TrainEpochM1_IRIS.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Impact of Barren Plateaus Mitigation Strategy on the Performance of Quantum Neural Networks\n",
    "*Training QNN treated with method 3 (Identity Blocks) of the BP mitigation strategy (IRIS)*\n",
    "\n",
    "**Authors:**\n",
    "- Jacob Cybulski and Thanh Nguyen\n",
    "\n",
    "**Date:**\n",
    "- June 2023\n",
    "\n",
    "**Data:**\n",
    "- Data = IRIS dataset with 99 examples\n",
    "- Features = 4\n",
    "- Target = {-1, 1}\n",
    "- Iterations = 150\n",
    "- Instance = 10\n",
    "\n",
    "**Description:**\n",
    "- This notebook aims to train 10 instances of a QNN initialised with identity blocks and produce averaged performance stats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import copy, math\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from matplotlib import pyplot as plt\n",
    "\n",
    "from sklearn.datasets import load_iris\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "import torch\n",
    "from torch import nn\n",
    "from torch.nn import functional as F\n",
    "from torch.utils.data import TensorDataset, DataLoader\n",
    "\n",
    "from qiskit.utils import algorithm_globals\n",
    "from qiskit.algorithms.optimizers import COBYLA, NELDER_MEAD, SLSQP, SPSA\n",
    "from qiskit.opflow import Z, X, I, StateFn, PauliExpectation, Gradient, NaturalGradient, PauliSumOp\n",
    "from qiskit.circuit.library import ZZFeatureMap, RealAmplitudes\n",
    "from qiskit_machine_learning.neural_networks import SamplerQNN, EstimatorQNN, OpflowQNN, CircuitQNN, TwoLayerQNN\n",
    "from qiskit_machine_learning.neural_networks import EffectiveDimension\n",
    "from qiskit_machine_learning.algorithms.classifiers import VQC, NeuralNetworkClassifier\n",
    "from qiskit.quantum_info import SparsePauliOp\n",
    "from qiskit_machine_learning.connectors import TorchConnector\n",
    "\n",
    "# Set seed for random generators\n",
    "# algorithm_globals.random_seed = 42\n",
    "\n",
    "from IPython.display import clear_output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from Components.circuits import preTrainedBlockGenerator, layerwise_training, featureMapGenerator, AnsatzGenerator\n",
    "from Components.data import iris, fetch_mnist, fetch_mnist_balanced\n",
    "from Components.utils import plot_loss, score, parity, classification_callback, plot_objfn_range, result_to_objfun_dataframes, save_results\n",
    "from Components.train import create_qnn, sampling_experiment, train, train_batch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "DATA_SIZE = 99\n",
    "FEATURE_DIM = 4\n",
    "MAX_ITER = 150\n",
    "MAX_INST = 10\n",
    "\n",
    "MAX_REPS = 9\n",
    "MIN_REPS = 1\n",
    "MAX_IDENTITIES_BLOCKS = 2  # <---- Number of identity blocks, depth values of identity blocks is close to that of normal ansatz (5 qubits)\n",
    "ENTANGLEMENT = 'linear'\n",
    "\n",
    "GLOBAL_OPERATOR = PauliSumOp.from_list([('Z'*FEATURE_DIM, 1)])\n",
    "LOCAL_OPERATOR = PauliSumOp.from_list([('I' * (FEATURE_DIM - 2)+'Z'*2, 1)])\n",
    "\n",
    "LOG_PATH = './Logs-IRIS-v4'\n",
    "METHOD_TAG = 'm1'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training set: 69 samples\n",
      "Testing set: 30 samples\n",
      "Number of features: 4\n",
      "Classes: [-1  1]\n"
     ]
    }
   ],
   "source": [
    "# X_train, X_val, y_train, y_val = fetch_mnist_balanced(PCA_n = FEATURE_DIM, data_size=DATA_SIZE)\n",
    "# X_train, X_val, y_train, y_val = fetch_mnist(PCA_n = FEATURE_DIM, data_size=DATA_SIZE)\n",
    "X_train, X_val, y_train, y_val = iris(pd=False, PCA_n=None)\n",
    "\n",
    "X_train_t = torch.from_numpy(X_train).to(torch.float32)\n",
    "y_train_t = torch.from_numpy(y_train).to(torch.float32)\n",
    "X_val_t = torch.from_numpy(X_val).to(torch.float32)\n",
    "y_val_t = torch.from_numpy(y_val).to(torch.float32)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot dataset (only the first two args)\n",
    "for x, y_target in zip(X_train_t, y_train_t):\n",
    "    if y_target == 1:\n",
    "        plt.plot(x[0], x[1], \"bo\")\n",
    "    else:\n",
    "        plt.plot(x[0], x[1], \"go\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_method_data(data, title='Instance Losses', dlabel='inst#', xlabel='Loss', ylabel='Iteration'):\n",
    "    # create figure and axis\n",
    "    fig, ax = plt.subplots()\n",
    "    fig.set_figheight(4)\n",
    "    fig.set_figwidth(8)\n",
    "    \n",
    "    # setting the axis' labels\n",
    "    ax.set_ylabel(xlabel)\n",
    "    ax.set_xlabel(ylabel)\n",
    "    \n",
    "    # Plot data\n",
    "    for i in range(len(data)):\n",
    "        data[i].T.plot(ax=ax, label=f'inst# {i}', figsize=(5, 3))\n",
    "    ax.legend([f'{dlabel} {i}' for i in range(len(data))])\n",
    "    plt.title(title)\n",
    "    plt.show()  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Method 1 (Local Loss and Shallow Ansatz)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from qiskit.primitives import Estimator # Statevector estimator\n",
    "from os.path import exists\n",
    "from os import makedirs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_map = featureMapGenerator(FEATURE_DIM)\n",
    "ansatz = AnsatzGenerator(FEATURE_DIM, reps=MIN_REPS, entanglement=ENTANGLEMENT)\n",
    "qc = feature_map.compose(ansatz)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Creating the first instance 0 of method m1\n",
      "\n",
      "Epoch      Loss                 Train Accuracy   Test Accuracy   \n",
      "[ 0 ]      0.9683899879455566   0.53623          0.56667         \n",
      "[ 10 ]     0.8918700218200684   0.56522          0.66667         \n",
      "[ 20 ]     0.8690999746322632   0.73913          0.8             \n",
      "[ 30 ]     0.8660200238227844   0.75362          0.8             \n",
      "[ 40 ]     0.8660699725151062   0.76812          0.8             \n",
      "[ 50 ]     0.8659999966621399   0.76812          0.8             \n",
      "[ 60 ]     0.8659800291061401   0.76812          0.8             \n",
      "[ 70 ]     0.8659800291061401   0.76812          0.8             \n",
      "[ 80 ]     0.8659800291061401   0.76812          0.8             \n",
      "[ 90 ]     0.8659800291061401   0.76812          0.8             \n",
      "[ 100 ]    0.8659800291061401   0.76812          0.8             \n",
      "[ 110 ]    0.8659800291061401   0.76812          0.8             \n",
      "[ 120 ]    0.8659800291061401   0.76812          0.8             \n",
      "[ 130 ]    0.8659800291061401   0.76812          0.8             \n",
      "[ 140 ]    0.8659800291061401   0.76812          0.8             \n",
      "[ 149 ]    0.8659800291061401   0.76812          0.8             \n"
     ]
    }
   ],
   "source": [
    "# To execute on multiple machines concurrently\n",
    "# Select separate sub-lists of instance numbers and run\n",
    "# Alternatively use range(MAX_INST) for all\n",
    "\n",
    "m = METHOD_TAG\n",
    "path = LOG_PATH\n",
    "\n",
    "# for i in [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]: \n",
    "for i in [0]: \n",
    "    \n",
    "    # Mark that the method's data has been saved\n",
    "    if (exists(f'{path}/{m}')):\n",
    "        print(f'Adding a new instance {i} of method {m}\\n')\n",
    "    else:\n",
    "        makedirs(f'{path}/{m}')\n",
    "        print(f'Creating the first instance {i} of method {m}\\n')\n",
    "        \n",
    "    # By default this will run as a local simulation\n",
    "    qnn = EstimatorQNN(\n",
    "        circuit=qc,\n",
    "        estimator=Estimator(),\n",
    "        observables=LOCAL_OPERATOR,\n",
    "        input_params=feature_map.parameters,\n",
    "        weight_params=ansatz.parameters\n",
    "    )\n",
    "\n",
    "    # initial weights are chosen uniformly at random from [-1, 1]\n",
    "    model = TorchConnector(qnn)\n",
    "\n",
    "    loss_function = nn.L1Loss() # nn.MSELoss()\n",
    "    optimizer = torch.optim.NAdam(model.parameters(), lr=0.05)\n",
    "\n",
    "    model, losses, accuracy_train, accuracy_test, weights = train(\n",
    "        model, \n",
    "        MAX_ITER, \n",
    "        X_train_t,\n",
    "        y_train_t,\n",
    "        X_val_t,\n",
    "        y_val_t,\n",
    "        optimizer = optimizer, \n",
    "        loss_function = loss_function\n",
    "        )\n",
    "\n",
    "    pd.DataFrame(losses).astype('float').to_csv(f'{path}/{m}/{m}-{i}-LossFunction.csv')\n",
    "    pd.DataFrame(accuracy_train).astype('float').to_csv(f'{path}/{m}/{m}-{i}-Scores-Train.csv')\n",
    "    pd.DataFrame(accuracy_test).astype('float').to_csv(f'{path}/{m}/{m}-{i}-Scores-Test.csv')\n",
    "    pd.DataFrame(weights).astype('float').to_csv(f'{path}/{m}/{m}-{i}-Weights.csv')\n",
    "    \n",
    "    # Mark that the method's data has been saved\n",
    "    if (exists(f'{path}/{m}/{m}-Method.csv')):\n",
    "        f = open(f'{path}/{m}/{m}-Method.csv', 'a')\n",
    "    else:\n",
    "        f = open(f'{path}/{m}/{m}-Method.csv', 'w')\n",
    "        f.write(f'{m},Instance,Max Inst,Examples,Features,Iterations\\n')\n",
    "    f.write(f',{i},{MAX_INST},{DATA_SIZE},{FEATURE_DIM},{MAX_ITER}\\n')\n",
    "    f.close()\n",
    "    \n",
    "    clear_output(wait=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1500x900 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "import matplotlib as mpl\n",
    "mpl.rcParams['figure.dpi'] = 300\n",
    "\n",
    "# Plot the loss of the last saved instance\n",
    "plot_method_data([pd.DataFrame(losses).astype('float').T])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABXsAAAOkCAYAAAD3EOdaAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8qNh9FAAAACXBIWXMAAC4jAAAuIwF4pT92AADD5klEQVR4nOzdebyt93j//9d15inJGSJIIjMZzUSlSFC0aghiCEWUmjvwqxprplT1S6tKSwUxC6KqKBJDDDHFFBGZJ5HknJPhzNP1++Ne21nn3uvee017rXut9Xo+Hvux1/257/tzf/Y6O8fD+1zr+kRmIkmSJEmSJEkabfOGvQBJkiRJkiRJUu8MeyVJkiRJkiRpDBj2SpIkSZIkSdIYMOyVJEmSJEmSpDFg2CtJkiRJkiRJY8CwV5IkSZIkSZLGgGGvJEmSJEmSJI0Bw15JkiRJkiRJGgOGvZIkSZIkSZI0Bgx7JUmSJEmSJGkMGPZKkiRJkiRJ0hgw7JUkSZIkSZKkMWDYK0mSJEmSJEljwLBXkiRJkiRJksaAYa8kSZIkSZIkjQHDXkmSJEmSJEkaA4a9kiRJkiRJkjQGDHslSZIkSZIkaQwY9kqSJEmSJEnSGDDslSRJkiRJkqQxYNgrSZIkSZIkSWPAsFeSJEmSJEmSxoBhryRJkiRJkiSNAcNeSZIkSZIkSRoDhr2SJEmSJEmSNAYMeyVJkiRJkiRpDBj2SpIkSZIkSdIYMOyVJEmSJEmSpDFg2CtJkiRJkiRJY8CwV5IkSZIkSZLGgGGvJEmSJEmSJI0Bw15JkqSai4hDIiKH9HXOsH/+SRcRD6z4s/nysNcmSZKkejHslSRJ0sSIiNNahKaHDHtds3hmxfgfRcQdBroSSZIk1ZphryRJklRTEbESeFzF6XnAMwa3GkmSJNXdgmEvQJIkSbPaBvy0w3tWAIeXxjYCF3c4T6fXq7+eDCyZ4fwzIuINmZmDWpAkSZLqy7BXkiSp5jLzWuBundwTEScBZ5eGf5iZJ/VlURqUcguHBKLp+BDgwcBXB7UgSZIk1ZdtHCRJkqQaioi7AfcoDf8/ikrvZn8+kAVJkiSp9gx7JUmSpHpqVdX7TuALpfHHRMSqwSxJkiRJdWbYK0mSJNVMRCwBnlIaPiczrwQ+WBpvda0kSZImkD17JUmS1JaIWAbcB7gdcBtgObAWuAE4PzMv6/Pz7gDcFTgA2BtYDGym2GjuWuBy4KLM3NLP59bEY4Byte6HGt//l+I9v03TuWcC75qLhTT+3O8N3B7YF9iH4s/hJuA3wAWZubaPz7sjcHTjWftSFKjcSvFnfiHFn/nOfj2vDiLiIODuwMHAXsBO4HfAJzJzUxv37wccRbEp40qK/zZvBdYB1wDnZeaGOVn87jXsQ/F7cltgDcXPsbGxhosofk9umcs1SJIkw15JkiTNICIWAKcBpwL3AxbNcO0lwCeBf8rMdV0+b3/ghY3nHdLGLdsi4nyKDco+lZnnt5jzdODpM8xxWUTMcBqA12Xma9tYT7+UWzhsBD4NkJnbI+KjwF83nb9bRNwjM3/cj4c3KoufBTwe+ANm+HMHMiJ+BnwR+EBm/qaL590V+CvgIcAdZrn8poj4OsXv2mczs9zDuHnec4ATm4a+0c0mhRFxGvCB0vChmXn5LPddThHgTvlgZp7WOLeQ4j1+HnDniinOpvhHjfK8ewGPAh4KnAQcNOMPADsj4ifAe4EPzfSedaIR8L4AeDRwT2D+LGv4AUUbktMz85rSXPOAS9jzv/vLgMMzM3tY499Q9LpudnJmntXtnJIk1ZltHCRJktRSRJwM/Ar4T+BBzBz4QVFV+HKK8PQFXTzvBcCvG3Mc0uZti4DjgVcAP2mElCMtIg6leL+bfbZUmVlu5QB92qgtIp4LXAr8K/AAZv9zD4oK7JcDv46Ix3XwrEMj4rPATyjWP1vQC0Xl6mOBjwM/bPdZddKoXv4x8G6qg96qe/8RuB44A3gaswe9UISw96L4b/mSiLh/Rwuevob5EfFKiiD6TRT/Dc4U9E6t4Q+ANwJXRMTxzSczcxfwntI9hwIP62WtwHNLx1czve+1JEljw7BXkiRJe4jC64DPAkd0McXewLsi4r0RMVsANPXMN1C0IVjRxfPGzTMoAtRmH2o+yMyfAD8vXfOUXsLuiFgSEWcA/07RsqGraSg+vt/O8x4I/AA4mek/b7v27vK+oYmIo4DvAsd1OcXxFH2au3Ug8LWIeGo3N0fEauBLFKHtyi7XMB9Y1mL8fUC5LcvzunwGEfEg4MjS8H+MWxsQSZKa2cZBkiRJZf8OPKfF+Drg/4AfUVQWbqIIe44F/pjpocqzKfq6vnSmh0XEScArW5y6ufG884GrKFoZLKYIEw9qPHeqh/BMrgR+2ni9munVo78CZvtY+3WznO+LxkfZTysNXwN8rcXlHwLe1nS8kqLi9aNdPHch8GWKSt6yXRR/5l+jeC/XUoSNayiqUu8DHNPh8/4U+Byt///IjRRtOX5I0Zt4C8XPdhBFdeoJtBko19Ay4PMU792UX1D0Yb4EWA/sB9yJooXGbJIi9P85xe/xDcAtFD1/9wIOo+ij+0BgYdN9C4H/jIhfNP7hoC2Ntg3nUvQHLttOEWKfQ/E7u56id/BtgLtRVPUeNuMPk7k2Ij7Onv8N/GlE3CEzr2p3nU3KQfEOikBZkqSxZdgrSZKk34uIZzA96F1H8RH9D1VthhZF09uTKT6GvV/Tqb+LiG9l5kwfm34V0ys7307RJ/fWWdYbwD2AJzK9zy0Amflq4NWN609jeu/Vh8/We3WAHsr0MPqMxkfcyz4CvIU9Pz7/TLoIeyl6mrYKej8DvCIzfz3TzRFxBEWf5VnbdzSuPYPp/1/kd8Brgfdl5o4Z7l9M8dH+59BhyFwDj2X3n9flwAsz839aXRgRL6J1xfMu4CsUrTy+kpk3zvbQiNiX4r+BFzbNuZjiHwzaaiPR+G/tw0wPehN4P/Dach/eFnPcFXgKxT8EVXkXe4a984G/aKy/bRFxO4pews0+l5m/7WQeSZJGTfTQ616SJEk11aiWPbs0POPmVBFxCPBL9vx49UXAgzPz6jafewfgOxQfFZ/yS+DOrTZZioi9KcLk5sDy9Mx8RjvPK821FNgy02ZO3W60NSgR8SnglNLwMZn5q4rrvwj8SdNQUmxodVkHz/xjisrSZgm8JDPf3u48jbmWACszs7ISOiK+R1EN3OyXwJ90Wr0ZEYfM9GdXww3apvwaeFBmXtvFWvbJzJs7va9x79OB00vDD8vMr7Rx73Mpqv6bbQVOy8yPd7iOvYEFVRs5RsR3KSqBp/wWOGimfwRoMcergDeUhh+cmV/vZK2SJI0ae/ZKkiRpykvYM+jdCPxxu0EvQCOse1Jp+FjgkRW3HMz0jZ3+s93nlZ69eaagt+4a1ZePKg3/sCrobShv1BYUPX870api8h86DXoBMnPLLEHvQ5ke9N4IPKSbj+nXJaTv0A7g1G6CXoBug97GvR8EPl0aftZs9zXafLy8xam/7DTobazjlqqgt+HfSse3p/jkQFsavcLL1cO/NuiVJE0Cw15JkiQREbdhekj4tk4qRKdk5rlM7zH7mIrLW/VeXdvpM8fEU4FFpbEPtbqwyVkUvY2bndbo/TuriLg/cN/S8C+B17Rzfxda9W/+ywn7aP2HO+mTOwfKv1N/2MY9T6Lomdzsy5nZ1T/MtOGTFH3Bm3WyUdufMr0dynt7WpEkSSPCsFeSJElQhCNLS2O9bGRU7kN6UsV1rYLd+/Xw3FFW7jm8HfjYTDc0eih/sjR8B4rev+0oVxID/HMnH5dvV+Oj+yeWhi8HPtXvZ9Xc+4f8/N+UjvePiHKQW9bq9+Qf+7SeaTJzG9Mr/B8YEXdqc4rnlo43M719hSRJY8kN2iRJkgTTQ7grZttsaRbliuBDImJlZt5UGv8NRc/e1U1jb42IizLzWz08f6RExH0o2l00+2I7m29RtHL4i9LYM4EvtXHvSaXj7UDHH8tv0/2Z3rLjo5m5c46eV0ebge/1c8LGhnX3A+4KHAfcBtgbWMH09xumV49DUbV7ZcX8wfS/H64ZQEuE91BUgk/9f9agCHFfPNNNEXEoxQZ+zT6Rmev7vkJJkmrIsFeSJEkw/aPc+0bE+T3Mt6LF2L7ATc0DmbkrIj4E/E3T8BrgmxHxf8AZtB96jrJyVS/M3sIBKNpmRMTFwBFNw4+KiDWZWdkSoxES3r00fH5mbmrnuV0ot4sAOHeOnlVXP+tXuB0RRwAvo9jQb58ep1s5w7nDKQLkZt/p8XmzysyrI+Is4HFNw6dFxCszc/MMtz6H6Z9gLW8sJ0nS2DLslSRJEsCBpePlFJWC/bQGuLjF+BspPiZ+WGn8IY2vjIhfUgRMPwC+lZm/7vPahiYiljN9U7t1wBc6mOZDwOubjhdR9AB+xwz37Mv0ys+fd/DMTt22xdhcPq+Oyn1ouxIRrwZeASzux3zMHBYP88/tXewZ9q4CnkhFS4aIWAT8eWn4x5l53pysTpKkGrJnryRJ0oSLiKVM79c7F1o+o1F9+kdA1aZVQfHx9GdT9PG8MCJ+GxHviYjyx8tH0eOZvlHdxxt9S9v1YSBLY62qhZutbjE2lx91H/Tz6uiWXieIiH8DXkf/gl6AhTOcG9qfW2aeQ7FhYLNyP95mj2N6FbJVvZKkiWLYK0mSpFXDXkBmXgbchyLIaadq93YUH9c+JyJ+EBEPnMv1zbGuWzhMyczLgW+Uho+LiONnuG3vFmMbOnluh1o9b+McPq+Oetr4LiL+DHh+i1PrKDZ++3OK3siHUPx3vTQzo/kLOLTDxw7696Ts30rH94mIcvuRKc8rHd/MLJscSpI0bgx7JUmS1Kr/5ffLIVEfvs6ZaRGZuT0z35uZRwH3Al4FfIXZqyHvBXwtIl7ZzQ8/TBFxJMXmWmXfi4js5Ivpm63B9I+0N2v1vrbqtdwvrZ63fA6fN1YiYiHwjy1OvQU4MDOflZkfyMxvZ+YVmXlTZm5pcX2nVfyD/j0p+zBFaNusHOoSEcdSBN173JuZk/YPCpKkCWfYK0mSpJuYXnHY6qPbA5OZP8rMN2XmwygqFO8CvAD4NK3DpwDeGBFPGeAy+2GmMLYfTo2IZRXnWm3eNpdV3oN+Xj/N1OZgUE4Ebl8a+9fMfPksG5aVdfrf9lD/3DJzA/DB0vCTI6JccTwtAAbeMzerkiSpvgx7JUmSJlxmJnBDafiAiKjFZr6ZuSszf56Z787MxwP7AU8GLmpx+Vvrsu7ZNNb5tDl+zN7AKRXnbmR6yH+XOVzLdS3G5vJ5ANtLx92Gtmt6XUgfPKR0vBN4UxfzlDdCnM0w/tzK/o09e1Ivp9iAEPj9JodPLd3zzcws9/uVJGnsGfZKkiQJ4Pul42XAPYexkNlk5tbM/BjF+sqbuh0A/MHgV9WVP6XoPdzsOuCnPX6VtdyorbEB3I9Lw3drBGdz4bstxv5wjp41pVwF3qr/bDuO6HUhfXCH0vFvMvN3Xcxz304uzsxLgfJzTujiuV3LzIuAr5aGmzdqezLT/2zdmE2SNJEMeyVJkgTwfy3GHjvwVXSg8fHu17U4NVPVYasNsub3Z0UdaxXCPiMz79bLF3Beac4HRERVWHlO6XgB8KTefqxK32b6+39qRMzl+39T6fiQiIgu5jmxD2vp1b6l43WdTtDo+3tyF88ub/63f0Q8uIt5evGu0vFxETHVo/e5pXPXA5+Z+yVJklQ/hr2SJEkC+B+Kj4U3e25ErBzCWjpxYYuxfWa4/tYWY4PcbAqAiLg98PDS8O9oHbp36owWY1W9gT/bYuxFc9EKIzNvBb5WGj4EeEK/n9Xk16XjFcAxnUwQEQ+kHpW95Y3GyuFvO57M9L6/7Wj1e/KSLubpxReAy0tjz42I44F7lMbf36hclyRp4hj2SpIkicy8gmLX+2Z7Ax/oshJyUFoFV+X+w83WtxjrtIdpPzyd6RXFH8vMcuDejY8zvYL26a0qaDPze8A3S8PH0rpiuh/e2mLsXxrh91z4UYuxJ7d7c6MS9i39W05Pfls6vlNEHNLuzRFxW+Cfunz2p4BLS2MPi4i/6HK+jmXmLqa3ZjgF+PvS2C7gPwayKEmSasiwV5IkSVNeD2wtjZ0M/EdELO5mwog4JCL+NSKOqzj/qIh4RrfzA3/dYqxV39opv2gxVq6wHYRWlbatKnI7lpk3AF8uDe8P/EnFLa9vMfbyiHhxp8+OiMURUe5D3Ly2s4FvlYb3Bb4SEQd28bxDZrnkO8DNpbG/ioiD25h7HkXrgOM7XdccKb9v0Do8nyYi1lBUxnZTDUzjHyFabQb3rxHxxE7ni4i9I2J1F0t5P7Cl6XgR8IjSNV/KzMu7mFuSpLFg2CtJkiQAMvMy4DktTj0L+F5EPKKdKt+I2CsinhwRnwMuBl4ILKm4/DDgv4ArI+KfI+J+jZBttmfsGxEfZHr/0Ysy8wdV92XmOqa3fnhGRLwoImZq/9A3EfEA4I6l4V9lZqsq1G61Co6rNmr7GvDPpeEA3h4Rn46IO832sIg4NCJeSfEx+z+e5fKnMr3f7HHADyPi2bO1kIiIRY3fxf8Gvj7TtZm5GfhoaXgF8NWIqGzn0Ohx/N/AsxtDW6quHaAvMb0NyRMi4n0zbaoXEQ+l2BzvXo2h8qZ1bcnM/wLOLA0vBj4WEe+NiANmmyMijouItwBXMnNv7ao1rAU+NstlbswmSZpokZnDXoMkSZL6LCJOAs4uDX8jM09q4963AC+tOH1lY96fAmuBTRQ9clcCdwLuCdyZouKu2b0z84ctnvU3wP8rDa8Ffgyc33jeTcBmYBlwMHBv4CHA0tJ9CTwsM2fsexsRLwH+scWpBK6mqAQtt1N4T2a+Z6Z529UIqZ9WGn5lZr65H/M3nrGUogfwXk3D24EDM/P6FtcvAL4CPLDFdLuAH1L0272CIqhdAqymCGnvDdy16fpnZObps6zvj4HPAwtbnL6Bonfxjxqvt1D8ft2Bojfr/djdl/mKzDxklmcdRFHRvVfp1A6KQPcc4LrGz3QAxXvwQIrN6gBuBN4JvKF0/6GzVZBGxOUUv7NTPpiZp810zyzzvQ54dYtTaylaLfyYolXJSop/SHkExX+PU3ZS/OPNB0r3z/pn1nj+3hTBcaugfDtFJfXZwLWNdSynqCa+C/AHFH9HTHlgZp4z2zNbrOGeFL+PrVwBHNZo+SBJ0kTq+8YLkiRJGm2Z+bKIuIqi2rMc2h5E0W92Lq2hCHMf0sE9Cbx4tqC34d8ofoZjS+NBESjeocU9la0JOtEIy04pDSfwkX7M//sJMzdHxJnAaU3DCylC5ml9WzNzR0Q8nKLK+tTS6XkUrQz61s4gM7/UqDg9kyI0bnYbir66bffWneVZV0bEXwKnl04tAB7T+KqygSIwPbofa+mDNwInAQ8oja8BnjvLvQk8nyLc7kpm3hIR96MIlh9cOr0QOLHxNWcy80cR8T2K8LjsPwx6JUmTzjYOkiRJmiYz/w24P/DtHqfaTLFh2JUV59cyfTOxTl0BPDoz39HOxZm5CXgYRfXooJ1KUaHc7NuNDfL6re1WDgCZuSUznwz8JTCt+rdNOyj+TGfVqOo8HvifLp8FRfVyO8/6IEWf5E5+1y4F7peZ3+9mYXMhM7cDj6Lov9uJm4AnZGbPG5dl5nqKVh2vZXpbiXZtoct2Eg3vajG2naKnryRJE82wV5IkSS1l5nmZeX+K0PcjFB/Nbse1wIcpqkhvn5mntmod0HjGhykqOU+lCGp+RVGBOJtdwDcpqhmPysz/bnNtU8+9JjMfStF24i0UG5pdThGK9Ro+z6RV2NqXjdlaOBu4pjR2VEScMNNNmfkuihYAfwecR/Fez2QX8H3g74FDOvmzyMxLMvMRFFWaH6Fo2zCb6xvXPgK4bwfP+gBFNfenmL4RYbNrgVcBd87MmTb7G4rMvJki8H0K8LNZLr8eeBtwZGZ+uo9r2JGZrwMOpWhv8TNm/+92G0VV8YuAAzLzxz0soVUbh89kZlvhvyRJ48yevZIkSWpbY7Ouoyk+Nr6G4qPbt1JU6V0GXFgV7HbwjH0oNjA7DNiPYkOt+Y3n3ARcBPw8Mzf28hy1JyJWUfTlvS1FML8U2EjRu/ci4ILM7KVKs/lZQdH/9/DGs1ZThO+3AldR/GPApdnj/4mJiGXAH1L0092XIqj8HUUv6vN7nX+QGj2J70vx57M3RdXstcAvgZ8N6meJiNtR9FTej+I9XUTRBuNG4NcUmxBu6tOz3gy8vDR8UmZ+ox/zS5I0ygx7JUmSJEkjISIWUrSFae6j/avMbLVpnCRJE8c2DpIkSZKkUfF4pm+Y+O5hLESSpDqysleSJEmSVHsRMY+i1cZxTcM3Awdm5obhrEqSpHqxsleSJEmSNApewJ5BL8B/GvRKkrSblb2SJEmSpNpqbKj3V8DrKTaFnHIrcFhm3jiUhUmSVEMLhr0ASZIkSZKmRMTrgUc1DvcB7gDMb3Hp6w16JUnak2GvJEmSJKlODgLuOss1ZwP/PIC1SJI0UuzZK0mSJEkaJf8HnJyZu4a9EEmS6sbKXkmSJElSnW0FrgN+AJyRmWcNeT2SJNWWG7RJkiRJkiRJ0hiwjYMkSZIkSZIkjQHDXkmSJEmSJEkaA4a9kiRJkiRJkjQGDHslSZIkSZIkaQwY9kqSJEmSJEnSGDDslSRJkiRJkqQxYNgrSZIkSZIkSWPAsFeSJEmSJEmSxoBhryRJkiRJkiSNAcNeSZIkSZIkSRoDhr2SJEmSJEmSNAYMeyVJkiRJkiRpDCwY9gKkURcR+wAnNg1dBWwb0nIkSZIkSZI0PIuAOzQdfyMzbx7Uww17pd6dCJw17EVIkiRJkiSpdh4NfH5QD7ONgyRJkiRJkiSNAcNeSZIkSZIkSRoDtnGQendV88HnPvc5jjjiiGGtRZIkSZIkSUNy8cUXc/LJJzcPXVVx6Zww7JV6t8dmbEcccQTHHnvssNYiSZIkSZKk+tg2+yX9YxsHSZIkSZIkSRoDhr2SJEmSJEmSNAYMeyVJkiRJkiRpDBj2SpIkSZIkSdIYMOyVJEmSJEmSpDFg2CtJkiRJkiRJY8CwV5IkSZIkSZLGgGGvJEmSJEmSJI0Bw15JkiRJkiRJGgOGvZIkSZIkSZI0Bgx7JUmSJEmSJGkMLBj2ArRbRBwOHA8cCCwC1gMXAt/JzC1DXNdK4N7AocBKin8kuBm4GvhBZl7Xx2ftBfwhcCdgb2AzcAXFe3Btv54jSZIkSZIkjRvD3hqIiJOBvwfuUXHJhog4HXhdZt44wHU9FnghcBIQM1z3E+A9wH9l5o4un3Uo8HrgCRRBd1lGxDeA12TmN7t5hiRJkiRJkjTObOMwRBGxOCLOAD5LddALsIIidL0gIh4wgHWtiYj/Ac4EHsgMQW/D3YH3At+LiCO6eN4TgF8Af0broJfGGk4CzomIt0TEbGuSJEmSJEmSJoph75BExDzgE8BTSqd2ApcB51O0Smh2G+B/I+K+c7iuvYGvAA9vcfoG4MfAj4BWrRvuCZwdEYd08LzHAx8DllU862ogm28BXgr8c7vPkCRJkiRJkiaBYe/wvAR4dGnsPcBBmXlYZt4dWA08Friy6ZplwCcjYp85WtebmV5l/HngHpm5X2beMzPvlZm3B44BPlK69kDgP9p5UKNH8QfY8/fwp8CDmp51B+Bo4DOl2/+m0WZCkiRJkiRJEoa9QxERa4BXloZfnpnPa96ELDN3ZeZngROAy5uuPRB48Rysaz/guaXhf8/MR2fmT8rXZ+avMvPPgFeXTj2kzerjNwDLm45/ADwgM88uPefXwClMD5H/MSLsOy1JkiRJkiRh2Dssfwfs1XT8TeCtVRdn5jXAs0rDL2qExv30CGB+0/ENwN+2cd+bgF+Vxh450w0RcSzwxKahbcDTM/OWVtdnZgJ/Dfymafhw4BltrE+SJEmSJEkae4a9A9bo1VsOKF/bCDMrZebXgG81De0FPKHPyzuydPzlzNw0202ZuYtik7lms23U9ufs+fv38cwsB8bl52wB3lIaLofgkiRJkiRJ0kQy7B28Eyg2WptyKXBOm/e+v3R8ch/W02x16fiqDu69snS8cpbrH1U6Lv9sVT4BbGw6vndE7N/mvZIkSZIkSdLYMuwdvD8tHf/fbFW9zdeWjk+KiOUtr+zOzaXjpR3cW772xqoLI+JI9qz83Qh8p52HZGb52mD6eypJkiRJkiRNHMPewbtb6bitkBOgsXnb5U1Di4Bjel/S751fOr53B/ceXzo+b4Zr71a+NjN3dPCsc2eZT5IkSZIkSZo4hr2Dd3Tp+IIO7y9fX56vF19gzxYJfxgR953tpog4Anhc09AW4KMz3FLn90CSJEmSJEkaSQuGvYBJEhFLgYNKw530xW11fXlTta5l5k0R8WbgTU3DZ0bEozPzB63uiYijgbMoqoynvCozr5/hUeU11+Y9kNSdzdt28pULruPnV9/Mjl3tdqaRJEmSJGlPJx15G046cr9hL2NkGfYO1r4UPWanbAdmCkVbuaZ03O/f/rcAxwJPbhzfHvhuRPwP8BXgCiCBA4AHAY8FFjbfn5lvn+UZ5TVf3eEay+/BbVpeJWkgbt68ndM+cB4/ufKmYS9FkiRJkjTiVi1bZNjbA8PewVpROt7UweZsUzaWjstz9iQzd0XEn1H0En4NRZA6H3hU46vKucBrMvNrbTymvObyzzSb8vULI2JxZm7tcJ5pImI/Og+PD+/1udIo+9QPrzLolSRJkiSpBgx7B6sccm7pYo7Ns8zZs0YA/W8RcRbw78AjZrnlXODtwNltPqLX96H8HkzN2XPYCzyfIuSW1KZv/ubGYS9BkiRJkiThBm2DtqR0vK2LOcqB5tIu11IpIpZHxD8DFzF70Avwh8BngF9GxB+0cX2v70OrULfv74Ok9tx4az/+nUWSJEmSJPXKsHewyhWsi1peNbPFs8zZk4jYH/gh8CJ2B6i/pqh4PYqignYZReuC04AfNd1+FPCtiDh5lsf0+j6U34NWc0oakHUbu/l3K0mSJEmS1G+2cRisDaXjcoVrO8oVrOU5uxYRSyg2YTuqafh9wAsys5zmXApcGhEfAt4AvLIxvgD4WETcIzN/VfGoXt+HVlW8/Xof3g18qsN7DgfO6tPzpZGSmZVh7z0PXsW+K7r5Ny1JkiRJ0qQ6fL/lw17CSDPsHaxyILksIqLDTdrKv/F9C3uBlwLHNh1/HXhOZu6quqGx9ldFxEHAUxvDSyh6+D684rbymjv9r7h8/Y7M7Etlb2ZeD1zfyT0R0Y9HSyNp47adbNvZ+q+IVzz8aO558KoBr0iSJEmSpMllG4fBuhFoDnYXAvt1OMcBpeOOgskqETEfeGFp+FUzBb0lrwSar/3jiLhDxbXlNR/Y5jOmlN+DGzq8X1KfrNtQ3cJhzXKreiVJkiRJGiTD3gHKzM3AlaXhgzqcpnz9hd2vaA93AfZtOr4R+F67N2fmVcBPm4YCuF/F5b8uHdflPZDUoXWbqsPeVYa9kiRJkiQNlGHv4JWDyWM6vP/oWebr1qGl48s7bC8BcFnpuFyBO6Wu74GkDq3buLXl+IJ5wd5L7BQkSZIkSdIgGfYO3vml4xPavTEibg8c0jS0Hbig9yUBsLh0vKOLObaXjudXXHd+6fjeEdFJKvSHs8wnaUDWVrRxWLV8kf2sJUmSJEkaMMPewftC6fiPov1E5KGl47Mzs18btK0tHe/fxRxt9dLNzAuBS5qGltNm6B0Ry4H7Nk/H9PdU0oCsr2jjYL9eSZIkSZIGz7B38L5D0Q93ymHASW3e+8zS8Vn9WFDD5aXjgyLi8HZvjoi9gHuXhi9pdW3D50vH5Z+tyhOBFU3HP8zMa9u8V1Kfrd3YOuxdbdgrSZIkSdLAGfYOWGbuAk4vDb9mtureiHgwcP+moVuBT/ZxXRcBV5eG/7aDKV7Mnq0gNjHzBm//RVGVO+VJEVHuxbuHiFgCvKw0/P4O1iipz9ZXhL1uziZJkiRJ0uAZ9g7HW4Hm9gsnAi+tujgiDgDeVxp+Z2be2Or6pvuy9HXSLOs6o3T8nIh42iz3EBGPBF5VGv54ZrbeuQnIzF+wZ1i9CPhgROxd8YwA3gHcsWn4UorQWNKQrKsIe23jIEmSJEnS4Bn2DkEjpH1zafgfIuLdEfH7XrkRMS8iTqZo/XBI07XXAm+fg6X9I7Cu6TgoAtgPRMSx5Ysj4oiI+Ffgc0DzBmubgNe38bxXNa6dcm/gm+VQOiLuBHwaeE7p/pdlZnlTOEkDVBX2rlpm2CtJkiRJ0qAtmP0SzZG3UmxK9oimsecBz46IK4CbgUOBlaX7NgNPyMyb+r2gzFwfEY8BvsKeLRlOA06LiOspWj0kxQZut28xzS7gyZl5RRvPuzgingl8lCJYBrgrcHZE3ABcCewHHNh0fsq/Zuan2v3ZJM2NysreFYa9kiRJkiQNmpW9Q9Lo3ft44OOlU/MpNm27O9OD3rXAwzPz3Dlc1zeBPwJahbX7AfcA7knroPd3wCMzs+2N4zLz48BTKELsZrdpPOcOTA96/wn463afIWnuVG3QZmWvJEmSJEmDZ9g7RJm5JTNPBU4Bzp/h0o3Au4FjMvOcAazr28CdgRcBF7Zxy+UULRmOzcwvdvG8jwHHUVT4ztSW4ZvASZn5kszMGa6TNADbd+7i1i07Wp6zZ68kSZIkSYNnG4cayMwzgTMj4gjgPsABFBuW3QT8Cjg3M7d0MW+5IraTe2+l2BDtHRFxO4p+uvtTVBsHRZuJ3wE/zMwru31O0/MuBZ4SEc8D7kexEdtewBaKdg7nZuY1vT5HUv+sr6jqBVhtGwdJkiRJkgbOsLdGMvNi4OJhr6MsM68D/ntAz7oF6Lg6WNLgrds0Q9hrGwdJkiRJkgbONg6SpK6s21Ad9q6yjYMkSZIkSQNn2CtJ6kpVZe9eSxawcL7/8yJJkiRJ0qD5/8YlSV1ZV9Gz183ZJEmSJEkaDsNeSVJX1la0cVht2CtJkiRJ0lAY9kqSurK+oo2DYa8kSZIkScNh2CtJ6sraijYOhr2SJEmSJA2HYa8kqSvrK8LeVYa9kiRJkiQNhWGvJKkrbtAmSZIkSVK9GPZKkrpS1cZh1TLDXkmSJEmShsGwV5LUscysbOOwZoVhryRJkiRJw2DYK0nq2C1bdrBjV7Y8t3r54gGvRpIkSZIkgWGvJKkLVVW9AKtt4yBJkiRJ0lAY9kqSOlbVrxdgtW0cJEmSJEkaCsNeSVLHqip7F82fx/JF8we8GkmSJEmSBIa9kqQurKsIe1cvX0REDHg1kiRJkiQJDHslSV2oauOwerktHCRJkiRJGhbDXklSx9ZvMuyVJEmSJKluDHslSR1bu8GwV5IkSZKkujHslSR1zMpeSZIkSZLqx7BXktQxe/ZKkiRJklQ/hr2SpI6t27i15fgqw15JkiRJkobGsFeS1LH1G7e3HF9j2CtJkiRJ0tAY9kqSOrJ1x042bN3R8pxtHCRJkiRJGh7DXklSR6qqesGwV5IkSZKkYTLslSR1ZG1Fv14w7JUkSZIkaZgMeyVJHZmpsnfl0oUDXIkkSZIkSWpm2CtJ6khVZe/KZQtZMN//WZEkSZIkaVj8f+WSpI6s27it5fjqZbZwkCRJkiRpmAx7JUkdWV8V9tqvV5IkSZKkoTLslSR1ZK1hryRJkiRJtWTYK0nqyPpNhr2SJEmSJNWRYa8kqSNrNxj2SpIkSZJUR4a9kqSOWNkrSZIkSVI9GfZKkjqyzp69kiRJkiTVkmGvJKltu3Yl6zdtb3nOsFeSJEmSpOEy7JUkte2WLdvZuStbnjPslSRJkiRpuAx7JUltW1vRwgEMeyVJkiRJGjbDXklS29Yb9kqSJEmSVFuGvZKktlVV9i5ZOI9lixYMeDWSJEmSJKmZYa8kqW3rKsLe1cus6pUkSZIkadgMeyVJbasMe1cY9kqSJEmSNGyGvZKktlWGvcsXD3glkiRJkiSpzLBXktS2qg3aVi9bOOCVSJIkSZKkMsNeSVLbqjZos7JXkiRJkqThM+yVJLVt/aaqsNfKXkmSJEmShs2wV5LUtrUbrOyVJEmSJKmuDHslSW2r3qBt0YBXIkmSJEmSygx7JUlt2bxtJ5u372x5zrBXkiRJkqThM+yVJLVlXUW/XjDslSRJkiSpDgx7JUltWV/RwgEMeyVJkiRJqgPDXklSW9ZWhL3zAvZZunDAq5EkSZIkSWWGvZKktlRV9q5ctoj582LAq5EkSZIkSWWGvZKktlRV9trCQZIkSZKkejDslSS1Zd3GrS3HDXslSZIkSaoHw15JUlvWbdzecnz1MsNeSZIkSZLqwLBXktSWysreFYa9kiRJkiTVgWGvJKkt663slSRJkiSp1gx7JUltWWvPXkmSJEmSas2wV5LUlnUbt7UcN+yVJEmSJKkeDHslSbPauSu5aXNFGwfDXkmSJEmSasGwV5I0q5s2bSOz9TnDXkmSJEmS6sGwV5I0q/WbWrdwAMNeSZIkSZLqwrBXkjSrtRsMeyVJkiRJqjvDXknSrKoqe5ctms+ShfMHvBpJkiRJktSKYa8kaVZrN7YOe63qlSRJkiSpPgx7JUmzWlfRxmGNYa8kSZIkSbWxYNgLkCT1385dya9+ewtXrdvUl/l+evVNLcdXGfZKkiRJklQbhr2SNGauvWkzT/uv87j4+g1z/izbOEiSJEmSVB+2cZCkMfPSM382kKAXYPUyw15JkiRJkurCsFeSxsgtW7bznUvWDux5q1cY9kqSJEmSVBeGvZI0Rq6/ZQs7d+XAnnfc/vsM7FmSJEmSJGlmhr2SNEbWbtg2sGcdedu9OOHwNQN7niRJkiRJmpkbtEnSGFm/qTrsXblsYV+esdeSBdz3sDW84uFHs2C+/2YoSZIkSVJdGPZK0hhZu7F12HvgqqV8+6UPGvBqJEmSJEnSIFmSJUljZF1FG4c1y91ITZIkSZKkcWfYK0ljZF1FG4dVhr2SJEmSJI09w15JGiPrKto4rDbslSRJkiRp7Bn2StIYqQx7lxn2SpIkSZI07gx7JWmMVIa9Kwx7JUmSJEkad4a9kjRGqsJeN2iTJEmSJGn8GfZK0pjIzMqwd5VtHCRJkiRJGnuGvZI0JjZt28nWHbtanltjGwdJkiRJksaeYa8kjYmqql6wsleSJEmSpElg2CtJY2KmsHfN8sUDXIkkSZIkSRoGw15JGhPrNrUOe+fPC/ZasmDAq5EkSZIkSYNm2CtJY2LdhurN2ebNiwGvRpIkSZIkDZphrySNiao2DmuW269XkiRJkqRJYNgrSWOiqo3DquULB7wSSZIkSZI0DIa9kjQmqto4uDmbJEmSJEmTwbBXksaElb2SJEmSJE02w15JGhNVPXtXW9krSZIkSdJEMOyVpDFRGfYus7JXkiRJkqRJYNgrSWOiMuxdYWWvJEmSJEmTwLBXksbA9p27uHnz9pbn1ixfNODVSJIkSZKkYTDslaQxcNOm1kEvwKplhr2SJEmSJE0Cw15JGgNVLRwA1qww7JUkSZIkaRIY9krSGJgp7F3pBm2SJEmSJE0Ew15JGgNVYe9eixeweMH8Aa9GkiRJkiQNw4JhL0B7iojDgeOBA4FFwHrgQuA7mbllmGuTVF/rNm5tOb7aFg6SJEmSJE0Mw96aiIiTgb8H7lFxyYaIOB14XWbeOIfruBw4uE/TfTAzT5vhWacBH+hh/m9k5kk93C+NjXUbW2/Q5uZskiRJkiRNDts4DFlELI6IM4DPUh30AqwAXghcEBEPGMjielfdRFRSX1VV9q5ZbtgrSZIkSdKkMOwdooiYB3wCeErp1E7gMuB84ObSudsA/xsR953zBfbuC8NegDQp1m2qqOw17JUkSZIkaWLYxmG4XgI8ujT2HuANmXkt/D4QfjTwDuCgxjXLgE9GxHGZWQ6De/UUYGkX950CPKfp+Hrgix3O8TbgKx1cv77D+aWxZWWvJEmSJEky7B2SiFgDvLI0/PLMfEvzQGbuAj4bEecB3wYOaZw6EHgx8Jp+riszz+3mvoh4fWnoI5m5o8NpLsjMr3bzfGnSVfbsNeyVJEmSJGli2MZheP4O2Kvp+JvAW6suzsxrgGeVhl/UCI2HKiLuBJTbSpw+hKVIE6uqsne1Ya8kSZIkSRPDsHcIGq0ZnlEafm1m5kz3ZebXgG81De0FPKHPy+vGaaXjn2Tmz4axEGkSZSbrNrbeD9E2DpIkSZIkTQ7D3uE4gWKjtSmXAue0ee/7S8cn92E9XWsE108tDZ8+hKVIE2vD1h1s39n634ps4yBJkiRJ0uQw7B2OPy0d/99sVb3N15aOT4qI5X1YU7ceTNE/eMp24KNDWos0kaqqesHKXkmSJEmSJolh73DcrXT8nXZvzMxrgcubhhYBx/S+pK49vXT8hcy8cSgrkSbUTGGvlb2SJEmSJE2OBcNewIQ6unR8QYf3XwAcUprvB70sqBsRsTfwmNLw6T3OOR84jKLNRQJrgd9m5q29zCuNs6qwd+H8YK/F/jUvSZIkSdKkMAUYsIhYChxUGr6qw2nK1x/Z/Yp68gRgWdPx74Av9jDfK4B/odh4rtmuiPg5RQuLd2XmFT08Qxo7ayvC3tXLFxERA16NJEmSJEkaFts4DN6+QHP6sh24vsM5rikd79fTirpXbuHwkczc0cN8d2R60AvF7+ldgb8FLo6If2+E5pKA9RVh76pltnCQJEmSJGmSWNk7eCtKx5s62JxtysZZ5pxzEXE4cL/S8OkDePQC4LnAH0bEwzLzt/2cPCL2o2gh0YnD+7kGqVNVbRzWrDDslSRJkiRpkhj2Dl45mN3SxRybZ5lzEE4rHf84M3/e5VznA1+g2KjuAuBGYBuwGrgz8DDgL4B9mu65M/DfEXFiZpbD7148H3hNH+eT5lxV2GtlryRJkiRJk8Wwd/CWlI5bpzQz21o6HmhLgyiagD61NHx6F1P9GLhnZv644vzvGl9fjYg3A/8FnNx0/p7A64H/r4tnS2OjsrJ3uWGvJEmSJEmTxJ69g1eu5O0mjVk8y5xz7YHAwU3H24CPdjpJZv5shqC3fO164HHAZ0qnnh8RB3T6bGmcrNtUUdlr2CtJkiRJ0kSxsnfwNpSOy5W+7ShX8pbnnGvljdm+kJlr5/qhmbkrIp4FPAhY2RheAjweeEefHvNu4FMd3nM4cFafni91zMpeSZIkSZIEhr3DUA5ml0VEdLhJ2/JZ5pwzEbGCosK22emDen5mro+I/wJe3DT8UPoU9mbm9cD1ndxTdLWQhmfdhtZh7+rl5Q8BSJIkSZKkcWYbh8G7EWgOdhcC+3U4R7ltQUfhZI8ez55h8++A/x3g8wG+Vjo+csDPl2pj245d3Lp1R8tzq5YvHPBqJEmSJEnSMBn2DlhmbgauLA0f1OE05esv7H5FHSu3cPhIZrZOmubOVaXj2wz4+VJtrK/o1wuwxspeSZIkSZImimHvcJTD2WM6vP/oWeabExFxKPCA0vAHBvHsku2lY8sXNbGq+vWClb2SJEmSJE0aw97hOL90fEK7N0bE7YFDmoa2Axf0vqS2PB1oblD7o8z8xYCe3ex2peMbhrAGqRZmDHuXuUGbJEmSJEmTxLB3OL5QOv6jaH+Xr4eWjs/OzDnfoK2xvqeVhk+f6+dWuF/puNzWQZoYayvC3r2XLGDhfP+KlyRJkiRpkpgEDMd3KDZqm3IYcFKb9z6zdHxWPxbUhgcAhzYdbwM+OqBn/15ELGB63+Dyhm3SxFhfEfauWWG/XkmSJEmSJo1h7xBk5i6mV8W+Zrbq3oh4MHD/pqFbgU/2d3WVygHrf2fmugE9u9nLgCNKY4MKvKXaqarsXb3cFg6SJEmSJE0aw97heSvQ3H7hROClVRdHxAHA+0rD78zMG1td33Rflr5O6nShEbEMOKU0fHqn85TmfF5EPKjDe14MvL40/PnM/FEva5FGWVVlr/16JUmSJEmaPIa9Q9IIad9cGv6HiHh3ROw/NRAR8yLiZIrWD4c0XXst8Pa5XmfDKcBeTcfXAV/qcc77AF+LiPMj4hURcY+ImPa584hYFhGPjIhvUPy8zdXPa4EX97gOaaRVbdC2xspeSZIkSZImzoJhL2DCvRU4AXhE09jzgGdHxBXAzRR9cleW7tsMPCEzbxrAGmF6C4ePZOaOPs1918bXm4AdEXEVxc+9DVhFEXAvbHHfrcAjM/OSPq1DGklVYe8qw15JkiRJkiaOYe8QZeauiHg88AHgSU2n5lNs2tbKWuCUzDx3rtcHEBEHAQ8sDZ8+R49bwJ6bwFX5PvAUg17Jyl5JkiRJkrSbbRyGLDO3ZOapFK0Szp/h0o3Au4FjMvOcASxtytPYs3XCjzLzF32Y99+BdwA/Bra3cf0WitYRjwTua9ArFdygTZIkSZIkTbGytyYy80zgzIg4gqKf7QHAIuAm4FfAuZm5pYt5Y/arZrz/jcAbe5mjYt7vU1ToEhGLgKMpqnr3p+gPvBC4BVgPXAT8JDNbp1rShMpM1m8y7JUkSZIkSQXD3prJzIuBi4e9jkFqhLg/bXxJatMtm3ewc1e2PGfYK0mSJEnS5LGNgySNqHUVVb1g2CtJkiRJ0iQy7JWkEbVu49bKc4a9kiRJkiRNHsNeSRpR6za23ttw0YJ5LFs0f8CrkSRJkiRJw2bYK0kjqqqyd83yRUT0tDejJEmSJEkaQYa9kjSi1m5s3bPXFg6SJEmSJE0mw15JGlHrDXslSZIkSVITw15JGlFW9kqSJEmSpGaGvZI0oqoqe1ctM+yVJEmSJGkSGfZK0ohaVxH2rrGyV5IkSZKkiWTYK0kjqrKNwwrDXkmSJEmSJpFhrySNqMoN2mzjIEmSJEnSRDLslaQRtGX7TjZu29nynBu0SZIkSZI0mQx7JWkErd/UuqoXDHslSZIkSZpUhr2SNILWbjDslSRJkiRJezLslaQRVFXZGwEr7dkrSZIkSdJEMuyVpBG0rmJztpVLFzJ/Xgx4NZIkSZIkqQ4MeyVpBFW1cbCFgyRJkiRJk8uwV5JGUFUbB8NeSZIkSZIml2GvJI2gtRVtHAx7JUmSJEmaXIa9kjSC1hv2SpIkSZKkEsNeSRpBVvZKkiRJkqQyw15JGkHrKsLeVcsMeyVJkiRJmlQLhr0ASRonG7fuYOuOXXP+nKqwd80Kw15JkiRJkiaVYa8k9cFPrlzP35/1C3557S1kDm8dq5cvHt7DJUmSJEnSUBn2SlKPrr1pM3/2vu+zcdvOYS+F1bZxkCRJkiRpYtmzV5J69JVfXleLoBdgtW0cJEmSJEmaWIa9ktSjS27YOOwlALB80Xz2NeyVJEmSJGliGfZKUo82b69HVe+f3Pn2LF4wf9jLkCRJkiRJQ2LPXknq0bDD3gXzgocee1ve8OjjhroOSZIkSZI0XIa9ktSjLRX9eh979wN4zomHz/nzD1i1lBWL/etckiRJkqRJZzogST3aVBH27r9yKUfebq8Br0aSJEmSJE0qe/ZKUo+q2jgsXWT/XEmSJEmSNDiGvZLUoy1VYe9Cw15JkiRJkjQ4hr2S1KOqNg5W9kqSJEmSpEEy7JWkHlW2cbCyV5IkSZIkDZBhryT1aEtFZe8Sw15JkiRJkjRAhr2S1KOqyt5ltnGQJEmSJEkDZNgrST3YtmMXO3Zly3P27JUkSZIkSYNk2CtJPaiq6gV79kqSJEmSpMEy7JWkHmyZKey1sleSJEmSJA2QYa8k9WBTxeZsYGWvJEmSJEkaLMNeSerBZsNeSZIkSZJUE4a9ktSDGXv22sZBkiRJkiQNkGGvJPWgqmdvBCxe4F+xkiRJkiRpcEwiJKkHVT17ly6cT0QMeDWSJEmSJGmSGfZKUg+q2jjYr1eSJEmSJA2aYa8k9WBLVWWv/XolSZIkSdKAGfZKUg82bdvRctzKXkmSJEmSNGiGvZLUg83bd7Uct7JXkiRJkiQNmmGvJPXAnr2SJEmSJKkuDHslqQdbqsJeK3slSZIkSdKAGfZKUg/s2StJkiRJkurCsFeSerB5W0XPXsNeSZIkSZI0YIa9ktQD2zhIkiRJkqS6MOyVpB7YxkGSJEmSJNVFrcPeiHhORKwY9jokqcpmK3slSZIkSVJN1DrsBf4d+G1E/GdE3GfYi5Gkss3bK3r2GvZKkiRJkqQBq3vYC7AM+HPgOxHxs4h4YUSsHPKaJAmALdsqKntt4yBJkiRJkgZsFMLeKQEcB7wTuDYiPhQRDxjymiRNuE3b7dkrSZIkSZLqoe5h7zuAdRRBL0A2Xi8BngKcHREXRsT/FxH7DmeJkibZ5m22cZAkSZIkSfVQ67A3M18MHAA8Gfh686nG9wDuBPwjcHVEfCIiHjLYVUqaZFuqNmizsleSJEmSJA1YrcNegMzclpkfz8w/Ao4A3gJcx/Rq30XAKcCXIuKSiHh5RNx+KIuWNBEyk81VYa+VvZIkSZIkacBqH/Y2y8zLMvMVwEHAY4D/AaY+Q91c7Xso8Ebgioj4bEQ8PCJi2oSS1INtO3exc1e2PGdlryRJkiRJGrSRCnunZObOzDwrMx8JHAK8BriS6dW+C4BHAf9NEfy+NiIOGsKSJY2hLRX9esHKXkmSJEmSNHgjGfY2y8xrMvMNwGHAHwNnAjumTje+B3Ag8PfApRHxvxHxmIgwjZHUtaoWDmBlryRJkiRJGryRD3unZOErmfl4imD3pcBFFEFvsrvadx7wUODTFJu6vclqX0nd2LRtR+U5K3slSZIkSdKgjU3Y2ywzb8jMtwHHAW9nz/YOzdW+twVeBlwSER+PiGMGvlhJI8vKXkmSJEmSVCdjGfZGxEER8TrgMuDF7A54f39J4/vU+Hzg8cD5EfEW2ztIaseWmcJeK3slSZIkSdKALRj2AvqlEdA+GvgL4CEUgW55wzYoWjt8ADgGOAVYyu7QdwHwEuBOwGMHsnBJI2tzxQZt8wIWzR/Lf0uTJEmSJEk1NvJpREQcERFvAa4GPkXRj3fq55oKcXcAnwQelJlHZeZbM/PpwP7AXwOXs7u3bwCPjohTB/dTSBpFVT17ly6cT0S0PCdJkiRJkjRXRjLsjYhFEfHkiDgb+DVFNe5t2V29S+P15cDLgQMz80mZeU7zPJl5c2b+K3Ak8Hp2B74Az5jTH0LSyKvq2bt00dh8aEKSJEmSJI2QkUokIuJYijYNfwasmhpm98ZrAewEvgC8JzO/0s68mbkDeG1E3Bl4TGP47n1cuqQxVNWzd+mikfx3NEmSJEmSNOJqH/ZGxFLgSRQh732mhhvfs+n4KuB9wPsy87ddPu4sdoe9K7ucQ9KE2LStIuxd6OZskiRJkiRp8God9kbEeyiC3r2mhtizijeBLwLvBb6Yma13S2rftU2vLc2TNKPKNg6GvZIkSZIkaQhqHfYCz2bPYJfG698C/wX8R2Ze1cfn5eyXSFJhS1Vl7yLDXkmSJEmSNHh1D3vL/o+iiveszGydsvTmUuB1czCvpDFkZa8kSZIkSaqTUQh7bwQ+QFHFe+lcPigzL8OwV1KbKnv2WtkrSZIkSZKGoO5h76nAZzJz+7AXIkll1ZW9df+rVZIkSZIkjaNaJxKZ+Ylhr0GSqmypCnsXub+jJEmSJEkaPBMJSepSZRsHe/ZKkiRJkqQhMOyVpC5tNuyVJEmSJEk1Uus2DhFxO+DNTUPvyczzupjnPsBzmob+NjPX9bo+SZOtuo1Drf9qlSRJkiRJY6ruicSzgdOABG4Ent/lPD8HHgmsbhz/Enh7r4uTNNmqN2jzQxOSJEmSJGnw6p5IPLHp9RmZuaWbSTJzE3AGEI2vJ/dhbZImXGXP3kW2cZAkSZIkSYNX27A3Ig4Ajm4aOrPHKT/d9PquEbFvj/NJmnC2cZAkSZIkSXVS27AXuEvT6+3AD3uc74eNeaCo7r1rj/NJmnBu0CZJkiRJkuqkzmHvoY3vCVyemdt6mSwztwKXNw0d3st8kiZbZrKpsmevYa8kSZIkSRq8Ooe9eze9vqlPc65ver1Pn+aUNIG27thFZutzSxfV+a9WSZIkSZI0ruqcSGxver2sT3MubXodfZpT0gSq6tcLsHShPXslSZIkSdLg1TnsvaHxPYD9+zTnAU2vb+zTnJIm0OaZwt5FtnGQJEmSJEmDV+ew95qm16si4u69TNa4f3XT0G97mU/SZNtUsTkb2LNXkiRJkiQNR53D3u8C2yg2aAN4Xo/zPb/p9U7gOz3OJ2mCbZ4p7LWyV5IkSZIkDUFtw97M3AScS9HGIYBnRMQDupkrIk4ETqMIjhP4Xmbe3KelSppAM/fsNeyVJEmSJEmDV9uwt+GfGt8TmA+cFRF/1MkEjes/R/GzTm3K9vZ+LVDSZKpq4zB/XrBwvvs/SpIkSZKkwat12JuZ/wt8iyKkTWAf4EsR8YGIuMdM90bEPSLidOBLjftozPGdzDxr7lYtaRJUbdC2dOF8Igx7JUmSJEnS4C0Y9gLa8ETgB8D+FGHtPOBpwNMi4nrgR8D1wAZgBbAfcM/Gd9gdFAdwFfD4QS5e0niqauNgv15JkiRJkjQstQ97M/O6iHg4cBZwCLs3bAvgtsCftLhtqqxuqkdvAJcCj8rM6+Z0wZImQtUGbfbrlSRJkiRJw1LrNg5TMvPnwN2BTzSGpqp1s+oWdoe8CZwB3CMzL5jjpUqaEFU9ew17JUmSJEnSsIxE2AuQmTdn5qnAMcB7KCp1Y4avi4F/BY7KzKdl5i1DWbiksVTZs9c2DpIkSZIkaUhq38ahLDN/DTwfICJuBxwOrAb2Am4F1gEXZ+bvhrbIHkTE4cDxwIHAImA9cCHFxnJbhrm2QYmIA4D7AgcDS4FbgIuAb2fmhmGuTZpS2bPXyl5JkiRJkjQkIxf2Nmv03x2LHrwRcTLw98A9Ki7ZEBGnA6/LzBvncB2XU4Ss/fDBzDytg2efCLwWOKnikm0R8Qng1Zl5ea+Lk3pR2cbByl5JkiRJkjQkI9PGYVxFxOKIOAP4LNVBL8AK4IXABRHxgIEsrnfb2rkoCv8InEN10AtFpfNTgV9ExON6Xp3UA9s4SJIkSZKkujHsHaKImEex6dxTSqd2ApcB5wM3l87dBvjfiLjvnC+wd19o87p/AV5SGkvgKuDHQLmSeTnwiYh4TG/Lk7q3xQ3aJEmSJElSzYx0G4cx8BLg0aWx9wBvyMxr4feB8KOBdwAHNa5ZBnwyIo7LzHIY3KunUPTJ7dQpwHOajq8HvjjbTRHxBIqK5WZnAi/PzN80Xfdg4J+BuzSG5gMfjIif2NJBw1BZ2WvYK0mSJEmShqTWYW9EHAa8r2no7zPz3C7muR/w+sZhAk/PzKv7sMSuRcQa4JWl4Zdn5luaBzJzF/DZiDgP+DZwSOPUgcCLgdf0c13dvL8AEfH60tBHMnPHLPcsAt5aGn4P8PzMzNK6vtZoX/FV4F6N4b2A1wFP72bNUi/s2StJkiRJkuqm7m0cnkvRw/VE4FDge13O8z3g8MY8JwHP6sPaevV3FGHllG8yPfj8vcy8hunrflEjNB6qiLgTUG4rcXobtz6T3eE1wG+AF5WD3imNKuans2cv4KdExFFtL1bqEyt7JUmSJElS3dQ97D2l6fWHM7N1ujKLRoXph4FofD2hD2vrWqM1wzNKw6+tCjmnZObXgG81De3FkH+WhtNKxz/JzJ+1cV85vP6HzNwy0w2ZeQFFn+Mp85n+XkpzbosbtEmSJEmSpJqpbdjbaOFwSNPQ53qc8jNNr4+MiP17nK8XJ1BstDblUuCcNu99f+n45D6sp2uN4PqppeHT27jvQOAeTUMbgE+2+djye1DueyzNuc1u0CZJkiRJkmqmtmEvcFzT6y3A+T3O99PGPFPuUnXhAPxp6fj/Zqvqbb62dHxSRCzvw5q69WCK/sFTtgMfbeO+8ntwbmZubPOZ5wKbmo6PjIg7tnmv1Bf27JUkSZIkSXVT57D34Mb3BK5obFTWtUYLiMubhg7tZb4e3a10/J12b8zMa9nz51gEHNP7krpW3hztC5l5Yxv33a103Ml7sAM4b5b5pDlV2cbByl5JkiRJkjQkdQ57mzcvu7lPczbPs3ef5uzG0aXjCzq8v3x9eb6BiIi9gceUhk9v8/axeA80udygTZIkSZIk1U2dw97mj+n3K5htDpC72uytVxGxFDioNHxVh9OUrz+y+xX15AnAsqbj3wFfbPPe8ppH9T3QBMrM6rDXNg6SJEmSJGlIFgx7ATO4ofE9gAMiIjroaztNYyOxOzQNtdNqYC7sS/EzTdkOXN/hHNeUjvfraUXdK7dw+EijxUI7blM6vrrDZ8/JexAR+zF9bbM5vB/P1ujYumMXVX8bGfZKkiRJkqRhqXPYe0XT672A+9JBX9cW/oA9K3s7DRf7ZUXpeFMXIXZ5I7PynHMuIg4H7lcaPr3Ne5cC5USs3c3Zqq7v13vwfOA1fZpLY2pzxeZsYBsHSZIkSZI0PHVu4/B9ilYOU0Hoi3ucr/n+bcC5Pc7XrXIouaWLOTbPMucgnFY6/nFm/rzNe1utt9P3oQ7vgSZUVQsHMOyVJEmSJEnDU9uwNzO3A1+naHkQwGMi4gndzBURTwQeSxEcJ3BOZpbDwkFZUjre1sUcW0vHS7tcS1ciIoCnloZP72CK8nsAnb8PQ30PNNk2zVDZu8w2DpIkSZIkaUjq3MYB4B+AR1AEtAF8MCJWZOZ/tTtBRDwTeFfTHNmYd1jKFayLuphj8SxzzrUHAgc3HW8DPtrB/a3Wu6hivMpcvQfvBj7V4T2HA2f16fkaAVtmqOxdYtgrSZIkSZKGpNZhb2Z+NyI+w+6q3MXAf0bEU4F/Ab6cmZvK90XEMuCPgb8C7s/ukDeBz2fmNwf0I7SyoXTcqsp1NuUq1vKcc628MdsXMnNtB/e3Wu8SOgts5+Q9yMzr6XDDvKLQWZPENg6SJEmSJKmOah32NpwGHAUcw+7q3Ac0vnZFxK8pwrkNFH1b9wOOZHeLimi67xfA0wa49lbKoeSyiIgON2lbPsuccyYiVgCPKw2f3skcmbk5Inay5yZty4GbOphmaO+BVNXGYcG8YOH82nbHkSRJkiRJY672YW9mboiIhwGfAe7N7g3bgiIsPAY4uumW5jLLqWreAL4HnJKZt875omd2I7vXBLCQIqD+XQdzHFA67qgStUePZ8+g9XfA/3Yxzw3A7ZqODwSu6eD+Yb4HmnCbK8LepbZwkCRJkiRJQzQSJWiZeQ1wP+CtwCZ2B6VTYe4el7NnILwBeCPwgMy8du5XO7PGxnBXloYP6nCa8vUXdr+ijpVbOHwkM3d0Mc+vS8ej9B5owlX17LWFgyRJkiRJGqaRCHsBMnN7Zr6cIuR7KfA1dge/5a+NwP8B/x9wUGa+ustAcq6Ug8ljOrz/6NLxQILOiDiUon1Gsw90Od1IvgcSVPfstbJXkiRJkiQNU+3bOJRl5nrgbcDbImI+RSuA1cBewK3AOuC6zKzeQWn4zgce1nR8AvDBdm6MiNsDhzQNbQcu6NfCZvF09myT8aPM/EWXc51fOj6h3RsjYgFw/CzzSXOmqmevlb2SJEmSJGmYRi7sbdYIdK+hs16vdfAFiurkKX/UwSZtDy0dn52Zc745WUQE0ze3O72HKf+ndHxCRCzPzI1t3PuHwLKm44sy86Ie1iJ1pLKNg5W9kiRJkiRpiEamjcOY+Q7FRm1TDgNOavPeZ5aOz+rHgtrwAODQpuNtwEe7nSwzrwJ+0jS0AnhCm7cP6z2QgBk2aLOyV5IkSZIkDZFh7xBk5i6mV8W+plE9WykiHgzcv2noVuCT/V1dpfLGbP+dmet6nPP9peOXRcSSmW6IiKOBJzYNtXovpTllGwdJkiRJklRHhr3D81aguf3CiezZ2mEPEXEA8L7S8Dsz88ZW1zfdl6WvkzpdaEQsA04pDZ/e6Twt/CdwZdPxnYD/VxV6R8TewIeARU3DH83MQfUslgA3aJMkSZIkSfVk2DskjZD2zaXhf4iId0fE/lMDETEvIk6maP1wSNO11wJvn+t1NpxCsQHelOuAL/U6aWZuA15WGn4u8MmIuGPzYEQ8CPgWcK+m4Q3Aq3tdh9Spyp69VvZKkiRJkqQhGrkN2iLiKOAE4O7AvsBKYHGH02RmPrjPS+vGWyl+lkc0jT0PeHZEXAHcTNEnd2Xpvs3AEzLzpgGsEaa3cPhIZu7ox8SZ+bGIuD/Fzz3lFOBxEXEVcANwMMWfdbNdwDMy87J+rEPqRGXPXit7JUmSJEnSEI1M2BsRj6aoAj2+16mA7H1FvcvMXRHxeOADwJOaTs2n2LStlbXAKZl57lyvDyAiDgIeWBo+vc+PeSGwBXhR86OBgxpfZZsogt5P93kdUls22cZBkiRJkiTVUO3bOETE/Ih4H/AZiqC3uZ9rNH21vH2W80OXmVsy81SKatbzZ7h0I/Bu4JjMPGcAS5vyNPZ8/36Umb/o5wMyc1dmvhiYatVQZRvwEeC4zBzUxnTSNFvcoE2SJEmSJNXQKFT2vgP488brqYrcKB03jzWb7XxtZOaZwJkRcQRwH+AAio3IbgJ+BZybmVu6mLennzsz3wi8sZc5OnjW2cDZEXEgRXuLg4AlwK3Ab4BvZ+Ytg1iLNJPKDdoMeyVJkiRJ0hDVOuyNiBOBF7A7tN0GvA04A7ic4qP/U+f+CPg2sBo4lqJK9M+B2zauuQB4Smb+dEDL70pmXgxcPOx1DFNmXg1Yuava2rStdctq2zhIkiRJkqRhqnsbh5c3vgewE3hkZr46My/KzG3lizNzW2Zel5lfy8xXUmzs9U+N00cD32xsBiZJXduyfVfLcSt7JUmSJEnSMNU27I2I5cCDKapyE/iPzPxqJ3M0wt+/A55PERjvBXw2Im7b7/VKmhyVbRys7JUkSZIkSUNU27AX+ANgPrt77b6n24ky873ApxqHq4A39LY0SZNssxu0SZIkSZKkGqpz2Htw0+ubM/MXs1y/eJbzb2p8D+ApEbGk65VJmli7dqWVvZIkSZIkqZbqHPaubnxP4MqKa5p3SZoxvM3MnwG/a7r2hJ5WJ2kibd3Rul8vWNkrSZIkSZKGq85h74Km1xsrrrmV3W0e9mtjzqubXh/ZzaIkTbaqql6wsleSJEmSJA1XncPem5ter6i45qam1wdXXNOs+edd2eF6JIlN23ZUnrOyV5IkSZIkDVOdw96p1g3B7pYOZRc2vb7PTJNFxHzgjhRtIQC29rQ6SRNpi5W9kiRJkiSppuoc9v6y6fX+EdGquvf8xvcA7hcR+88w32MoKoSn2j78tucVSpo4m7fZs1eSJEmSJNVTbcPezLwcuK5p6F4tLvvc1OUUPX7/PSIWlC+KiMOAd7K7qhfg3L4sVNJEmbFnr2GvJEmSJEkaotqGvQ1fb3r98PLJzPwB8POmoUcAP4qIv4yIh0bEIyLiLcCPgdtRVPUm8M3MvLI8nyTNpqpn76L581gwv+5/pUqSJEmSpHE2rQq2Zj4DPLnx+vER8dLMzNI1LwTOabwO4M7AO0rXTIW8ANuBl/Z9pZImQlXP3iULDXolSZIkSdJw1T2d+B+KqtwLgA3ACeULMvNbwPOnDhtf0fRF09h24NmZed7cLlvSuKpq4+DmbJIkSZIkadhqXdmbmVtp3au3fN17I+Ii4O3A3UqnpwLf7wJ/m5nf7esiJU2UTdtah73LFtX6r1NJkiRJkjQBxiadyMyzgXtExB2BPwBuSxH0Xgd8JzMvGeb6JI2HzRVh7xI3Z5MkSZIkSUM2NmHvlMz8DfCbYa9D0niq6tm71J69kiRJkiRpyEwnJKkD9uyVJEmSJEl1VduwNyJOjYh1TV8PG/aaJKmqZ+/ShWP3QQlJkiRJkjRi6pxOHAysbLzeAHx1eEuRpEJlGwcreyVJkiRJ0pDVtrIX2NL4nsAVmdk6YZGkAaraoM2evZIkSZIkadjqnE78tun11qGtQpKaVLVxWLaozh+UkCRJkiRJk6DOYe/Fje8BHDDMhUjSlKoN2pYstI2DJEmSJEkartqGvZn5I+CyxuF+EXHcMNcjSTBDz17DXkmSJEmSNGS1DXsbTm96/fJhLUKSplRV9i5dVPe/TiVJkiRJ0rirezrxNuBXFK0cnhQRLxzyeiRNuKqevUvt2StJkiRJkoas1mFvZm4B/gS4iCLwfWdEnBERhw13ZZIm1ZaqsNc2DpIkSZIkachqXYoWEU9rvPw34BXAbYFTgVMj4gfAeRR9fW8Btncyd2Z+qI9LlTQhKts4GPZKkiRJkqQhq3XYS9GzN5uOk6LCF+B44N49zG3YK6ljVWHvskWGvZIkSZIkabjqHvZOCXaHvtniXLumwuLyHJI0q127ki3bd7U8t8TKXkmSJEmSNGSjEPZG6Xs/5pKkjm3Z0bqqF2Cplb2SJEmSJGnI6h72PmPYC5CkKZsrNmcDe/ZKkiRJkqThq3XYm5kfHPYaJGnKphnCXnv2SpIkSZKkYZs37AVI0qjYUrE5G9izV5IkSZIkDZ9hryS1afMMYa89eyVJkiRJ0rAZ9kpSm+zZK0mSJEmS6sywV5LatKmisnfRgnnMnxcDXo0kSZIkSdKeDHslqU1bKip7reqVJEmSJEl1sGDYC5hJRBw0V3Nn5pVzNbek8VTVs9ewV5IkSZIk1UGtw17gciDnYN6k/j+7pJrZVFHZu8zN2SRJkiRJUg2MSuBpM0xJQ7elorJ3iZW9kiRJkiSpBkYl7O1GuSLYwFhSTzZX9ey1sleSJEmSJNVA3cPeD3ZxzzLgNsA9gL0bYwlcCnyrT+uSNIGqevbaxkGSJEmSJNVBrcPezHxGt/dGRAAPB14L3BM4FPhEZr6yP6uTNGmqevbaxkGSJEmSJNXBvGEvYK5k4X+A+1JUCAfwsoh4y3BXJmlUVfXsXWrYK0mSJEmSamBsw94pmbkDeBbwfYrA9yUR8dDhrkrSKKpq42DYK0mSJEmS6mDsw16AzNwJvLpp6B+GtRZJo6uqjYMbtEmSJEmSpDqYiLC34evArRTVvXeLiKOGvB5JI6ayjYNhryRJkiRJqoGJCXsb1b2XNw0dP6SlSBpRm6sqe23jIEmSJEmSamBiwt6GrU2vbz+0VUgaSVU9e5dZ2StJkiRJkmpg0sLeg5pe7xjaKiSNpKrK3iVW9kqSJEmSpBqYmLA3Ik4E9msaun5Ya5E0mqoqe23jIEmSJEmS6mAiwt6IWAO8G8im4fOGtBxJI6oy7LWNgyRJkiRJqoGxDnsjYmVE/AVwPnBUYziBX2bmr4e2MEkjaVPVBm2GvZIkSZIkqQYWDHsBM4mIr3d561LgthQ9eqPxlU3fX9aXBUqaGDt3Jdt27Gp5zjYOkiRJkiSpDmod9gInsWfrhU5E0+tsOn5NZn6xl0VJmjxbKlo4gGGvJEmSJEmqh7qHvb2YComnKnuvAf46Mz8zvCVJGlVV/XoBltnGQZIkSZIk1cAohL0x+yXTbAFuAa4Afgx8EfhiZlanNZI0g80V/XoBlljZK0mSJEmSaqDWYW9mjvUGcpJGx0yVvW7QJkmSJEmS6sAwVZLaMFNlrz17JUmSJElSHRj2SlIbNhn2SpIkSZKkmjPslaQ2bKlo47B4wTzmzeumtbgkSZIkSVJ/GfZKUhuqevbar1eSJEmSJNWFYa8ktaGqZ+8yWzhIkiRJkqSaqHXYGxF3iYhLm75O7HKek5rmuCQi7tTvtUoab5sqKnuXWNkrSZIkSZJqotZhL/A84JDG16bM/EY3k2TmOcDWprme04/FSZocWyoqe92cTZIkSZIk1UXdw95HN74ncEaPc32o8T2Ax/Q4l6QJU9Wzd5mVvZIkSZIkqSYWDHsBVSLiaOB2TUNn9TjlWcCbGq8PjohDM/OyHueUNGY2bt3Bl395HT+7+mZ27Nr1+/Hzr7qp5fVLrOyVJEmSJEk1UduwFzim6fWGzPxVj/P9CtgArGgcHwcY9kr6vZs2beOp7z+Pn19zc9v32MZBkiRJkiTVRZ3bOBzQ+J7AVb1OlpkJXNk0dFCvc0oaLx8776qOgl6ApbZxkCRJkiRJNVHnsHdF0+tb+jTnrU2v9+rTnJLGxLd+c0PH96xYXOcPSEiSJEmSpElS57C3OZhd1ac5Vza93tanOSWNiRtu3drxPfc5bM0crESSJEmSJKlzdQ57b2x8D+AOEbGwl8kiYhFwh6ahzkv4JI219Zs6+zeg4w9dzUOOvu0crUaSJEmSJKkzdf788SVNr5cCJwJf7WG+E4FlTcdX9DCXpDGza1eyftP2lueOP3Q1t1mx+PfHixfO414Hr+Yxdz/Anr2SJEmSJKk26hz2/hC4Gdi7cfxyegt7X9b0eiPw3R7mkjRmbtmynZ27suW5V/3p0dzlwJWDXZAkSZIkSVKHatvGITN3AV+kaOMQwEkR8aJu5oqIFwMPBLLx9eXMbF3CJ2kird1Y3cJh9fJFA1yJJEmSJElSd2ob9ja8CdhFEdAG8LaIeH1EtPW56YiYHxFvAP6xaY4E3jBH65U0otbNEPauWb648pwkSZIkSVJd1DrszcwLgP9gd0g7D3glcGFEvCgijmp1X0Qc1ajmvRB4Bbt/zgTen5k/m/PFSxopVWHvkoXz7MsrSZIkSZJGQp179k75K+BY4P7srs49HPgn4J8iYiNwI7ABWAHsCyxv3BuN71P3nQO8YFALlzQ6qsJeq3olSZIkSdKoqHVlL0Bm7gAeCXyO3RW+U+FtUAS8hwDHNb6vaDrXfO2ngUc35pOkPVSFvauWLxzwSiRJkiRJkrpT+7AXIDNvyczHAs8DrmLPit2qLxrXXQE8KzOfkJm3DnThkkZGVdi72speSZIkSZI0IkahjcPvZeZ7I+J9wGOBh1K0djiMPX+OHcDFwLeALwFnZeauQa9V0mipDHuXWdkrSZIkSZJGw0iFvQCZuRP4VOMLgIjYC9gLuNXqXUndsLJXkiRJkiSNupELe1tpBLyGvJK6VrlB24pFA16JJEmSJElSd0aiZ68kzbXKDdqWGfZKkiRJkqTRYNgrSczUxsGwV5IkSZIkjQbDXkkTb/O2nWzevrPlOcNeSZIkSZI0Kmod9kbE/SJiZ9PXA7uc50FNc+yIiHv2e62SRte6Ta2resGwV5IkSZIkjY5ah73Ac4BofP0gM8/uZpLM/Drwk8Y884C/6NsKJY28dRuqw941hr2SJEmSJGlE1DbsjYh5wMOBbHx9pMcpP9T4nsCjepxL0hipquydF7DP0oUDXo0kSZIkSVJ3ahv2AncGVlFU4wL8T4/zTd0fwG0j4sge55M0JtZt3NpyfNWyRcybFy3PSZIkSZIk1U2dw96jm17flJmX9jJZZl4C3NQ0dGwv80kaH+s2bm85vsoWDpIkSZIkaYTUOey9XeN7Atf0ac6rm14f0Kc5JY24qspeN2eTJEmSJEmjpM5h77Km1xv7NGfzPCv6NKekEVdV2bt6mWGvJEmSJEkaHXUOe29uer2mT3Oubnq9qU9zShpxlZW9Kwx7JUmSJEnS6Khz2HtD43sAd4iIpb1MFhHLgIMp2kI0zy9pwq3buK3l+BrbOEiSJEmSpBFS57D3wqbXi4CH9jjfwxrzROP4kh7nkzQmqsLeVbZxkCRJkiRJI6S2YW9m/gy4nqISN4C/73HKV7G7qvcm4Lwe55M0Jiore23jIEmSJEmSRkhtw96Gz7G7EvfuEfHP3UzSuO/ujcMEPpeZOcMtkibEzl3JTZtbb9BmZa8kSZIkSRolC4a9gFm8CTgNWEgR+v51ROwLvDAzb5nt5ojYG/g34MnsrhDeBrxxrhbcq4g4HDgeOJCi7cR6ipYW38nMLcNcG0BEzAfuCRwD7EfxZ7MBuBr4FXBhZu4a3gqlzty0aRtV//Sz2p69kiRJkiRphNQ67M3MqyLiH4DXsDusfQrwqIj4EPBF4IeZeePUPY0w+F7Aw4GnAnuzuzo4gbdm5mWD+ynaExEnU7SquEfFJRsi4nTgdc0/76BExKHAS4BTgZUzXHpLRJwN/EdmfnGWOU8Czu5hWVdk5iE93C9VtnAAw15JkiRJkjRaah32AmTm6yLiOOBx7A589wZe0PgiIhLYBCxjd7ALe4a8AXwyM18zoKW3JSIWA++nCLFnsgJ4IfDEiDglM78554sDImIe8FKKwH1xG7fsDTwa2EIRxku1ZtgrSZIkSZLGRd179k45FXgnRWCb7N5oLRpf8yjC0HlNYzRdB/B24M8Gsdh2NYLUTzA96N0JXAacD9xcOncb4H8j4r4DWN9C4OPAm5ke9N5M0V7iPIr2DZvmej3SXKgKe5cvms+ShfMHvBpJkiRJkqTu1b6yFyAzdwAviogvUVSY/kHz6Ra3NAe+36ZoffC1uV1lV15CUQXb7D3AGzLzWvh9IPxo4B3AQY1rlgGfjIjjMrMcBvfT+4HHNx3vAN4LfJCifcbv3/vGOu8EPAx4Aq3/XGbzYeBDHVy/uYtnSHtYt6l12LvKql5JkiRJkjRiRiLsnZKZXwa+HBH3Bh4K3B84HFgN7AXcCqwDfgN8C/hSZv5kSMudUUSsAV5ZGn55Zr6leaCx2dlnI+I8iuD6kMapA4EXU4Tfc7G+P6PoeTzlWuBPMvNnra5vrPPCxtc7I2JVF4+9NDO/2sV9UtfWbWgd9q4x7JUkSZIkSSNmpMLeKZn5A+AH3d4fEfMzc2cfl9SNv6MIqKd8E3hr1cWZeU1EPAtoDkNfFBH/kplr+7mwxiZ3/69p6GbgxMy8uN05MnN9P9ckzRUreyVJkiRJ0rgYlZ69fRERd42IfwauHvI65gHPKA2/trktQiuNVhTfahrai6JlQr+9Eti36fgVnQS90iip6tnr5mySJEmSJGnUjH3YGxH7RsTfRMRPgB8Dfw3sN+RlnUCx0dqUS4Fz2rz3/aXjk/uwnt+LiMXA05qGrqPo0yuNpaqw1zYOkiRJkiRp1IxkG4fZRMQC4JHAacAfU/yc0XRJN5uH9dOflo7/b7aq3uZrS8cnRcTyzNzYh3UBPIaiB/KUj9eg5YU0Z6rCXts4SJIkSZKkUTNWYW9E3BN4OnAquwPLqZA3S8fDdLfS8XfavTEzr42Iy9m9Udsi4Bh66GFcUg6iz+7TvFItWdkrSZIkSZLGxciHvRFxW+CpFCHvMVPDje/Z+IrG1ybgC8DHBrzMsqNLxxd0eP8F7A57p+brV9h779LxT6HY1A54KMX7fHfgQGA7cAPwE+B/gU9k5qZeHh4RARxK0WpjPrAOuM4N3zQXMrO6sneZYa8kSZIkSRotIxn2RsQi4NEUbRoeQhEKlts0TAW824AvAx8Hzuo1jOxVRCwFDioNX9XhNOXrj+x+RbtFxD7AnZqGdmbmFRFxGHAGcN8Wt+0DHAE8HnhjRLwsMz/c5RKeDvwle7aRmFrbhcDXgXdn5i+7nF/aw6ZtO9m6Y1fLc2tWGPZKkiRJkqTRMlJhb0TchyIQfCKwcmq48b3cpuFrFBW8n8nMmwa0xHbsy57B9Hbg+g7nuKZ03K8N5w5jz7XdGhHHULSZ2KeN+/cHPhQRx2bmy7p4/iEznDuq8fW8iDgTeE5mruviGdLvVVX1gpW9kiRJkiRp9NQ+7I2I/YGnUYS8U1WnVW0afr/JWWY+ZIDL7MSK0vGmDjZnm1LejK08Z7dWlo6Tou3FVNC7Cfgo8E1gLbAGOBF4MrC06b6XRsQ1mfmvfVpXswBOAY6PiD/JzE5bYMw8ecR+wG06vO3wfq5BgzNT2Ltm+eIBrkSSJEmSJKl3tQx7I2IJ8BiKNg0PAuZRHfBuAD4LXAq8ZtBr7UI5mN3SxRybZ5mzWytLx6saXwA/Ah6bmVeWrvlwRLwROAu4S9P42yLiy5l5URvPvYgiVP4G8EuKSufNjWffiaJVx3OA2zXdcxDwxYi4T2b+ro1ntOv5jMbvkfqgKuydPy/Ye2kt/3qUJEmSJEmqVKs0IyL+kKKC9/HA3lPDje/NAe8O4CvAR4DPZebmiHjwgJfbrSWl4+rSwmpbS8dLW17VuarQ+GrgIVWbpGXm5Y33/+fsDmQXA38LPHuG510OPDAzz6k4f0Pj69yIeAvwDorQd8rBwLuBx83wDKnSTJuzFXsFSpIkSZIkjY55w15ARNwhIl4VERdRtAd4JkXbgFa9eM8D/grYPzMfkZkfy8xylWvdlSt5u2kMWv58eTfVwa1UzfOSqqB3SmbeCJT79D61sSFd1T2XzxD0lq/dkpnPpQh8mz02Iu7VzhxSWVXYu2a5/XolSZIkSdLoGWplb0R8jaLn61TFLkxv03AJRQXvGZl58TDW2WcbSsflSt92lAPU8pzdajXPOuDMNu//BPBOdvf4XQIcT9GeoV9eAjwCOKJp7M+AH/Zp/ncDn+rwnsMp2lhoxKzbVFHZu3zhgFciSZIkSZLUu2G3cXhg0+vmgPdGiuDwjMz8/jAWNofKgeqyiIgON2lbPsuc3Wo1z3czc3s7N2fmlog4j6LH7pR70cewNzN3RMS/AP/SNPzQPs5/PUXP4Lb5cf/RtW5DVWWvm7NJkiRJkqTRM/Q2Duxu0wDwbYqqzdtn5l+OYdALRZDd/DMvBPbrcI4DSscdhZMzaLXRWTsbrDX7dem405+tHV8rHd8xTFzVhbUVbRxW28ZBkiRJkiSNoDqEvbC7qvcE4KXAMyJin5lvGU2NHsNXloYP6nCa8vUXdr+iPVzC9A3jbulwjvL1q7pfTqWrSscL5ug5GnPrK9s4GPZKkiRJkqTRU4ewt7lX7zzgfsB7gesi4syIeExEjFsDzXI4e0yH9x89y3xdycydTK/k7fTz7OUexJu6X1GlVm0lxu13RAPgBm2SJEmSJGmcDDvs/RPgk8BWdoe+NF4vBk4GPg38LiLeGxEPGPgK58b5peMT2r0xIm4PHNI0tB24oPcl/d6PS8e37fD+ctuGtT2spcrtSsc5R8/RmKsKe63slSRJkiRJo2ioYW9mfjkzn0QR3j0P+B57VvrSOF4JPAs4OyIuj4g3RUSn1bB18oXS8R910HO2vBnZ2ZnZrw3aAD5fOr5nh/eXry/38O2H+5WOf5uZO+bgORpj23fu4ubNrfcetLJXkiRJkiSNomFX9gKQmbdk5nsz8wTgSOAtwDW0Dn4PAl4G/DwifhIRL25Uu46S71Bs1DblMOCkNu99Zun4rH4sqMmXgC1Nx3eJiDu2c2NEHMv0FhPn9GldzcrvQXnDNmlWN21qHfQCrFpm2CtJkiRJkkZPLcLeZpn5m8x8BXAwRRXrxyjCx+bgNxpfdwXeRrHh2b8MfrXdycxdwOml4dfMVt0bEQ8G7t80dCtFG4x+rm0jcEZp+FVt3v7q0vE3MvP63le1W0T8GdOD8c/18xmaDFUtHADWrDDslSRJkiRJo6d2Ye+ULHw1M59C0ebh2cC32TP0nQp+5wNHsbsCmIgof9S/bt4KNLdfOBF4adXFEXEA8L7S8Dsz88ZW1zfdl6Wvk9pY2+vYs7r3aRHx57M85/nAE0rD/zDD9U+KiMd20L6CiDiV6e/B+cBn251DmrJ249bKc1b2SpIkSZKkUVTbsLdZZt6ame/LzAcARwBvoqjmLbd5mHodwDci4qqIeFtE3GuwK55dI6R9c2n4HyLi3RGx/9RARMyLiJMpWj8c0nTttcDb52htV1OE0c3eFxHviog7NA9GxEER8e/Au0rXfywzvzzDY44CzgQuavRgvm9ELC9fFBGLIuJBEXEW8FGKjfumbAGel5lZvk+azfqNrds47LV4AYsWjMRfjZIkSZIkSXsYuUQjMy/NzL/PzEOBB1O0HNjE7tYOsDvwPQB4MfD9iLgoIl5fs43d3sr0zdqeB1wZEZdExI+BtRSVqwc1XbMZeEJm3jSHa3tDaW0BvAC4orG28yLiEuAK4Lnsfu8BfkxRid2OI4BXUITZt0TEFRHx04j4XkT8CriZoifvo0r3bQeekpnf6/QHkwDWVVT2rraFgyRJkiRJGlEjF/Y2y8yzM/NpFG0engl8o3Eq2LPNQ1CEiq+k2Njtp0NY7jSN3r2PBz5eOjWfYtO2uwMrS+fWAg/PzHPneG07gVOAD5ZORWNt9258L/s8cGJmbmhxbjbzKELtuwD3oaj+XdLiuouA+2bmZ7p4hgTAuorKXls4SJIkSZKkUTXSYe+UzNyYmR/IzAcChwOvBy6ndX/fAI4bxjpbycwtmXkqRbB6/gyXbgTeDRyTmecMYGlk5tbMPA34E2CmcDmB7wOPzMxHtxn0fpKip+93KSqVZ7MD+BbwZOC4zPxRG/dIlaoqe9csN+yVJEmSJEmjacGwF9BvmXk58FrgtRHxAOA04HHAXsNb1ewy80zgzIg4gqKq9QBgEXAT8Cvg3MzcUj1D5bxtb4A2wxxfAr7U2CTuvsDBFBW364HfNtZ2fYdzXkDRvoGImA8cSVEpfCCwN8XPvqHxjMuAH2RmO6Gw1Ja1G7e1HF9l2CtJkiRJkkbU2IW9zTLzm8A3I+IFFO0SngY8cLirmllmXgxcPOx1tJKZ1wCfnoN5dwIXNL6kgVi/qXXYa2WvJEmSJEkaVWMd9k5pVIR+CPhQRBwEPHXIS5I0ZGs3tA57Vxv2SpIkSZKkETURYW+zzLwSeNOw1yFpuKoqe23jIEmSJEmSRtVYbNAmSZ3ITNZV9Oy1jYMkSZIkSRpVhr2SJs6GrTvYvjNbnrOyV5IkSZIkjSrDXkkTp6qqF6zslSRJkiRJo8uwV9LEWTtD2OsGbZIkSZIkaVQZ9kqaOOsrwt6F84MViydu30pJkiRJkjQmDHslTZyqyt7VyxcREQNejSRJkiRJUn8Y9kqaOFWVvauW2cJBkiRJkiSNLsNeSROnaoO2NSsMeyVJkiRJ0ugy7JU0caraOFjZK0mSJEmSRplhr6SJU9XGYc1yw15JkiRJkjS6DHslTZzqDdoWD3glkiRJkiRJ/WPYK2nirN9UFfYuHPBKJEmSJEmS+sewV9LEWbfByl5JkiRJkjR+DHslTZRtO3Zx69YdLc+tsrJXkiRJkiSNMMNeSROlqoUDwBoreyVJkiRJ0ggz7JU0UdZWtHAAK3slSZIkSdJoM+yVNFFmquxdtWzRAFciSZIkSZLUX4a9kibK2o2tw959li5k4Xz/SpQkSZIkSaPLZEPSRFlfEfauXm5VryRJkiRJGm2GvZImSlVlr2GvJEmSJEkadQuGvQBJ6tauXckFv72Fy27c2PY95191U8tx+/VKkiRJkqRRZ9graSTduGErp33gPH5xzS19mW+Nlb2SJEmSJGnE2cZB0kh67ed/2begF2D1CsNeSZIkSZI02gx7JY2cHTt38bVfXd/XOVfbxkGSJEmSJI04w15JI2f9pu1s3r6zr3Med8A+fZ1PkiRJkiRp0Ax7JY2c9Zu29XW+Y/ffm+MPXd3XOSVJkiRJkgbNDdokjZy1G6rD3n076L2795KF3PfwNfzdHx/F/HnRj6VJkiRJkiQNjWGvpJFTVdm71+IF/PBVDxnwaiRJkiRJkurBNg6SRs7aja3D3tUdVPVKkiRJkiSNG8NeSSNnXUUbh1XLDHslSZIkSdLkMuyVNHKq2jisWW7YK0mSJEmSJpdhr6SRU9nGwbBXkiRJkiRNMMNeSSNnvWGvJEmSJEnSNIa9kkaOlb2SJEmSJEnTGfZKGjlVlb2rDHslSZIkSdIEM+yVNFIyk3UVYa8btEmSJEmSpElm2CtppGzYuoNtO3e1PGcbB0mSJEmSNMkMeyWNlPUbt1eeM+yVJEmSJEmTzLBX0khZu3Fr5TnDXkmSJEmSNMkMeyWNlPWbWvfrXTg/WLF4wYBXI0mSJEmSVB+GvZJGytoNrcPe1csXEREDXo0kSZIkSVJ9GPZKGilVlb2rltnCQZIkSZIkTTbDXkkjZe3G1mHvmhWGvZIkSZIkabIZ9koaKesq2zgsHvBKJEmSJEmS6sWwV9JIqWrjsHrZwgGvRJIkSZIkqV4MeyWNlKo2Dlb2SpIkSZKkSWfYK2mkrK8Me63slSRJkiRJk82wV9JIsbJXkiRJkiSpNcNeSSNj245d3LplR8tzq6zslSRJkiRJE86wV9LIuKliczaANVb2SpIkSZKkCWfYK2lkVLVwAFi9fNEAVyJJkiRJklQ/hr2SRkbV5mwAK5fZxkGSJEmSJE02w15JI6OqsnefpQtZON+/ziRJkiRJ0mQzHZE0MtZX9Oy1hYMkSZIkSZJhr6QRsnaDYa8kSZIkSVIVw15JI2NdRRuHVcsMeyVJkiRJkgx7JY2MdRVtHNZY2StJkiRJkmTYK2l0rKtq47DCsFeSJEmSJMmwV9LIqNygzTYOkiRJkiRJhr2SRsfaip69btAmSZIkSZJk2CtpRGQm6w17JUmSJEmSKhn2ShoJt2zZwY5d2fKcYa8kSZIkSZJhr6QRsa6iqhcMeyVJkiRJksCwV9KIMOyVJEmSJEmamWGvpJFQFfYuXjCPZYvmD3g1kiRJkiRJ9WPYK2kkzLQ5W0QMeDWSJEmSJEn1Y9graSSsnSHslSRJkiRJkmGvpBGxbuPWluOGvZIkSZIkSQXDXkkjYd3G7S3HDXslSZIkSZIKhr2SRkJVZe+qZYa9kiRJkiRJYNgraUSs29S6sneNlb2SJEmSJEmAYa+kEVHZs3eFYa8kSZIkSRIY9koaEeurevbaxkGSJEmSJAkw7JU0Arbu2MmGrTtannODNkmSJEmSpIJhr6TaW7dxW+U5w15JkiRJkqSCYa+k2jPslSRJkiRJmp1hr6Taqwp7I2ClPXslSZIkSZIAw15JI6Aq7F25dCHz58WAVyNJkiRJklRPhr2Saq8q7LWFgyRJkiRJ0m6GvZJqz7BXkiRJkiRpdoa9kmrPsFeSJEmSpP+/vTuPj6uu9z/+/iRN2ibdUyptobQUqYCytYBSloIsYhEQ9w1ZrnqVK+IPARFFEZcLiNcFcQNBQFwQ72VRQEDK0opsghulQGkLLQhNS/c2afL5/XEm6ZmTOTPnTCYzycnr+XjMg54z3y2TLyczn/mezxcojWAvgH6PYC8AAAAAAEBpBHsB9HsEewEAAAAAAEoj2Aug34sL9o5tItgLAAAAAADQhWAvgH5v9cbCwd6WEQR7AQAAAAAAuhDsBdCvdXa6Vm9sL/jcuOahVR4NAAAAAABA/0WwF0C/tmZTuzo6veBz40jjAAAAAAAA0I1gL4B+bVVMCgdJGkcaBwAAAAAAgG4EewH0a3Gbs0ms7AUAAAAAAAgj2AugX4sL9g5vqNfwxvoqjwYAAAAAAKD/ItgLoF+LC/aOa2ZVLwAAAAAAQBjBXgD9GsFeAAAAAACAZAj2AujXCPYCAAAAAAAkQ7AXQL9GsBcAAAAAACAZgr0A+jWCvQAAAAAAAMkQ7AXQrxHsBQAAAAAASGZIrQeAfGY2XdL+knaQ1ChptaSFkha4++Zajk2SzKxe0kxJu0uaIKlB0npJL0p6StJCd+/sZR8tkmZLmi6pWdIGSc9Jmu/urb1pGwMPwV4AAAAAAIBkCPb2E2Z2gqQvSdo3psh6M7tG0oXuvrJa4+piZtMknS3pA5LGFCm61szulfQTd/9Dyj72kvRVSceq8KrzDjP7vaQvufvf0rSNgSsu2Du2iWAvAAAAAABAGGkcaszMhprZ9ZL+V/GBXkkaIem/JP3LzA6pyuAkmVmdmZ2nYNXuJ1U80CtJoyQdL+mklP18RtKjko5T/Lyszz3/mJl9Ok37GJg2tXVoU3tHwedaRhDsBQAAAAAACGNlbw2ZWZ2kXysIjoZ1SFomaY2kaZJGh57bTtLtZnaEu/+5j8fXIOkXkt5T4Ok1kl6StFbSSEk7SWoqs5//J+myAk+9JGmFpEmSJobOD5H0PTMzd/9eOX1iYFi1sfCqXok0DgAAAAAAAFGs7K2ts9Uz0PsjSVPcfWd330fSOEknKgj+dmmS9BszG62+dZXyA71bJf1AQU7hse6+m7sf4O67Kwj47ibpTEkLJHmSDszsQEmXRE7PkzTT3Se5+yx3nyRpP0n3RcpdZmb7p/uRMJCsWl8k2EsaBwAAAAAAgDwEe2sktwnZ+ZHT57n7J919RdcJd+909/+VdKCkJaGyO0j6f304vg9L+kjo1AoFAdj/cvdH3D0vmJsb50J3/667z5b0qYRdXaogPUOXWyUd7e6PR9p/VNJRkn4fOj0kVx8ZFbeyt86k0cMbqjwaAAAAAACA/o1gb+2co2A1bJf7JV0cV9jdl0v6j8jpz+aCxhVlZuMl/U/o1BpJh6bZFM3dVyfo5xgFQewurZJOc/eCEb7c+VNz5bocYmZHJh0XBpZVG7YUPD+2qVF1dVbl0QAAAAAAAPRvBHtrIJer95TI6a9EV8tGufs9kh4InRop6b0VHp4UrDgeHzr+grs/2wf9RIPXP3D3V4tVcPdXJF1Roh1kxKoN7QXPk68XAAAAAACgJ4K9tXGggo3WuixWkKc2iasixydUYDzdzGyopJNCp16W9ONK9hHq5+jI6Z8lrB4td4yZEf3LoLiVvQR7AQAAAAAAeiLYWxtzI8d3lVrVGy4bOZ5jZs0VGFOXdyrYFK7Lr9y9o4Ltd5kjKTzup919aZKK7r5E0jOhUyMlHVqxkaHfWLWhcM5egr0AAAAAAAA9Eeytjb0jxwuSVsxt3rYkdKpR0u69H1K3aCD63gq2HbZ35Djxa5Azv0R7yACCvQAAAAAAAMkR7K2N3SLH/0pZP1o+2l5v7Bc5flKSzKzezI4xs1+Z2dNmtsHMXjOzZ8zsN2Z2ipk1peinP78G6CcI9gIAAAAAACRHsLfKzGy4pCmR0y+kbCZafkb5I9rGzEZL2jV0qsPdl5rZzgo2hvuDpPflyjRJGi1pF0nvUZBH9xkz+0jC7qJj7hevAXpvc3uHkmclKY5gLwAAAAAAQHJDaj2AQWi8JAsdt0t6JWUbyyPHE3o1om12Vv7Y1pnZ7gpSLIxOUH+SpGvNbA93/3yJstExv5h8mJL67jVAmZa1btRZNz6hx5e9ppbmRn38kJ31Hwfv3Ks2CfYCAAAAAAAkR7C3+kZEjjem2Jyty4YSbZZrTOTYJd2mbYHejZJukHS/pFZJLQo2RvugpOGheuea2XJ3/36RvqJjjv5MpfTJa2BmEyRtl7La9Er0PZC1d3TqfT/5s15as1mS9Mq6Lfra75/S6OENes+sHctqs6PT9dqm9oLPEewFAAAAAADoiWBv9UWDkpvLaGNTiTbLNSZyPDb3kKTHJJ3o7ssiZa4zs69JulnSnqHzl5rZne6+KKav3r4OffUafErSlyvU1qCx4LnW7kBv2G8fe7HsYO9rG9sU9zUIwV4AAAAAAICeyNlbfcMix4XvUy9uS+R4eMFS6cUFTF+UdGSBQK8kyd2XSHqrpJdDp4dK+lyRvnr7OvTVa4AyPPPvdQXPP/fq+rLbjEvhIBHsBQAAAAAAKIRgb/VFlz+WE7UaWqLNcsW1c7a7ry5W0d1XSorm6f1IbkO6JH2lfR366jVAGVZvLByYXb2xXZ2d5W3WVizYO7aJYC8AAAAAAEAUaRyqL7rUMbrCNYloALX85ZOl21kl6aaE9X8t6bvaluN3mKT9Jd0X01dT6Djt69BXr8EVkm5MWWe6gjQWg1ZcYLaj07V2c7vGlBGcjWuzubFewxrqU7cHAAAAAACQdQR7qy8alGwyM0u5SVtziTbLVaidP7t74V2yItx9s5k9LOnI0OlZig/2TggdR3+mUvrkNXD3VyS9kqaOmVWi6wGt2CrcVRvaygv2xqwWHjeCVb0AAAAAAACFkMah+lZKCgd2G5Qf9ExicuQ4VXCyiH8XOBe3wVqcpyPHcT9bdMw7pOynr14DlKFUsLesNtfHBHtJ4QAAAAAAAFAQwd4qc/dNkqIbnU1J2Uy0/MLyR5TnOfXcKG1tyjai5cfGlIsGhfvLa4Ay9EWwtzWmHpuzAQAAAAAAFEawtzaigcndU9bfrUR7ZXH3DvVcyRvdCK2UaO7djTHl+uVrgPL0RbA3btO3cc1ppyQAAAAAAMDgQLC3Np6IHB+YtKKZTZQ0NXSqXdK/ej+kbo9Hjl+Xsn40bUNrTLknIseJX4Oc2SXaQ5V0dLpe2xSf1jluhW4pcUHicc0NZbUHAAAAAACQdQR7a+O2yPERlnyXr6Mix/e6e6U2aJOkWyLHM1PWj5aPpmvoMk/ShtDxrma2U5IOzGyqpNeHTq3LtYcaeG1jm4ptL7i64sFeVvYCAAAAAAAUQrC3NhYo2Kity86S5iSse1rk+OZKDCjkDkmbQ8d7mtnr4wqHmdke6pleYV6hsu6+WdIfI6dPTTjGaLk73L28iCJ6rVSahrI3aGNlLwAAAAAAQCoEe2vA3TslXRM5/eVSq3vN7K2SDg6dWifpNxUe2wZJ10dOfzFh9Qsix/e5+ytFyl8VOT7dzLYr1oGZTZD0qRLtoIpKBntjcu8W4+6s7AUAAAAAAEiJYG/tXCwpnH7hUEnnxhU2s8mSroyc/q67ryxUPlTPI485CcZ2ofJX955kZkVX3ZrZpyS9N3L6m8XquPvvJT0UOtUi6SozK7h008waFQR2W0KnH3D3O4v1g77VFyt7N7Z1aMvWzoLPsbIXAAAAAACgMIK9NZIL0n4jcvqbZnaFmU3qOmFmdWZ2goLUD1NDZVdIuqyPxvaigmB02JVmdrmZ7Rg+aWZTzOyHki6PlP9lwiDs2ZLCUb13SPqjme0b6WemgrQPx4ZOd0g6J0Ef6EOlNmArJ9hbrA4rewEAAAAAAAobUusBDHIXSzpQ+QHMT0r6uJktlbRG0jRJYyL1Nkl6r7u/1odju0jBZmtdYzNJp0v6lJk9L6lVwQrbnQvUfVzSx5N04u4Pmtl5yg8uz5H0mJmtkPSSpEmSJhaofo67P1TgPKqo1AZslQ/2NqZuDwAAAAAAYDBgZW8N5XL3vkfSryJP1SsIou6jnoHeVklvd/f5fTy2DknvlvTzyFOWG9t+KhzovUXSoe6+vsBzcX1dIulzClbqhk1SEHCOBno7JH3W3b+dtA/0nVIreze2dWhze/RXW1xcsHdInWnUML6jAgAAAAAAKIRgb425+2Z3/4CCwOoTRYpukHSFpN3dfV4VhiZ33+LuJ0s6RlKx4LJL+oukd7j78WkCvaG+LpM0S9LvlZ/WIaxT0m2SZrr7d9L2gb6xOsEGbGlX98aVH9vcqBL7GAIAAAAAAAxaLJHrJ9z9Jkk3mdkukg6QNFlSo6TXJD0lab67b45vIbbdXkfG3P0OSXfkNol7i6SdJA2TtFpBmoX57v5KBfp5QtKxZjZe0kEKVg43Kwh0P5frp+iGdKi+JIHcVRvaNGnM8F63Oa6JFA4AAAAAAABxCPb2M+7+rKRnaz2OQtx9uaTfVqGflZL+r6/7QWUkDfamajNmtTD5egEAAAAAAOKRxgFAr/RJsHc9wV4AAAAAAIC0CPYCKJu7l9ygTUof7I1rk2AvAAAAAABAPNI4ACjbxrYOtW2N209vm7TB3rhN3wj2AgAAAAOfu6uzs1PuXuuhABgkzEx1dXWDYtN3gr0AypY0iBuXgzdtuwR7AQAAgIHH3bV582atW7dO69atU1tbus8HAFApjY2NGjlypEaOHKlhw4ZlMvhLsBdA2RIHe2Ny8KZtl2AvAAAAMLBs3LhRK1asUHt7e62HAgBqa2tTa2urWltb1dDQoEmTJqmpqanWw6oocvYCKFviYG+KNA7tHZ1as6nwG0GCvQAAAMDAsXHjRi1btoxAL4B+qb29XcuWLdPGjRtrPZSKItgLoGx9kcYhLl+vRLAXAAAAGCi6Ar3k5QXQn7l75gK+pHEAULa+WNm7ekP8t/4EewEAAID+z921YsWKHoHehoYGjRo1SiNGjFBDQ0Mmc2UC6J/cXe3t7Vq/fr3Wrl2bd8dB1zVr+vTpmbguEewFULbWhEHc1za2qaPTVV9X+qLZumFL7HNjmwj2AgAAAP3d5s2be6RuGDlypCZPnpyJQAqAgamhoUFNTU3abrvttHz5cq1bt677ufb2dm3ZskXDhg2r4QgrgzQOAMq2OmGwt9MVm4e3Z5uFy40cNkSNQ7hkAQAAAP1dOIAiBQEWAr0A+gsz0+TJk9XQ0JB3fu3atTUaUWUROQFQtqQre6U0KR8Kr+wlhQMAAAAwMESDvaNGjSLQC6BfMTONGjUq71z02jVQEewFULZim6lFJQ/2Fl7ZS7AXAAAA6P/cXW1t+e/9R4wYUaPRAEC86LWpra0tE5tKEuwFULY0G6/FrdhNWm4c+XoBAACAfq+zs7PHueit0gDQHwwZ0nMrs0LXsIGGYC+AsrWuTxbAleJX7PZoMyaAzMpeAAAAoP8rtCqOFA4A+qO6up5hUVb2Ahi02js6tXbz1sTlk67sjUsNQbAXAAAAAACgOIK9AMqSJl+vlGJl73qCvQAAAAAAAOUg2AugLKsTBm+7sLIXAAAAAACgbxHsBVCW1oTB2y6rNpYODrt77KZvBHsBAAAAAACKI9gLoCx9sbJ3/Zatau8onAydYC8AAAAAAEBxBHsBlCVpWobu8jG5ePPbjC9DsBcAAAAAAKA4gr0AytJaJDBbyKoEG7oVa5NgLwAAAAAAQHEEewGUZXVMYHb08IaC5ze3d2pj29ay2myoN40YOiTdAAEAAAAAAAYZgr0AyhK3Cvf1E0bE1imWpqFYm+OaG2VmyQcHAAAAAOjXlixZIjPrfpx88sm1HhKQCSyVA1CW1TFpGXaZMEKPLl1d8LlVG9q0w9im+DZjg71D0w8QAAAAAAD0e62trZo/f76ee+45bdiwQc3NzZo+fbpmz56tlpaWWg9vwCHYC6AsrTEbru3U0qz6OlNHp/d4rtTK3rjnxzUXTg0BAAAAAADyrVixQpMnT+4+fvLJJ7Xnnnv2KDdlyhS98MILkqRf/OIX+uAHP1i1MXaN64ILLtBtt92mzs7OHs/X19dr7ty5uuiiiwqOH4WRxgFAWeICs+NHNGpsU+HgbPlpHFjZCwAAAAAYuL7yla/kpa2YN29en/U1f/787n+PGTNGb3zjG3uUWbp0aXegV5IOOuigPhtPId/97nc1a9Ys3XLLLQUDvZLU0dGhW265RTNnztT3v//9qo5vICPYCyA1d49N49AyolHjmhsLPlcq2BubxiEmeAwAAAAAAPKFg71vectbVFfXM/z3wAMPdP97hx120JQpU6oyNkn69re/rTPPPFNbt+Zv4j5x4kTNnDlTEydOzDu/detWnXHGGfre975XtTEOZAR7AaS2bstWtXf0TNMgSWObGjW2qbxgLyt7AQAAAGBwmDp1qty9+3HNNdfUekiZEQ72xq3YDQd7Z8+e3edj6rJgwQKdc845eefmzJmjxx57TCtWrNCjjz6qFStW6JFHHtGhhx6aV+6ss87Sww8/XLWxDlQEewGkFrcCV5JamoeqZUSZK3tjVguTsxcAAAAAgNI2btyoJ554ovu4vwV7zz77bHV0dHQfv+Md79Cdd96pfffdN6/crFmz9Mc//lFz587tPrd161adffbZVRvrQEWwF0BqcStwJWlsc0PZK3tXxWz6xspeAAAAAABK+8tf/tKdHqGxsVH7779/jzIrV67UwoULu4+rFey9/fbbtWDBgu7jlpYWXXXVVWpsLBxDaGxs1M9+9jO1tLR0n7v//vt111139flYBzKCvQBSi1vZ21hfpxFDh6iljJy9bVs7tW7L1oLPxeUABgAAAAAA24RTOMycOVPDhg3rUebBBx+Ue5CaccSIEdprr72qMrYrr7wy7/j000/XdtttV7TOhAkT9KlPfapoO8g3pNYDADDwxK3sHdvcIDPT2Lhgb0yaBik+hYNEsBcAAAAAUNyaNWt0//3369lnn9WmTZvU0tKiXXbZRQcddJCGDi3/btFVq1bp8ccf17PPPqs1a9Zo69atampq0vjx4zVt2jTtscceGjt2bAV/kt5Jm6/3gAMOUH19fZ+Pa8uWLbrzzjvzzp166qmJ6p566qm66KKLuo9vv/12tbW1xa4IHuwI9gJILW6Fble6hbjgbLGVva0xKRyKtQcAAAAAGJiWLFmiadOmdR9/9KMfLbpJ25w5c3Tfffd1H3etTH3ppZd0/vnn6xe/+IXa2np+rmxubtaZZ56p8847T83NzYnHd8899+jiiy/WPffco87OzthyZqYZM2bo+OOP1+mnn64dd9yx+7l58+bpsMMOK1gv7nyXrp8vjc7OTv35z3/uPu5P+XrnzZunDRs2dB/PmDFDO+20U6K6U6dO1etf/3o988wzkqR169bpvvvu05FHHtknYx3oSOMAILW4NA5d6RtaYnLsvraxXVs7Cv+RLLayd0wTG7QBAAAAAPI98MADeuMb36irr766YKBXkjZs2KCvf/3rOuyww7Rq1aqSbbq7Pv3pT+uII47QXXfdVTTQ21V+4cKFuvjii3usXO0L11xzjcys4KO+vl5r1qzpLnv88ccXLPfII490l/nqV79asMycOXMqOu7wpnGSdOCBB6aqHw1KR9vDNqzsBZBafBqHxtx/44Ozr21q1/gRPYPBcW2OHt6ghnq+lwIAAAAAbPPYY4/pmGOO6V4tWldXp2nTpqmlpUWrV6/Ws88+m7c69pFHHtHJJ5+sW265pWi7F1xwgS6//PIe58eNG6cdd9xRw4cP14YNG7Ry5Uq99NJLlf2hMuypp57KO959991T1Y+Wj7aHbQj2Akit3JW9UpDKoVCwN65NUjgAAAAAAKLe+973asOGDRo3bpy++MUv6qSTTlJLS0v38ytWrND555+flxri1ltv1e23365jjjmmYJvLly/XxRdfnHfuE5/4hD7zmc9ot91261F+9erVWrBggW677Tb98pe/7PH8XnvtpbvuukuSdO211+q6667rfu5b3/pW1TZG6w+efvrpvONwuoskouWj7WEbgr0AUotd2dtUemVvXN7euDYJ9gIAAACDw9aOTr20ZnOthzEoTBw9TEMG+B2Uixcv1tSpU3X33Xdr+vTpPZ6fNGmSrr76ajU0NOinP/1p9/mf/OQnscHem2++We3t7d3HF1xwgS688MLYMYwdO1Zz587V3Llzdemll2r16tU9nj/iiCMkSQ8++GDeczNnziwrVcLRRx/dHUCOeuc736n169dLks4///yC7X/2s5/VP/7xD0nSRz7yEZ100kkF26r0pnOvvPJK3vEOO+yQqv7kyZOLtodtCPYCSC0uv+64EUFgduiQeo0YOkTrt2ztUSYu2MvKXgAAAGBwe2nNZh18yb21Hsag8MA5h2nHcU21Hkav1NfX68YbbywY6A27+OKLdd1112nz5uCLhD/+8Y/q7OxUXV3PYPeiRYvyjj/1qU8lHs+IESM0YsSIxOXLNXHiRE2cOLHH+YULF3YHeiXpYx/7WI8N0Nrb27V48eLu4w984APdwei+Fh6bpFSb5RUqH20P2wzsr3EA1MSq9TGB2aZtgdm41b1xwd648+E2AQAAAACQpHe9612aNWtWyXJjx47VUUcd1X28ceNGLVy4sGDZTZs25R03NAyczcLvvXfbFyXTpk3rEeiVpEcffVQbN26UFATLo5ue9aVocHbYsGGp6g8fPrxoe9iGYC+AVLZs7dC6Ait2pfxVuONi8vbGp3HYUrjNEQR7AQAAAAD53ve+9yUuu/fee+cdv/DCCwXLTZo0Ke/4+uuvTz2uWgkHe+PSQ9x3333d/95nn300atSovh5Wt66V1V0aG9N91h86ND/GEA3MYxuCvQBSeW1je+xzLaHAbEtM+oX4NA6F22VlLwAAAAAgKsmq3i4TJkzIO16zZk3BckceeWTe8VlnnaUvfvGLevnll9MPsMrCgdzDDjusZJlDDz20z8cUFl3J29ZWODYQZ8uW/AViaVcGDyYEewGk0hqTwkHatkFb9N9hbNAGAAAAAOitaAC3mGi+17hVoQceeGBewHfr1q36+te/rsmTJ+vggw/WV77yFd1zzz1at25deYPuI//85z/zNiwrFOzt6OjQ/Pnzu4+rHeyN5jOOrvQtJfo7q0Z+5IGKDdoApBK3OZskjW3als+oJSb9QqFgr7uX3PQNAAAAAIAuvVnZ6e6xz91www16xzveoYceeqj7XGdnpx588EE9+OCDkqQhQ4Zo1qxZOvbYY/WhD31IU6dOLXsslRBO4TB9+nTtsMMOPco8/vjj3UHquro6HXTQQVUbnxQEZ8MB6Q0bNqSqHy1PsDcewV4AqcStwB09vEFD6rfdLJBmZe/aTVvV0Vn4jy1pHAAAAIDBYeLoYXrgnMK3n6OyJo7mFvg448eP1/33368rr7xS3/72t/Xss8/2KLN161Y99NBDeuihh3TBBRfogx/8oL71rW/pda97XZ+Na86cOXlpGOI899xzMrOiZTo7OzVu3LiCz1199dU6+eSTyxliURMmTNDixYu7j1988cVU9ZcvX96jPRRGsBdAKqvWF95ILZqjN03O3rjN2STSOAAAAACDxZD6Ou04rqnWwwDU0NCgT37yk/rkJz+pRx99VPfcc4/mzZunBQsWaO3atXllOzs7df311+vuu+/WvHnzNGPGjBqNun+bMWNG3mrpZcuWpaofLf+GN7yhIuPKInL2AkhlVcwGbWMjQdno8bb6bT1umSmWGoJgLwAAAACgVmbNmqVzzz1Xt99+u1avXq3HH39c3/jGN7TXXnvllXv55Zf17ne/W52dnTUaaf8WDc7+61//SlX/qaeeKtoetmFlL4BUVsWswo0GZeOCtG1bO7WhrUMjhm67/MRt+jZ0SJ2aGuvLHCkAAAAAAJVTV1enffbZR/vss4/OO+88/e53v9OHP/zh7s3D/vGPf+jOO+/UMcccU/G+L7vsMq1evbrH+TvuuEOXXXaZJGmHHXbQ1Vdf3aPMwoUL9elPf7r7+Oabb1ZTU+FV9HvssUeFRpxv7733zjtesGBBqvrhzeUKtYdtCPYCSGX1hsIre6O5dYutyF29oS0v2Bu7OVtzY8lcQwAAAAAA1MKJJ56os846S1/72te6zz344IMFg711dfk31xfbJK6QmTNnFjx/4403dv/70EMP1RFHHNGjzNNPP9397ze84Q067rjjUvVdCXPmzFFzc3P3RmuLFi3S0qVLtdNOO5Wsu2TJEj3zzDPdxyNHjtScOXP6aqgDHmkcAKQSl1933Ijkwd7oJm9xm76RwgEAAAAA0J/Nnj0773jlypUFyzU3N+cdb9y4sSL933///d3/PvjggwuWeeCBB7r/fcghh1Sk37SGDRumo446Ku/cz372s0R1o+Xe9ra3qbGReEEcgr0AUolb2RvdkG3UsCEaUld4VW40FcRqgr0AAAAAgAEoGtwdO3ZswXLjxo3LO37++ed73ferr76qhQsXdh/352CvJJ122ml5xz/4wQ/06quvFq3zyiuv6IorrijaDvIR7AWQStwq3LGRNA5mFr9JWyRgzMpeAAAAAECtnX766br11lsTp1jYsmWLvve97+Wdi0u3EM2Fe9NNN5U3yJBwELelpUW77bZbjzKLFy/WihUruo/jAsLVMHfuXL35zW/uPm5tbdVpp52m9vbCi8ra2tp02mmnqbW1tfvcwQcfrKOPPrrPxzqQEewFkFhnp8fn1x3RMzAbXe3bJbqydxXBXgAAAABAjc2fP1/HHXecpk2bps997nOaN2+e1q5d26Nce3u77rjjDs2ePVuPPPJI9/ntt99exx57bMG299lnH40fP777eN68eTr88MP1ox/9SH/4wx9099135z2SCKdwOOiggwrueRMOCO+0006aMmVKorb7yqWXXpqXv/jWW2/VUUcdpccffzyv3GOPPaajjjpKt912W/e5+vp6XXLJJVUb60DFBm0AElu3eas6Ogt/wxndoE3qudq3S3Rlb2wah5j6AAAAAAD0laVLl+qyyy7TZZddJjPT5MmT1dLSouHDh2vt2rVavHixNm/enFenvr5eV155pYYPH16wzYaGBp1xxhm64IILus/de++9uvfeewuWT7K6eKDk6w076KCD9M1vflPnnntu97l58+Zp5syZmjRpkiZOnKgVK1bopZde6lH3kksuyVsZjMII9gJILG5zNqnwKtxCq32lnit7Y9M4xNQHAAAAAKAa3F0vvviiXnzxxdgyY8eO1dVXX625c+cWbesLX/iCFi1apOuvv77X41q7dq2efPLJ7uOBEuyVpHPOOUf19fU699xz1dHR0X1+xYoVeSknutTX1+tb3/qWzjzzzCqOcuAijQOAxOJSOEgxwV5W9gIAAAAABohbb71Vl19+uebOnasxY8aULD9p0iSdffbZWrRokY4//viS5evr63XdddfpwQcf1Omnn679999f48eP19ChQ1OPdf78+ers7JQkNTU1ad999+1R5pVXXtGiRYu6j/tLsFeSzjrrLD366KOaO3duXlqHsLq6Oh177LF67LHHCPSmwMpeAIm1ri8clB06pE5NjfU9zsfl3A2v7N3c3qENbR0Fy5GzFwAAAACyaerUqYk3QpOCW/3LdfLJJ+vkk08uWW7HHXfU6aefrtNPP13urkWLFumZZ57RsmXLtHbtWnV0dGjkyJHafvvtteeee2rXXXeNDVQWM3v2bM2ePbuMn2SbY445puTrN2HChFSvcbXtvffeuu2227Ry5Uo9+OCDWrx4sTZs2KDm5mZNnz5ds2fPzstzjGQI9gJILG5lb0tzY8FE8C2xaRy2tVNstXBcfQAAAAAA+pKZacaMGZoxY0ath5J548eP1wknnFDrYWQGaRwAJBaXW3dszArc+A3atrUTt1q4WH0AAAAAAAD0RLAXQGKrYgKzcekWWmLOr928Ve0dQW6hVTEBZDNpDMFeAAAAAACAxAj2AkhsVUzKhbhgb9yKX2lb+oa4NA5jhjeovq5naggAAAAAAAAURrAXQGJxq3DTruwNtxWXxoHN2QAAAAAAANIh2AsgsdVxwd6YdAvF0jB0BXvjVvYS7AUAAAAAAEiHYC+AxOI2aBs3onBgtnFInUYOG1Lwue6VvSlXCwMAAAAAAKAwgr0AEotL41AsXUPcc11txW/6NjTl6AAAAAAAAAY3gr0AEtnc3qGNbR0FnxtbJF1D3CZt3cHe2DQODSlHCAAAAAAAMLgR7AWQSNyqXklqiUnjICVY2RubxoGVvQAAAAAAAGkQ7AWQSLFgb9GVvTHPdW/QFhvsZWUvAAAAAABAGoV3TgIwKH3vnmf0f08sL/jc5pgUDmbSmCLB3rjN2+556hUdftm8Ihu0sbIXAAAAAAAgDYK9ALq1rt+ixa9uSFVnzPAG1ddZ7PPjYgLBm9o7ivYVVw8AAAAAAACFkcYBQK+Mi8nJm/T52HpF8gADAAAAAACgJ4K9AHpl+9HDij4/cfTw1G0OqbPYjd0AAAAAAABQGMFeAL1yxG6vK/r8ftPGauSwdBljDn79eA1rqO/NsAAAAAAAAAYdgr0Ayva2PbbXBw+YUrTM0CH1+v4H9tHwhMHbqS1N+urxb6zE8AAAAABUmVnP/TzcvQYjAYDiOjs7e5wrdA0baNigDUC3Y/eapNe/bmTJckPqTG/aYbR2236U6opsztZlzowJmv/5w/Xw861aub4tttzO45u1705jWdULAAAADFB1dT3XlLW3t6uhoaEGowGAeFu3bu1xrtA1bKAh2Aug235Tx2m/qeP6pO1xzY162xsn9knbAAAAAPoHM1NjY6Pa2rYt8li/fr2amppqOCoA6Gn9+vV5x42NjZlY2Tvww9UAAAAAAKDfGDky/27BtWvXksoBQL/i7lq7dm3euei1a6Ai2AsAAAAAAComGjBpb2/X8uXLCfgC6BfcXcuXL1d7e3ve+VGjRtVoRJVFGgcAAAAAAFAxw4YNU0NDQ14gZd26dXruuec0atQojRgxQkOGDMlEbkwAA0NnZ6e2bt2q9evXa+3atT0CvQ0NDRo6dGiNRldZBHsBAAAAAEDFmJkmTZqkZcuW5a3mbW9vV2trq1pbW2s4OgDI13XNykK+Xok0DgAAAAAAoMKampo0ZcqUzARPAGSTmWnKlCmZ2kSSYC8AAAAAAKi4roBvQ0NDrYcCAD00NDRkLtArkcYBAAAAAAD0kaamJk2fPl1btmzR2rVrtW7dOrW1tdV6WAAGqcbGRo0cOVKjRo3S0KFDM3n3AcFeAAAAAADQZ8xMw4YN07BhwzRhwgS5uzo7O/Py+QJAXzIz1dXVZTK4G0WwFwAAAAAAVI2Zqb6+vtbDAIBMImcvAAAAAAAAAGQAwV4AAAAAAAAAyACCvQAAAAAAAACQAQR7AQAAAAAAACADCPYCAAAAAAAAQAYQ7AUAAAAAAACADCDYCwAAAAAAAAAZQLAXAAAAAAAAADKAYC8AAAAAAAAAZADBXgAAAAAAAADIAIK9AAAAAAAAAJABBHsBAAAAAAAAIAMI9gIAAAAAAABABhDsBQAAAAAAAIAMINgLAAAAAAAAABkwpNYDADKgMXzw7LPP1mocAAAAAAAAqKECcaHGQuX6irl7NfsDMsfMjpN0c63HAQAAAAAAgH7neHe/pVqdkcYBAAAAAAAAADKAYC8AAAAAAAAAZABpHIBeMrPRkg4NnXpBUluNhlMp05WfmuJ4Sc/VaCwYGJgzKAfzBuVg3qAczBuUg3mDcjBvUA7mTbY0StoxdHyfu6+pVuds0Ab0Uu5/2KrlXqkGM4uees7d/1mLsWBgYM6gHMwblIN5g3Iwb1AO5g3KwbxBOZg3mfTXWnVMGgcAAAAAAAAAyACCvQAAAAAAAACQAQR7AQAAAAAAACADCPYCAAAAAAAAQAYQ7AUAAAAAAACADCDYCwAAAAAAAAAZQLAXAAAAAAAAADKAYC8AAAAAAAAAZADBXgAAAAAAAADIAIK9AAAAAAAAAJABBHsBAAAAAAAAIAOG1HoAAPqlVyVdGDkGimHOoBzMG5SDeYNyMG9QDuYNysG8QTmYN6gYc/dajwEAAAAAAAAA0EukcQAAAAAAAACADCDYCwAAAAAAAAAZQLAXAAAAAAAAADKAYC8AAAAAAAAAZADBXgAAAAAAAADIAIK9AAAAAAAAAJABBHsBAAAAAAAAIAMI9gIAAAAAAABABhDsBQAAAAAAAIAMINgLAAAAAAAAABlAsBcAAAAAAAAAMoBgLwAAAAAAAABkAMFeAAAAAAAAAMiAIbUeAID+xcymS9pf0g6SGiWtlrRQ0gJ331zLsaFvmJlJmirpTQp+72MkbVHwu39G0iOV/t2b2UhJsyXtKmmUpE2SliqYZysq2ReywcyGSTpQ0hskjZXUJulFSX9x98W1HBt6x8xmSNpLwfWnScH14N+SFkl60t239KJt5k2GmNlQSftI2k3B73O4pLWSXpH0uKRn3d0r0M8QSQdIeqOkFkkdkl6S9Ji7/7O37aP/q+a1g/feA5+ZjZC0h4L50iJpmKTXFFybHnX3JX3Q5x6SZkqaKKleUqukfyiYo1sr3R8GPq41g4tV4P0QgAwwsxMkfUnSvjFF1ku6RtKF7r6ySsNCHzGzsZJOkPQ2SYdLGl+keLuk30v6jrvf18t+p0n6qqT3KniTEeWS7pP0ZXe/vzd9oXbM7JeS3h85vdTdp5bR1naSvizpZEnNMcUek3SRu9+ctn3URu4Ln09L+g9J04oUbZP0sKTfuvt3U7TPvMkQM5sp6bOS3i1paJGiyyVdJem77r6qjH5GSPq8pE9KGhdT7GlJF0u6phKBZSRjZpMVBCkOyP13lqSRoSJl/Y0p0E/Vrh289+57fTlvzOwABe+l36og6Frsrumlkn4k6cfuvrqc/nJ9mqRTJJ2rYMFEIa2Sfijpv919Q7l9DVbVutYU6f9NCq4xDZGnTnH3a8ps8wRxrRl0CPYCg1xulcxVkj6UsMqrkt5NIG7gMrMfKAiwFAq2lnKtpE+7+9oy+n2vpKsVrNwrxSVdIuk8PkwPLGb2Dkm3FHgq9ZtjM5sj6UYV/zIi7FpJH3P3tjT9oLrM7FhJV0p6XYpq/3b37RO2P0fMm0wwszpJ35B0ttKln/u3pJPd/Y4Ufb1J0s0q/uVD2J2S3ufua1KMCymY2WxJZykIukwqUbzXAZhqXTt47923+nremNnekm6StHMZw3tZQdAu8bUp1O8YSb+RdGTCKoslHcfdCKVV+1pTZBz1kv4sab8CT6cO9nKtGdzI2QsMYrkPUb9Wzz8AHZKel/SEpOiHmO0k3W5mb+nzAaKvHKDCgd4OBbcoPibpb+r5u5ekkyTdlVv9lJiZvUfSL9Uz0PuqgltvX1QQ4O2uomDVwrfT9IPaMrPRClaTVKKtgyT9QT0/dL8m6a+SliiYs2EnSfplbuUL+iEz+6yCLwOigd7NCj6YPizp75LKWlnCvMmcHyv4WxD9zLJRwTx5WNJzyv/7IQXz62YzOyZJJ7lUIn9Sz0DvegV/D59RcJdL2NEK3g8NS9IHyrKfpHeqdPCl16p17eC9d1X09bzZQfGB3jUKVv8/rOBvWvTatL2k35tZ9O6nosxsuIIvmKKB3jYFKY/+Lim6indnSfea2S5p+hqkqnatKeGzKhzoTY1rDQj2AoPb2ZKOj5z7kaQp7r6zu++j4DbGEyUtC5VpkvSbXGAHA9trkq6QNFfSWHff0d1nufteCnKOHSbpgUid/RXc6pNILj/U1cr/m/OkpMPdfYK7z3T3HRXkYPxdpPqZZnZiip8HtXWppMm5f5d962AuzcivFeTj7LJUwe2S49x9X3efpiDX9I8j1U9U8GYZ/YyZnabgC5xwYOR2ScdIGuPu0939AHff0923UzCXPqJgBVXJlXPMm2wxs3cruAsl7F8K/l6Nzs2TA9x9FwXB3S8rf540Svp5bl4U62eIeq7mXCXpowrmzV7uvquCIM3XJXWGyr1FwV0oqL71lWqoytcO3nvXVsXmTchDkv5L0h7uPsbd35C7Nk1XcG06X8EXVF3qJF1rZvuk6OPbCt5/d+mUdJGk7d19hrvvqWDenKIgD2uX7RTMm/rUPxW69MWc6SH3eemroVO9TcHBtWawc3cePHgMwoeCQN5aBd84dz0+X6T8ZAXfAobLX1jrn4NHWb/7R3O/y9MkDU9Qvl7BhxuPPA5L2N8NkXoPSxoVU9YK9PWspCG1ft14lPw9z1Hw4cMVrBo4O/J7XJKirW9E6i6WNKlI+S9Eyr+m4MuLmr8uPLp/R7so2Hit63fUJukDKeqX/H0yb7L1ULBSLfz7eURSc4k6hytYgRuud16JOh+PlF8lafci5T8YKd8u6fW1fr2y+JB0Zu41XivpXgWB9XdL2in3N6esvzEF+qnKtUO8987EvJF0rIL3OdcpCPAmqbOngly64b7vS1j3DZK2RurG/v1UsFHc6kj5U2r9e+nPj2pda4r0bwruLunq4xZJ8yL9npyiPa41PAj28uAxWB8KNhfJe8OhXB7vInXeGqmzVlJLrX8WHql/93MlNaasU6/gg3b49/+LBPX2yL0h7qqzRdJuJeoMU3BLWrivj9X6deNR9Hc2XEFQvuv39Z1y3xwrWIWyLlL3rSXqWO4aFq7z9Vq/Ljzyfkd/ivx+3lPh9pk3GXoouP3YI4/9Etb9UaTegiJlGxWsaAqXPzVBH9el/XvIo6x5MF3S7pLqCjxX1t+YAu1U7doh3ntnYt4o2BgtUZA3Uu+4Ate1XRLU+3WkzrUJ6pwW/TklNdT6d9NfH9W41pToP/yl4zpJO6p3wV6uNTxI4wAMRrkcPqdETn/Fc1f6OO5+j/Jv6R8p6b0VHh76mLv/3lNuJuLuHep5q+rRCaqeqvz0Db9y96dK9LVZ0n9HTkdv5UX/cpGCN8pSEDj5Yi/aer+kcE7o+3PXnli5a9eFkdOnkoO1fzCz4xWkhOlyo7vfWOFumDfZMiNy/KK7P5Kw7k2R42L5Ko9W8KG6yxIFaYdK+YqCD8Nd3sMtr5Xn7s+5+7/cvbN06bJV5drBe+/q6et54+6LvIxNz9z9FgWpaMLeVqxOLsVIOJ2ZK7j+lHK1glQkXXaSdESCeoNSla41BZnZZOV/xjrf3V/oRXtcayCJnL3AYHWggpUMXRYr+PYwiasixydUYDwYGKK5e1vMLLrhWtRxkePo/Inza+XnqtrPzGq9aQIKMLP9FNz+1uV0d+9NfrPjI8dJ58y9Cm5B67K9pDf3YhyonI9HjqNBkkpg3mTLuMhxmg++yyLHY4qUjc6bq0t9IJaCwICClVJdGiS9PdHo0N9U69rBe29IPd9LTylRfq6kIaHjee6+uFQnuaBl9IurE0qODrVwhaSuLwsflnR5L9vjWgNJBHuBwWpu5PiuJB9uuspGjueYWXMFxoT+b3WBc7ErmXK7m4dXVG2QtCBJR+4eLWvqOW9RY2bWoOCNYdfGHze6+229aG+EpEMip/+YpG7uGnZ35PSx5Y4FlZFbsRK+C+CJclZEleiDeZM90R3ChxcsVVi07MoiZaN/VxLNm5zo+yHmzQBT5WsH770h9XwvXeqOAK5RGWZm79e2hTFbFaSt6+3qYq41kESwFxis9o4cJwrASZK7r1Bwm2OXRgU5jpB9kwucay1Sfu/I8cPuvjVFf/NLtIfaO0/Sm3L/fk3SGb1sbw8FK+S6PO/uL6eoz5zpf96mbV8GSMFquEpj3mTPE5Hj3VJ84Nw/cvxwoUJm9joFqzG7bJH0eMI+JOZNFlTz2hF9jvfeg1P0vXSx99FSL+aNpMcUXNe6TDKz7eIKo7rMrEXS90KnLnP3v1Wg6b0jx1xrBimCvcDgtFvkOJo/qpRo+Wh7yKaDI8dLS+T+ZZ5lmJntLun80KlzU35ILoQ5kz37RY6f7PqHme1jZt8zsyfNbLWZbTSzJWZ2l5l9LrcqOAnmTca4+4vK/4A6VAm+TDKzocpPKyPF35Yf/T0/mzKffXTe7GJmQwqWRH9VzWsH16lBLpfT+aDI6UVFyjeoZ87xxPPG3bdIei5ymnnTf3xX29ItPKfKpbjiWgNJBHuBQcfMhqtnfqi0SeCj5aMbqSCbTo0c/6FE+ei8YJ5lRG7zh6sUfOMvBTnoflqBpis9Z3Yys2G9GA96LxrsXWxmI8zsKgWrKD8taU8FeVWHa9smMpdKesbMvpH7wFsM8yabzpUUvp31q2b20bjCZjZG0m+V/8H0Vne/NaZKr+aNu78qaXPoVKOkaWnaQM1V5drBe2/kzFH+NcIl3VGk/M7Kz9e7yd2LpaUphHnTD5nZ2yV9KHTqP919UwXa5VqDbgR7gcFnvIL8p13aJb2Sso3lkeMJvRoR+r3cm5JoXrtrSlSLzosXU3YbnWfcetZ/nKFtG9G0Sfp4inxgxfR2zvxbQc6zLnWSWno1IvRWdFVSp6T71fPLo0KGK0gV8gczG1mkHPMmg9z9QUn/pSAgIgVBj2vM7GEz+7yZvdPM3mZmHzaz7ytYGRXOSXmXpA8U6aK380aSVpRoE/1bta4dvPce5HJfkn8zcvqOEndERX/H0TmQBPOmn8m9n/lR6NR17h7N/10urjXoxq1GwOAzInK8sYwgzYYSbSJDzGycpB9HTv+fuxfMgxgSnRfReVNKtHyDmQ3N3ZaGGjGzaZK+Fjr1TXdfWKHmezVn3N3NbJOkcGCQ61ON5D7cRoO035O0T+7fLuk2BXcJvCipOffcRyRNCtU5QsGXS++K6Yp5k1Hu/kMze1rBvNkjd3o/9VwxHrZY0iWSflpio5ve/o0qVId5M7BU69rBe298TtIBoeNO5afCKoRrVDZdLGnH3L9XSvp/FWybaw26sbIXGHyiF+zNBUsVF73NhD8CGZUL1lwvaYfQ6TVKthFXb+daoduZmGu19xMFQTlJWijpGxVsm+tTtoxW/goTSdo3999WSYe6+3Hu/iN3v83df+3un1dwy+ANkXonmtlJMf0wbzLM3f+kILj7LUkdJYovy5W7IcGO5swbVGsOMNcGMTM7WNLXI6e/4+5/LVGVeZMxubnwn6FTZ5WRmqMY5gy6EewFBp9oLrE0m5F0ia6sHF7mWND/XSrpmMi5T7h7kvxPvZ1rhVbwMtdqyMxOU7DKUgpWZX485YZGpXB9ypa4Dwgdkua6+wOFnnT39QpW9/4x8tQXchvcRDFvMszM/lNBiobPSaovUXyKpCskLTGzUqlCmDeo1hxgrg1SZrazpN8p/47qJyR9IUF15k2G5PJ5X6VtX4Lf7e7XVrgb5gy6EewFBp/oN3yNBUsVN7REm8gAMztDPW8tusTdf52wid7Oteg8K9QmqsTMJipYMdflyrhgXS9wfcqWuNf+Snf/S7GKuVWZn1T+Bl0zJB2aoB/mTQaYWYOZ/VbSDyVNzJ1eJemrkvaXNFbB73qSpOMk/a+25fcdJ+kqM7u0SBfMG1RrDjDXBiEzGy/pdgV5VLv8W9KJCVOSMW+y5UJJr8/9e5PyV/hWCnMG3Qj2AoPP+shxOTuOR7/hi7aJAc7MPijpO5HT10j6fIpmejvXCn2TzFyrnR9IGpP798uSzumDPrg+ZUvca//TJJXdfbGk6KYlhYK9zJts+qHy8zQ/LGkPd/+yuz/i7q+5e7u7v+Tut7r7iZJOUP4H08+Z2Skx7TNvUK05wFwbZHKbcN0uadfQ6TWSjnb35xM2w7zJCDPbV9JZoVNfdffn+qAr5gy6EewFBp/oBbsp5rbYYpojx/wRyBAzO1bSz5Wfa/N3kv4jZZL/6LyIzptSouW3ujvfLteAmb1H0jtDpz7j7q/1QVe9mjO5axlvUvsJd9+knjlW10kqlacw7L7I8awCZZg3GWNmcySdFjr1iqRjS+xcL3e/RdLpkdOXmlmSLw/T/o0qVId5M7BU69rBe+9BJHe7/i3K/3u1UUH6oidTNMU1KgPMbIikn2lbGqK/Kf9OuUriWoNuBHuBwWeltt3mKEkNkiakbGNy5PiVXo0I/YaZHSbpRuXnFrtL0gfcvdTGOFHRebFDwVLxovPs1ZT1UTnhW6F/7+6/6aN+ejtnXqf8udup4JqH2on+Tp9NsHFW2NOR40J/r5g32RPdBPQ77p70b8A1khaFjlsknVigXG/njRSkkCjWJvq3al07eO89SJhZg6TfSJoTOt2mIHXD/JTNRX/H0TmQBPOm9j4saa/cvzsV7HextY/64lqDbgR7gUEmt9JqWeT0lJTNRMsvLH9E6C/M7AAFKxHCt/wskPTOMjfhigZpmGcD15jQv+eamZd6SLo30sZOBcrtHSlT6TmzlNXgNfdU5HhtyvrR8mMLlGHeZEhuFdLhkdO3Jq2f+zLh95HThxQo2qt5Y2YTlP/3sk3S4jRtoOaqcu3gvffgYGZ1kq6V9I7Q6Q5JH3T3O8tocrGkcFBwuJltl7IN5k3tjQn9u07SQwnfR0fTVl0dKfN/0Y641iCMYC8wOEUv2runrL9bifYwwJjZngpyi40Inf6rpLe7+4Yym2WeIS3mTPb8K3JcaOPFYqL55jYWKMO8yZaxkkZHziXNcRlXvtCKuOjvebqZpdnMJjpvnuvD1VroG9W8dnCdyrDcl1Q/kfT+0GlXkALtpnLadPd2SdG8ronnjZkNlbRz5DTzJvu41kASwV5gsHoicnxg0opmNlHS1NCpdvX8MI8BxMxmKEjVEF4x95SCTSTW9KLpJyLH++XyViU1u0R7yJ5/KrimdJmau+YkxZzpfx6PHL8uZf3o7YetBcowb7Kl0BcCaYOo7ZHj+miBXP7fcA7goZJmpuiDeTPwVfPaEX2O997Z8j/KzzMuSWe4+zW9bPeJyHHieaPgeha+nr7k7tySn31PRI651gxSaT50A8iO2ySdGzo+wsws4eZbR0WO73V3ErcPUGa2k4Ld7sMBleclHZkiP2JB7r7QzJ6TND13qlnBG477E4yrWdJbws0pmLeojeMV5P1KYy/lb0DxbwV5y8KeDR+4+zozu1/SW0Onj1RwW2RRuVU1R0ROJ771G33m9wpy1HUtMJhmZuPcfVXC+tHgW/S2a+ZN9hQK6E9SutW9SXO+/175AZojJf05YR9HRo6ZNwNMla8dvPfOKDO7SNJnIqe/4O6XV6D52yS9L3R8pKRvJqzLNap/uEnSP8qod5mkPUPHl0r6Y+g4LnDPtQaSCPYCg9UCBQncx+eOd1awkUA0x2Yh0W+tb67csFBNuW9v71H+hiTLJb3V3ZdXqJtbJH02dHyaEgR7FbyxDaeUeNTdV1RoTEjJ3e9LW8fMoivxNrv73Qmq3qL8D96nKcEHb0mHSZoWOv63pL8kqIc+5O6vmNl8SQeHTp8o6cpSdXN3ArwzcnpeTHHmTUa4e5uZvSQpvMLycElXpWjmrZHj6K3QXW5R/vuaU8zsolIfis1suvLzKbZL+kOK8aH/qNa1g/feGWRmZ0v6YuT0N909aUC2lD8ouLOhK24zx8x2dvei+cFzX0acHDnNvKkBd39B0gtp65nZ6sipfyV8H821BpJI4wAMSrnNS66JnP5y7o1BLDN7q/I/sK9TsOMsBhgzG6cgdcP00OlXFazoTZsbsZifKX9X2PebWTQXVHRswyR9PnI6zYd8DGy/khTOE32ImUU3a8qTu3Z9OXL66ty1DrX348jx2blcgqV8TNL2oeO1kuI2uWHeZMs9keMzk6YBMrNDlX9nSKH2utwp6cXQ8VRJpyTo5iuSwu+Zbupl2iPUTlWuHbz3zh4z+4SkSyKnL3f3L1Sqj9xdMP8X7lbB9aeUU5V/O/5SBXfyIeO41qALwV5g8LpYUvi2jEOVf8tHHjObrJ4rsb7r7iv7YGzoQ2Y2UtIdkvYInX5N0lHu/lQl+3L3fyj/jUKjpJ+b2aiYsZmk70h6fej0YgVBYwwCuXxy0VsfrzSzSUWqnSfpkNDxGgW3u6F/+KWkv4eOd5X049zO5QWZ2QHq+SH6iriAGvMmc66PHL9R0hXF5owkmdkukm6InH5GMakZ3H2LpK9HTn/LzGI3tDGzDyo/JU2Hegb+MEBU+drBe++MyF0HroicvlrSGX3Q3ZcVpEPq8hEz+0CRse2u/DRaknSRu7f1wdjQP3GtgSxZ6g4AWWRm50n6RuT0DyV9reuW+dwHq+MkfVfSlFC5FZL2cPfXqjBUVJCZ3avgdp6wC5Q8T2HYY+4evc0o2t8ukp6U1BQ6/aSkM919XqjcrgrykJ0YaeK97n5jGWNDDZnZHOXfMrbU3acmrDtOwcY54VWdSxV8iLq16xZrM9tBwe2Tn4g0cY67E7TrR3IrRu5S/mrIuyV93t0fC5UbreA2wguVn8plkaRZ7r6uSB/Mmwwxsz8puFU+7EEFq9ruc/etobItCm5Z/pKk0ZE673H33xbpp0HSX5X/BegqBSmIbujqJze/PivpC8pfMHOFu5+e+AdDKmY2W9LwAk8lyQvfZYW7x24yVM1rB++9q6Mv542ZHSHpduWnxFwo6UwFX/6ksTr8NzCOmf1Y0sdDpzoVfFH1P13vw3PXsg9J+rbyN13+m6SZ4WsmeqrGtSbleOYpP13QKWk2/ONaA4K9wCCWu8DfLOnYyFMdCt7krlGQj2xM5PlNCm73n9/XY0TlmVklL/yHhQO2Rfp8v4LVVtFbiF6VtEzBBnE7FHj+++7eF6sk0Md6E+zN1T9EwS3WwyJPvaZgo6YxCt6Y1keev1nSOxNuRIEqMrNzJf13gadeVnArfbOC1DKNkedbFVxr/h6tWKAP5k1GmNn2CnIPTivw9HoFv89NkloU5CQsdIvqZe7+uQR97aYgkDyuQD/PKQgATFPPjSofljTH3TeV6gPlMbMlknbqZTM/d/eTS/RTlWsH772roy/njZl9RZVbzX+fu88pVcjMmiTdJ2lW5Kk2BfNzi4Lr4IjI8yslzXb3Rb0farZV61qTVAWCvVxrBjnSOACDWC6nz3sU5CsLq1fwhmEf9fwD0Crp7fwBQBru/isFqw2iH4i3kzRT0o7q+UH9W+q5uzEGCXe/X9JcBSvswsYouDZNU88P3TdIeh8Bu/7J3S9WsFKuPfLU9go+wO6mnoHepyW9JUmgN9cH8yYj3P1lBR905xV4eoSkN0naX8EXBNG/H+0Kcr+fnbCvpxRsAre0QD97KUg9Eg303i3paAK92VCtawfvvVEOd98o6WhJf4o81ShphqQ91TPQu0TS4QR6ByeuNSDYCwxy7r7Z3T8g6d2SnihSdIOC3FS7J1nJCUS5+y8V5F28QT2DPWH3K1gpdTbBl8HN3f8kaXcFt51tLFL0r5Le5e4fyuXgRD/l7t9X8KH01yp+HXhewZc9e7r7Myn7YN5kRG4X87dKeq+CoG+pzfPWKPi9v8ndL04ZhHtSQQD5m5KKpSd6RsHmgUdxi2u2VOvawXtvlCO3WduRCtI5PFuk6CoFt++/KekXpcgmrjWDG2kcAOTJ5Vc9QNJkBd8WvybpKUnz3X1zDYeGDMlt0HaQgo3YRkrarCCdw3x3X17LsaF/MrPhkg5UsPpzjIJbF5dL+ou7F/vQg34qdx04UMF1YLSCW+b/Lelxd3+6Qn0wbzIkt8HoLAWrksYouOV+rYLVSH+T9K/caqbe9tOg4L3QGxWkieiQ9JKCuUnwZBCo5rWD994oh5m9SdK+kiYqWK3ZKukfCuZosS9TMUhxrRlcCPYCAAAAAAAAQAaQxgEAAAAAAAAAMoBgLwAAAAAAAABkAMFeAAAAAAAAAMgAgr0AAAAAAAAAkAEEewEAAAAAAAAgAwj2AgAAAAAAAEAGEOwFAAAAAAAAgAwg2AsAAAAAAAAAGUCwFwAAAAAAAAAygGAvAAAAAAAAAGQAwV4AAAAAAAAAyACCvQAAAAAAAACQAQR7AQAAAAAAACADCPYCAAAAAAAAQAYQ7AUAAAAAAACADCDYCwAAAAAAAAAZQLAXAAAAAAAAADKAYC8AAAAAAAAAZADBXgAAAAAAAADIAIK9AAAAAAAAAJABBHsBAAAAAAAAIAMI9gIAAAAAAABABhDsBQAAAAAAAIAMINgLAAAAAAAAABlAsBcAAAAAAAAAMoBgLwAAAIABy8zmmZl3PWo9HgAAgFoi2AsAAAAAAAAAGUCwFwAAAIOamS0Jrww1szm1HhMAAABQDoK9AAAAQEKkDKgsM5saCbRfU+sxAQAADGQEewEAAAAAAAAgA4bUegAAAAAAUC53n1PrMQAAAPQXrOwFAAAAAAAAgAwg2AsAAAAAAAAAGUCwFwAAAAAAAAAygJy9AAAAQI2ZWb2kWZJ2kTRB0lBJr0p6XtJ8d99S4f5aJL1Z0iRJ20laL+kOd19UpM4ISXtIeoOkFknDJa2RtFLS48Xq9ndmNkHSgZK2lzROwc/1b0l/cfcX+qC/Kbn+pkgyBb/rJyT91d290v0BAIDBg2AvAAAAUISZfUXSl2OeKxaYu6/U5mFmNlXSlySdoCDIWMhGM/udpAvc/fkSw+1qd4mknXKHS919au78WySdJ+ltkhoi1T4rKS9ga2a7SvpArvwsFfn8YGYvS/qhpO+7++oU4wv7qJl9tEjVU9z9mkhb8yQd2nXs7las70jdEyR9XtL+CoKuhcr8XdJlkq5z986E7YbnRfc8MLP9JP23pMNjqr5gZhdEf0YAAICkSOMAAAAA1ICZfVHS05JOVXygV5KaJH1Y0kIzO60X/Z0jab6kd6hnoLdQ+WNz4/uKglXApRaKbC/pQkl/zwU1+y0zG21md0r6X0kHKCbQm/MmSddIesjMJvaizzMk/VnxgV5J2lHS1Wb2EzPjsxoAAEiNlb0AAABAFeVSNlwlqdAK1lZJyyW1SZooaXLouUZJV5rZKHf/n5R9fkLSxaFTbZKWKEhXMCn3iBpW4NwmSS9IWifJFaRz2En5i0gmS5pnZrPc/ak046wGMxsn6R5Jexd4epmkVxQE36cq/+faT9J8MzvM3Zem7PM/JX03dGqTgtd/vYLXK/r6f0zSQknfTtMPAAAAwV4AAACguGslPZj792WS9gw9d2SRenGpDC5QfqC3XdIVkn7i7v8KFzSz6ZLOlvRxbVt9eomZ/cXdFyQbvsZJ6goOvyTpi5JudPd1oX6mSWouULdD0h2SbpH0J0mLo6kMzGykpONyP9euudNNkm4ws31jctB+SEHO39dJuj50/o+SLi3ys/yzyHNJ/UQ9A70/lXSxuz/XdcLMJkn6tKTPadvnpmkKfq5D3L0jYX+7aFug9ykFr/8f3H1zqK99JV0u6S2heheZ2c/dvTVhPwAAADLy/wMAAGAwK5A/9jB3nxdTdp7KzA+bq3+gpAe0bcXoSknHuPujJeq9T9INoXp/c/e9ipRfop45cZ+SdLi7v5xwrNMkdSZdxWpmwyTdJOntodNvd/fbi9SZqmATui4/d/eTk/QXamOeEv5OzOxdkn4bOf0f7n5VkTpvl/R/yk998Vl3/06ROoU+ZN0p6UR33xhTp0nSAknh3+uZ7v7dQuUBAAAKIQ8UAAAAUD0XaNt78E5Jx5cK9EqSu/9awariLnua2REp+m2X9N6kgd5cn8+nSVeQW6n6EQWpIbqcknyIVXFW5PjyYoFeSXL3PyjYRC/szFw6jqSWSXpfXKA3189GBZvFhR2Tog8AAACCvQAAAEA1mNluko4Onfp1ilQMUpBzd2vo+F0p6v7a3f+RonxZ3H2VpPBK3gP7us+kzGx35adJ2KCeQdw431aQq7jLTpKOStH9f7v7mtLFdJfy03/sm6IPAAAAgr0AAABAlURXaV6XpnIud+tjoVMHp6j+yzR99VI4LcNkM9uuin0Xc2jk+Hfu/lqSiu7erp6/r0MS9uuSfpOwnw5Jfw+d2s7MhibsBwAAgGAvAAAAUCXR4GzJ9A0FLAv9+w1mljRn8MNl9NXNzMaY2WlmdrWZPW5mL5nZBjPz6EPSeZHq43vTdwUdEDn+U8r690SO35yw3pKUm6y9EjkenaIuAAAY5IaULgIAAACgAnaLHL+SPFZbUL2kUcrPkVvIendfWU4HZtYs6cuSzpBU7grTMWXWq7TohnV/S1n/ycjxlIT1osHbUjZEjoenrA8AAAYxgr0AAABAdbT0QZujVTrYu7achs1svILVr28qp35If0lDMDZynDYAvkrBpnpdd0dG24uzOWU/Ub36RgAAAAwuBHsBAACA6hjTB20mScvWXmbbN6pnoPcFSfdK+pekFyWtl7RJQRC0y0mSPlJmn31pROQ4uoK2KHd3M9skqTl3amRFRgUAAFBBBHsBAACA6tioIO1Cl2Mkbe1lmy/3sn5BZnacpDmhU+skfVLSL929s2ClbXXf2hdjqoD1keNmSYlz6ebyI4dTKqyrxKAAAAAqiWAvAAAAUB0rlR/sfdzd0+ZzrZb3R44/4e6/TFh3XKUHUyGrI8ctyt/wrpRxyl9JHW0PAACg5pLc9gUAAACg956PHO9Sk1Ek8+bQv1sl/SZF3T0qPJZKWRo53itl/Wj5aHsAAAA1R7AXAAAASC4vhUHu1v6k7o0cH9774fSZ14X+/ay7dySpZGajJM1M0U80JURfbkb2UOQ47esfLR9tDwAAoOYI9gIAAADJRTf1akpR947I8cfNrKGX4+kr4aBrY4p6p0oalqJ8b17PtO6LHL/TzEYnqZj7PUU3nYu2BwAAUHMEewEAAIDkVkWOpyWt6O6PKX91746SvlaJQfWB8MZve5jZmFIVzGyypC+n7GetpPCq4cSvZ1ru/pSkBaFTIyRdmLD6ZyRNCR0vkXRXZUYGAABQOQR7AQAAgOT+GTl+d8r6X1J+6oJzzOyCNOkgzGwHM7vUzPZL2Xca4aBoo6RvlhjTdpJukzQmTSfu3i5pUejU3mY2PU0bKV0WOT7DzE4qVsHMjpb09cjp77h7NAUFAABAzRHsBQAAAJKLrub8kpldbWYfMbO3mdkRoUeP3LXuPl/S+ZHTF0p6xMzeb2Zjo3XMrN7MdjOzj5vZnQo2evucpOYK/UyFXBs5/k8z+7mZ7RQZ20gzO03S3yTtnTv9VMq+/hj6d72k+83sy2b2TjM7MvKaTkzZdh53/52km0KnTNI1ZvYjM9s5XNbMJprZNxQEscOpLBZIurw34wAAAOgr5u61HgMAAABQM2a2RFI4iHmYu88rUv4eJdvc6z53nxPTxrclfbbAU52SlklqzR2PkTRRhXPZxo4z8jMtdfepCcYbbeP3kt5e4KnFkl7NjW2a8gOhN0h6RvnpHEq9nrtKelLJcv2e4u7XROrPk3Ro17G7F10lbWbjJP1J0l4Fnl6i4Gcbp+Bniy6OeV7Bz7O0RB/hD1mx8yCm7jWSPho6Nc3dlyStDwAABjdW9gIAAADpfETS471pwN3/n4KA3muRp+okTZU0M/eYrsKB3nUF6lbahyQ9XOD8zpIOkDRD+YHeX0k6JW0n7r5IwWu6vowxpubuqxQEhwvl3J0qaT8Fr3v0s9IjkmaXCvQCAADUEsFeAAAAIAV3XyHpzZLeo2Al6z8VBF63pmznWgXBxS8pP29tnNWSfivpJEnbu/sTafpLy91fk3SIpIskrSlS9J+SPujuH3D3tjL7+q2kXSV9XtKdkl5QEPztk9sQ3X2Nux8l6UQFAe1i/fxDQRD7ze7+Ul+MBwAAoFJI4wAAAAD0A2Y2WcGq0gmSWhSkdFgrabmCPLjP1WpTMDMbJuktknaTNFZSm6QVkh7Jrcwd0MzsdZIOlLS9gp9vraR/S/qLuy+r5dgAAADSINgLAAAAAAAAABlAGgcAAAAAAAAAyACCvQAAAAAAAACQAQR7AQAAAAAAACADCPYCAAAAAAAAQAYQ7AUAAAAAAACADCDYCwAAAAAAAAAZQLAXAAAAAAAAADKAYC8AAAAAAAAAZADBXgAAAAAAAADIAIK9AAAAAAAAAJABBHsBAAAAAAAAIAMI9gIAAAAAAABABhDsBQAAAAAAAIAMINgLAAAAAAAAABlAsBcAAAAAAAAAMoBgLwAAAAAAAABkAMFeAAAAAAAAAMgAgr0AAAAAAAAAkAEEewEAAAAAAAAgAwj2AgAAAAAAAEAGEOwFAAAAAAAAgAwg2AsAAAAAAAAAGUCwFwAAAAAAAAAygGAvAAAAAAAAAGQAwV4AAAAAAAAAyACCvQAAAAAAAACQAQR7AQAAAAAAACADCPYCAAAAAAAAQAYQ7AUAAAAAAACADCDYCwAAAAAAAAAZQLAXAAAAAAAAADKAYC8AAAAAAAAAZADBXgAAAAAAAADIAIK9AAAAAAAAAJABBHsBAAAAAAAAIAMI9gIAAAAAAABABvx/PojpWlSjKBkAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1500x900 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot the loss of the last saved instance\n",
    "plot_method_data([pd.DataFrame(accuracy_test).astype('float').T],\n",
    "                title='Test Accuracy', dlabel='inst#', xlabel='Accuraccy', ylabel='Iteration')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "qiskit                        0.43.0\n",
      "qiskit-aer                    0.12.0\n",
      "qiskit-dynamics               0.4.0\n",
      "qiskit-experiments            0.5.2\n",
      "qiskit-finance                0.3.4\n",
      "qiskit-ibm-experiment         0.3.1\n",
      "qiskit-ibm-runtime            0.9.4\n",
      "qiskit-ibmq-provider          0.20.2\n",
      "qiskit-machine-learning       0.6.1\n",
      "qiskit-nature                 0.6.0\n",
      "qiskit-optimization           0.5.0\n",
      "qiskit-terra                  0.24.0\n"
     ]
    }
   ],
   "source": [
    "!pip list | grep qiskit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch                         2.0.1\n",
      "torchsummary                  1.5.1\n"
     ]
    }
   ],
   "source": [
    "!pip list | grep torch"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.15"
  },
  "vscode": {
   "interpreter": {
    "hash": "cd40c0a4cc788d6c69a9e14a194c403ce75abdbd749a2a56d805605a5a1393dd"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}