDsz8jNbDJpDyvc/Ycd3Zm7rMT3NnPzp17LYZhGIiIiIiIiMhtcTA7gIiIiIiISGGg4kpERERERMQOVFyJiIiIiIjYgYorERERERERO1BxJSIiIiIiYgcqrkREREREROxAxZWIiIiIiIgdqLgSERERERGxAxVXIiIiIiIidqDiSsREAwcOJDQ09Ja2HT9+PBaLxb6BRESkSAkPD8disTBlyhSzo4gUCiquRK7CYrHc1GPFihVmRy3wfvrpJz755BOzY4iI2M2UKVOynSvc3NwoVaoUnTt35rPPPiM+Pt7siAK8/fbbzJkzJ8/bPXPmDOPHj2fHjh153rbkPothGIbZIUTymx9++CHb82nTprF48WKmT5+ebXnHjh0pWbLkLbeTnp6O1WrF1dU1x9tmZGSQkZGBm5vbLbefH9x9993s2bOH8PBws6OIiNjFlClTGDRoEK+//jrly5cnPT2dqKgoVqxYweLFiylbtix//vkntWvXNjsq4eHhlC9fnu+//56BAweaHSdPeXl50bt37zy/ardlyxYaNWpUJI95UeBkdgCR/Khfv37Znm/YsIHFixdfsfy/kpKS8PDwuOl2nJ2dbykfgJOTE05O+l9YRCS/6tq1Kw0bNrQ9Hz16NMuWLePuu+/mnnvuYf/+/bi7u5uYUETsTd0CRW5R27ZtqVWrFlu3bqV169Z4eHjw8ssvA/DHH39w1113UapUKVxdXalYsSJvvPEGmZmZ2fbx33uusvq+f/DBB3z99ddUrFgRV1dXGjVqxObNm7Nte7V7riwWC8OGDWPOnDnUqlULV1dXatasyYIFC67Iv2LFCho2bIibmxsVK1bkq6++uun7uA4fPsy9995LUFAQbm5ulClThr59+xIbG5ttvR9++IEGDRrg7u6Ov78/ffv25eTJk9mO4bx58zhx4oSt+8yt3oMmIlIQ3HHHHbz66qucOHHiil4SBw4coHfv3vj7++Pm5kbDhg35888/ba9v2bIFi8XC1KlTr9jvwoULsVgszJ0717bs9OnTPPLII5QsWdJ2Pvjuu+9uKueyZcto1aoVnp6e+Pn50b17d/bv359tnaxzxoEDB7j//vvx8fEhICCAZ599lpSUlGzrZp2ffvvtN2rUqIG7uzvNmjVj9+7dAHz11VdUqlQJNzc32rZte9XeDBs3bqRLly74+vri4eFBmzZtWLt27VUzHTlyhIEDB+Ln54evry+DBg0iKSkpW57ExESmTp1qO//c6CrSxIkTqVmzJh4eHhQrVoyGDRvy008/ZVvnRsd8xYoVNGrUCIBBgwbZ2tY9b4WHfvYWuQ0XLlyga9eu9O3bl379+tm6CE6ZMgUvLy9GjBiBl5cXy5YtY+zYscTFxfH+++/fcL8//fQT8fHxPP7441gsFt577z169erFsWPHbni1a82aNcyePZunnnoKb29vPvvsM+69914iIiIICAgAYPv27XTp0oXg4GBee+01MjMzef311ylRosQNs6WlpdG5c2dSU1N5+umnCQoK4vTp08ydO5eYmBh8fX0BeOutt3j11Ve5//77GTJkCOfPn2fixIm0bt2a7du34+fnxyuvvEJsbCynTp3i448/Bi530xARKcwefvhhXn75ZRYtWsSjjz4KwN69e2nRogWlS5fmpZdewtPTk19//ZUePXowa9YsevbsScOGDalQoQK//vorAwYMyLbPGTNmUKxYMTp37gzA2bNnadq0qa2oKVGiBPPnz2fw4MHExcUxfPjwa+ZbsmQJXbt2pUKFCowfP57k5GQmTpxIixYt2LZt2xU/gt1///2EhoYyYcIENmzYwGeffcalS5eYNm1atvVWr17Nn3/+ydChQwGYMGECd999N6NGjeKLL77gqaee4tKlS7z33ns88sgjLFu2zLbtsmXL6Nq1Kw0aNGDcuHE4ODjw/fffc8cdd7B69WoaN258Raby5cszYcIEtm3bxjfffENgYCDvvvsuANOnT2fIkCE0btyYxx57DICKFSte85j873//45lnnqF379624nHXrl1s3LiRBx988KaPefXq1Xn99dcZO3Ysjz32GK1atQKgefPm12xbChhDRG5o6NChxn//d2nTpo0BGJMnT75i/aSkpCuWPf7444aHh4eRkpJiWzZgwACjXLlytufHjx83ACMgIMC4ePGibfkff/xhAMZff/1lWzZu3LgrMgGGi4uLceTIEduynTt3GoAxceJE27Ju3boZHh4exunTp23LDh8+bDg5OV2xz//avn27ARi//fbbNdcJDw83HB0djbfeeivb8t27dxtOTk7Zlt91113ZjoGISEH3/fffG4CxefPma67j6+tr1KtXz/a8ffv2RlhYWLZzhNVqNZo3b25UrlzZtmz06NGGs7NztnNEamqq4efnZzzyyCO2ZYMHDzaCg4ON6OjobO327dvX8PX1tZ2nss4733//vW2dunXrGoGBgcaFCxdsy3bu3Gk4ODgY/fv3ty3LOg/dc8892dp46qmnDMDYuXOnbRlguLq6GsePH7ct++qrrwzACAoKMuLi4rK9R8C2rtVqNSpXrmx07tzZsFqttvWSkpKM8uXLGx07drwi07+PhWEYRs+ePY2AgIBsyzw9PY0BAwYYN6N79+5GzZo1r7vOzR7zzZs3X3HMpfBQt0CR2+Dq6sqgQYOuWP7vPvTx8fFER0fTqlUrkpKSOHDgwA3326dPH4oVK2Z7nvXL1rFjx264bYcOHbL9+la7dm18fHxs22ZmZrJkyRJ69OhBqVKlbOtVqlSJrl273nD/WVemFi5cmK2Lxb/Nnj0bq9XK/fffT3R0tO0RFBRE5cqVWb58+Q3bEREpzLy8vGyjBl68eJFly5Zx//33284Z0dHRXLhwgc6dO3P48GFOnz4NXD4/pKenM3v2bNu+Fi1aRExMDH369AHAMAxmzZpFt27dMAwj2+dw586diY2NZdu2bVfNFRkZyY4dOxg4cCD+/v625bVr16Zjx478/fffV2yTdSUqy9NPPw1wxbrt27fPdtWrSZMmANx77714e3tfsTzrvLVjxw4OHz7Mgw8+yIULF2zvJTExkfbt27Nq1SqsVmu2tp544olsz1u1asWFCxeIi4u76vu+ET8/P06dOnVFF/0st3PMpXBRt0CR21C6dGlcXFyuWL53717GjBnDsmXLrvgg/+99SVdTtmzZbM+zCq1Lly7leNus7bO2PXfuHMnJyVSqVOmK9a627L/Kly/PiBEj+Oijj/jxxx9p1aoV99xzD/369bMVXocPH8YwDCpXrnzVfdzOQB4iIoVBQkICgYGBABw5cgTDMHj11Vd59dVXr7r+uXPnKF26NHXq1KFatWrMmDGDwYMHA5e7BBYvXpw77rgDgPPnzxMTE8PXX3/N119/fc39Xc2JEycAqFq16hWvVa9enYULF5KYmIinp6dt+X8/6ytWrIiDg8MV90399/yUdc4ICQm56vKs89bhw4cBrugK+W+xsbHZfpS83nnUx8fnmvu5lhdffJElS5bQuHFjKlWqRKdOnXjwwQdp0aIFcHvHXAoXFVcit+FqozzFxMTQpk0bfHx8eP3116lYsSJubm5s27aNF1988Ypf167G0dHxqsuNm5g54Xa2vVkffvghAwcO5I8//mDRokU888wztr72ZcqUwWq1YrFYmD9//lXz6L4qESnKTp06RWxsrO0HrazzwsiRI233TP3Xv3/86tOnD2+99RbR0dF4e3vz559/8sADD9hGkM3aX79+/a5ZkOTmMPDXGhjpWuenG523st7P+++/T926da+67n/PK/Y+F1avXp2DBw8yd+5cFixYwKxZs/jiiy8YO3Ysr732munHXPIPFVcidrZixQouXLjA7Nmzad26tW358ePHTUz1/wIDA3Fzc+PIkSNXvHa1ZdcSFhZGWFgYY8aMYd26dbRo0YLJkyfz5ptvUrFiRQzDoHz58lSpUuW6+7mZ0QlFRAqTrDkTswqpChUqAJev6nfo0OGG2/fp04fXXnuNWbNmUbJkSeLi4ujbt6/t9RIlSuDt7U1mZuZN7e/fypUrB8DBgweveO3AgQMUL14821UruHxlqXz58rbnR44cwWq12m3016yu7j4+Pjl+P9eT0/OPp6cnffr0oU+fPqSlpdGrVy/eeustRo8enaNjrvNe4aZ7rkTsLOvXsn//OpaWlsYXX3xhVqRsHB0d6dChA3PmzOHMmTO25UeOHGH+/Pk33D4uLo6MjIxsy8LCwnBwcCA1NRWAXr164ejoyGuvvXbFr4SGYXDhwgXbc09Pz5vqKikiUhgsW7aMN954g/Lly/PQQw8Bl3/0atu2LV999RWRkZFXbHP+/Plsz6tXr05YWBgzZsxgxowZBAcHZ/sxz9HRkXvvvZdZs2axZ8+eG+7v34KDg6lbty5Tp04lJibGtnzPnj0sWrSIO++884ptJk2alO35xIkTAW7qPt6b0aBBAypWrMgHH3xAQkLCFa9f7/1cj6enZ7b3eD3/Pm8BuLi4UKNGDQzDID09PUfHPKs4vdm2pWDRlSsRO2vevDnFihVjwIABPPPMM1gsFqZPn27Xbnm3a/z48SxatIgWLVrw5JNPkpmZyeeff06tWrXYsWPHdbddtmwZw4YN47777qNKlSpkZGQwffp024kFLv/K+OabbzJ69GjCw8Pp0aMH3t7eHD9+nN9//53HHnuMkSNHApdPmjNmzGDEiBE0atQILy8vunXrltuHQEQk182fP58DBw6QkZHB2bNnWbZsGYsXL6ZcuXL8+eefuLm52dadNGkSLVu2JCwsjEcffZQKFSpw9uxZ1q9fz6lTp9i5c2e2fffp04exY8fi5ubG4MGDcXDI/nv5O++8w/Lly2nSpAmPPvooNWrU4OLFi2zbto0lS5Zw8eLFa+Z+//336dq1K82aNWPw4MG2odh9fX0ZP378FesfP36ce+65hy5durB+/Xp++OEHHnzwQerUqXN7B/AfDg4OfPPNN3Tt2pWaNWsyaNAgSpcuzenTp1m+fDk+Pj789ddfOd5vgwYNWLJkCR999BGlSpWifPnytsE0/qtTp04EBQXRokULSpYsyf79+/n888+56667bINx3Owxr1ixIn5+fkyePBlvb288PT1p0qRJtqt/UoDl/QCFIgXPtYZiv9awrGvXrjWaNm1quLu7G6VKlTJGjRplLFy40ACM5cuX29a71lDs77///hX7BIxx48bZnl9rKPahQ4desW25cuWuGG526dKlRr169QwXFxejYsWKxjfffGM8//zzhpub2zWOwmXHjh0zHnnkEaNixYqGm5ub4e/vb7Rr185YsmTJFevOmjXLaNmypeHp6Wl4enoa1apVM4YOHWocPHjQtk5CQoLx4IMPGn5+fgagYdlFpMDLGoo96+Hi4mIEBQUZHTt2ND799NNsw47/29GjR43+/fsbQUFBhrOzs1G6dGnj7rvvNmbOnHnFuocPH7btf82aNVfd39mzZ42hQ4caISEhhrOzsxEUFGS0b9/e+Prrr23rXG0odsMwjCVLlhgtWrQw3N3dDR8fH6Nbt27Gvn37sq2TdR7at2+f0bt3b8Pb29soVqyYMWzYMCM5OTnbulc7P13rnLd8+fKrTvmxfft2o1evXkZAQIDh6upqlCtXzrj//vuNpUuXXpHp/Pnz2bbN+pv8eyj4AwcOGK1btzbc3d0N4LrDsn/11VdG69atbW1XrFjReOGFF4zY2Nhs693MMTeMy1Os1KhRwzYFioZlLzwshpGPfk4XEVP16NGDvXv32kZmEhERuZbx48fz2muvcf78eYoXL252HJF8QfdciRRRycnJ2Z4fPnyYv//+m7Zt25oTSERERKSA0z1XIkVUhQoVGDhwIBUqVODEiRN8+eWXuLi4MGrUKLOjiYiIiBRIKq5EiqguXbrw888/ExUVhaurK82aNePtt9++5sS/IiIiInJ9uudKRERERETEDnTPlYiIiIiIiB2ouBIREREREbED3XN1FVarlTNnzuDt7Y3FYjE7johIkWEYBvHx8ZQqVeqKSVGLOp2bRETMkZNzk4qrqzhz5gwhISFmxxARKbJOnjxJmTJlzI6Rr+jcJCJirps5N6m4ugpvb2/g8gH08fExOY2ISNERFxdHSEiI7XNY/p/OTSIi5sjJuUnF1VVkdbfw8fHRCUxExATq9nYlnZtERMx1M+cmdWgXERERERGxAxVXIiIiIiIidqDiSkRERERExA5UXImIiIiIiNiBiisREZEiwmpYzY4gIlKoqbgSEREp5LZEbeGRhY/w5oY3zY4iIlKoqbgSEREp5DKNTDZHbWbRiUWkZ6abHUdEpNBScSUiIlLINSzZkAC3AGJTY1kfud7sOCIihZaKKxERkULO0cGRjuU6ArAwfKHJaURECi8VVyIiIkVA1/JdAVgWsYzUzFST04iIFE4qrkRERIqAuoF1CfQIJCE9gbWn15odR0SkUFJxJSIiUgQ4WBzoHNoZgAXhC0xOIyJSOKm4EhERKSK6hHYBYMXJFSRnJJsbRkSkEFJxJSIiUkSEFQ+jtFdpkjOSWXVqldlxREQKHRVXIiIiRYTFYrF1DdSogSIi9qfiSkREpAjJ6hq46tQqEtMTTU4jIlK4qLgSEREpQqr5V6OcTzlSM1NZcXKF2XFERAoVFVciIiJFyL+7Bi44rlEDRUTsScWViIhIEZPVNXDNmTXEpcWZnEZEpPBQcSUiIlLEVC5WmUp+lciwZrAsYpnZcURECg0VVyIiIkWQJhQWEbE/FVciIiJFUFbXwI1nNnIp5ZLJaURECgcVVyIiIkVQqG8o1fyrkWFksCRiidlxREQKBRVXIiIiRZRtQuHjmlBYRMQe8nVx9eWXX1K7dm18fHzw8fGhWbNmzJ8//7rb/Pbbb1SrVg03NzfCwsL4+++/8yitiIhIwZLVNXDz2c1EJ0ebnEZEpODL18VVmTJleOedd9i6dStbtmzhjjvuoHv37uzdu/eq669bt44HHniAwYMHs337dnr06EGPHj3Ys2dPHicXERHJ/8p4lyGseBhWw8riE4vNjiMiUuBZDMMwzA6RE/7+/rz//vsMHjz4itf69OlDYmIic+fOtS1r2rQpdevWZfLkyTfdRlxcHL6+vsTGxuLj42OX3CIicmP6/L223Do2U/dO5YMtH1A/sD5Tu061235FRAqLnHz+5usrV/+WmZnJL7/8QmJiIs2aNbvqOuvXr6dDhw7ZlnXu3Jn169dfd9+pqanExcVle4iIiJgpr85NWfddbT+3najEqFxpQ0SkqMj3xdXu3bvx8vLC1dWVJ554gt9//50aNWpcdd2oqChKliyZbVnJkiWJirr+yWLChAn4+vraHiEhIXbLLyIicivy6twU5BlEvcB6GBgsCl+UK22IiBQV+b64qlq1Kjt27GDjxo08+eSTDBgwgH379tm1jdGjRxMbG2t7nDx50q77FxERyam8PDdlDWyxMFyjBoqI3I58X1y5uLhQqVIlGjRowIQJE6hTpw6ffvrpVdcNCgri7Nmz2ZadPXuWoKCg67bh6upqG5Ew6yEiImKmvDw3dQrthIPFgV3RuzidcDrX2hERKezyfXH1X1arldTU1Ku+1qxZM5YuXZpt2eLFi695j5aIiIhAcffiNCzZENDVKxGR25Gvi6vRo0ezatUqwsPD2b17N6NHj2bFihU89NBDAPTv35/Ro0fb1n/22WdZsGABH374IQcOHGD8+PFs2bKFYcOGmfUWRERECoSsgS0WHF9gchIRkYIrXxdX586do3///lStWpX27duzefNmFi5cSMeOHQGIiIggMjLStn7z5s356aef+Prrr6lTpw4zZ85kzpw51KpVy6y3ICIiUiB0LNcRR4sj+y/u50TcCbPjiIgUSAVunqu8oHlWRETMoc/fa8uLY/PE4idYe2Ytw+oO4/E6j+dKGyIiBU2hnOdKREREcpeta2C4ugaKiNwKFVciIiICwB1l78DJwYkjMUc4GnPU7DgiIgWOiisREREBwNfVlxalWgC6eiUicitUXImIiIjNv0cN1G3ZIiI5o+JKREREbNqFtMPV0ZXwuHAOXjpodhwRkQJFxZWIiIjYeLl40ap0K0BzXomI5JSKKxEREcmmc/n/HzVQXQNFRG6eiisRERHJpnXp1rg7uXM64TR7L+w1O46ISIGh4kpERESy8XD2oE2ZNgDMPz7f5DQiIgWHiisRERG5QpfyXQBYGL4Qq2E1OY2ISMGg4kpERESu0LJ0S7ycvTibdJad53eaHUdEpEBQcSUiIiJXcHV0pV1IO0CjBoqI3CwVVyIiInJVWV0DF51YRKY10+Q0IiL5n4orERERuapmwc3wcfEhOjmabee2mR1HRCTfU3ElIiIiV+Xs6Ez7su0BjRooInIzVFyJiIjINWV1DVxyYgkZ1gyT04iI5G8qrkREROSaGgc1xt/Nn0upl9gUucnsOCIi+ZqKKxEREbkmJwcnOpTtAMCCcI0aKCJyPSquRERE5LpsXQMjlpCemW5yGhGR/EvFlYiIiFxX/cD6FHcvTnxaPOvOrDM7johIvqXiSkRERK7L0cGRzqGdAXUNFBG5HhVXIiIickNdQi93DVx+cjmpmakmpxERyZ9UXImIiMgN1S5RmyDPIBLTE1lzao3ZcURE8iUVVyIiInJDDhYHOpdT10ARketRcSUiIiI3JWvUwJWnVpKUnmRyGhGR/EfFlYiIiNyUmgE1KeNVhuSMZFadWmV2HBGRfEfFlYiIiNwUi8Viu3qlroEiIlfK18XVhAkTaNSoEd7e3gQGBtKjRw8OHjx43W2mTJmCxWLJ9nBzc8ujxCIiIoVb1qiBq0+tJiEtweQ0IiL5S74urlauXMnQoUPZsGEDixcvJj09nU6dOpGYmHjd7Xx8fIiMjLQ9Tpw4kUeJRURECrcqxaoQ6hNKmjWN5SeXmx1HRCRfcTI7wPUsWJC9y8GUKVMIDAxk69attG7d+prbWSwWgoKCbrqd1NRUUlP/f86OuLi4nIcVERGxo/x6bsrqGjh552QWhC+gW8VuZkcSEck38vWVq/+KjY0FwN/f/7rrJSQkUK5cOUJCQujevTt79+697voTJkzA19fX9ggJCbFbZhERkVuRn89NWV0D151ZR2xqrMlpRESub+3ptWw/tz1P2iowxZXVamX48OG0aNGCWrVqXXO9qlWr8t133/HHH3/www8/YLVaad68OadOnbrmNqNHjyY2Ntb2OHnyZG68BRERkZuWn89NFf0qUrlYZTKsGSyLWGZ2HBGRa9p3YR/PrXiOIQuHsPP8zlxvL193C/y3oUOHsmfPHtasuf6s8M2aNaNZs2a2582bN6d69ep89dVXvPHGG1fdxtXVFVdXV7vmFRERuR35/dzUJbQLhy8dZkH4AnpW7ml2HBGRK5xJOMPQpUNJzkimaXBTavjXyPU2C8SVq2HDhjF37lyWL19OmTJlcrSts7Mz9erV48iRI7mUTkREpOjJ6hq4MXIjF1MumpxGRCS72NRYnlzyJNHJ0VQuVpmP2n6Es6Nzrrebr4srwzAYNmwYv//+O8uWLaN8+fI53kdmZia7d+8mODg4FxKKiIgUTWV9ylLdvzqZRiZLTiwxO46IiE1aZhrDlw/nWOwxAj0C+aL9F3i7eOdJ2/m6uBo6dCg//PADP/30E97e3kRFRREVFUVycrJtnf79+zN69Gjb89dff51FixZx7Ngxtm3bRr9+/Thx4gRDhgwx4y2IiIgUWl3LdwU0obCI5B9Ww8qYNWPYcnYLns6efNH+C4I8b34U8duVr4urL7/8ktjYWNq2bUtwcLDtMWPGDNs6ERERREZG2p5funSJRx99lOrVq3PnnXcSFxfHunXrqFEj9/tYioiIFCWdQzsDsCVqC+eTzpucRkQEPt32KfPD5+NkceKjth9R1b9qnrafrwe0MAzjhuusWLEi2/OPP/6Yjz/+OJcSiYiISJZSXqWoXaI2u87vYtGJRTxU/SGzI4lIETbjwAy+2/MdAOObj6d5qeZ5niFfX7kSERGR/C1rYIuF4QtNTiIiRdmKkyt4e9PbAAytO5TulbqbkkPFlYiIiNyyTuU6YcHC9nPbiUqMMjuOiBRBe6L3MGrVKKyGlV6Ve/F47cdNy6LiSkRERG5ZSc+S1C9ZH9DVKxHJeyfjT9rmsmpRqgVjmo7BYrGYlkfFlYiIiNyWrK6BC45r1EARyTsxKTE8teQpLqZcpJp/NT5s+yHODrk/l9X1qLgSERGR29KhXAccLA7subCHk/EnzY4jIkVAamYqzyx/hvC4cII9g5nUfhKezp5mx1JxJSIiIrenuHtxGgU1AtQ1UERyn9Ww8vLql9l+bjvezt580f4LAj0CzY4FqLjKNRmZVrMjiIiI5BmNGigieeWjLR+x6MQinByc+KTdJ1QqVsnsSDYqruwsMTWD8X/uZeD3m29qni4REZHCoEPZDjhZnDhw8QDHY4+bHUdECqkf9//I1H1TAXizxZs0Dm5scqLsVFzZWXRCKr9sjmDNkWhmbFa/cxERKRr83PxoWqopAAvCNbCFiNjf0hNLeXfTuwA8W/9Z7qpwl8mJrqTiys7KBXgyslNVAN6at5+o2BSTE4mIiOQNW9fA4+oaKCL2tfP8Tl5c/SIGBvdVuY/BtQabHemqVFzlgkEtylM3xI/41AzGzNmt7oEiIlIk3FH2DpwdnDkae5TDlw6bHUdEComIuAieXvo0qZmptC7TmpebvGzqXFbXo+IqFzg6WHivd22cHS0s2X+OP3eeMTuSiIhIrvN28aZF6RaAugaKiH1cTLnIk0ue5FLqJWoE1OD91u/j5OBkdqxrUnGVS6qU9GZYu8oAvPbXPi4kpJqcSEREJPd1De0KXJ5QWD03ROR2JGck8/Syp4mIj6C0V2kmtZ+Eh7OH2bGuS8VVLnqybUWqBXlzMTGN8X/tMzuOiIhIrmsb0hY3Rzci4iPYf3G/2XFEpIDKtGYyevVodp3fhY+LD190+ILi7sXNjnVDKq5ykYuTA+/1ro2DBf7aeYbF+86aHUlERCRXeTh70KpMK0BdA0Xk1hiGwftb3mdpxFJcHFz47I7PqOBbwexYN0XFVS6rXcaPR1tf/sfwyu+7iU1ONzmRiIhI7soaNXBR+CJ1DRSRHJu2bxo/7v8RgLdavUWDkg1MTnTzVFzlgec6VKF8cU/Oxacy4W91kRARkcKtVZlWuDu5czrhNLujd5sdR0QKkIXhC/lgywcAPN/geduPNQVFrhVXJ0+e5NSpU7bnmzZtYvjw4Xz99de51WS+5ebsyLv31gbgl80nWXsk2uREIiIiucfdyZ12Ie0AdQ0UkZu37ew2Xl79MgAPVHuAATUHmJwo53KtuHrwwQdZvnw5AFFRUXTs2JFNmzbxyiuv8Prrr+dWs/lW4/L+9G9WDoCXZu8iKS3D5EQiIiK5xzahcPhCrIbV5DQikt8djz3OM8ufIc2aRruQdrzY6MV8O5fV9eRacbVnzx4aN24MwK+//kqtWrVYt24dP/74I1OmTMmtZvO1UV2qUdrPnZMXk3l/4UGz44iIiOSaFqVb4O3szbmkc2w/t93sOCKSj0UnR/PkkieJTY2ldvHavNv6XRwdHM2OdUtyrbhKT0/H1dUVgCVLlnDPPfcAUK1aNSIjI3Or2XzNy9WJt3uFATBlXThbT1w0OZGIiEjucHF0oV3Zf7oGHlfXQBG5uqT0JIYtHcbphNOEeIcwsf1E3J3czY51y3KtuKpZsyaTJ09m9erVLF68mC5dLncPOHPmDAEBAbnVbL7XpkoJ7q1fBsOAUTN3kZKeaXYkERGRXJHVNXDxicVkWnW+y88Mw2DdmXVsO7vN7ChShGRYMxi1ahR7L+zFz9WPLzt8ib+bv9mxbkuuFVfvvvsuX331FW3btuWBBx6gTp06APz555+27oJF1at3V6e4lytHzycycdlhs+OIiIjkiqalmuLr6suFlAtsObvF7DhyDZujNtNvfj8eX/w4gxYOYse5HWZHkiLAMAze2fQOK0+txNXRlYl3TKScTzmzY922XCuu2rZtS3R0NNHR0Xz33Xe25Y899hiTJ0/OrWYLBD8PF97sUROAySuPsfdMrMmJRERE7M/ZwZkOZTsAMP/4fJPTyH8duHiAJ5Y8wSMLH2HX+V0AWA0rY9aOITkj2eR0Uth9v/d7ZhycgQUL77R6h7qBdc2OZBe5VlwlJyeTmppKsWLFADhx4gSffPIJBw8eJDAwMLeaLTC61ArmzrAgMq0Go2buIj1TIymJiEjh06X85a6BSyKWkG5NNzmNAJyMO8moVaO476/7WHt6LU4WJ/pU7cMfPf4g0COQE3En+HTbp2bHlELs72N/8/HWjwEY1WgUHcp1MDmR/eRacdW9e3emTZsGQExMDE2aNOHDDz+kR48efPnll7nVbIHy2j218PNwZu+ZOL5edczsOCIiInbXsGRD/N38iU2NZWPkRrPjFGnRydG8teEt7plzj+1KYtfyXfmzx5+MaTqGCr4VeL355elyftz/I5siN5kZVwqpzVGbGbN2DAAP13iYfjX6mZzIvnKtuNq2bRutWrUCYObMmZQsWZITJ04wbdo0Pvvss5vax4QJE2jUqBHe3t4EBgbSo0cPDh688RDmv/32G9WqVcPNzY2wsDD+/vvv23ovuaWEtytj764BwKdLD3PkXILJiUREROzLycGJjuU6Aho10CzxafFM3D6RO2ffyS8HfyHDyKBF6Rb81u033mv9HiE+IbZ1W5RuQe8qvQEYu24siemJZsWWQuhozFGeXf4s6dZ0OpbryMiGI82OZHe5VlwlJSXh7e0NwKJFi+jVqxcODg40bdqUEydO3NQ+Vq5cydChQ9mwYQOLFy8mPT2dTp06kZh47f/R161bxwMPPMDgwYPZvn07PXr0oEePHuzZs8cu78veetYrTZsqJUjLsPLirF1kWg2zI4mIiNhV1qiByyKWkZaZZnKaoiM1M5Wpe6dy5+w7+XrX1yRnJFO7eG2+6/wdkztMppp/tatuN7LhSEp7leZ0wmk+2PJBHqcuGk7GneSN9W/w19G/isz9beeTzvPkkieJT4unbom6vN3ybRwsuVaKmMZiGEaufJuvXbs2Q4YMoWfPntSqVYsFCxbQrFkztm7dyl133UVUVFSO93n+/HkCAwNZuXIlrVu3vuo6ffr0ITExkblz59qWNW3alLp16970QBpxcXH4+voSGxuLj49PjnPm1OmYZDp9tJLEtEzGdavBoBblc71NEZH8KK8/fwuSgnxsrIaVjr915FzyOT5r95lt/ivJHRnWDP46+hdf7PyCqMTL37fK+5bn2XrPckfZO7BYLDfcx+aozTyy8BEAvuzwJS1Lt8zVzEVJUnoSD857kKOxRwHwcvaia/mu9KzUk1rFa93U36egSUpPYuCCgey/uJ9yPuWY3nU6xdyKmR3rpuXk8zfXysWxY8cycuRIQkNDady4Mc2aNQMuX8WqV6/eLe0zNvbyqHr+/tce/379+vV06JD9prjOnTuzfv36a26TmppKXFxctkdeKu3nzkt3VgfgvQUHOXkxKU/bFxGR/Mfsc5M9OVgc6BTaCYAF4eoamFsMw2BpxFLu/fNexq4bS1RiFCU9SvJ689eZfc9s2pdrf9Nf3BsFNaJf9cv3woxbN47YVI1sbC/vbHqHo7FH8Xfzp7RXaRLSE/jt0G88+PeD9PqzF1P3TuVC8gWzY9pNhjWD51c+z/6L+/F38+fL9l8WqMIqp3KtuOrduzcRERFs2bKFhQsX2pa3b9+ejz/+OMf7s1qtDB8+nBYtWlCrVq1rrhcVFUXJkiWzLStZsuR1r5RNmDABX19f2yMkJOSa6+aWhxqXpXF5f5LTMxk9eze5dEFRREQKiPxwbrKnrFEDV5xcQUpGirlhCqGsuaqGLx/Osdhj+Lr6MrLhSOb1mkfPyj1xcnDK8T6fqf8M5XzKcS7pHO9uejcXUhc9fx39i9+P/I6DxYEP2nzA373+5rvO39GtQjfcHN04EnOED7Z8QIffOjB8+XBWnlxJhjXD7Ni3zDAM3tzwJmtOr8HN0Y3P7/g82z1+hVGudnQMCgqiXr16nDlzhlOnTgHQuHFjqlW7eh/f6xk6dCh79uzhl19+sXdMRo8eTWxsrO1x8uRJu7dxIw4OFt69tzauTg6sORLNb1tO5XkGERHJP/LDucmeahevTSnPUiRlJLH69Gqz4xQa/52ryt3JnUfDHmV+r/kMqDkAV0fXW963u5M7b7Z4EweLA38d+4ulEUvtmLzoCY8N540NbwDwRO0naBTUCAeLA42CGvF2q7dZdv8yXm36KmHFw8gwMlgasZRhy4bRaWYnPtn6CeGx4ea+gVvwv93/Y9bhWThYHHiv9XuElQgzO1Kuy7Xiymq18vrrr+Pr60u5cuUoV64cfn5+vPHGG1itOZvTadiwYcydO5fly5dTpkyZ664bFBTE2bNnsy07e/YsQUFB19zG1dUVHx+fbA8zlC/uyfOdqgDwxrx9nI3TL3siIkVVfjk32YvFYqFzaGdAowbaw8n4k7y46sUr5qqa13Mez9R/Bm8Xb7u0UzewLgNrDgTg9fWvcynlkl32W9SkZqYycuVIkjOSaRTUiMdqP3bFOt4u3txf9X5+uusnZt8zm/41+uPv5s/55PN8u+dbus3pxoD5A/j98O8kpef/W0j+OvoXE7dPBOClxi8VmXstc624euWVV/j8889555132L59O9u3b+ftt99m4sSJvPrqqze1D8MwGDZsGL///jvLli2jfPkbD/TQrFkzli7N/svK4sWLbfd85XePtChPnTK+xKdkMGbOHnUPFBGRQiOra+CqU6sKxJfD/Mg2V9Xv9/D38ctTzXQt35U/evzBmKZjKOFRwu5tDq07lEp+lbiYcpE3Nryh7ya34IPNH3Dw0kH83fx5p9U7ODo4Xnf9ysUq80KjF1jSewmftP2ENmXa4GBxYNu5bYxdN5a2v7Zl7NqxbD+3PV/+PTZGbmTsurEADKo5iAeqPWByoryTa6MFlipVismTJ3PPPfdkW/7HH3/w1FNPcfr06Rvu46mnnuKnn37ijz/+oGrVqrblvr6+uLu7A9C/f39Kly7NhAkTgMtDsbdp04Z33nmHu+66i19++YW3336bbdu2XfderX8ze0SmA1FxdJu4hvRMg4kP1KNbnVJ5nkFExAxmf/7mZ4Xh2BiGwd2/301EfATvtnqXOyvcaXakAiM+LZ4pe6cwfd9029DdLUq34Nl6z1I9oHqut7/vwj4emvcQGUYG77V+j67lu+Z6m4XF4hOLGbFiBHB7Iy+eSzrHn0f/ZM6ROZyI+/9pjUJ9QulRqQf3VLwnV4rrnDp06RAD5g8gIT2BrqFdeaf1OwV+yPV8MVrgxYsXr3pvVbVq1bh48eJN7ePLL78kNjaWtm3bEhwcbHvMmDHDtk5ERASRkZG2582bN+enn37i66+/pk6dOsycOZM5c+bcdGGVH1QL8uGptpUAGP/nXi4mak4QEREp+LJ1DdSogTflRnNV5UVhBVAjoIatK9tbG9/ifNL5PGm3oDsVf4pxa8cB8EitR25rSPtAj0CGhA3hrx5/MbXLVHpU6oG7kzvhceF8su0TOs7syNNLn2ZpxFLSren2egs5cjbxLE8teYqE9AQalGzAmy3fLPCFVU7l2pWrJk2a0KRJEz777LNsy59++mk2bdrExo0bc6NZu8gPvw6mZVjpNnENB8/G071uKT7te2vD14uIFCT54fM3vyosx+bQpUPc++e9ODs4s7LPSrvdG1TYZFoz+fPon7c1V5W9pVvTeWjeQ+y/uJ82Zdow8Y6JhXJOJntJz0xnwIIB7I7eTZ0Sdfi+y/c4OzjbtY3E9EQWhS/i9yO/s/3cdttyfzd/ulXoRs/KPanoV9GubV5LQloCAxYM4NClQ5T3Lc/0rtPxdfXNk7ZzW04+f3OtuFq5ciV33XUXZcuWtd3vtH79ek6ePMnff/9Nq1atcqNZu8gvJ7CdJ2Po+cVarAZ8O6Ah7auXvPFGIiIFWH75/M2PCsuxMQyDHn/04FjsMd5q+Rb3VLznxhsVIYZhsOzkMiZum2ibZLakR0mG1h1Kt4rdbmlIdXs6fOkwfeb2Id2azhst3qBHpR6m5snPPtzyIVP2TsHHxYffuv1GKa/cvc3jeOxx5hyZw59H/yQ6Odq2vHbx2vSo3IMuoV1y7ceMdGs6Q5cMZX3kegLcAvjxrh8p7VU6V9oyQ77oFtimTRsOHTpEz549iYmJISYmhl69erF3716mT5+eW80WKnVC/BjSqgIAr/y+h7gUcy7xioiI2IvFYrENbDH/+HyT0+Qv/56r6mjsUXxcfHi+wfPM7Tn3lueqsrfKxSoztO5QAN7d9K7tqppkt+rUKqbsnQLA6y1ez/XCCi5f2XyuwXMs7r2YiXdM5I6QO3CyOLErehevr3+dO369g5dXv8zmqM12HQTDMAxeW/ca6yPX4+7kzqQOkwpVYZVTuXbl6lp27txJ/fr1yczMzMtmcyQ//TqYnJZJ109XEX4hiQcal2VCr8I/P4CIFF356fM3vylMx+Z47HHumXMPThYnlt+/HD83P7MjmerAxQN8uu1T1pxeA1yeX6pf9X4MrDUQH5f897fOtGbSf0F/dp3fRdPgpnzd8Wt1D/yXqMQo7vvrPmJSY3io+kO81Pgl07JEJ0cz79g8Zh+ezbHYY7blId4htkEwgjyvPV3Rzfhixxd8ufNLHC2OfHbHZ7Qu0/p2Y+c7+eLKldiHu4sj795bG4CfN0Ww7mj0DbYQERHJ38r7lqdqsaq2iVKLqn/PVbXm9Jor5qrKj4UVgKODI2+1eAs3Rzc2RG7g14O/mh0p38iwZvDiqheJSY2hun91RjQYYWqe4u7FGVBzAHO6z+HHO3+kd5XeeDp7cjL+JBO3T6TzrM48seQJFoYvJC0z5wOo/X74d77c+SUArzR9pVAWVjml4qoAaFIhgH5NywLw0qzdJKVlmJxIRETk9mR1DSyKowZeda6q0Nydq8reQn1DGd5gOAAfbv2Qk3EnzQ2UT3y580u2nduGp7MnH7T5ABdHF7MjAZe749YuUZtxzcax7L5lvN3ybRqWbIjVsLL29FpGrhzJHb/dwTub3uHgxYM3tc91p9fx+vrXAXg07FHuq3Jfbr6FAkPFVQHxYpdqlPJ1I+JiEh8uOmR2HBERkduSNST7pqhNXEi+YHIa+8nMzGTFihX8/PPPrFixItttEPFp8UzcPpE7Z9/JLwd/IcPIoEWpFvx696+81+Y9yvqUNTF5zj1Q7QEaBTUiOSOZMWvHYDWsZkcy1YbIDfxv1/8AGNdsXL79e3o4e9CtYje+7/I983rO49GwRwn0CCQ2NZYf9/9I779602duH3458AuxqbFX3ceBiwd4bsVzZBgZ3F3hbp6u93Qev4v8y+73XPXq1eu6r8fExLBy5Urdc3ULlh88x6DvN2OxwKwnm1O/bDGzI4mI2FV+/fzNDwrjsXlg7gPsubCHMU3G0KdaH7Pj3LbZs2fz/PPPEx4eblsWGhrKO++/Q0r1FL7Z/Q0xqTEAhBUPY3j94TQObmxOWDs5FX+Ke/+8l6SMJF5o+AL9a/Y3O5IpopOj6f1nby6kXODeyvcyvvl4syPlSKY1k/WR6/n98O8sO7mMDOvlXlIuDi60L9eenpV60iS4CQ4WB6ISo3ho3kOcSz5H46DGTO4wGWdH+w4xn9+Yes+Vr6/vdR/lypWjf/+i+T/e7WpXNZBe9UpjGDBq5i5SM/JvgSoiInIjtlEDwwv+qIGzZ8+md+/ehIWFsX79euLj41mzdg3FKxSn7/19eXXyq8SkxlDetzyftP2EH+/8scAXVgBlvMswstFIAD7d9mm2QROKCqthZfTq0VxIuUAlv0q82PhFsyPlmKODIy1Lt+TDth+y7L5lvNT4JaoUq0KaNY35x+fz2OLH6DqrK1/s+IInlzzJueRzVPKrxMftPi70hVVO5flogQVBfv518FJiGh0/Xkl0QhpP31GJ5ztVNTuSiIjd5OfPX7MVxmMTlRhFx5kdsWBhyX1LCPQINDvSLcnMzKRSpUqEhYUxZ84cLBaLba6qI5eOEPFZBOln0pm+cjo9qvTIF0Oq25NhGDy55EnWnllLWPEwpnWdVuje4/X8b9f/+Gz7Z7g7ufPzXT/n2aS9uc0wDPZd3Mfvh3/n7+N/E58Wb3st0D2QH+/68bZHGiwoNFpgIVbM04XXu9cC4MsVR9l3Js7kRCIiIrcmyDOIuiXqYmCwKHyR2XFu2erVqwkPD2fUS6OYHz6fPnP72Oaq8nXz5ZnnnyHlXArFo4oXyqLDYrEwvvl4vJ292R29m+/3fG92pDyz7ew2Ju2YBMDLTV4uNIUVXP671gyoyZimY1h23zLebfUuTYObEuoTyqQOk4pMYZVTKq4KoDvDgulSM4gMq8GLs3aRkVm0byAVEZGCqzCMGnjs5OWucGOOjuGl1S+x/+J+3J3ceTTsUebfO58Xu1/uJhYZGWlmzFwV5BnE6CajAfhi5xc3PeJcQRaTEsOoVaPINDK5u8LddK/Y3exIucbNyY07K9zJ/zr9j796/kU1/2pmR8q3VFwVUK/3qImPmxO7T8fyv9XHzY4jIiJySzqV64QFCzvP7+RMwhmz4+RIZEIk729+nw8OfgDAyUMnCXALYFjdYSy8d6Ftrqo9e/YAEBwcbGbcXHd3hbtpF9KODGsGr6x5hfTMdLMj5RrDMBizdgxnk84S6hPKmKZjNJGyACquCqxAbzdevbsGAB8vOcTR8wkmJxIREcm5Eh4laBjUEICF4QtNTnNz9kbvZdTKUXSd3ZVp+6bhUNEBj0AP/Nf5M7/XfB6v8zjF3C6P6Gu1WpkwYQLly5enVatWJifPXRaLhbHNxuLn6sfBSwf5atdXZkfKNdP3TWflqZW4OLjwfpv38XT2NDuS5BMqrgqw3g3K0LpKCdIyrLw0axdWq8YmERGRgqdLaP7vGmg1rKw4uYKBCwbSd15f5ofPJ9PIpElwE77s+CXTvpjG1uVb6XNvH9togevXr6dHjx7MnTuXDz74AEdHR7PfRq4r7l6cMU3HAPDN7m/YE73H5ET2tyd6Dx9v+xiAUY1GqYucZKPiqgCzWCy83bMWni6ObA6/xPQNJ8yOJCIikmMdynXA0eLIvgv7iIiLMDtONikZKfx68Fe6z+nO08ueZuvZrThZnLi7wt381u03vun0Da3KtOLee+9l5syZ7N69m+bNm+Pj40Pz5s3Zs2cPM2fOvOE8oIVJ59DOdA3tSqaRyStrXiE1M9XsSHYTnxbPyJUjybBm0LFcR+6ver/ZkSSfKXxD1hQxZYp58GLXaoz9Yy/vLjjAHdUCCfH3MDuWiIjITfN386dxUGPWR65nYfhCHq39qNmRuJB8gV8O/sKMAzO4lHoJAG9nb3pX7c2D1R686khpvXr1onv37qxevZrIyEiCg4Np1apVkbhi9V8vN3mZzWc3cyz2GJ9v/5znGz5vdqTbZhgG49aN43TCaUp7lWZ88/G6z0quoOKqEOjXpBxzd0ayKfwiL/++m2mPNNb/7CIiUqB0Ld+V9ZHrWRC+wNTi6ljsMabtncZfR/8izZoGQCnPUjxc42F6Vu55w3trHB0dadu2bR4kzd/83PwY12wcTy97mql7p9IupB31S9Y3O9Zt+e3Qbyw+sRgnixPvt34fH5fCMd+c2Je6BRYCDg4W3rk3DFcnB1Yfjmbm1lNmRxIREcmRO8regZODE4cuHeJYzLE8bdswDDZFbmLo0qF0n9OdWYdnkWZNI6x4GO+3eZ95vebRr0Y/DVqQQ21D2tK9YncMLo+sl5SeZHakW3bw4kHe3fQuAMMbDCesRJjJiSS/UnFVSFQo4cVzHasA8MbcfZyLSzE5kYiIyM3zdfWleanmQN4NbJFuTWfesXn0mduHwYsGs+rUKixYuCPkDqZ2mcqPd/5Il9AuhXLi37zyYuMXCfIM4mT8ST7e+rHZcW5JUnoSI1eOJM2aRpsybehfo7/ZkSQfU3FViAxpWZ6w0r7EpWTw6h97MAyNHigiIgXHv0cNzM1zWHxaPFP2TOHO2XfaJv11c3SjT9U+/NXzLz6941Pql6yvLvZ24O3izWvNXwPgl4O/sCFyg8mJcu6tjW8RHhdOSY+SvNniTf27kOtScVWIODk68O69tXFysLBw71n+3h1ldiQREZGb1i6kHS4OLhyPPc6hS4fsvv+sSX87zuzIh1s/JCoxCn83f4bWHcqi3osY03QM5XzK2b3doq55qeb0qdoHgLFrx5KQVnDm5vzjyB/8efRPHCwOvNv6Xfzc/MyOJPmciqtCpkYpH55qWxGAcX/u4VJimsmJREREbo6XixetylyeaNeeEwr/d9LfxPREKvpW5LXmr7Go9yKeqPOEbdJfyR0jGoygjFcZIhMjeX/L+2bHuSnHYo7x1sa3ABhadygNSjYwOZEUBCquCqGhd1SicqAX0QlpvD53n9lxRKQIWXc0msembSEpLcPsKFJAZXUNnH98/m11DbzepL9ftP+C2d1n06tyL1wdXe2UXK7Hw9mDN1q8gQULsw/PZtWpVWZHuq6UjBRGrhpJckYyTYObMrjWYLMjSQGh4qoQcnVy5L3etXGwwO/bT7P8wDmzI4lIEbDrVAyPTt3Con1n+XLFUbPjSAHVukxr3J3cOZVwin0Xcv4D4c1O+utg0VegvNYwqCH9avQDYPy68cSmxpqc6Nre2/wehy8dJsAtgAmtJuDoUPTmKpNbo0+WQqpe2WI80qI8AC//vpv4lHSTE4lIYXbkXDwDvttEYlomzSsGMLRdJbMjSQHl4exB6zKtgZyNGngh+QKTdkyi08xOvLHhDcLjwvF29mZQzUHMv3c+E1pNoJp/tdyKLTfpmXrPEOoTyvnk87y98W2z41zVguML+O3Qb1iwMKHVBIq7Fzc7khQgKq4Ksec7VaVcgAeRsSm8M/+A2XFEpJA6HZPMw99u4lJSOrXL+PJ1/4a4OetXXrl1WV0DF4YvvGHXwGOxxxi/bjydZnZi8s7JXEq9RCnPUoxqNIrF9y1mRMMRBHkG5UVsuQluTm681fItHCwO/H38bxafWGx2pGxOxp1k/PrxAAwJG0KzUs3MDSQFTr4vrlatWkW3bt0oVaoUFouFOXPmXHf9FStWYLFYrnhERRW9kfPcXRx5p1dtAH7cGMH6oxdMTiQihU10QioPf7ORyNgUKpbwZMqgxni5ak4guT0tS7fE09mTyMRIdp7fecXrhmGwOWrzFZP+1gqoZZv09+EaD2vS33yqdonatnuY3lj/BheS88f3k7TMNEauGklieiL1A+vzVN2nzI4kBVC+L64SExOpU6cOkyZNytF2Bw8eJDIy0vYIDAzMpYT5W7OKATzYpCwAL83eRXJapsmJBOBcXArrjkRrLjIp0OJS0hnw3SaORSdS2s+dH4Y0wd/TxexYUgi4ObnRLqQdkH3UwH9P+vvIwkdsk/62C2nHlC5T+OmunzTpbwHxRJ0nqFKsCpdSL/Hmhjfzxfnw460fs+/CPvxc/Xi39bv6dyS3JN//q+natStdu3bN8XaBgYH4+fnZP1AB9FLXaizbf44TF5L4aPFBXrmrhtmRiizDMJiz4zRj5+wlPjWD5ztW4en2lc2OJZJjKemZDJm6hb1n4gjwdGH64MYE+7qbHUsKkS6hXZh7bC4LwxfyRJ0n+P3w7/x44EeiEi/3RHF1dKVHpR70q96PUN9Qc8NKjrk4uvBWy7d4YO4DLIlYwrzj87i7wt2m5VkesZwf9v8AwFst31JXUrll+f7K1a2qW7cuwcHBdOzYkbVr11533dTUVOLi4rI9ChMfN2fe7lULgG/XHGfHyRhzAxVRsUnpPP3zdp6bsZP41MvDVH+05BArD503OZlIzqRnWhn20zY2Hb+It6sTUx9pTIUSXmbHKnQK+7npRpqXao63izfnk8/T/rf2V0z6u7j3YsY0HaPCqgCr5l+NJ+o8AcDbG9/mXJI5oxtHJkQyZu0YAPrX6G8bUEXkVhS64io4OJjJkycza9YsZs2aRUhICG3btmXbtm3X3GbChAn4+vraHiEhIXmYOG/cUa0kPeqWwmrAqJk7Sc1Q98C8tPZINJ0/WcXcXZE4OlgY0bEKfRuFYBjw7C/bOXkxyeyIIjfFajV4ceYuluw/h6uTA98MaEit0r5mxyqUisK56XqcHZ3pULYDAKmZqVTwraBJfwuhwWGDqRlQk/i0eMatG5fn3QPTremMWjWKuLQ4agXUYnj94XnavhQ+FiM/dHK9SRaLhd9//50ePXrkaLs2bdpQtmxZpk+fftXXU1NTSU1NtT2Pi4sjJCSE2NhYfHx8bidyvnIxMY2OH63kQmIaz7avzHMdq5gdqdBLSc/k/YUH+XbNcQDKF/fk4z51qRviR2pGJvdPXs/OU7HUKu3DzCeaa4Q1ydcMw+C1v/YxZV04jg4WvurXgA41Stq1jbi4OHx9fQvd5++tKCrnpuu5kHyBqfum0rBkQ1qWbqm5qQqpozFHuf+v+0mzpvFa89foVblXnrX96bZP+Wb3N3g7e/Nrt18p410mz9qWgiMn56Yi8SnVuHFjjhw5cs3XXV1d8fHxyfYojPw9XXite00AJi0/woGootXFJK/tj4yj++drbYXVQ03KMu+ZltQN8QMuT/b8Rb8G+Hu6sOd0HOP+2GtiWpEb+2zpEaasCwfgg/tq272wkuyKyrnpegLcAxjRYASty7RWYVWIVfSryNP1ngYuT957JuFMnrS79vRavtn9DQDjm49XYSV2USQ+qXbs2EFwcLDZMfKFu8KC6VSjJBlWg1Ezd5GRaTU7UqFjtRp8s/oY3T9fy8Gz8RT3cuHbAQ15q2cYHi7Zx5Ap7efOxAfq4WCBGVtO8vOmCJNSi1zf1HXhfLzkEADjutWgZz19CRER+3m4xsPUC6xHYnoiY9eOxWrk7veT80nneXnNywD0qdqHTqGdcrU9KTryfXGVkJDAjh072LFjBwDHjx9nx44dRERc/hI6evRo+vfvb1v/k08+4Y8//uDIkSPs2bOH4cOHs2zZMoYOHWpG/HzHYrHwZo9aeLs5setUrO2qitjHmZhk+n27kTfn7Sct00qH6oEsGN6a9tWv/Qt/i0rFGdm5KgDj/tjLTg04IvnMnO2nGffn5Surz7avzKAW5U1OJCKFjaODI2+2eBN3J3c2Rm3klwO/5FpbmdZMXlr9EhdTLlK1WFVeaPRCrrUlRU++L662bNlCvXr1qFevHgAjRoygXr16jB07FoDIyEhboQWQlpbG888/T1hYGG3atGHnzp0sWbKE9u3bm5I/Pwr0cePVf4Zj/2jxIY6dTzA5UeHw584zdPlkFeuOXsDd2ZEJvcL4X/+GFPdyveG2T7apSKcaJUnLtPLkD1u5mJiWB4lFbmzZgbM8/9vlSVwHNCvH8A6aOkBEckdZn7I81+A5AD7Z9gkn4k7kSjtf7/6aTVGbcHdy5/027+PqeOPztMjNKlADWuSVonBDtWEY9P9uE6sPR9M41J9fHmuKg4PF7FgFUmxyOuP+2MOcHZf7iNcJ8eOTPnUpX9wzR/uJS0mn++drOR6dSMtKxZn6SGMc9TcRE206fpGHv91IaoaVHnVL8dH9dXP9c6IofP7eKh0bKQqshpXHFj3GxqiN1C1RlyldpuDoYL/BnjZHbWbIoiFYDStvt3ybbhW72W3fUnhpQAu5IYvFwts9w/BwcWRT+EV+3Jg7vw4VduuPXqDrJ6uYs+MMjg4Wnm1fmZlPNMtxYQWX5yOb3K8B7s6OrDkSzUeLD+ZCYpGbs+d0LIOnbCY1w0r7aoG8f18d/QAjIrnOweLA6y1ex9PZkx3ndzB939VHer4VF1Mu8tKql7AaVnpU6qHCSnKFiqsiLMTfg1H/3OvzzvwDnLqkuZZuVmpGJhPm7+fBbzZwJjaFcgEe/Pp4M57rWAVnx1v/36pqkDfv9q4NwKTlR1m0N8pekUVu2vHoRAZ+v4n41Awah/oz6aH6t/XvWkQkJ0p5lWJUo1EATNw+kaMxR297n1bDyitrXuFc8jkq+FZgdOPRt71PkavR2bKI698slIblipGYlskrv+/J88n7CqJDZ+PpMWkdX608hmFA30Yh/P1MKxqUs8+ElvfUKcUj/wwY8PyvO3VPnOSpyNhk+n2zkeiENGoE+/DNwIaaf01E8lzPSj1pVboVadY0XlnzCunW9Nva39S9U1lzeg2ujq580OYDPJw97JRUJDsVV0Wcg4OFd3vXxsXJgZWHzvPtmuManv0arFaD79ce5+6Ja9gfGYe/pwtfPdyAd+6tjaer0413kAOj76xGo9BixKdm8MQPW0lKy7Dr/kWu5lJiGv2/3cTpmGTKF/dk6iON8XFzNjuWiBRBFouF8c3H4+Piw94Le/l297e3vK+d53fy2bbPAHip8UtULqaBeST3qLgSKpbwso0A9ua8/bR+bzmTlh/hQkKqycnyj7NxKQz4fhOv/bWPtAwrbauWYMHwVnSuGZQr7Tk7OjDpwfqU8Hbl0NkEXpq1W1cVJVclpGYwcMpmDp9LIMjHjemDG1PCWyNoiYh5Aj0CebnJ5bmovtr5FQcuHsjxPmJTYxm1chQZRgZdQ7tyb+V77R1TJBsVVwLAY60q8Gz7yhTzcOZMbArvLzxIswnLGPHrDnYU8XmX/t4dSedPVrH6cDRuzg680b0m3w9sRKC3W662G+jjxhcP1cfJwcKfO88wZV14rrYnRVdqRiaPT9/CzpMx+Hk4M31wY8oUU5cZETHfneXvpEPZDmQYGby85mXSMm9+qhLDMBi3bhxnEs8Q4h3C2GZjsVg0MI/kLhVXAoCTowPPdazC+tHt+fC+OtQp40tappXZ207TY9Jaun++hllbT5GSnml21DwTn5LO87/u5KkftxGTlE5YaV/mPt2Kh5uF5tmHc6NQf16+szoAb83bz+bwi3nSrhQdGZlWnv15B2uPXMDTxZEpgxpTuaS32bFERIDL3QPHNB2Dv5s/hy8d5sudX970tj8f+JmlEUtxcnDi/Tbv4+XilYtJRS5TcSXZuDk7cm+DMvwxrCVzhragV/3SuDg6sPNULM//tpPm7yzj3QWFf2TBzeEX6frpamZtO4WDBYa1q8SsJ5tTKTDvP5gHtQilW51SZFgNnvpxG+fiUvI8gxROhmHw8u+7WbA3ChdHB77u35C6IX5mxxIRySbAPYBXm74KwHd7vmPX+V033Gb/hf18sOUDAEY2HEnNgJq5mlEkiyYRvgpN1JjdhYRUftl8kh83nOBM7OUv9g4WaF+9JAOahdKiUkChucyelmHl06WH+HLFUawGhPi78/H9dWkY6m9qrsTUDHp+sZZDZxNoHOrPj4820dDYclsMw2DC/AN8veoYDhb44qEGdKmVO/cQ5oQ+f69Nx0aKupdWv8S8Y/MI9Qnlt26/4eZ09e75iemJ3P/X/UTER9AupB2ftvu00HxPEXNoEmGxqwAvV4a2q8SqUe346uEGtKgUgNWAxfvO0u/bjbT/aCVT1h4nPuX2hkk125FzCdz75TomLb9cWPVuUIa/n2llemEF4OnqxOR+DfB2dWJT+EUm/J3zm3pF/u3LlUf5etUxAN7pVTtfFFYiItczuvFoSriXIDwunM+2f3bVdQzD4LX1rxERH0GwZzBvtHhDhZXkKRVXctOcHB3oXDOIH4c0ZcmI1gxoVg4vVyeOnU9k/F/7aPr2UsbM2c2hs/FmR80RwzCYvj6cuyeuZvfpWPw8nPniofp8cF8dvPPRMNQVSnjx4f11APhu7XH+3HnG5ERSUP20MYL3FhwE4JU7q3N/oxCTE4mI3Jivqy/jm48H4Id9P7AlassV6/x+5HfmH5+Po8WR91q/h6+rbx6nlKJOxZXckkqB3rzWvRYbXm7PG91rUinQi8S0TH7YEEGnj1fxwNcbmL87Mt/PmXUuPoVBUzbz6h97SUm30qpycRYOb82dYcFmR7uqTjWDeKptRQBenLmrwBWyYr65u87wypzdADzVtiKPtq5gciIRkZvXukxrelXuhYHBmLVjSEr//3vAj1w6woSNEwB4ut7T1A2sa1JKKcpUXMlt8XJ14uFmoSx+rjU/PdqELjWDcLDA+mMXePLHbbR6bzmfLztMdD6cM2vh3ii6fLKaFQfP4+LkwLhuNZg6qDElfXJ3iPXb9XynqrSsVJzk9EyemL6VuALeHVPyzspD53luxg4MAx5sUpYXOlc1O5KISI690PAFgj2DOZ1wmg+3fAhAckYyI1eOJCUzhRalWjCo1iCTU0pRpQEtrkI3Dd+e0zHJ/LTxBL9sOsmFxMvzUbg4OnBnWBD9m4dSL8TP1P7PiakZvP7XPmZsOQlAjWAfPulblyoFaPjpCwmpdJu4hjOxKXSqUZKvHm6gPuVyXVtPXKLfNxtJTs/krtrBfNa3Ho4O+e/fjD5/r03HRuT/bYzcyJBFQwD4qsNXLDyxkNmHZ1PCvQS/dfuNAPcAkxNKYZKTz18VV1ehE5h9pGZk8vfuSKatP8H2iBjb8lqlfejfLJR76pTCzdkxTzNtPXGJEb/u4MSFJCwWeLx1RZ7rWBlXp7zNYQ87T8Zw3+T1pGVaebFLNZ78p7ugyH8diIqjz1cbiE1Op3WVEnzTvyEuTvmz44I+f69Nx0Yku7c3vs3PB37Gy9mLhPQEHCwOfNPpGxoFNTI7mhQyKq5uk05g9rfrVAzT1p/gz51nSMu4fB+Wn4czfRqF0K9JOUL8PXK1/fRMKxOXHeHzZYexGlDaz50P769D0woF+5etnzdFMHr2bhwsMH1wE1pUKm52JMlnIi4kce/kdZyPT6V+WT9+GNIEDxcns2Ndkz5/r03HRiS7pPQk7vvrPiLiIwB4qs5TPFn3SZNTSWGk4uo26QSWey4mpvHrlpNMX3+C0zHJAFgs0L5aIP2bhdKyUnEc7NxV6Xh0IsNn7GDnyRgAetYrzWvda+KTj0YCvFWGYfDirF38uuUU/p4u/PV0S0r7uZsdS/KJc3Ep9J68noiLSVQt6c2Mx5vi5+Fidqzr0ufvtenYiFxpx7kdPLroURqUbMCk9pNwdCh4PVEk/1NxdZt0Ast9mVaD5QfOMXV9OKsPR9uWly/uycNNy3FvgzL4ut9e8WMYBj9vOskbc/eRnJ6Jj5sTb/YM4546pW43fr6Skp5J78nr2HM6jjplfPn1iWYFspuj2FdsUjp9vl7Pgah4yvp7MPOJZgTm88FaQJ+/16NjI3J1iemJuDm6qbCSXKPi6jbpBJa3jp1PYPqGE8zccor41AwA3J0d6Vm/NP2blaNaUM7/BtEJqbw0axdL9p8DoHnFAD64rw6lCulVnZMXk+j2+RpiktJ5sElZ3u4ZZnYkMVFSWgb9vtnItogYSni7MuuJ5pQNyN2ut/aiz99r07ERETGHiqvbpBOYORJTM/h9+2mmrz/BwX/N39S4vD8DmoXSqWZJnB2z34SfmZnJ6tWriYyMJDg4mFatWrHiUDQvztpFdEIaLo4OjOpSlUdalLd7d8P8ZsXBcwyashnDgPd61+b+hpoYtihKy7AyZNoWVh06j4+bE78+0eyWfqAwiz5/r03HRkTEHCqubpNOYOYyDIONxy8yff0JFuyNItN6+Z9oSR9XHmxcjgeahBDo7cbs2bN5/vnnCQ8Pt23rG1gKl+YD8ajanGpB3nzcpy7Vg4vO3/CzpYf5aPEhXJwcmP1kc2qV1sz0RUmm1eDZX7Yzd1ck7s6O/DCkMQ3K+ZsdK0f0+XttOjYiIubIyedv/hyLV4o0i8VC0woBTHqoPmtfvINn7qhEcS9Xzsal8vGSQ7R4Zxl3PvcBvXv3JiwsjPXr17Nu/0nqDZtEmk8Zzs+ZQDOHI8wZ2qJIFVYAw9pVon21QNIyrDzxw1Yu/TPPmBR+hmEw9o89zN0VibOjhckPNyhwhZWIiEhBp+JK8rUgXzdGdKrKupfu4NO+dWlQrhhp6Rks+v4D3Co2Ir3d88yL8qLftF1c9CxH2IA3aXFHJzb+8inORfBft4ODhY/61KVcgAenLiXz7Iwdtit/Urh9sOggP26MwGKBj+6vS5sqJcyOJCIiUuQUwa+fUhC5ODnQvW5pZj3ZnPGNHMiMPUvxln3Zf/byYBiZVoO7awezaERb3n9zHMePH2f16tVmxzaFr7szXz7UADdnB1YdOs+nSw+bHclUZ2KSeW/BAT5ZcogNxy6Qkp5pdiS7+2b1MSYtPwrAmz1q0a2QjYgpIiJSUOTfmSRFrsEt4/JgF+veHcD8AzEsO3COPo1C6F63FBaLhVq1agEQGRlpZkxT1Sjlw4ReYTw3YyefLT1MnTK+tK9e0uxYeSoxNYOvVh7l69XHSEm3/rP0MC5ODtQv60fTCgE0rRBA3RA/3JwL7vC9v245yZvz9gPwQueqPNSknMmJREREii4VV1LgBAcHA3Dq2CEebd2UR1tXyPb6nj17sq1XVPWsV4btETFMW3+C52bs4K+nW1IuwNPsWLku02owa+sp3l90kPPxqQA0Ci1GkK87G45d4Hx8KhuOXWTDsYsU9GJrwZ4oXpq1C4BHW5XnqbYVTU4kIiJStOX70QJXrVrF+++/z9atW4mMjOT333+nR48e191mxYoVjBgxgr179xISEsKYMWMYOHDgTbepEZnyt8zMTCpVqkRYWBhz5szBweH/e7darVZ69OjBnj17OHz4MI6OBeNLcm5Jy7DS5+v1bI+IoXqwD7OfbI67S+E9JuuORPPGvP3sj4wDoFyAB6O7VqNzzSAsFguGYXAsOpENxy78U2BdsBVgWQpKsbXuSDQDv99MWqaV+xqU4b3etbFYCv50A/r8vTYdGxERc+Tk8zffX7lKTEykTp06PPLII/Tq1euG6x8/fpy77rqLJ554gh9//JGlS5cyZMgQgoOD6dy5cx4kltzm6OjIhx9+SO/evenRowejR4+mVq1a7NmzhwkTJjB37lxmzpxZ5AsruFwofPlQA+6euJr9kXG88vtuPry/TqH4Ev5vR88nMOHvAyzZfxYAbzcnnm1fmYeblcPV6f//HVgsFiqW8KJiCS8ealLumsVWfr+ytfNkDI9O20JappXONUsyoVdYofubioiIFET5/srVv1kslhteuXrxxReZN2+erWsYQN++fYmJiWHBggU31Y5+HSwYrjbPVfny5fnggw9uqhAvStYfvUC/bzeSaTV4o3tNHm4WanYku7iUmManSw/zw4YTZFgNHB0s9GtSlmc7VMHf0yXH+7uZK1uuTg7UL1vsn2LLnzp5XGwdORfPfZPXcykpneYVA/huYCPTiz170ufvtenYiIiYo1Bducqp9evX06FDh2zLOnfuzPDhw6+5TWpqKqmp//8FKi4uLrfiiR316tWL7t27s3r1aiIjIwkODqZVq1a6YnUVzSoG8FKXarz1935en7uPGqV8aVCumNmxbllahpXpG07w2dLDxCanA9C+WiCj76xOpUCvW97vzV7ZWn/sAuuPXQCuLLbqlvXLdrXMnk5dSqLfN5u4lJROnTK+fN2/YaEqrCQ7nZtERAqeQldcRUVFUbJk9lHRSpYsSVxcHMnJybi7u1+xzYQJE3jttdfyKqLYkaOjI23btjU7RoEwpFV5tp+8xN+7o3jqx63MfboVJbxdzY6VI4ZhsGjfWSb8vZ/wC0kAVAvyZsxdNWhZubjd27tRsbX+6AWiE/Km2IpOSOXhbzcRFZdCpUAvvh/UGC/XQvcRLv+ic5OISMGjMzMwevRoRowYYXseFxdHSEiIiYlE7M9isfBe7zocjIrn6PlEnv55Gz8MboKTY8GY7m7P6VjenLfvn3uhoLiXKyM7VeG+hiE4OuTN/UZXK7aOns8qti4XXLlRbMWlpDPgu00cj06ktJ870wc3vqVuj1Kw6NwkIlLwFLriKigoiLNnz2ZbdvbsWXx8fK561QrA1dUVV9eC9Qu+yK3wcnXiq4cb0P3ztWw4dpH3Fx5k9J3VzY51XVGxKXyw6CCztp3CMC4XK4+2qsATbSuafuXGYrFQKdCLSoFe9GuaO8VWSnomQ6ZsYe+ZOAI8XZg+uDHBvlf/LJPCRecmEZGCp9AVV82aNePvv//Otmzx4sU0a9bMpEQi+UulQG/ev68OT/24ja9WHaNuiB9dw/LfnGBJaRl8veoYX608RnJ6JgDd65ZiVJdqlPbLn8WFvYut9EwrQ3/cxqbwi3i7OjH1kcZUKHHr95SJiIhI7sr3xVVCQgJHjhyxPT9+/Dg7duzA39+fsmXLMnr0aE6fPs20adMAeOKJJ/j8888ZNWoUjzzyCMuWLePXX39l3rx5Zr0FkXznzrBgHm9dga9WHWPkbzupXNKLSoHeZscCwGo1+H37ad5feJCouBQAGpQrxpi7qlOvbMEahON2i61D5+JZeuAcrk4OfDOgIbVK+5r8jkREROR68v1Q7CtWrKBdu3ZXLB8wYABTpkxh4MCBhIeHs2LFimzbPPfcc+zbt48yZcrw6quvahJhkf/IyLTS79uNbDh2kYolPPljWEvTu9ltOHaBN+ftY8/py6OilSnmzktdq3FXWHChnMfpWsXWvzk6WPj64Qa0r17yGnspXPT5e206NiIi5sjJ52++L67MoBOYFBXn41PpNnENUXEp3BkWxKQH65tSxIRHJzJh/n4W7r18v6SXqxPD7qjEwOahRWqo8f8WWwej4nmmfWW61SlldrQ8o8/fa9OxERExR5Ge50pEbl4Jb1cmPVSfvl+v5+/dUXyz+jiPtq6QZ+3HJqUzcdlhpq4PJz3TwMECDzQuy3Mdq1Dcq+jdyP/fboQiIiJSsKi4EiniGpQrxti7a/DqH3t5Z8EBapX2pVnFgFxtMz3Tyo8bTvDJ0sPEJF2eBLhNlRK8cld1qpTMH/d+iYiIiOSUiisRoV/TcmyPiGH29tM8/fM25j7diiBfN7u3YxgGyw6c462/93PsfCIAlQO9eOWu6rStGmj39kRERETykoorEcFisfBWzzD2R8WzPzKOJ3/cyozHmuHiZL8JhvedieOtv/ex9sjlUfECPF14rmMV+jYKKTATGYuIiIhcj77RiAgA7i6OTO5XHx83J7ZHxPDmvH122e+5+BRenLmLuyauZu2RC7g4OvBEm4osf6Et/ZqWU2ElIiIihYauXImITbkATz7pW5dHpmxh2voT1A3xo1f9Mre0r5T0TL5ZfYwvVhwlKe3yJMB31Q7mpS7VCPH3sGdsERERkXxBxZWIZHNHtZI8074yny09zMu/76ZakA81St38sM9Wq8GfO8/w3oIDnIm9PAlwnRA/xt5dnQbl/HMrtoiIiIjpVFyJyBWebV+ZnSdjWHnoPE/8sJW/hrXE18P5htttCb/IG/P2s/NkDAClfN14sWs1utUuhYND4ZsEWEREROTfdLODiFzB0cHCp33rUqaYOxEXkxjx6w6s1mvPN37yYhJDf9xG78nr2XkyBk8XR17oXJVlI9vSvW5pFVYiIiJSJKi4EpGr8vNwYXK/Brg6ObD0wDk+X37kinXiUtKZMH8/7T9cybzdkVgs0LdRCMtfaMvQdpVwc3Y0IbmIiIiIOdQtUESuqVZpX97sUYsXZu7i4yWHqF3Gl7ZVA8nItPLz5pN8vPgQFxPTAGhRKYBX7qyRo/uzRERERAoTFVcicl33NQxh+8kYftoYwbO/7ODVu2vw1cqjHD6XAECFEp6Muas67aoGYrGo+5+IiIgUXSquROSGxnWrwd7Tsew8FcvI33YC4OfhzHMdqvBgk7I4a64qEREREd1zJSI35urkyJf9GlDC2xVnRwtDWpZn5ch2DGgeqsJKRERE5B+6ciUiN6WUnztLnmtDpmHg7+lidhwRERGRfEfFlYjctJuZ60pERESkqFJ/HhERERERETtQcSUiIiIiImIHKq5ERERERETsQMWViIiIiIiIHai4EhERERERsQMVVyIiIiIiInagodivwjAMAOLi4kxOIiJStGR97mZ9Dsv/07lJRMQcOTk3qbi6ivj4eABCQkJMTiIiUjTFx8fj6+trdox8RecmERFz3cy5yWLo58ErWK1Wzpw5g7e3NxaLJcfbx8XFERISwsmTJ/Hx8cmFhLlDufNOQcwMyp3XCmLu281sGAbx8fGUKlUKBwf1XP83nZsKTu6CmBmUO68VxNwFMTPk7blJV66uwsHBgTJlytz2fnx8fArUP7wsyp13CmJmUO68VhBz305mXbG6Op2bCl7ugpgZlDuvFcTcBTEz5M25ST8LioiIiIiI2IGKKxERERERETtQcZULXF1dGTduHK6urmZHyRHlzjsFMTMod14riLkLYuaioqD+bQpi7oKYGZQ7rxXE3AUxM+Rtbg1oISIiIiIiYge6ciUiIiIiImIHKq5ERERERETsQMWViIiIiIiIHai4EhERERERsQMVVyIiIiIiInag4kpERERERMQOVFyJiIiIiIjYgYorERERERERO1BxJSIiIiIiYgcqrkREREREROxAxZWIiIiIiIgdqLgSERERERGxAxVXIiIiIiIidqDiSkRERERExA5UXImIiIiIiNiBiisRERERERE7UHElIiIiIiJiByquRERERERE7EDFlYiIiIiIiB2ouBIREREREbEDFVciIiIiIiJ2oOJKRERERETEDlRciYiIiIiI2IGKKxERERERETtQcSUiIiIiImIHKq5ERERERETsQMWViIiIiIiIHai4EhERERERsQMVVyIiIiIiInag4kpERERERMQOVFyJiIiIiIjYgYorERERERERO1BxJSIiIiIiYgcqrkREREREROzAyewA+ZHVauXMmTN4e3tjsVjMjiMiUmQYhkF8fDylSpXCwUG///2bzk0iIubIyblJxdVVnDlzhpCQELNjiIgUWSdPnqRMmTJmx8hXdG4SETHXzZybVFxdhbe3N3D5APr4+JicRkSk6IiLiyMkJMT2OSz/T+cmERFz5OTcpOLqKrK6W/j4+OgEJiJiAnV7u5LOTSIi5rqZc5M6tIuIiIgUYaGhoXzyySdmx8gV48ePp27dumbHkCJExZWIiIiInWRmZrJixQp+/vlnVqxYQWZmptmRTLVixQosFgsxMTGmtD9y5EiWLl1qSttSNKm4EhEREbGD2bNnU6lSJdq1a8eDDz5Iu3btqFSpErNnzzY72lWlpaWZHSHXeXl5ERAQYHYMKUJUXImIiIjcptmzZ9O7d2/CwsJYv3498fHxrF+/nrCwMHr37n3bBZbVamXChAmUL18ed3d36tSpw8yZM22vZ2ZmMnjwYNvrVatW5dNPP822j4EDB9KjRw/eeustSpUqRdWqVa9o55FHHuHuu+/Otiw9PZ3AwEC+/fbbq2Y7ceIE3bp1o1ixYnh6elKzZk3+/vtvwsPDadeuHQDFihXDYrEwcOBAAFJTU3nmmWcIDAzEzc2Nli1bsnnzZts+s654zZs3j9q1a+Pm5kbTpk3Zs2ePbZ0pU6bg5+fHnDlzqFy5Mm5ubnTu3JmTJ0/a1vlvt8CsY/DBBx8QHBxMQEAAQ4cOJT093bZOZGQkd911F+7u7pQvX56ffvqpUHedFPvSgBYiIiIityEzM5Pnn3+eu+++mzlz5tjmwWnatClz5syhR48ejBw5ku7du+Po6HhLbUyYMIEffviByZMnU7lyZVatWkW/fv0oUaIEbdq0wWq1UqZMGX777TcCAgJYt24djz32GMHBwdx///22/SxduhQfHx8WL1581XaGDBlC69atiYyMJDg4GIC5c+eSlJREnz59rrrN0KFDSUtLY9WqVXh6erJv3z68vLwICQlh1qxZ3HvvvRw8eBAfHx/c3d0BGDVqFLNmzWLq1KmUK1eO9957j86dO3PkyBH8/f1t+37hhRf49NNPCQoK4uWXX6Zbt24cOnQIZ2dnAJKSknjrrbeYNm0aLi4uPPXUU/Tt25e1a9de81guX76c4OBgli9fzpEjR+jTpw9169bl0UcfBaB///5ER0ezYsUKnJ2dGTFiBOfOncvBX0uKNEOuEBsbawBGbGys2VFERIoUff5em45N/rV8+XIDMNavX28YhmHEzp9vnHp+pJERH28YhmGsW7fOAIzly5ff0v5TUlIMDw8PY926ddmWDx482HjggQeuud3QoUONe++91/Z8wIABRsmSJY3U1NRs65UrV874+OOPbc9r1KhhvPvuu7bn3bp1MwYOHHjNdsLCwozx48df9bWsY3Pp0iXbsoSEBMPZ2dn48ccfbcvS0tKMUqVKGe+991627X755RfbOhcuXDDc3d2NGTNmGIZhGN9//70BGBs2bLCts3//fgMwNm7caBiGYYwbN86oU6dOtmNQrlw5IyMjw7bsvvvuM/r06ZNt+82bN9teP3z4sAFkO0ZStOTk81fdAkVERERuQ2RkJAC1atUi48IFzrz8CnFz5xI75w/b8n+vl1NHjhwhKSmJjh074uXlZXtMmzaNo0eP2tabNGkSDRo0oESJEnh5efH1118TERGRbV9hYWG4uLhct70hQ4bw/fffA3D27Fnmz5/PI488cs31n3nmGd58801atGjBuHHj2LVr13X3f/ToUdLT02nRooVtmbOzM40bN2b//v3Z1m3WrJntv/39/alatWq2dZycnGjUqJHtebVq1fDz87tiP/9Ws2bNbFcQg4ODbVemDh48iJOTE/Xr17e9XqlSJYoVK3bd9ySSRcWViIiIyG3I6j63Z88eLnz9P4ykJAASVq60Lf/3ejmVkJAAwLx589ixY4ftsW/fPtt9V7/88gsjR45k8ODBLFq0iB07djBo0KArBq3w9PS8YXv9+/fn2LFjrF+/nh9++IHy5cvTqlWra64/ZMgQjh07xsMPP8zu3btp2LAhEydOvKX3mheyuhRmsVgsWK1Wk9JIYaPiSkREROQ2tGrVitDQUN4cO5YLP/1kW560cSMZCQm2gSiuV6BcT40aNXB1dSUiIoJKlSple4SEhACwdu1amjdvzlNPPUW9evWoVKlStqtaOREQEECPHj34/vvvmTJlCoMGDbrhNiEhITzxxBPMnj2b559/nv/9738Atqtk/x6SvmLFiri4uGS7Lyo9PZ3NmzdTo0aNbPvdsGGD7b8vXbrEoUOHqF69um1ZRkYGW7ZssT0/ePAgMTEx2dbJiapVq5KRkcH27dtty44cOcKlS5duaX9S9GhACxEREZHb4OjoyIcffkjve+8lydOLYS1aUCE1lf3HjzOiUycWbNjAzJkzb3kwC29vb0aOHMlzzz2H1WqlZcuWxMbGsnbtWnx8fBgwYACVK1dm2rRpLFy4kPLlyzN9+nQ2b95M+fLlb6nNIUOGcPfdd5OZmcmAAQOuu+7w4cPp2rUrVapU4dKlSyxfvtxW3JQrVw6LxcLcuXO58847cXd3x8vLiyeffJIXXngBf39/ypYty3vvvUdSUhKDBw/Otu/XX3+dgIAASpYsySuvvELx4sXp0aOH7XVnZ2eefvppPvvsM5ycnBg2bBhNmzalcePGt/S+q1WrRocOHXjsscf48ssvcXZ25vnnn8fd3R2LxXJL+5SiRcWViIiIyG26u359PikTwntno7h30ULb8rJxscycOZNevXrd1v7feOMNSpQowYQJEzh27Bh+fn7Ur1+fl19+GYDHH3+c7du306dPHywWCw888ABPPfUU8+fPv6X2OnToQHBwMDVr1qRUqVLXXTczM5OhQ4dy6tQpfHx86NKlCx9//DEApUuX5rXXXuOll15i0KBB9O/fnylTpvDOO+9gtVp5+OGHiY+Pp2HDhixcuPCKe5veeecdnn32WQ4fPkzdunX566+/st0z5uHhwYsvvsiDDz7I6dOnadWq1TWHjL9Z06ZNY/DgwbRu3ZqgoCAmTJjA3r17cXNzu639StFgMQzDMDtEfhMXF4evry+xsbH4+PiYHUdEpMjQ5++16djkb6efH0ncvHm4t2pF+EMPcnz5chy+/54moeWpunJFgbvqkZCQQOnSpfn+++9vuzC8FStWrKBdu3ZcunQJPz+/q64zZcoUhg8fTkxMTK5mOXXqFCEhISxZsoT27dvnaluSP+Xk81dXrkRERERuQ8qBA8TNmwdA0IjnCK1endbNmnHoz78wzp0j9cAB3G7xHqC8ZrVaiY6O5sMPP8TPz4977rnH7Eh5btmyZSQkJBAWFkZkZCSjRo0iNDSU1q1bmx1NCgANaCEiIiJyG85/+hkAPnd2tRVRDq6ueP4zjHjWqIEFQUREBCVLluSnn37iu+++w8mp6P0On56ezssvv0zNmjXp2bMnJUqUsE0oLHIj6hZ4Fep6ISJiDn3+XpuOTf6UtH07Jx54EBwdqTD3L1z/NYDEpRm/EjVuHO516hA64xcTU4rI7cjJ56+uXImIiIjcAsMwOP/xJwD49uyRrbAC8GrbBoDkXbvIuHgxr+MVCWmnTnOoZSui3nzL7CgigIorERERkVuStH49SZs2YXF2psRTT13xunPJkrhWrw6GQeLq1SYkLPxi/5hDZnQ0MTNnYk1NNTuOSP4oriZNmkRoaChubm40adKETZs2XXPdtm3bYrFYrnjcddddtnUGDhx4xetdunTJi7ciIiIiRYBhGJz756qV3wN9cb7GcOVebS4PghC/YkUeJSta4pcuBcBISSF561aT04jkg+JqxowZjBgxgnHjxrFt2zbq1KlD586dOXfu3FXXnz17NpGRkbbHnj17cHR05L777su2XpcuXbKt9/PPP+fF2xEREZEiIGHpUlJ278bi4UHxxx675nrebdsCkLhmLUZ6eh6lKxrST58mdd9+2/OENWtNTCNymenF1UcffcSjjz7KoEGDqFGjBpMnT8bDw4Pvvvvuquv7+/sTFBRkeyxevBgPD48riitXV9ds6/13UjoRERGRW2FkZnL+008B8O//ME7Fi19zXbewMByLFcMaH0/S9u15FbFIyLpqZflnUuHENWvMjCMCmFxcpaWlsXXrVjp06GBb5uDgQIcOHVi/fv1N7ePbb7+lb9++eHp6Zlu+YsUKAgMDqVq1Kk8++SQXLly45j5SU1OJi4vL9hARETGTzk35V9y8eaQePoKDjw8Bjzxy3XUtjo54/TM/UsKKgjMke0EQv+RyceU/+BGwWEg9dIj0s1fv+SSSV0wtrqKjo8nMzKRkyZLZlpcsWZKoqKgbbr9p0yb27NnDkCFDsi3v0qUL06ZNY+nSpbz77rusXLmSrl27kpmZedX9TJgwAV9fX9sjJCTk1t+UiIiIHejclD8ZaWmcn/g5AAFDhuB4E8PiZ40aWJDmu8rvMi5dImnLFgD87u2NW61aACSuVddAMZfp3QJvx7fffktYWBiNGzfOtrxv377cc889hIWF0aNHD+bOncvmzZtZcY2bSUePHk1sbKztcfLkyTxILyIicm06N+VPMbNnk37yJI7Fi+Pf76Gb2sazRQtwdCTt6FHS9He0i4TlK8BqxbV6dVzKlMazZQtAXQPFfKYWV8WLF8fR0ZGzZ89mW3727FmCgoKuu21iYiK//PILgwcPvmE7FSpUoHjx4hw5cuSqr7u6uuLj45PtISIiYiadm/Ifa0oK0ZO+AKD4E0/g4OFxU9s5+vjg0aABoK6B9pJ1v5V3+/YAeLVsCUDiunUY1+ipJJIXTC2uXFxcaNCgAUv/+R8EwGq1snTpUpo1a3bdbX/77TdSU1Pp16/fDds5deoUFy5cIDg4+LYzi4iISNF06cefyDh/HudSpfC7/74bb/AvXm3UNdBerElJtitU3h0v37fvXrs2Dl5eZMbEkLJvn5nxpIgzvVvgiBEj+N///sfUqVPZv38/Tz75JImJiQwaNAiA/v37M3r06Cu2+/bbb+nRowcBAQHZlickJPDCCy+wYcMGwsPDWbp0Kd27d6dSpUp07tw5T96TiIiIFC6ZCQlc+PprAIoPG4bDPyPU3ays+66SNm7Empho93xFScLatRipqTiXKYNrlSoAWJyd8WzWFFDXQDGX6cVVnz59+OCDDxg7dix169Zlx44dLFiwwDbIRUREBJGRkdm2OXjwIGvWrLlql0BHR0d27drFPffcQ5UqVRg8eDANGjRg9erVuLq65sl7EhERkcLl4vdTyIyNxaVCBXzv6Zbj7V0qVMA5JAQjPZ3EDRtyIWHRkbBkCXC5S6DFYrEt92xxuWug5rsSMzmZHQBg2LBhDBs27KqvXW0QiqpVq2IYxlXXd3d3Z+HChfaMJyIiIkVYxqVLXPz+ewBKPPMMFqecf32yWCx4tWnDpR9+IGHFStu9QpIzRno68f/ct5bVJTBL1qAWyTt2kBkfj6O3d57nEzH9ypWIiIhIfnbh6/9hTUrCrUYNvDt1vOX9/Pu+q2v9SCzXl7R1K9bYWBz9/XGvVy/bay5lyuBSrhxkZurqoJhGxZWIiIjINaRHRXHpxx8BKPHccCwOt/7VyaNxIyweHmScO0fq/v32ilikxC++3CXQ6452WBwdr3jdM2vUQHUNFJOouBIRERG5hugvJ2OkpeHesIHti/utcnB1xfOf0ZA1amDOGYZxxRDs//Xv+a50dVDMoOJKRERE5CrSTpwgZtYsAAKHD882eMKtyho1MP4q95TL9aXs2UtGVBQWDw88mze/6jqejRuDszPpp0+TFh6etwFFUHElIiIiclXnP58EGRl4tm6FR8OGdtmnV+vLxVXKrt1kXLhgl30WFfFL/+kS2KoVDtcYAdrB0xOP+vUBdQ0Uc6i4EhEREfmPlIOHiJs7F4ASzz5rt/06lwzEtUZ1MAwSVq22236LgoSsLoEdrj/Soq1r4FoVV5L3VFyJiIiI/Mf5zz4Dw8C7Sxfca9a0676927YFdN9VTqSFh5N6+Ag4OdlGXbwWr6xBLTZtwkhLy4t4RUbS5s2c+/BDMmNizI6Sb6m4EhEREfmX5J07L18lcXCgxDNP233/WcVB4po1GOnpdt9/YZQ1kIVn48Y4+vhcd13XqlVxLF4cIymJpG3b8yJekWAYBmfGjOHC/74hvO8DuqftGlRciYiIiPzLuU8+AcC3Rw9cK1Sw+/7dwsJw9PfHmpBA0tZtdt9/YWQbgv0GXQIBLA4OeLW4POBF4to1uZqrKEk9fJj0ExHA5SuJ4X36krhpk8mp8h8VVyIiIiL/SFy/nqT1G8DZmRJDn8qVNiwODni1bg2oa+DNSD93juSdO4FrD8H+X1nD5idoUAu7ybrnzb1hA9xq1yYzNpaIwUOI+X2OucHyGRVXIiIiIlzu9pR11apYnz44ly6da21lDcmu4urGEpavAMPArXZtnEuWvKltsoZqT92/n4zo6FxMV3RkXT3069mTctOm4t2lC6SnEzl6NOc+/gTDajU5Yf6g4kpEREQESFi+nJSdu7C4u1P88cdytS3PFi3AyYm0Y8dIi4jI1bYKuvgll7/Ue3focNPbOAUE4FajBqBRA+0h/cwZUvbtAwcHvNq1w8HNjdIffUjAE48DcOGrrzj93AisyckmJzWfiisREREp8gyrlfMffwKA/8MP41SiRK625+jtjUeDBgAkrNDVq2vJTEggccMG4MZDsP+XugbaT/ySf7oE1q+Hk78/cLl7a+Dw4QRPmADOzsQvXMiJ/gPIOH/ezKimU3ElIiIiRV7cvL9JPXwYB29vAgY/kidtZo0aqK6B15awciWkp+NSvnyOBxf593xX6rJ2e+Jtc4xdefXQr2cPyn33LY6+vqTs3s3xPn1IOXgwryPmGyquREREpEgz0tM5P3EiAAGDB+Po65sn7Xr9M99V0qZNWBMT86TNgibhOl/qb8Sjbl0cPDzIvHiRlP377R2tyMi4dImkLVuAa/8dPBo1InTGL7iEhpJxJpITDzxYZH80UHElIiIiRVrM7N9Jj4jAMSAA/4f75Vm7LuVDcS5bFiM9ncT16/Os3YLCmpZGwspVQM67BAJYXFzwaNIEgER1DbxlCStWQmYmrtWq4VKmzDXXcwkNJfSXn/Fo0gRrUhInn3yKi9N/yMOk+YOKKxERESmyrCkpRH/xBQDFH38cB0/PPGvbYrHYugbGr1iRZ+0WFEkbNmBNTMQpMBC3sLBb2oeta+AazXd1q+KX/jOgyE0Mg+/o50fZ/32N7729wGrl7FtvEfX6GxgZGbkdM99QcSUiIiJF1qWffyHj7FmcgoPx69snz9vPGpI9ceUqDMPI8/bzs6xBFLza34HF4da+snr9M6hF0vbtZCao62VOWZOTbVf9bvbqocXFheA33yRw5PMAXPrpJ04++RSZCQm5ljM/UXElIiIiRVJmQgIXvvoKgBLDhuLg4pLnGTwaNcLi4UHG+fOXh7oWAIzMTOKXLQPAu33O77fK4lKuHM4hIZCRQdKmjfaKV2Qkrl2LkZKCc+nSuFardtPbWSwWAoYMofRnn2JxcyNx9WpOPPAg6adP52La/CFfFFeTJk0iNDQUNzc3mjRpwqZNm665btu2bbFYLFc87rrrLts6hmEwduxYgoODcXd3p0OHDhw+fDgv3oqIiIgUEBenTiUzJgaX0FB8u3c3JYODiwuezZsBkKCugTbJO3eRGR2Ng7c3no0b3da+1DXw1mVdPfTu0B6LxZLj7X06daLc9Ok4lShB6uHDHL+/D8k7d9o7Zr5ienE1Y8YMRowYwbhx49i2bRt16tShc+fOnDt37qrrz549m8jISNtjz549ODo6ct9999nWee+99/jss8+YPHkyGzduxNPTk86dO5OSkpJXb0tERETysYxLl7j43fcAlHj2GSxOTqZl8f5n1MCswRvk/+/z8WrTBsttXlH00nxXt8TIyCBh+XLg1kZrzOIeVovQX2fgWq0amRcucKL/AOLmz7dXzHzH9OLqo48+4tFHH2XQoEHUqFGDyZMn4+HhwXfffXfV9f39/QkKCrI9Fi9ejIeHh624MgyDTz75hDFjxtC9e3dq167NtGnTOHPmDHPmzMnDdyYiIiL51YVvvsGamIhr9ep4d+5sahbP1q0BSNm9m4zoaFOz5AeGYRC/5J9BFG5hlMD/8mjSBJycSI+IIC0i4rb3V1QkbdlKZmwsjsWK4V6v3m3tyzk4mNAff8CrbVuM1FROPzeC6MmTC+V9hqYWV2lpaWzdupUO/6qGHRwc6NChA+tvckjSb7/9lr59++L5z+g+x48fJyoqKts+fX19adKkyTX3mZqaSlxcXLaHiIiImXRuyj3pZ89x6YcfAQgc/uwtD5ZgL86BgbjVqAGGQcKq1aZmyQ/Sjhwh/UQEFhcXPFu2uu39OXp54VG3LnD5HiK5OVkFrle7dna5suvg6UmZSZ/jP2AAAOc/+ZTIl0ZjTUu77X3nJ6Z+mkRHR5OZmUnJkiWzLS9ZsiRRUVE33H7Tpk3s2bOHIUOG2JZlbZeTfU6YMAFfX1/bIyQkJKdvRURExK50bso90ZO/xEhNxb1+fdtVI7N52boGFs2JV/8t60u9Z7NmOHrZZ2h8T3UNzBHDMIi/jQmcr8Xi6EjJ0S8RNH4cODoS+8cfRDzyCBmXLtmtDbOZ3i3wdnz77beEhYXRuHHj29rP6NGjiY2NtT1Onjxpp4QiIiK3Ruem3JF28iQxv80EIPC54bd0k35usA3JvmYNRiH7JT+nbIModLTfl/qs4ippwwaM9HS77bewStm3j4zISCweHrYBV+ypWN++hHz1FQ5eXiRv2Up4n76kHjtu93bMYGpxVbx4cRwdHTl79my25WfPniUoKOi62yYmJvLLL78wePDgbMuztsvJPl1dXfHx8cn2EBERMZPOTbkj+vPPISMDz5Yt8Wh0e6PQ2ZNbrVo4BgRgTUwkads2s+OYJj0ykpS9e8FiwatdO7vt161GdRz9/bEmJpK8Y4fd9ltY2boEtmyJg5tbrrTh1bIFob/8jHPp0qRHRBDety+JGzbkSlt5ydTiysXFhQYNGrD0n8uOAFarlaVLl9Ks2fWr5N9++43U1FT69euXbXn58uUJCgrKts+4uDg2btx4w32KiIhI4ZV6+DCxf/4FQInhw80N8x8WBwe8/umimLCi6HYNzLpq5V6/Pk4BAXbbr8XBAc/mzQF1DbwZCf8agj03uVaqROivM3CvWxdrXBwRQx4lZubMXG0zt5neLXDEiBH873//Y+rUqezfv58nn3ySxMREBg0aBED//v0ZPXr0Fdt9++239OjRg4D//I9nsVgYPnw4b775Jn/++Se7d++mf//+lCpVih49euTFWxIREZF86Pxnn4Fh4N2pE+61apod5wpebS53DSzK813lxn0+WTTf1c1JO3GC1MOHwcnJ9m8yNzkFBFB26hR87roLMjKIHPMqZ99/H8NqzfW2c4N5kzr8o0+fPpw/f56xY8cSFRVF3bp1WbBggW1AioiICBz+M4rPwYMHWbNmDYsWLbrqPkeNGkViYiKPPfYYMTExtGzZkgULFuCWS5c1RUREJH9L3r2b+MVLwMGBEs88bXacq/Js2QKcnEgLDyftxAlcypUzO1Keyrh0iaTNm4HcuWLi1eJycZWybx8ZFy/i5O9v9zYKg6yrh56NG+Ho65snbTq4ulLqg/dxCQ0letIkLn77HekREZR6910cPDzyJIO9mH7lCmDYsGGcOHGC1NRUNm7cSJMmTWyvrVixgilTpmRbv2rVqhiGQceOHa+6P4vFwuuvv05UVBQpKSksWbKEKlWq5OZbEBERkXzs/MefAOB7zz24VqpkbphrcPTywqNhQ6BojhqYsHIlZGbiWrUqLrkwOqZTiRK4VqsGhkHi2nV2339hkXX10Kt97nYJ/C+LxUKJp4dR6v33sDg7E794CSf6PUz62XN5muN25YviSkRERCS3JG7YSOK6deDsTPFhQ82Oc11FuWtgQlaXwFz8Uu+lroHXlREdTfL27UDu/h2ux7dbN8pOnYJjsWKk7NtH+P33k7J/vylZboWKKxERESm0DMPg/CefAFDsvvtwKVPG3EA3YBuSffMWMhMSTU6Td6zJySSsvlzw5OYgCp7/dA1MWLcWwzByrZ2CKn7ZMjAM3MLCcL7ByN25yaN+fUJ/nYFLxYpknD1L+EP9iF+23LQ8OaHiSkRERAqthBUrSN6xA4ubGwFPPG52nBtyLV8e53JlIT2dxPVFp+ta4rp1GCkpOJcqhWv16rnWjnuDBljc3ck8H03qwYO51k5BlTUEu1lXrf7NJSSE0J9/wrN5M4ykJE4NHcqFKVPyfVGs4kpEREQKJcNq5fwnnwLg/3A/nAMDTU50c7zbtgWKVtfA+MX/zKvUoX2uTuzs4OKCR+PL85upa2B2mQkJJK2/PM+UPSdwvh2OPj6EfPUVfn36gGFw7p13iRr/Wr6eCFrFlYiIiBRKcfPnk3rwIA5eXgQMHmx2nJtmu+9q1aoCOxx1ThgZGSQsv9zlKzeGYP8vrxYtAc139V+Jq1djpKfjEhqKS4UKZsexsTg7EzR+HIEvvggWCzEzZnDy8SfIjIszO9pVqbgSERGRQsdIT788rxXg/8ggHP38zA2UAx4NG+Lg4UHm+WhS9u4zO06uS9q6jczYWBz9/PCoXz/X2/Nsebm4St66FWtSUq63V1BkXT30zuWrh7fCYrEQMGggZSZ9jsXDg8R16wh/4EHSTp40O9oVVFyJiIhIoRMzZw7pJyJwLFYM//4DzI6TIxYXl/8feKEIDMmedZ+PV7t2WJxyfwpWl/KhOJcqhZGeTuKmTbneXkFgTUuz/VvLi6uHt8r7jjsI/WE6TiVLknb0KOH39yFp23azY2Wj4kpEREQKFWtqKtGTvgAg4PHHcPTyNDlRzmWNGljYiyvDMIhf+s8Vkzy6z8disdiuXmm+q8uSNm7EmpiIU4kSuNWubXac63KrUYPQX3/FrUYNMi9dImLgQGL/mmt2LBsVVyIiIlKoxPzyCxlRUTgFBVHsgQfMjnNLvFq3BiBl924yzp83OU3uSdm3j4wzkVjc3fFs3jzP2vXUfFfZxC/Jmjj4DiwO+b88cC4ZSLkfpuPVoT1GWhpnXniB859PyhcjCeb/oyciIiJykzITEon+6msAij/1JA6uriYnujVOJUrgVqsWAAmrVpucJvdkTRzs1bIlDm5uedauZ7Nm4OhI2vHjpJ8+nWft5keG1Ur8sqwJnPNvl8D/cvDwoMxnnxEw5PJgNdGff86ZF0ZhTU01N5eprYuIiIjY0aXp08i8eBHncmXx69nT7Di3xTZqYCHuGph1xSQ3Jw6+Gkdvb9zr1AE0amDyzp1kno/GwcsLzyaNzY6TIxYHBwJHjiTojdfByYm4uXOJGDiIjIsXTcuk4kpEREQKhcyYGC58+x0AJZ5+Bouzs8mJbk/WfVeJa9dipKWZnMb+0iIiSD10CBwdbYVkXlLXwMtsVw/btMHi4mJymltT7L77KPvN/3Dw8SF5+3bC7+9D6pEjpmRRcSUiIiKFwoVvv8WakIBrlSr43NnV7Di3za1mTRyLF8eamEjS1q1mx7G7rKtWHo0bmTJUvlfWoBYbNmBkZOR5+/mBYRjZhmAvyDybNiX0l59xLluW9FOnCO/7AAlr8/6qpIorERERKfDSz53j4vQfACgxfHiBuCn/RiwODraBLRJWrDA3TC7IGoLdrPt83GrWxNHXF2t8PMm7dpmSwWxpR4+SduIEFmdnPFu1NjvObXOt8H/t3Xd8U2X7x/FPku500FI6mGXvDZZSkL1UlsiyAoLCI6Lig/5UXCgquB7kUREEmQ5AeFgq4igbyp4KsqGsllHopCs5vz9OG6gUbEqSk7TX+/XKyzScnHxboSdX7vu+7mpELF6Ed4vmmNPSODvqX1xbtMihGVz/N48QQgghSr2rM75EyczEu3FjfDu01zqOzVhasq8vWeuucq9c4cZedX8iv04dNcmgMxgwRqsdCkvr1EDL6GHrKJfcsqAwboGBVJ4zh4DevcBkIuGtt0mc/D6KyeSQ17e6uIqIiGDixInEx8fbI48QQgghhFWyz53j2pIlAJT797/R6XQaJ7IdY+tocHcn+8wZsk6d0jqOzaSuWweKgleDBriHh2uWwxitTg0srU0tbo4euvaUwL/Te3gQ/v77lHt+LABJ8+dz7plnMaen2/+1rX3C888/z7Jly6hWrRpdunRh0aJFZGnc8lAIIYQQpdeVz6dBTg7G1lEYW0VqHcemDL5GfFo0B0pW10DLm3qN1/nkN7XIPHiQ3GvXNM3iaDkXL5L5xx+g0+HXUZvRQ3vS6XQEP/UUFT6Zgs7Tk/StW8k+c8bur1us4mrfvn3s2LGDunXr8uyzzxIeHs4zzzzDnj177JFRCOEs0i5DcuneD0QI4Vyyjh8nedUqQF1rVRL5tW8PlJziypSWTsbWOAD8Omu7r5J7aCieNWuCopARF6dpFkdLjV0LgHfTprgFB2ucxn78e/SgyoL5VPjPx3jVq2f31yv2mqtmzZrx6aefcuHCBSZMmMBXX31Fy5YtadKkCXPmzHGKHZKFEDYUvx3+2xg+aw4X92udRgghALj86WdgNuPbuRPejRppHccu8tuUZ+zajSktTeM09y5900aUnBw8qlTBo3p1reNgjFZHr0rb1MDU2PzRQ9fZOLi4vBs3dtj3WeziKicnh++//55evXrxwgsv0KJFC7766iv69evHq6++SkxMTJHOM23aNCIiIvDy8iIyMpIdO3bc9fjr168zZswYwsPD8fT0pFatWqxevdry52+99RY6na7ArU6dOsX9NoUQAOf3wLePQE465N6A74dBZrLWqYQQpdyNg3+Q+uuvoNNR7rnntI5jNx4REXhUqQI5OaRv2ap1nHuW30TBt3Mnp1gfZ8xvyb55c6kZHDAlJ5OxYyegXUORksrN2ifs2bOHuXPnsnDhQvR6PUOHDuWTTz4pUMD07duXli1b/uO5Fi9ezLhx45gxYwaRkZFMnTqVbt26ceTIEUJCQm47Pjs7my5duhASEsLSpUupUKECZ86coczf9kaoX78+v+fN5QVwc7P62xRC5Es4CF/3hawUqNwaUs7BtVOwcgwM+Bqc4MIohCidLv/3vwD493wIr1q1NE5jX77t25M0fz5pGzbg362r1nGKTcnOtkxvdJYRE58WzdF5epJ76RJZx46V+L9LkNfa32TCs2ZNtXAXNmP1yFXLli05duwY06dP5/z583z88ce3jQxVrVqVQYMG/eO5pkyZwsiRIxk+fDj16tVjxowZ+Pj4MGfOnEKPnzNnDklJSaxYsYLo6GgiIiJo164djRs3LnCcm5sbYWFhlltwCZ5HKoRdXfoLFvSBzOtQsSXEfA/954HBAw7/ANumaxxQCFFape/YobbPdnOj3DPPaB3H7iwt2TduRDGbNU5TfOnbd2BOS8NQLhjvv71/04reywufvEGB9FIyNTB/9NCvi3MUuCWJ1cXVyZMnWbNmDf3798fd3b3QY4xGI3Pnzr3rebKzs9m9ezedb/nUQq/X07lzZ+LusKBw1apVREVFMWbMGEJDQ2nQoAGTJk3C9Le+9ceOHaN8+fJUq1aNmJiYf2wbn5WVRUpKSoGbcA0mk4n169ezcOFC1q9ff9vfBXEPrp6ABb0g4wqEN4aYpeDpBxWaQ7dJ6jG/vQFnd2qbU4gSSq5Nd6YoCpenqqNWZR7ph0flyhonsj+f5s3RG42Yrlwh888/tY5TbJZ1Ph07OdVGz/ldA0vDflfmzEzS8r5P3xLWgt0ZWP23+tKlS2zfvv22x7dv386uXbuKfJ4rV65gMpkIDQ0t8HhoaCgJCQmFPufkyZMsXboUk8nE6tWreeONN/jPf/7Du+++azkmMjKSefPmsWbNGqZPn86pU6do27Ytqampd8wyefJkAgICLLdKlSoV+fsQ2lm2bBk1atSgQ4cOPProo3To0IEaNWqwbNkyraO5vmunYX5PSEuEkPowZAV4l7n55y2fhPp9wZwLSx6HjCRtcgpRgsm16c7SN27kxp496Dw9CR49Wus4DqHz8LjZeMFFNxRWzGbS8jrUad2C/e9889ZdZezahfnGDY3T2Ff61q0oN27gVj7cId3zShuri6sxY8Zw9uzZ2x4/f/48Y8aMsUmoOzGbzYSEhDBz5kyaN2/OwIEDee2115gxY4blmB49etC/f38aNWpEt27dWL16NdevX+f777+/43nHjx9PcnKy5VbY9yecy7Jly3jkkUdo2LAhcXFxpKamEhcXR8OGDXnkkUekwLoXyedhfi9IOQ/BtWDoSvAJKniMTge9PoOyNdQ1WMtGgQtPUxHCGcm1qXCK2cylvFGrwJgY3P/2IW1Jlt81MG39em2DFFPmgQPkXr6M3mjEJ9K59iPzqF4dt7AwlOxsMnbt1jqOXVmmBHbq7BQNRUoaq4urQ4cO0axZs9seb9q0KYcOHSryeYKDgzEYDCQmJhZ4PDExkbCwsEKfEx4eTq1atTAYDJbH6tatS0JCAtnZ2YU+p0yZMtSqVYvjx4/fMYunpyf+/v4FbsJ5mUwmXnjhBR566CFWrFhBq1at8PX1pVWrVqxYsYKHHnqIF198UaYIFkdqgjpidf0MBFaFoavAt1zhx3r6Qf/54OYFx3+DzVMcm1WIEk6uTYVL/eUXsg4fRm80Unbkk1rHcSjfdvcDkPnnn+RcuqRxGuvlbxzs264deg8PjdMUpNPpSsXUQCU3l7S1eaOHMiXQLqwurjw9PW8riAAuXrxoVVc+Dw8PmjdvTmxsrOUxs9lMbGwsUVFRhT4nOjqa48ePY77lE/KjR48SHh6Oxx3+kaalpXHixAnCw8OLnE04t02bNnH69GleffVV9Od3wcpn4Kzawl+v1zN+/HhOnTrFpk2bNE7qYtKvwILekHQCAirDsB/A/x/+3YQ1gAf/o95f9x6ckp+5EMJ+lNxcLv/3UwCChg/HLTBQ40SO5RYcjFfDhgCku9g1TlEUUn/L31fJOd/U508NTNtScourjD17MF2/jiEgAJ8WzbWOUyJZXVx17drVMlUh3/Xr13n11Vfp0qWLVecaN24cs2bNYv78+Rw+fJjRo0eTnp7O8OHDARg6dCjjx4+3HD969GiSkpIYO3YsR48e5aeffmLSpEkFpiO++OKLbNiwgdOnT7N161b69u2LwWBg8ODB1n6rwkldvHgRgAYXl8DsrrD3a5j3IOxfrD7eoEGB40QRZCTB133g8l/gVx6GrYQyRVzf0fQxaBIDihmWjoDU2z98EUIIW0heuZLs06cxlClD0OPDtI6jCVedGph98iTZZ86gc3fHeP/9WscplDEqCvR6so+fIKeEvoewjB526IBOtiqyC6t/qh9//DH3338/VapUoWnTpgDs27eP0NBQvv76a6vONXDgQC5fvsybb75JQkICTZo0Yc2aNZYmF/Hx8ehv6SRTqVIlfvnlF/7973/TqFEjKlSowNixY3n55Zctx5w7d47Bgwdz9epVypUrR5s2bdi2bRvlyt1hapNwOeFcBuCPn2bQqqIbBNeGK0dg+Si4eow/PDuox8loZdFkJsM3/dT9rIwhMGwVBFWz7hwPfAwX9sKlQ/C/J9R1WnrDPz9PCCGKyJydzeXPpwFQdtQoDL6+GifShm/79lz5/HPSt2zFnJ3tdNPr7iR/1MonqpXT/r8zBATg3bAhN/bvJ33LFso88ojWkWxKURTS8tdbOenoYUmgU4qxFXV6ejrffvst+/fvx9vbm0aNGjF48OA7tmZ3NSkpKQQEBJCcnCxz3J1JVir8/ham7bOo8VkaDcsbWbHsf+hrdobYt2HLVMyKQp+fAvgjyY1jx44XWJ8nCpGVphZWZ7eBdxA8/hOEFrNz0JVjMLM9ZKdB2xeh0xs2jSpKB/n9e2el/WeTtOBrEidNwi0khOq//oLey0vrSJpQzGaOtWuH6fIVKs+ZjbF1a60jFcmp/gPIPHiQsLffJnDgAK3j3NHlzz7nyrRp+HXvTsWpn2gdx6YyDx/mVN+H0Xl5UStuK3pvb60juQxrfv8Wa4MBo9HIqFGjmDZtGh9//DFDhw4tMYWVcFLHY+GLKNj5FQa9jv889QA/Hk6nz/99Ttz27aS2epG4iOfpsziTH3ef4+Nuvhgyrmid2rllZ8DCQWph5RUAQ1cUv7ACCK4JvdS1EGz6GI79ZpOYQghhTk/nypdfAhD89NOltrAC0On1+OZNq0t1kamBOQkJZB48CDodfh07aB3nrixNLeLiUEpYY6z80UNjm2gprOyo2JMtDx06RHx8/G1d+nr16nXPoYSwuHEdfn1dXVcFUKYy9PqMh6u1Z2nLZbzwwgu0vuVTu6qVwln6WC4Ph8bDV51g8CK16YIoKDcLFsfA6U3g4QePLVM3Cr5XDfrBma2w8yu1PftTmyCg4r2fVwhRqiV9/Q2mq1dxr1SJMv0e1jqO5nzbtyf5f8tIW78BZfx4p2+nnZrXvMy7SRPcnHyZhnfDhuj9/TEnJ5N58CDeTZpoHclm8v8/+HXurHGSks3q4urkyZP07duXgwcPotPpyJ9VmP8PW9pfC5s5+gv88DykXlC/vu9f0OlN8FTnaj/88MP07t2bTZs2cfHiRcLDw2nbti2G66fhuwFw9TjM6QaPzIFa3TT7NpxObjZ8PwxOrAV3H4hZAhVb2O783SbBuV1wcZ+6wfDjq8HNNdYECCGcjyk5mauzZwNQ7rln0clMGYxRrcHdnZz4eLJPncazWlWtI91Vmgu9qde5uWGMiiL1l19I27ylxBRX2WfPknXkCBgMlqYowj6snhY4duxYqlatyqVLl/Dx8eHPP/9k48aNtGjRgvUuMjwtnFxGkjrq8d0AtbAKqg7Df4YHPrQUVvkMBgPt27dn8ODBtG/fXl1jVbY6PPEbRLRV1/8sHATbpoP1ywtLHlMuLHsSjv6s7k81eBFUKXzrg2Jz84QB89Wphud2wu9v2fb8QohS5ersOZhTU/GsWRP/Bx7QOo5TMPgaMbZUPxRL27BB4zR3Z0pOJn3HTgD8OnXUOE3RlMT9rvI3DvZp0aLUbWHgaFYXV3FxcUycOJHg4GD0ej16vZ42bdowefJknnvuOXtkFKXJoVUwLRIOLAadHlo/C09thipWLtj1CYIhy6HZULVF+JpX4KcXwJRjn9yuwGyCFaPh0ErQu8PAb6GanT69CoyAPjPU+9umweEf7PM6QogSLffyZZLyOhGXe34sOmlSZOHbvj3g/MVV2oYNkJuLZ80aeEREaB2nSHyj1eLqxoEDmG7ZesiVpcbm7zHm/KOHrs7q4spkMuHn5wdAcHAwFy6oU7aqVKnCkSNHbJtOlB5pl9UpZN8PgfRLanv1J36Dru+Ch0/xzmlwh56fqudAB7tmw7f91XVcpY3ZDD+MhYPfg95NHVmqaedfsHUeUItjgBVPQ9JJ+76eEKJEybl0ibOjn0a5cQOvRo3w7egaox6Okj+1K2PXLkypqRqnubP8Jgq+LvSm3r18eTyqVQOzmfS4bVrHuWe5V69yY/cewHVGD12Z1cVVgwYN2L9/PwCRkZF8+OGHbNmyhYkTJ1KtmpV74wihKHBwKXwRCX8uB50B2r6gNkKwxTognU59gz/oW3V90cl16sbDSafu/dyuQlHg55fUpiA6PTw8C+o86JjX7jQBKrWCrBR1nVdOpmNeVwjh0jL/+ovTAwaS+ccfGMqUIfytCU7ftMHRPKpUwaNqVcjNJX3LVq3jFMqcmUla3tQ6v06uU1zBLVMDt7j+1MC0detAUfCqXx/38uW1jlPiWV1cvf7665jNZgAmTpzIqVOnaNu2LatXr+bTTz+1eUBRgqUmwKIYddPZjKsQ2gBGrlWbVrh52va16jwII9aAX3l1w+GvOkG8638a9Y8URe22uHMWoIM+06GBAzttGdzVhiI+ZSHhgDo9Uwgh7iJ13TpOPxpDbkICHtWqEfH9Yrzq3cM2ESVY/uiVs04NTN8ah3LjBm7h4XjVd63/h75t2gCQtnkLxdgS1qnkjx7KxsGOYXVx1a1bNx5+WH1zVqNGDf766y+uXLnCpUuX6ChD9qIoFAX2fQfT7oMjP6nrf9q/CiPXQfkm9nvd8MZq8RbeRC3m5veE/Yvt93rOYN17EPe5er/nVGg8yPEZAiqoo2XoYPdcOPC94zMIIZyeoigkzZ/PuTHPoGRk4BPVioiF3+FRubLW0ZyWb/u84mrjRpS8D76dServeW/qO3VyuZFHn5Yt0Xl4kHvxItknXXdauyktnfS4OAB8O0lx5QhWFVc5OTm4ubnxxx9/FHg8KCjI5f7RCI0kn1PXPa0YDZnJaqHzrw3Q/mXHtOv2D4fhq6FuTzBlw/JRsPZddU1SSbPxI/UG0OMjaP64dllqdIJ2L6n3f3geLsv6TCHETUpuLgkTJ5I4+X0wmynTvz+VZ87EEBCgdTSn5tOsGXpfX0xXr5L5t/dmWlNyc0lbuxZwzRETvbc3Pi2aA67dNTB982aU7Gzcq1TGs2ZNreOUClYVV+7u7lSuXFn2shLWUxTYPQ+mtYLjv4HBU12P82QshNZ3bBYPI/RfAG3+rX698SP43wjIueHYHPa09TO1aATo8g5EjtI2D0C7l6FqO8hJh++HQna61omEEE7AlJrK2X89xfWFi0CnI+Sllwib+LbsZ1UEOg8PjHmd7dLWO9fUwIw9ezBdv44+IACfFjbcS9GBjNF5UwO3bNE4SfHdHD3sLAMhDmL1tMDXXnuNV199laSkJHvkESXRtdOwoLfarS47FSq2VBtWtB0HBqv3sbYNvR46vwW9p6nTEv9cDvMehNREbfLY0o5Z6jorgA6vQbSTbJGgN0C/r8A3DC7/BT+Ok73HhCjlss+d4/TgwaRv2YLO25uKn39G2RHD5U2gFSzrrpxsr1HLxsHt26Nz0+haf4+MeeuuMnbsxJyVpXEa6ynZ2Zb1eNKC3XGsLq4+//xzNm7cSPny5alduzbNmjUrcBPCwmyG7TPhi9ZwagO4eUO3STDiFyhXW+t0qqaPwdAV4B0I53erjS4SnGtqhVX2LIDVL6r3274A9/+ftnn+zjdEbXChM8CBRbBnvtaJhBAaydi7l9MDBpJ9/ARuISFU+eZr/GRNiNV8728LOh2Zhw6Rk3hJ6ziAun4uf9NaXxecEpjPs1ZN3EJCUDIzubF7t9ZxrJa+Yyfm1FQMwcF4N26kdZxSw+qPEvr06WOHGKLEuXoCVj4D8XntYatEQ6/PoGx1bXMVJqKNOj3xuwFw9TjM6aYWALW6aZ3MOvsXw6q8UapWY6DjG2oremcTEQ2d3oDf34LVL0H5ZhAuv/SFKE2Sf/yJi6++ipKdjWe9ulSaPh330FCtY7kkt+BgvBo2JPPAAdI2biCwf3+tI5H111/knD+PzsvL0nXPFel0OozR0SQvX07a5i0YW7fWOpJVLBsHd+ggG3A7kNXF1YQJE+yRQ5QUZhNs+0Jd75ObCe5G6PI2tHhCnYrnrMpWVzct/n4onN4ECwepo2yRTzlngfJ3fy6HFU8Bivqz7vaec+duPVZthX90jfoz/9cG8JKF60KUdIqicOWLL7jymdrF1LdjRyp89CF6o1HjZK7Nt939anG1wTmKq/xRK2N0NHpvb43T3BtjG7W4St+8GV5ystkgd6GYzaTF5jUU6SJTAh3Jid/tCpdz6S91g95fX1cLq2rt4ek4uG+kcxdW+XyCYMhyaDYUFLO6J9NPL4ApR+tkd/fXavjfk2rmpo/BAx87d2EF6t+HPtMhoDJcO6WOcsr6KyFKNHNWFhdeetlSWAUNH07Fzz6VwsoGfNu3B9R9pczZ2dqG4ZYmCiVgnY+xdWvQ6cg6etRppl0WRebBg+ReuoTeaMSnVSut45QqVr/j1ev1GAyGO95EKWTKhU3/gS/bwvld4OkPPT+FISsgsIrW6axjcFezd30X0MGu2Wrr+BvXtU5WuOO/w5JhYM6Fhv3V7K5QyIJazPafpzYUObwKts/QOpEQwk5yk5KIHz6ClB9+AIOBsLffJvTll2Sqko141auHW7lyKBkZZOzYqWmW7LNnyTpyBPR6yz5crswtMBCvBg0ASHehroGWNW/t7kfv4YCtboSF1dMCly9fXuDrnJwc9u7dy/z583n77bdtFky4iIQ/YOXTcHG/+nXNrvDQVHXjWFel00HrZyGomjoidHKdOiL36GIIqqp1uptObYRFMep+XXV7QZ8Zakc+V1KxuTqF8eeX1BHPCi2gUkutUwkhbCjrxAnO/uspcs6dQ+/nR8X/TnW5tSvOTqfT4du+HdeXLCVtwwZ820RrliU1r0ugT4sWuAUGapbDloxtosk8eJD0zZsp83BfreMUSf7ooWwc7HhWf8Tdu3fvArdHHnmE9957jw8//JBVq1bZI6NwRrnZsG4yzGynFlZeZaDvl/Do965dWN2qzoMwYg34lYcrR9ROgvHbtE6lit8G3w1Sp1/W6g79ZmvX1v5e3TcK6vVRR9+WPA4Zss2DECVF+tatnB40mJxz53CvVImIRQulsLKTW1uyKxpOsy5JUwLz5TflSN+6FcUF9nrNOnmS7FOnwN0d3/vv1zpOqWOz+UOtWrUiNu/TCmtMmzaNiIgIvLy8iIyMZMeOHXc9/vr164wZM4bw8HA8PT2pVasWq1evvqdzCitd2Asz28OG99U3xHUegjHbofEg51/rY63wxjByLYQ3gYyrML+n2pVPS+d2wzePqJvxVu8I/eeDmwsP+et0aifJoOqQcg6W/0tt4y+EcGnXFn9P/MhRmFNT8W7WjIjFi/Cs7oQdY0sIY1QUOnd3cs6eVd9YayD36lVu7NkLgF+njppksAfvRo3Q+/piun6dzEOHtI7zjywNRVq1wuDnp3Ga0scmxdWNGzf49NNPqVDBuhGLxYsXM27cOCZMmMCePXto3Lgx3bp149KlwhcMZmdn06VLF06fPs3SpUs5cuQIs2bNKvC61p5TWCEnE35/G2Z1gkt/gk9ZtWX5wG/AL0zrdPbjHw7DV0PdnuoUvOWj1G6IWhQAFw/AN33VzZgj2sLAb8Hdy/E5bM3LHwbMBzcvOPYrbPlE60RCiGJSTCYSP/iQhAkTwGTCv2dPKs+bi1tQkNbRSjS90YjPffcBkLZ+gyYZ0tatA7MZr3r1cLfyPaEz07m749MqEkDtGujkLKOHMiVQE1YXV4GBgQQFBVlugYGB+Pn5MWfOHD766COrzjVlyhRGjhzJ8OHDqVevHjNmzMDHx4c5c+YUevycOXNISkpixYoVREdHExERQbt27WjcuHGxzymK6OxO+PJ+2DwFFBPUfxjG7IAG/UreaFVhPIzQfwG0+bf69caP4H8jIOeG4zJcOgxf94HMZKgUCYMXgYeP417f3sIawgN5v0PWvgunnf8CJoQoyJyezrlnnyNp7lwAgp97lvIffiAL6h3k1qmBWigJGwffSf7UwLTNzt3UIicxkcwDB0CnK1Gjh67E6kUan3zyCbpb3kzr9XrKlStHZGQkgVYsXMzOzmb37t2MHz++wLk6d+5MXFxcoc9ZtWoVUVFRjBkzhpUrV1KuXDkeffRRXn75ZQwGQ7HOCZCVlUVWVpbl65SUlCJ/HyVedgasew/ipgEKGEPgoSnqKE5po9dD57egbA344Xl1b6nr8TBoIfjZefPLK8dhfi91amL5phCzBDx97fuaWmg6BM7Ewf7vYOkI+Ncm+/9shXBSrnZtyklI4Ozop8k6fBidhwfhkycR8OCDWscqVXzbtyNx0iQy9uzBlJrq0Clh5vR00rduBcCvU8lZb5XPmFdc3di3z+E/W2vkNxTxbtwYt3LlNE5TOlldXD3++OM2eeErV65gMpkI/duO7KGhofz111+FPufkyZOsXbuWmJgYVq9ezfHjx3n66afJyclhwoQJxTonwOTJk6XTYWFOb4FVz0DSSfXrxoPVjXV9SvnUjqaPQWAELH4Mzu9WG10MXgRhDezzekmn1LVe6ZcgtAE8tqzkbrir08GDH6vr+i4fhv89AUNXul4XRCFswJWuTTf+/JNzo58m99IlDEFBVJz2OT5Nm2odq9TxqFwZj2rVyD55kvQtW/Dv3t1hr522aTNKdjbulSvjWaumw17XUTwqVsSjShWyz5whfds2/Lt00TpSodLyRg/9SuDooauwelrg3LlzWbJkyW2PL1myhPnz59sk1J2YzWZCQkKYOXMmzZs3Z+DAgbz22mvMmHFv++OMHz+e5ORky+3s2bM2SuyistLgpxdh3gNqYeVXHh5dAn1nSGGVL6INPBmrjmIln4U53eDoL7Z/neRzsKAXpF6A4NpqoVHS/x94GGHAAnA3wulNsH6y1omE0ISrXJtSf/+dM48NIffSJTxqVCfi+8VSWGnIMjVw3XqHvm7+iIlfp04FZjiVJPmjV+lOOjXQlJJCel4Tt5LUrdHVWF1cTZ48meDg4NseDwkJYdKkSUU+T3BwMAaDgcTExAKPJyYmEhZWeHOE8PBwatWqVWCz4rp165KQkEB2dnaxzgng6emJv79/gVupdWIdfBEFO2epXzcbBmO2Qa2u2uZyRmWrwxO/qY0lstNg4SDYNh1s1QI3NUEdsboer+65NWwVGG//t1cilasFvT5V72/8CI79rm0eITTg7NcmRVG4OnsO5559DuXGDYzR0UQsXIhHxYpaRyvVLMXVpk0oDmq8pGRnW9Z5+XUpuW/qjXn7h6Vv3qxpu/s7SduwAXJz8ahRHY+ICK3jlFpWF1fx8fFUrXr7RqpVqlQhPj6+yOfx8PCgefPmBdq3m81mYmNjiYqKKvQ50dHRHD9+HPMtvyyOHj1KeHg4Hh4exTqnyJObBT+MVRsmJMdDQGUYskJ9g1tSp6DZgk8QDFkOzYaCYoY1r8BPL4Ap597Om3ZZXWOVdBLKVIZhP5TsjoyFafgItHhCvb9spDqKJ4RwCkpODglvvsmljz4CRaHM4EFU+nKG065DKU18mjdT24YnJZF58KBDXjN9507MqakYypbF+5YmYyWN8b77wN2dnPPnyTlzRus4t8lvKFIS17y5EquLq5CQEA4cOHDb4/v376ds2bJWnWvcuHHMmjWL+fPnc/jwYUaPHk16ejrDhw8HYOjQoQWaU4wePZqkpCTGjh3L0aNH+emnn5g0aRJjxowp8jlFIUw5avOA3fPUr1uOhKfjoHoHTWO5DIM79PwUur4L6GDXbPi2P9y4XrzzZSSpRe6VI+BfQS2sAkrpJ8HdJql7jd1IgiXD771oFULcM1NyMvGjRnF9yVLQ6Qh9dTxhb76Jzs1FNzIvYXTu7pbpa6kO6hqYlj8lsGMHdIaSu0ZWbzTi06wZ4HxdA82ZmaRt2gTIlECtWV1cDR48mOeee45169ZhMpkwmUysXbuWsWPHMmjQIKvONXDgQD7++GPefPNNmjRpwr59+1izZo2lIUV8fDwXL160HF+pUiV++eUXdu7cSaNGjXjuuecYO3Ysr7zySpHPKf7GbIJlo+CvH8HgCTFL1YYCJbETnT3pdND6WRj0Lbj7wMl1MLur2ozCGpnJ8HVfSPwDfENh6Cq1eUZp5e6lbpLsGQDndsDvb2mdSIhSLTs+ntODBpMRtw2djw8Vv5hG0NChJXaNjavybZ83NXCD/fe7UszmmyMmpeBN/a1TA51JelwcSkYGbmFheDWor3WcUk2nWDlpNDs7myFDhrBkyRLc8j6lMpvNDB06lBkzZuBRAvaySElJISAggOTkZKeb425TZjOsfBr2LwS9Owz6TtZW2cLF/fDdILUJhU9ZdaPfKkWYlpqVCl8/rBYRPmXh8Z8gpK7987qCwz/C4hj1/sBvSudWAKVEqfn9Wwxa/2wydu3i3DPPYrp+HbewMCrNmI5XnToOzyH+WW5SEsei24CiUGPDetzt+AHzjQMHOD1gIHofH2rGbUXv6Wm313IGmYcPc6rvw+h8fKi9LQ6dk7zvvfD66yQv/R+BMTGEvfG61nFKHGt+/1o9cuXh4cHixYs5cuQI3377LcuWLePEiRPMmTOnRBRWpYaiwI/Pq4WVzgD950phZSvhjWHkWghvou5LtaAX7F989+dkZ6gF2bkd6hq3ISuksLpV3Ycg6hn1/oox1o8ICiHuSfKqVcQPH4Hp+nW8GjQg4vvFUlg5MbegILwaNQTsP3qV+pvacMh4//0lvrAC8KxdG0NwMEpGBhl79modBwDFZCJt7TpAWrA7A6uLq3w1a9akf//+PPTQQ1SpUsWWmYS9KQr8/DLsmQ86PfSbJSMBtuYfDsNXqz9XUzYsHwVr31VHC/8uJ1MdlTmzGTz81AYZ4Y0cn9nZdX4LKkVCVjIsGab+3IQQdqWYzVz673+58NLLKDk5+HXpQpWvF+AeEqJ1NPEP/Nq3ByBtw0a7vo6lBXspmBIIoNPr8Y1uDUD6FueYGnhj715MSUnoAwLwadFC6zilntXFVb9+/fjggw9ue/zDDz+kf//+Ngkl7EhR4Lc3YceX6te9p0GDftpmKqk8jNB/AbT5t/r1xo/gfyMg58bNY3Kz1ULhxFp1X6fHlkKF5trkdXYGd3hkLngHqVMvfxn/z88RQhSbOTOTCy++yNXp6l6SZUc+SYX/TkXv7a1xMlEU+S3Z07duxZyVZZfXyDp5kuyTJ8HdHd9299vlNZxRfsMQZ2lqkT966Ne+HTp3d43TCKuLq40bN/LAAw/c9niPHj3YuNG+n44IG1g/Gbbm7R/00FRo8qimcUo8vV4dcen9hbqu7c/lMO9BSE0EUy787wk4ugbcvODRxVC5ldaJnVtABXWkFR3smgMHbt/QXAhx73KvXCF+2OOkrP4Z3NwIf+89Ql54AZ2+2BNehIN51q2LW0gIyo0bZOzYaZfXyG9kYYyMLFVt+I2t1ZGrrMOHyb1yRdMsiqJYRg99O8mUQGdg9W/JtLS0QtdWubu7k5KSYpNQwk42/Qc25I06dv8AWkh7eodpGgNDV4B3IJzfDbM6wvdD4fAqMHioXQarttU6pWuo0Rnu/z/1/g9j4fIRbfMIUcJkHj3K6QEDubF/P/qAACrPnk2Zfg9rHUtYSafT3dxQ2E7rrlJj80ZMStk6H7eyZfGqVw+A9C3ajl5lHT1Kzrlz6Dw98c0bURPasrq4atiwIYsX3744f9GiRdTL+4smnFDcFxA7Ub3f+W1o9ZS2eUqjiDbwZCyUrQEp5+DIT6B3gwEL1IJBFF37V6Dq/ZCTDt8Pg+x0rRMJUSKkbdrMmcGPknPhAu5VKhOxaCHGyPu0jiWKydKSff16rGwO/Y9yEhPJ3K/ue+rboaNNz+0KnGVqoKWhSHQ0eh8fTbMIldU7/r3xxhs8/PDDnDhxgo4d1X9MsbGxfPfddyxdutTmAYUN7Pzq5vqU9q9Cm+c1jVOqla0OT/4OS5+A+Djo8wXU7qF1KtejN0C/2TCjDVw+DD+9AH2mq/uNCSGKJem770h8bxKYTPi0aEGFzz7FLTBQ61jiHhhbtULn4UHOuXNknzyJZ/XqNjt32tq1AHg3box7aOlrcGJsE83VmTNJ37IFxWzWbMqspaGITAl0Glb/TejZsycrVqzg+PHjPP3007zwwgucP3+etWvXUqNGDXtkFPdiz9fqG09QGyu0e0nbPEKdGjhkGbx8Gur31TqN6/INgUfmqB0v9y+EvV9rnUgIl6SYTCRMmkTixHfAZCKgTx8qz5kthVUJoDca8blPHXlMW2/bqYH56618S9mUwHw+TZqg9/HBlJRE5uHDmmTIPneOrMOHQa/Ht2MHTTKI2xWrzH7wwQfZsmUL6enpnDx5kgEDBvDiiy/SuHFjW+cT9+LAElj1rHq/1dPQaYJ8su9M3Er+fiB2F9EGOr6h3l/9f5BwUNs8QrgYU1o6554ew7UF6ocT5f79b8InT3KajVHFvbOsu1q/3mbnNKWkkL59O1B6WrD/nc7DA5/ISADSNZoamJY3auXTvLl8GOJEij2GuXHjRoYNG0b58uX5z3/+Q8eOHdm2bZsts4l7cWglLP8XoECLEdBtkhRWomSKfh5qdoXcTLVJSKYLN9ZRFHWD5EOrYO17sPwpSDykdSpRQuVcuMCZmBjSNmxA5+lJhalTCf7XKHRyrShR8tddZezZg8lGjcfSNmyE3Fw8qlfHs2pVm5zTFRnbRAOQvlmb/a7yRw/9upTOAtdZWbXmKiEhgXnz5jF79mxSUlIYMGAAWVlZrFixQppZOJMja2DpCFBM0CQGHviPFFai5NLroe+X8OX9kHQSVj0D/ec7/9/53Cy4/Jc62nbrLetvb37O7YSnNoO77C0kbOfGwYOcffppTJevYAgOptIX0/BuJJuXl0QelSrhUb062SdOkL55M/6FbKdjrdTf87oElvJ1Pr5t2pAIZOzdiyktHYOv0WGvnZuURMbu3WqOjqX7/4OzKfLIVc+ePalduzYHDhxg6tSpXLhwgc8++8ye2URxHI+F74eAORcaPAK9PlPffApRkvkEQf956l5ih1bCjplaJyroxjU4tUnt2rl8NEyPhknl1YJw5RjYPgPObFELK4MHhDeGpo+BXzhcPQ7r3tP6OxAlSMovv3JmyFBMl6/gWasWVb9fLIVVCWfLluzmrCzSNm0CZMTEo0oV3CtVgtxcMnZsd+hrp61bD2YznnXr4lGxgkNfW9xdkUeufv75Z5577jlGjx5NzZo17ZlJFNfpzbAoBkzZULcn9J2hdlUTojSo2AK6vgtrXoZfXoMKzdXHHElRIPmsOgJ18cDN0ajk+MKP9yoDYQ0hrFHefxtCcC1wy1vvcvQX+G4AbP0c6vaCStISWxSfoihcnTmLy598AoDx/rZUmDIFg6+vxsmEvfm2b0fSnDmkbdyEYjKhMxT/vUF6XBxKRgZuoaF41a9vw5SuydgmmusLF5G+eTN+HR3Xkt4yelhKG4o4syIXV5s3b2b27Nk0b96cunXrMmTIEAYNGmTPbMIa8dvh2wGQewNqdoN+c8DgrnUqIRwr8l/qCNDhVbDkcfjXRnVUyx5ys+HKkb9N6zsAmcmFH1+mcsEiKqwRBFS8+/TFWt2g8aOw/ztY8TQ8tUmmB4piUbKzuTjhLZKXLwcg8LHHCH3lZXRuVu/IIlyQT9Om6P38MF27xo0DB/Bp2rTY57p1SqBW7cediW+bNlxfuIg0B24mbE5Pt2xeXFobijizIv9WbdWqFa1atWLq1KksXryYOXPmMG7cOMxmM7/99huVKlXCz8/PnlnFnZzfA98+om6oWq29uimtm3R6EqWQTge9P4fEP9T1V8ufgsGL7n1qbGYyJPxRsIi6/Jc6Svx3ejcoVxfCbymkQhuAd5nivXb3SXBiLVw9BusmQdd37ulbEaVT0tdfq4WVXk/oq68S9FiM1pGEA+nc3TG2iSb15zWkbdhQ7OJKMZlIW7sOkBGTfD6RkeDmRs6ZeLLPnsWjUiW7v2bali0o2dm4V6qEZ61adn89YR2rP7IyGo2MGDGCESNGcOTIEWbPns3777/PK6+8QpcuXVi1apU9coo7STgIX/dV12pUiYZBC8HdS+tUQmjHK0BtaPFVZzj2C2z9r7rHW1EoCqScL1hEXTwA188UfrxnwC0jUXm3crVt22bfOxB6ToWFgyAuf3pgS9udX5QKQUOGkLF7D4GDBuJ7//1axxEa8GvfPq+42kjI888X6xw39u7FlJSE3t8fn5byewjA4OuLT5MmZOzaRfrmzXgMHmz31ywweujszZtKoXuaD1C7dm0+/PBDJk+ezA8//MCcOXNslUsUxaW/YEEfyLwOFVvCo4vBw0frVEJoL7wRPPAR/PAcxL4DFe+DiOiCx5hy4Mqxm0VUQt4aqRvXCj9nQKXbC6kyVRzTlbB2D2g0CA4sgpVPw782yYcowio6Dw8qfTFN6xhCQ8a2bUGnI+vwYXISEnAPC7P6HJaNg9u3Q+cuSw/yGdu0IWPXLtI2byHQzsWVkpNj2RBaRg+dk00mWxsMBvr06UOfPn1scTpRFFdPwIJekHFF7SwWsxQ8ZVqmEBbNhsKZrWpBsnQE9PlC/XeTX0RdOgymrNufpzNAuTo3C6jwRuq0Pnut3Sqq7pPh5Dq4chTWT4IuE7XNI4RwKW5BQXg3bsyNfftI27CRwIEDrHq+oiik5m1a69dJ1vncytimDZenTiVj2zaUnBy7Fp4ZO3diTknBEBSE9z2snRP2IytZXdG10zC/J6QlQkh9GLKi+Os5hCipdDp4aApc3Keuj/rm4duP8fCDsAYFG02Uq+Oco0I+QdDzv+r0wK2fqdMDHd0NUQjh0nzbt8srrjZYXVxlHT1Kztmz6Dw98W3bxk4JXZNXvboYgoIwJSVxY98+u06ZtIweduxwT10fhf04RZuXadOmERERgZeXF5GRkezYseOOx86bNw+dTlfg5uVV8I3Q448/ftsx3bt3t/e34RjJ52F+L3VdSHAtGLpS+0/UhXBWHka1wYt/BfArD7W6w/3/pz723F54JR5GrIEHPoRmQ6B8E+csrPLV7gGNBoJiVrsH5mRqnUgI4ULy97tKj4vDnFXIyP1d5K/zMbZujd5HliDcSqfXY2zdGoC0zfbrGlhg9FC6BDotzUeuFi9ezLhx45gxYwaRkZFMnTqVbt26ceTIEUJCQgp9jr+/P0eOHLF8Xdhivu7duzN37lzL156eNlxgrpXUBHXE6voZCKwKQ1eBbzmtUwnh3MrVhnGHtE5hO93fhxPr1DbwG96Hzm9pnUgI4SI869TBLTSU3MREMnbswLdt2yI/N3/ERNb5FM7YJpqUH38kffNm+PfzdnmNzD/+IDcxEZ2PD8aoKLu8hrh3mo9cTZkyhZEjRzJ8+HDq1avHjBkz8PHxuWtzDJ1OR1hYmOUWGhp62zGenp4FjgkMDLTnt2F/6VdgQW9IOgEBlWHYD+AfrnUqIYSj+QSp3QMBtvwXzu/WNI4QwnXodDrL6FXauvVFfl72ufNkHT4Mej2+HTrYKZ1r841WmyZlHjpEblKSXV7DMiWwbVv0JWHQoITStLjKzs5m9+7ddL5laFOv19O5c2fi4uLu+Ly0tDSqVKlCpUqV6N27N3/++edtx6xfv56QkBBq167N6NGjuXr16h3Pl5WVRUpKSoGbU8lIUrsCXv5Lndo0bBWUsf8+CkIIJ1XnQWjY/+b0wFzrpvcI1+D01ybhknzb5xVXGzagKEqRnpO2Vn1T79OsGW5BshShMG7lyuFZpw4oCulbttrlNVJj81qwy5RAp6ZpcXXlyhVMJtNtI0+hoaEkJCQU+pzatWszZ84cVq5cyTfffIPZbKZ169acO3fOckz37t1ZsGABsbGxfPDBB2zYsIEePXpgMpkKPefkyZMJCAiw3Co5YAO4IstMhm/6QeJBMIaohVVQVa1TCSG01uND9XfC5b9g/ftapxF24NTXJuGyjK1aofPwIOf8ebJPnCjSc1J/U9/U+8qUwLsyRqvrrtI3b7b5ubNOnSL7+Alwc8O3nexV58w0nxZoraioKIYOHUqTJk1o164dy5Yto1y5cnz55ZeWYwYNGkSvXr1o2LAhffr04ccff2Tnzp2sX7++0HOOHz+e5ORky+3s2bMO+m7+QVYafNsfLuwB7yC1eUVwTa1TCSGcgU8QPPSJen/LVJkeWAI57bVJuDS9jw8+kZEApN3hfdGtcq9dI2O3+vtFRkzuzreN2kUxbeuWIo8KFlVaXiML4333YfD3t+m5hW1pWlwFBwdjMBhITEws8HhiYiJhRdzczt3dnaZNm3L8+PE7HlOtWjWCg4PveIynpyf+/v4FbprLzlBbLp/dDl4BMHQFhNbTOpUQwpnUfQgaPJI3PXCMTA8sYZzy2iRKBMu6q7zNaO8mbe06MJvxrFMHj4oV7R3NpXk3b47O2xvT5Stk3dJ4zRbyRw/9ukiB6+w0La48PDxo3rw5sXnVOIDZbCY2NpaoInZBMZlMHDx4kPDwOzd3OHfuHFevXr3rMU4lNwsWx8DpTeo+PI8tVzcKFkKIv+vxIRjLweXDsOFDrdMIIVxA/rqrjL17MSUn3/VYaf1ddHoPD3zuU/e4suXUwJxLl7ixfz8Avh072uy8wj40nxY4btw4Zs2axfz58zl8+DCjR48mPT2d4cOHAzB06FDGjx9vOX7ixIn8+uuvnDx5kj179vDYY49x5swZnnzySUBtdvF///d/bNu2jdOnTxMbG0vv3r2pUaMG3bp10+R7tEpuNnw/DE6sBXcfiFkCFZtrnUoI4ayMZeHBKer9zZ/Ahb3a5hFCOD2PihXxqFEdTCbS7lIEmDMySN+i7tskLdiLxjc6b2qgDfe7Slu7DgCvxo1wL6RDtnAumhdXAwcO5OOPP+bNN9+kSZMm7Nu3jzVr1liaXMTHx3Px4kXL8deuXWPkyJHUrVuXBx54gJSUFLZu3Uq9euqUOYPBwIEDB+jVqxe1atXiiSeeoHnz5mzatMn597oy5cKyJ+Hoz+DmBYMXQRXZx0AI8Q/q9YL6D4Niku6BQogi8WvfHlC7Bt5J2ubNKFlZuFeogGft2g5K5tqMeeuubuzejTkjwybnzN/A2a+TjB66Ap1i6xV3JUBKSgoBAQEkJyc7bo672QTLn4KD34PBAwYthJryj0gIUUTpV2HafZBxBe7/P+j4utaJikWT378uQn42wpYydu7kzJChGMqUoeaWzegMhtuOufDyyySvXEXQsGGEjn9Fg5SuR1EUTnTqTM6FC1T6coZlfVtxmVJTOdo6GnJyqLb6JzyrVbNRUmENa37/aj5yJQCzGX4YqxZWejfoP08KKyGEdYxl4aG86YGbpsCFfZrGEUI4N++mTdH7+2O6fp0b+w/c9udKTg6peRsNy5TAotPpdJbRK1tMDUzbsBFycvCoVk0KKxchxZXWFAV+/j/Y+zXo9PDwLHWDUCGEsFa93lC/7y3TA7O1TiSEcFI6N7ebrcMLmRqYsWsX5pQUDIGBeDdr5uh4Ls3YJhqwTVMLy8bBnaTAdRVSXGlJUeDX12HnV4AO+syABg9rnUoI4coe+Bh8ysKlP2HTx1qnEUI4sfyugYXtd5X6u9ol0Ldjh0KnDIo7M0ZFgcFA9qlT5Jw/X+zzmLOzSd+wEZAW7K5EiistrX0X4j5X7/f8LzQeqG0eIYTrMwbDg/9R72/6D1zcr20eIYTTMrZtCzodWUeOkHNL8zBFUW62YJcmClYz+Pnh3VjdQudepgZmxMVhzsjALSQErwYNbBVP2JkUV1rZ8NHNT5V7fATNh2mbRwhRctTvC/X6gDlXpgcKIe7ILTAQ7yZNgLy1PXky//iT3IQEdD4+GFtL1+LisMXUQMvoYaeO6PTylt1VyP8pLWz5FNa9q97v8g5EjtI2jxCi5MmfHpj4hzqCJYQQhcjvZnfr1MD81t++bdqg9/LSIpbLy1/Plr5tG0purtXPV0wmUteuBWQDZ1cjxZWjbZ8Jv72h3u/wOkQ/p20eIUTJ5FtOLbBAHSW/eHs3MCGEyF93lb5tG+bMTOCWJgqyzqfYvOrXxxAQgDk1lRsHrP/9e2P/fkxXr6L388PYsqUdEgp7keLKkXbPVzsDArR9Adr9n7Z5hBAlW/2+ULfXzemBphytEwkhnIxn7dq4hYWhZGaSsX07WadOkX38BLi54Xv//VrHc1k6gwFjdGugeFMDLVMC27dH5+Fh02zCvqS4cpT9i9W9rACinoGOb2ibRwhR8ul0anML7yBIPKjufyWEELfQ6XQ3pwZu2EBaXiML430tMQQEaBnN5Rmji7fflaIolqmZ0oLd9Uhx5Qh/LocVTwEKtHwSur6rvukRQgh78w2BBz5S72/8EBIOaptHCOF0brZk30Dqb3nrrWSdzz3LH7nKPHiQ3GvXivy8rGPHyImPR+fhgW/bNvaKJ+xEiit7++sn+N+ToJih6WNqZ0AprIQQjtSgH9TtmTc9cLRMDxRCFGBs1Qqdpyc5Fy5wY7+6fYOMmNw797AwPGvWAEUhIy6uyM/LH7Uytm6N3mi0VzxhJ1Jc2dOx32HJ4+obmob9oeenIK00hRCOptPBg1PAO1Adudr8idaJhBBORO/tjU/kfZavvRo1wj00VMNEJUdxpgam5a238ussBa4rknf69nJyAyyOAVO2uqC8zwzQyw7nQgiN+Ibc7B644UNI+EPbPEIIp+Lbvr3lvoxa2Y4xvyX75s0oivKPx+dcuEDmoUOg1+PbsaO94wk7kOLKHs7EwcJBkJsJtXpAv9lgcNM6lRCitGvQD+o8BOYcWCndA4UQN/nlNbUAGTGxJZ8WzdF5epJ76RJZx4794/H5XQK9mzXFLSjI3vGEHUhxZWuXDsO3/SEnA6p3hP7zwE1aaAohnMCt0wMv7octU7VOJIRwEu4VKhD65huEjn8Fz+rVtY5TYui9vPDJ26cqvQhTA1PzujX6dZKGIq5KiitbC6oG1dpBRFsY+C24y87mQggn4heqNtYBWP8BJP6pbR4hhNMIevRRgoYN0zpGiWNsEw1A+pa7F1e5166RsWsXIKOHrkyKK1tz81RHqx5dDB4+WqcRQojbNXwEaj+oTg+U7oFCCGFXvnnrrjJ27cKcmXnH49LWbwCTCc/atfGoVMlR8YSNSXFlDwZ38JDWmUIIJ6XTwUNTwKtM3vTA/2qdSAghSiyP6tVxCwtDycoiY+euOx6XGisbB5cEUlwJIURp5BcGPT5U769/HxIPaZtHCCFKKJ1Od3Nq4ObNhR5jvnHDsibLr4ust3JlTlFcTZs2jYiICLy8vIiMjGTHjh13PHbevHnodLoCNy+vguuaFEXhzTffJDw8HG9vbzp37syxInRoEUKIUqXRALWjqaV7YK7WiYQQokTKnxqYtqXw4ip9yxaUzEzcy5fHs04dR0YTNqZ5cbV48WLGjRvHhAkT2LNnD40bN6Zbt25cunTpjs/x9/fn4sWLltuZM2cK/PmHH37Ip59+yowZM9i+fTtGo5Fu3bqReZd5rkIIUerodPDQJ+AVABf2wlaZHiiEEPZgjIoCvZ7s4yfIuXjxtj9P/S1vSmCXzuh0OkfHEzakeXE1ZcoURo4cyfDhw6lXrx4zZszAx8eHOXPm3PE5Op2OsLAwyy30ll3EFUVh6tSpvP766/Tu3ZtGjRqxYMECLly4wIoVKxzwHQkhhAvxDy84PfDSYW3zCCFECWQICMC7YUPg9q6BSm4uqevXA+Ar661cnqbFVXZ2Nrt376Zz55tzS/V6PZ07dyYuLu6Oz0tLS6NKlSpUqlSJ3r178+efN1sJnzp1ioSEhALnDAgIIDIy8o7nzMrKIiUlpcBNCCFKjUYDoVZ3MGXDCpke6Czk2iREyWLMnxr4t/2uMnbtxpycjKFMGXyaNdMimrAhTYurK1euYDKZCow8AYSGhpKQkFDoc2rXrs2cOXNYuXIl33zzDWazmdatW3Pu3DkAy/OsOefkyZMJCAiw3CpJ+0shRGmi08FDU/OmB+6BuM+0TiSQa5MQJY2lqUVcHIrJZHk89Xd1SqBvx47o3Nw0ySZsR/NpgdaKiopi6NChNGnShHbt2rFs2TLKlSvHl19+Wexzjh8/nuTkZMvt7NmzNkwshBAuwD8cur+v3l83CS79pW0eIdcmIUoY74YN0fv7Y05OJvPgQUBdzpIaGwvIxsElhabFVXBwMAaDgcTExAKPJyYmEhYWVqRzuLu707RpU44fPw5geZ415/T09MTf37/ATQghSp3Gg6FmV3V6oHQP1Jxcm4QoWXRubhhbtQJuTg3M/PMQuRcvovP2xti6tZbxhI1oWlx5eHjQvHlzYvMqdgCz2UxsbCxRUVFFOofJZOLgwYOEh4cDULVqVcLCwgqcMyUlhe3btxf5nEIIUSrpdNDzv+AZAOd3Q9znWicSQogS5e/7XeVvHOzbpg36v20tJFyT5tMCx40bx6xZs5g/fz6HDx9m9OjRpKenM3z4cACGDh3K+PHjLcdPnDiRX3/9lZMnT7Jnzx4ee+wxzpw5w5NPPgmonQSff/553n33XVatWsXBgwcZOnQo5cuXp0+fPlp8i0II4Tr8y0P3yer9dZPg8hFt8wghRAmSv9/VjQMHMCUnk/a7TAksaTRfNTdw4EAuX77Mm2++SUJCAk2aNGHNmjWWhhTx8fHo9TdrwGvXrjFy5EgSEhIIDAykefPmbN26lXr16lmOeemll0hPT2fUqFFcv36dNm3asGbNmts2GxZCCFGIJo/Cn8vh+G9q98AnfgW9QetUQgjh8tzLl8ejWjWyT57k2uLvyTp2DAwGfNu31zqasBGdoiiK1iGcTUpKCgEBASQnJ8scdyFE6ZR8Hr5oBVkp0OUdiH7OIS8rv3/vTH42QpQMCZMmcW3B1+h8fFAyMvCJakWVuXO1jiXuwprfv5pPCxRCCOGEAipAt0nq/bXvwuWj2uYRQogSIn9qoJKRAYDfLXuzCtcnxZUQQojCNX0ManQGUxasHANm0z8/RwghxF35tGyJzsPD8rVfx44aphG2JsWVEEKIwlm6B/rDuR2w7QutEwkhhMvTe3vj06I5AF4NGuCe1/FalAxSXAkhhLizgIrQ7T31/tp34coxbfMIIUQJUOaRRwAIjInROImwNSmuhBBC3F3TIVC9I+RmyvRAIYSwAf8HHqD2/n2U6dtH6yjCxqS4EkIIcXc6HfT8FDz84Ox22D5D60RCCOHy9J6eWkcQdiDFlRBCiH9WphJ0e1e9HzsRrhzXNo8QQgjhhKS4EkIIUTTNhkG1DjI9UAghhLgDKa6EEEIUjU4HvT7Lmx64DbZ/qXUiIYQQwqlIcSWEEKLoylSCru+o92MnwtUT2uYRQgghnIgUV0IIIazT/HGo1h5yb+RNDzRrnUgIIYRwClJcCSGEsI6le6AvxMfBjplaJxJCCCGcghRXQgghrBdYBbpMVO///pZMDxRCCCGQ4koIIURxtRgBVe/Pmx74jEwPFEIIUepJcSWEEKJ4dDro9Tm4GyF+K+ycpXUiIYQQQlNSXAkhhCi+wCrQ9ZbpgUknNY0jhBBCaEmKKyGEEPem+QiIaAs5GbDyWZkeKIQQotSS4koIIcS90euhd970wDObYedXWicSQgghNCHFlRBCiHsXGAFd3lbv/z4Bkk5pGkcIIYTQghRXQgghbKPFEzenB/71o9ZphBBCCIdziuJq2rRpRERE4OXlRWRkJDt27CjS8xYtWoROp6NPnz4FHn/88cfR6XQFbt27d7dDciGEEBZ6PfT6DB5dAq2f1TqNEEII4XCaF1eLFy9m3LhxTJgwgT179tC4cWO6devGpUuX7vq806dP8+KLL9K2bdtC/7x79+5cvHjRclu4cKE94gshhLhVUFWo1VXrFEIIIYQmNC+upkyZwsiRIxk+fDj16tVjxowZ+Pj4MGfOnDs+x2QyERMTw9tvv021atUKPcbT05OwsDDLLTAw8I7ny8rKIiUlpcBNCCGE0JJcm4QQwvVoWlxlZ2eze/duOnfubHlMr9fTuXNn4uLi7vi8iRMnEhISwhNPPHHHY9avX09ISAi1a9dm9OjRXL169Y7HTp48mYCAAMutUqVKxfuGhBBCCBuRa5MQQrgeTYurK1euYDKZCA0NLfB4aGgoCQkJhT5n8+bNzJ49m1mzZt3xvN27d2fBggXExsbywQcfsGHDBnr06IHJZCr0+PHjx5OcnGy5nT17tvjflBBCCGEDcm0SQgjX46Z1AGukpqYyZMgQZs2aRXBw8B2PGzRokOV+w4YNadSoEdWrV2f9+vV06tTptuM9PT3x9PS0S2YhhBCiOOTaJIQQrkfT4io4OBiDwUBiYmKBxxMTEwkLC7vt+BMnTnD69Gl69uxpecxsNgPg5ubGkSNHqF69+m3Pq1atGsHBwRw/frzQ4koIIYQQQggh7pWm0wI9PDxo3rw5sbGxlsfMZjOxsbFERUXddnydOnU4ePAg+/bts9x69epFhw4d2Ldv3x3no587d46rV68SHh5ut+9FCCGEEEIIUbppPi1w3LhxDBs2jBYtWnDfffcxdepU0tPTGT58OABDhw6lQoUKTJ48GS8vLxo0aFDg+WXKlAGwPJ6Wlsbbb79Nv379CAsL48SJE7z00kvUqFGDbt26FSmToigA0plJCCEcLP/3bv7vYXGTXJuEEEIb1lybNC+uBg4cyOXLl3nzzTdJSEigSZMmrFmzxtLkIj4+Hr2+6ANsBoOBAwcOMH/+fK5fv0758uXp2rUr77zzTpHnrqempgJIZyYhhNBIamoqAQEBWsdwKnJtEkIIbRXl2qRT5OPB25jNZi5cuICfnx86nc7q56ekpFCpUiXOnj2Lv7+/HRLah+R2HFfMDJLb0Vwx971mVhSF1NRUypcvb9UHa6WBXJtcJ7crZgbJ7WiumNsVM4Njr02aj1w5I71eT8WKFe/5PP7+/i71Fy+f5HYcV8wMktvRXDH3vWSWEavCybXJ9XK7YmaQ3I7mirldMTM45tokHwsKIYQQQgghhA1IcSWEEEIIIYQQNiDFlR14enoyYcIEl9v8UXI7jitmBsntaK6Y2xUzlxau+v/GFXO7YmaQ3I7mirldMTM4Nrc0tBBCCCGEEEIIG5CRKyGEEEIIIYSwASmuhBBCCCGEEMIGpLgSQgghhBBCCBuQ4koIIYQQQgghbECKKxvauHEjPXv2pHz58uh0OlasWKF1pH80efJkWrZsiZ+fHyEhIfTp04cjR45oHesfTZ8+nUaNGlk2g4uKiuLnn3/WOpbV3n//fXQ6Hc8//7zWUe7qrbfeQqfTFbjVqVNH61j/6Pz58zz22GOULVsWb29vGjZsyK5du7SOdVcRERG3/ax1Oh1jxozROtpdmUwm3njjDapWrYq3tzfVq1fnnXfeQXomaU+uTY5TEq5Ncl2yP7k2OY4W1yY3u525FEpPT6dx48aMGDGChx9+WOs4RbJhwwbGjBlDy5Ytyc3N5dVXX6Vr164cOnQIo9Godbw7qlixIu+//z41a9ZEURTmz59P79692bt3L/Xr19c6XpHs3LmTL7/8kkaNGmkdpUjq16/P77//bvnazc25f31cu3aN6OhoOnTowM8//0y5cuU4duwYgYGBWke7q507d2IymSxf//HHH3Tp0oX+/ftrmOqfffDBB0yfPp358+dTv359du3axfDhwwkICOC5557TOl6pJtcmx3H1a5Ncl+xPrk2Opcm1SRF2ASjLly/XOobVLl26pADKhg0btI5itcDAQOWrr77SOkaRpKamKjVr1lR+++03pV27dsrYsWO1jnRXEyZMUBo3bqx1DKu8/PLLSps2bbSOcc/Gjh2rVK9eXTGbzVpHuasHH3xQGTFiRIHHHn74YSUmJkajRKIwcm1yPFe5Nsl1yTHk2uRYWlybZFqgKCA5ORmAoKAgjZMUnclkYtGiRaSnpxMVFaV1nCIZM2YMDz74IJ07d9Y6SpEdO3aM8uXLU61aNWJiYoiPj9c60l2tWrWKFi1a0L9/f0JCQmjatCmzZs3SOpZVsrOz+eabbxgxYgQ6nU7rOHfVunVrYmNjOXr0KAD79+9n8+bN9OjRQ+NkoiSQa5P9yXXJMeTa5FhaXJucf/xUOIzZbOb5558nOjqaBg0aaB3nHx08eJCoqCgyMzPx9fVl+fLl1KtXT+tY/2jRokXs2bOHnTt3ah2lyCIjI5k3bx61a9fm4sWLvP3227Rt25Y//vgDPz8/reMV6uTJk0yfPp1x48bx6quvsnPnTp577jk8PDwYNmyY1vGKZMWKFVy/fp3HH39c6yj/6JVXXiElJYU6depgMBgwmUy89957xMTEaB1NuDi5NtmfXJccR65NjqXJtcluY2KlHC449eKpp55SqlSpopw9e1brKEWSlZWlHDt2TNm1a5fyyiuvKMHBwcqff/6pday7io+PV0JCQpT9+/dbHnOF6Rd/d+3aNcXf39+pp7q4u7srUVFRBR579tlnlVatWmmUyHpdu3ZVHnroIa1jFMnChQuVihUrKgsXLlQOHDigLFiwQAkKClLmzZundTRxC7k22Z+rXZvkuuRYcm1yLC2uTVJc2YmrXcDGjBmjVKxYUTl58qTWUYqtU6dOyqhRo7SOcVfLly9XAMVgMFhugKLT6RSDwaDk5uZqHbHIWrRoobzyyitax7ijypUrK0888USBx7744gulfPnyGiWyzunTpxW9Xq+sWLFC6yhFUrFiReXzzz8v8Ng777yj1K5dW6NEojBybXI8Z782yXXJseTa5FhaXJtkWmAppygKzz77LMuXL2f9+vVUrVpV60jFZjabycrK0jrGXXXq1ImDBw8WeGz48OHUqVOHl19+GYPBoFEy66SlpXHixAmGDBmidZQ7io6Ovq1189GjR6lSpYpGiawzd+5cQkJCePDBB7WOUiQZGRno9QWX8RoMBsxms0aJhCuTa5PjyHXJseTa5FhaXJukuLKhtLQ0jh8/bvn61KlT7Nu3j6CgICpXrqxhsjsbM2YM3333HStXrsTPz4+EhAQAAgIC8Pb21jjdnY0fP54ePXpQuXJlUlNT+e6771i/fj2//PKL1tHuys/P77Y1A0ajkbJlyzr1WoIXX3yRnj17UqVKFS5cuMCECRMwGAwMHjxY62h39O9//5vWrVszadIkBgwYwI4dO5g5cyYzZ87UOto/MpvNzJ07l2HDhrlEa2GAnj178t5771G5cmXq16/P3r17mTJlCiNGjNA6Wqkn1ybHccVrk1yXHEuuTY6lybXJbmNipdC6desU4LbbsGHDtI52R4XlBZS5c+dqHe2uRowYoVSpUkXx8PBQypUrp3Tq1En59ddftY5VLK4wt33gwIFKeHi44uHhoVSoUEEZOHCgcvz4ca1j/aMffvhBadCggeLp6anUqVNHmTlzptaRiuSXX35RAOXIkSNaRymylJQUZezYsUrlypUVLy8vpVq1asprr72mZGVlaR2t1JNrk+OUlGuTXJfsS65NjqPFtUmnKHbcolgIIYQQQgghSgnZ50oIIYQQQgghbECKKyGEEEIIIYSwASmuhBBCCCGEEMIGpLgSQgghhBBCCBuQ4koIIYQQQgghbECKKyGEEEIIIYSwASmuhBBCCCGEEMIGpLgSQgghhBBCCBuQ4kqIUkyn07FixQqtYwghhBCAXJeE65PiSgiNPP744+h0uttu3bt31zqaEEKIUkiuS0LcOzetAwhRmnXv3p25c+cWeMzT01OjNEIIIUo7uS4JcW9k5EoIDXl6ehIWFlbgFhgYCKhTI6ZPn06PHj3w9vamWrVqLF26tMDzDx48SMeOHfH29qZs2bKMGjWKtLS0AsfMmTOH+vXr4+npSXh4OM8880yBP79y5Qp9+/bFx8eHmjVrsmrVKsufXbt2jZiYGMqVK4e3tzc1a9a87aIrhBCi5JDrkhD3RoorIZzYG2+8Qb9+/di/fz8xMTEMGjSIw4cPA5Cenk63bt0IDAxk586dLFmyhN9//73ARWr69OmMGTOGUaNGcfDgQVatWkWNGjUKvMbbb7/NgAEDOHDgAA888AAxMTEkJSVZXv/QoUP8/PPPHD58mOnTpxMcHOy4H4AQQginItclIf6BIoTQxLBhwxSDwaAYjcYCt/fee09RFEUBlKeeeqrAcyIjI5XRo0criqIoM2fOVAIDA5W0tDTLn//000+KXq9XEhISFEVRlPLlyyuvvfbaHTMAyuuvv275Oi0tTQGUn3/+WVEURenZs6cyfPhw23zDQgghnJpcl4S4d7LmSggNdejQgenTpxd4LCgoyHI/KiqqwJ9FRUWxb98+AA4fPkzjxo0xGo2WP4+OjsZsNnPkyBF0Oh0XLlygU6dOd83QqFEjy32j0Yi/vz+XLl0CYPTo0fTr1489e/bQtWtX+vTpQ+vWrYv1vQohhHB+cl0S4t5IcSWEhoxG423TIWzF29u7SMe5u7sX+Fqn02E2mwHo0aMHZ86cYfXq1fz222906tSJMWPG8PHHH9s8rxBCCO3JdUmIeyNrroRwYtu2bbvt67p16wJQt25d9u/fT3p6uuXPt2zZgl6vp3bt2vj5+REREUFsbOw9ZShXrhzDhg3jm2++YerUqcycOfOezieEEMJ1yXVJiLuTkSshNJSVlUVCQkKBx9zc3CyLc5csWUKLFi1o06YN3377LTt27GD27NkAxMTEMGHCBIYNG8Zbb73F5cuXefbZZxkyZAihoaEAvPXWWzz11FOEhITQo0cPUlNT2bJlC88++2yR8r355ps0b96c+vXrk5WVxY8//mi5iAohhCh55LokxL2R4koIDa1Zs4bw8PACj9WuXZu//voLUDsmLVq0iKeffprw8HAWLlxIvXr1APDx8eGXX35h7NixtGzZEh8fH/r168eUKVMs5xo2bBiZmZl88sknvPjiiwQHB/PII48UOZ+Hhwfjx4/n9OnTeHt707ZtWxYtWmSD71wIIYQzkuuSEPdGpyiKonUIIcTtdDody5cvp0+fPlpHEUIIIeS6JEQRyJorIYQQQgghhLABKa6EEEIIIYQQwgZkWqAQQgghhBBC2ICMXAkhhBBCCCGEDUhxJYQQQgghhBA2IMWVEEIIIYQQQtiAFFdCCCGEEEIIYQNSXAkhhBBCCCGEDUhxJYQQQgghhBA2IMWVEEIIIYQQQtiAFFdCCCGEEEIIYQP/DznCQEHgugecAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1000x600 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "fig, ((ax_tl, ax_tr), (ax_bl, ax_br)) = plt.subplots(2, 2, sharex=True, sharey='row', figsize=(10, 6))\n",
    "ax_tl.set_title('Training set')\n",
    "ax_tr.set_title('Development set')\n",
    "ax_bl.set_xlabel('Epochs')\n",
    "ax_br.set_xlabel('Epochs')\n",
    "ax_bl.set_ylabel('Accuracy')\n",
    "ax_tl.set_ylabel('Loss')\n",
    "\n",
    "colours = iter(plt.rcParams['axes.prop_cycle'].by_key()['color'])\n",
    "range_ = np.arange(1, len(trainer.train_epoch_costs)+1)\n",
    "ax_tl.plot(range_, trainer.train_epoch_costs, color=next(colours))\n",
    "ax_bl.plot(range_, trainer.train_eval_results['acc'], color=next(colours))\n",
    "ax_tr.plot(range_, trainer.val_costs, color=next(colours))\n",
    "ax_br.plot(range_, trainer.val_eval_results['acc'], color=next(colours))\n",
    "\n",
    "# mark best model as circle\n",
    "best_epoch = np.argmax(trainer.val_eval_results['acc'])\n",
    "ax_tl.plot(best_epoch + 1, trainer.train_epoch_costs[best_epoch], 'o', color='black', fillstyle='none')\n",
    "ax_tr.plot(best_epoch + 1, trainer.val_costs[best_epoch], 'o', color='black', fillstyle='none')\n",
    "ax_bl.plot(best_epoch + 1, trainer.train_eval_results['acc'][best_epoch], 'o', color='black', fillstyle='none')\n",
    "ax_br.plot(best_epoch + 1, trainer.val_eval_results['acc'][best_epoch], 'o', color='black', fillstyle='none')\n",
    "\n",
    "ax_br.text(best_epoch + 1.4, trainer.val_eval_results['acc'][best_epoch], 'early stopping', va='center')\n",
    "\n",
    "# print test accuracy\n",
    "model.load(trainer.log_dir + '/best_model.lt')\n",
    "val_acc = acc(model(val_circuits), val_labels)\n",
    "print('Validation accuracy:', val_acc.item())"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {
    "raw_mimetype": "text/restructuredtext"
   },
   "source": [
    ".. rubric:: See also:\n",
    "\n",
    "- `Training: Classical case <./trainer-classical.ipynb>`_\n",
    "- `Training: Hybrid case <./trainer-hybrid.ipynb>`_\n",
    "- `Advanced: Manual training <../manual-training.rst>`_"
   ]
  }
 ],
 "metadata": {
  "language_info": {
   "name": "python"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}