TestConvMixGRU.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The autoreload extension is already loaded. To reload it, use:\n",
      "  %reload_ext autoreload\n",
      "Loading data pickle...\n",
      "Data pickle loaded.\n",
      "(6960, 250, 22)\n",
      "(6960, 4)\n",
      "(1772, 250, 22)\n",
      "(1772, 4)\n",
      "(1500, 250, 22)\n",
      "(1500, 4)\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "from models import *\n",
    "import datetime\n",
    "from utils import *\n",
    "import random\n",
    "import glob\n",
    "from os.path import join, getctime, basename\n",
    "from keras.models import load_model\n",
    "from shutil import rmtree\n",
    "\n",
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "# Create result folders \n",
    "save_path = join(get_save_path(), 'ConvMixGRU')\n",
    "new_path = join(save_path, 'best_val')\n",
    "time = str(datetime.datetime.now()).replace(' ', '_')\n",
    "workpath = join(save_path, time)\n",
    "ensure_dir(new_path)\n",
    "ensure_dir(workpath)\n",
    "ensure_dir(save_path)\n",
    "\n",
    "# Load preprocessed data\n",
    "aug_data = load_data_pickle(get_save_path())\n",
    "x_train = aug_data['x_train']\n",
    "y_train = aug_data['y_train']\n",
    "x_valid = aug_data['x_valid']\n",
    "y_valid = aug_data['y_valid']\n",
    "x_test = aug_data['x_test']\n",
    "y_test = aug_data['y_test']\n",
    "x_train = np.transpose(x_train, (0, 2, 1))\n",
    "x_valid = np.transpose(x_valid, (0, 2, 1))\n",
    "x_test = np.transpose(x_test, (0, 2, 1))\n",
    "print(x_train.shape)\n",
    "print(y_train.shape)\n",
    "print(x_test.shape)\n",
    "print(y_test.shape)\n",
    "print(x_valid.shape)\n",
    "print(y_valid.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "config = {\n",
    "    # Network\n",
    "    'num_inputs': x_train.shape[0],\n",
    "    'input_shape': (x_train.shape[1],x_train.shape[2],1),\n",
    "    'epochs': 50,\n",
    "    'dropout': 0.5,\n",
    "    'batch_size': 32,\n",
    "    'l2': 0.03,\n",
    "    'LSTM': False,\n",
    "    'lr': 0.001\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential_4\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " conv1d_4 (Conv1D)           (None, 241, 22)           4862      \n",
      "                                                                 \n",
      " batch_normalization_4 (Bat  (None, 241, 22)           964       \n",
      " chNormalization)                                                \n",
      "                                                                 \n",
      " max_pooling1d_4 (MaxPoolin  (None, 120, 22)           0         \n",
      " g1D)                                                            \n",
      "                                                                 \n",
      " gru_4 (GRU)                 (None, 120, 44)           8976      \n",
      "                                                                 \n",
      " dropout_8 (Dropout)         (None, 120, 44)           0         \n",
      "                                                                 \n",
      " flatten_4 (Flatten)         (None, 5280)              0         \n",
      "                                                                 \n",
      " dense_8 (Dense)             (None, 64)                337984    \n",
      "                                                                 \n",
      " dropout_9 (Dropout)         (None, 64)                0         \n",
      "                                                                 \n",
      " dense_9 (Dense)             (None, 4)                 260       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 353046 (1.35 MB)\n",
      "Trainable params: 352564 (1.34 MB)\n",
      "Non-trainable params: 482 (1.88 KB)\n",
      "_________________________________________________________________\n",
      "Model compiled.\n",
      "Epoch 1/50\n",
      "11/11 [==============================] - 7s 540ms/step - loss: 3.6081 - accuracy: 0.3471 - val_loss: 3.2022 - val_accuracy: 0.4093\n",
      "Epoch 2/50\n",
      "11/11 [==============================] - 5s 501ms/step - loss: 2.8446 - accuracy: 0.4588 - val_loss: 2.8079 - val_accuracy: 0.4960\n",
      "Epoch 3/50\n",
      "11/11 [==============================] - 6s 508ms/step - loss: 2.5330 - accuracy: 0.5395 - val_loss: 2.5382 - val_accuracy: 0.5253\n",
      "Epoch 4/50\n",
      "11/11 [==============================] - 6s 523ms/step - loss: 2.2828 - accuracy: 0.6095 - val_loss: 2.3253 - val_accuracy: 0.5653\n",
      "Epoch 5/50\n",
      "11/11 [==============================] - 6s 535ms/step - loss: 2.0666 - accuracy: 0.6536 - val_loss: 2.1607 - val_accuracy: 0.5880\n",
      "Epoch 6/50\n",
      "11/11 [==============================] - 6s 517ms/step - loss: 1.8873 - accuracy: 0.6856 - val_loss: 2.0520 - val_accuracy: 0.5893\n",
      "Epoch 7/50\n",
      "11/11 [==============================] - 6s 522ms/step - loss: 1.7416 - accuracy: 0.7182 - val_loss: 1.9439 - val_accuracy: 0.6287\n",
      "Epoch 8/50\n",
      "11/11 [==============================] - 5s 484ms/step - loss: 1.6036 - accuracy: 0.7457 - val_loss: 1.8440 - val_accuracy: 0.6220\n",
      "Epoch 9/50\n",
      "11/11 [==============================] - 6s 503ms/step - loss: 1.4934 - accuracy: 0.7530 - val_loss: 1.7856 - val_accuracy: 0.6287\n",
      "Epoch 10/50\n",
      "11/11 [==============================] - 5s 495ms/step - loss: 1.3915 - accuracy: 0.7736 - val_loss: 1.7490 - val_accuracy: 0.6367\n",
      "Epoch 11/50\n",
      "11/11 [==============================] - 5s 504ms/step - loss: 1.3127 - accuracy: 0.7888 - val_loss: 1.6851 - val_accuracy: 0.6467\n",
      "Epoch 12/50\n",
      "11/11 [==============================] - 5s 495ms/step - loss: 1.2459 - accuracy: 0.7864 - val_loss: 1.6274 - val_accuracy: 0.6427\n",
      "Epoch 13/50\n",
      "11/11 [==============================] - 6s 521ms/step - loss: 1.1546 - accuracy: 0.8190 - val_loss: 1.5984 - val_accuracy: 0.6447\n",
      "Epoch 14/50\n",
      "11/11 [==============================] - 5s 503ms/step - loss: 1.1182 - accuracy: 0.8157 - val_loss: 1.5806 - val_accuracy: 0.6327\n",
      "Epoch 15/50\n",
      "11/11 [==============================] - 6s 526ms/step - loss: 1.0510 - accuracy: 0.8355 - val_loss: 1.5522 - val_accuracy: 0.6213\n",
      "Epoch 16/50\n",
      "11/11 [==============================] - 6s 524ms/step - loss: 1.0020 - accuracy: 0.8399 - val_loss: 1.5221 - val_accuracy: 0.6320\n",
      "Epoch 17/50\n",
      "11/11 [==============================] - 6s 505ms/step - loss: 0.9709 - accuracy: 0.8430 - val_loss: 1.5043 - val_accuracy: 0.6373\n",
      "Epoch 18/50\n",
      "11/11 [==============================] - 6s 543ms/step - loss: 0.9186 - accuracy: 0.8546 - val_loss: 1.4755 - val_accuracy: 0.6420\n",
      "Epoch 19/50\n",
      "11/11 [==============================] - 6s 517ms/step - loss: 0.8809 - accuracy: 0.8595 - val_loss: 1.4659 - val_accuracy: 0.6367\n",
      "Epoch 20/50\n",
      "11/11 [==============================] - 6s 510ms/step - loss: 0.8453 - accuracy: 0.8675 - val_loss: 1.4867 - val_accuracy: 0.6393\n",
      "Epoch 21/50\n",
      "11/11 [==============================] - 6s 545ms/step - loss: 0.8515 - accuracy: 0.8550 - val_loss: 1.4815 - val_accuracy: 0.6273\n",
      "Epoch 22/50\n",
      "11/11 [==============================] - 6s 551ms/step - loss: 0.8384 - accuracy: 0.8578 - val_loss: 1.4853 - val_accuracy: 0.6407\n",
      "Epoch 23/50\n",
      "11/11 [==============================] - 6s 546ms/step - loss: 0.8111 - accuracy: 0.8642 - val_loss: 1.4673 - val_accuracy: 0.6333\n",
      "Epoch 24/50\n",
      "11/11 [==============================] - 6s 563ms/step - loss: 0.7622 - accuracy: 0.8784 - val_loss: 1.4823 - val_accuracy: 0.6200\n",
      "Epoch 25/50\n",
      "11/11 [==============================] - 6s 558ms/step - loss: 0.7483 - accuracy: 0.8769 - val_loss: 1.4289 - val_accuracy: 0.6193\n",
      "Epoch 26/50\n",
      "11/11 [==============================] - 6s 561ms/step - loss: 0.7053 - accuracy: 0.8937 - val_loss: 1.4574 - val_accuracy: 0.6440\n",
      "Epoch 27/50\n",
      "11/11 [==============================] - 6s 526ms/step - loss: 0.6876 - accuracy: 0.8918 - val_loss: 1.4432 - val_accuracy: 0.6300\n",
      "Epoch 28/50\n",
      "11/11 [==============================] - 6s 535ms/step - loss: 0.6870 - accuracy: 0.8920 - val_loss: 1.4527 - val_accuracy: 0.6407\n",
      "Epoch 29/50\n",
      "11/11 [==============================] - 6s 561ms/step - loss: 0.6689 - accuracy: 0.8993 - val_loss: 1.4364 - val_accuracy: 0.6460\n",
      "Epoch 30/50\n",
      "11/11 [==============================] - 6s 544ms/step - loss: 0.6408 - accuracy: 0.9003 - val_loss: 1.4431 - val_accuracy: 0.6547\n",
      "Epoch 31/50\n",
      "11/11 [==============================] - 6s 520ms/step - loss: 0.6343 - accuracy: 0.9004 - val_loss: 1.4450 - val_accuracy: 0.6513\n",
      "Epoch 32/50\n",
      "11/11 [==============================] - 6s 539ms/step - loss: 0.6255 - accuracy: 0.9045 - val_loss: 1.4166 - val_accuracy: 0.6540\n",
      "Epoch 33/50\n",
      "11/11 [==============================] - 5s 459ms/step - loss: 0.6023 - accuracy: 0.9068 - val_loss: 1.4829 - val_accuracy: 0.6447\n",
      "Epoch 34/50\n",
      "11/11 [==============================] - 6s 537ms/step - loss: 0.6067 - accuracy: 0.9034 - val_loss: 1.4553 - val_accuracy: 0.6407\n",
      "Epoch 35/50\n",
      "11/11 [==============================] - 6s 523ms/step - loss: 0.5773 - accuracy: 0.9131 - val_loss: 1.4743 - val_accuracy: 0.6547\n",
      "Epoch 36/50\n",
      "11/11 [==============================] - 6s 516ms/step - loss: 0.5747 - accuracy: 0.9115 - val_loss: 1.5102 - val_accuracy: 0.6340\n",
      "Epoch 37/50\n",
      "11/11 [==============================] - 6s 555ms/step - loss: 0.5828 - accuracy: 0.9066 - val_loss: 1.5855 - val_accuracy: 0.6353\n",
      "Epoch 38/50\n",
      "11/11 [==============================] - 6s 565ms/step - loss: 0.5605 - accuracy: 0.9144 - val_loss: 1.5015 - val_accuracy: 0.6500\n",
      "Epoch 39/50\n",
      "11/11 [==============================] - 6s 529ms/step - loss: 0.5336 - accuracy: 0.9177 - val_loss: 1.5399 - val_accuracy: 0.6493\n",
      "Epoch 40/50\n",
      "11/11 [==============================] - 6s 529ms/step - loss: 0.5283 - accuracy: 0.9162 - val_loss: 1.5302 - val_accuracy: 0.6533\n",
      "Epoch 41/50\n",
      "11/11 [==============================] - 6s 508ms/step - loss: 0.5260 - accuracy: 0.9185 - val_loss: 1.4723 - val_accuracy: 0.6527\n",
      "Epoch 42/50\n",
      "11/11 [==============================] - 6s 532ms/step - loss: 0.5003 - accuracy: 0.9274 - val_loss: 1.5255 - val_accuracy: 0.6667\n",
      "Epoch 43/50\n",
      "11/11 [==============================] - 6s 534ms/step - loss: 0.4967 - accuracy: 0.9239 - val_loss: 1.5246 - val_accuracy: 0.6493\n",
      "Epoch 44/50\n",
      "11/11 [==============================] - 6s 517ms/step - loss: 0.4914 - accuracy: 0.9250 - val_loss: 1.5428 - val_accuracy: 0.6427\n",
      "Epoch 45/50\n",
      "11/11 [==============================] - 6s 542ms/step - loss: 0.4780 - accuracy: 0.9290 - val_loss: 1.5110 - val_accuracy: 0.6593\n",
      "Epoch 46/50\n",
      "11/11 [==============================] - 6s 531ms/step - loss: 0.4623 - accuracy: 0.9341 - val_loss: 1.4976 - val_accuracy: 0.6613\n",
      "Epoch 47/50\n",
      "11/11 [==============================] - 6s 521ms/step - loss: 0.4591 - accuracy: 0.9297 - val_loss: 1.5507 - val_accuracy: 0.6560\n",
      "Epoch 48/50\n",
      "11/11 [==============================] - 6s 528ms/step - loss: 0.4524 - accuracy: 0.9323 - val_loss: 1.5694 - val_accuracy: 0.6527\n",
      "Epoch 49/50\n",
      "11/11 [==============================] - 6s 532ms/step - loss: 0.4543 - accuracy: 0.9307 - val_loss: 1.5689 - val_accuracy: 0.6340\n",
      "Epoch 50/50\n",
      "11/11 [==============================] - 6s 505ms/step - loss: 0.4486 - accuracy: 0.9328 - val_loss: 1.5283 - val_accuracy: 0.6580\n"
     ]
    }
   ],
   "source": [
    "ConvMixGRU = ConvMixGRU()\n",
    "ConvMixGRU.build_model(config)\n",
    "history = ConvMixGRU.train(x_train, y_train, x_valid, y_valid, config, workpath)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "56/56 [==============================] - 1s 9ms/step - loss: 1.6530 - accuracy: 0.6242\n",
      "Raw Acc result: 0.6241534948348999\n",
      "Old model exists. Comparing performance.\n",
      "New model is worse than the old one. Will not update the old model\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "raw = ConvMixGRU.evaluate(x_test, y_test)\n",
    "print(\"Raw Acc result: {}\".format(raw[1]))\n",
    "replace_model_if_better('ConvMixGRU.pickle', np.mean(raw[1]), ConvMixGRU, config)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhgAAAGHCAYAAADyXCsbAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8pXeV/AAAACXBIWXMAAA9hAAAPYQGoP6dpAACLyUlEQVR4nO3dd1gUVxfA4d/SQYooSlEEbKhgib33rlGTGE3sUZMYo9EYU0wzMX7RmERN0zRborEklpjEhr13sfeGCIhYAOmw8/0x7CLSYZelnPd59mGZnZm9O5Q9e++552oURVEQQgghhDAgM1M3QAghhBAljwQYQgghhDA4CTCEEEIIYXASYAghhBDC4CTAEEIIIYTBSYAhhBBCCIOTAEMIIYQQBicBhhBCCCEMTgIMIYQQQhicBBjCJDQaTa5uO3fuLNDzfPLJJ2g0mnwdu3PnToO0oagbMWIE3t7eWT5+9+5drKyseOGFF7LcJyoqCjs7O/r06ZPr5128eDEajYYbN27kui2P02g0fPLJJ7l+Pp2QkBA++eQTAgMDMzxWkN8XQ0lKSsLNzQ2NRsNff/1l0rYIURAWpm6AKJ0OHDiQ7vvPPvuMHTt2sH379nTb69SpU6DnGT16NN27d8/XsQ0bNuTAgQMFbkNxV6FCBfr06cO6det48OABzs7OGfZZsWIFcXFxjBo1qkDP9dFHHzFhwoQCnSMnISEhfPrpp3h7e9OgQYN0jxXk98VQ/v33X+7cuQPAggUL6N+/v0nbI0R+SYAhTKJ58+bpvq9QoQJmZmYZtj8pNjYWOzu7XD9P5cqVqVy5cr7a6OjomGN7SotRo0axevVqli1bxrhx4zI8vnDhQlxdXenVq1eBnqdatWoFOr6gCvL7YigLFizAysqKdu3asWXLFoKDg03epsykpKSQnJyMtbW1qZsiiigZIhFFVvv27fH392f37t20bNkSOzs7Ro4cCcDKlSvp2rUr7u7u2NraUrt2bd577z1iYmLSnSOzLm9vb2969+7Npk2baNiwIba2ttSqVYuFCxem2y+zIZIRI0Zgb2/PlStX6NmzJ/b29nh6evLWW2+RkJCQ7vjg4GD69++Pg4MDZcuWZfDgwRw5cgSNRsPixYuzfe13795l7Nix1KlTB3t7eypWrEjHjh3Zs2dPuv1u3LiBRqPhq6++Yvbs2fj4+GBvb0+LFi04ePBghvMuXrwYX19frK2tqV27Nr/99lu27dDp1q0blStXZtGiRRkeO3/+PIcOHWLYsGFYWFgQEBBA3759qVy5MjY2NlSvXp1XX32ViIiIHJ8nsyGSqKgoXn75ZcqXL4+9vT3du3fn0qVLGY69cuUKL730EjVq1MDOzo5KlSrx9NNPc/r0af0+O3fupEmTJgC89NJL+qE43VBLZr8vWq2WWbNmUatWLaytralYsSLDhg0jODg43X6639cjR47Qpk0b7OzsqFq1KjNnzkSr1eb42kHtXdm0aRNPP/00b7/9NlqtNsvflT/++IMWLVpgb2+Pvb09DRo0YMGCBen22bRpE506dcLJyQk7Oztq167NjBkz0rW5ffv2Gc795M9B93s2a9Yspk+fjo+PD9bW1uzYsYP4+HjeeustGjRogJOTE+XKlaNFixb8/fffGc6r1Wr57rvvaNCgAba2tpQtW5bmzZuzfv16QA1ky5UrR2xsbIZjO3bsiJ+fXy6uoigqJMAQRVpoaChDhgxh0KBBbNiwgbFjxwJw+fJlevbsyYIFC9i0aRMTJ05k1apVPP3007k678mTJ3nrrbd48803+fvvv6lXrx6jRo1i9+7dOR6blJREnz596NSpE3///TcjR45kzpw5fPHFF/p9YmJi6NChAzt27OCLL75g1apVuLq6MnDgwFy17/79+wBMnTqV//77j0WLFlG1alXat2+faU7IDz/8QEBAAHPnzmXZsmXExMTQs2dPIiMj9fssXryYl156idq1a7N69Wo+/PBDPvvsswzDUpkxMzNjxIgRHD9+nJMnT6Z7TBd06IK/q1ev0qJFC+bPn8+WLVv4+OOPOXToEK1btyYpKSlXr19HURT69evH77//zltvvcXatWtp3rw5PXr0yLBvSEgI5cuXZ+bMmWzatIkffvgBCwsLmjVrxsWLFwF12EvX3g8//JADBw5w4MABRo8enWUbXnvtNd599126dOnC+vXr+eyzz9i0aRMtW7bMEDSFhYUxePBghgwZwvr16+nRowdTpkxh6dKluXq9ixcvJiUlhZEjR9K5c2e8vLxYuHAhTy56/fHHHzN48GA8PDxYvHgxa9euZfjw4dy8eVO/z4IFC+jZsydarZYff/yRf/75hzfeeCNDYJQX3377Ldu3b+err75i48aN1KpVi4SEBO7fv8/kyZNZt24dy5cvp3Xr1jz77LMZAtgRI0YwYcIEmjRpwsqVK1mxYgV9+vTR5+FMmDCBBw8e8Mcff6Q77ty5c+zYsYPXX389320XJqAIUQQMHz5cKVOmTLpt7dq1UwBl27Zt2R6r1WqVpKQkZdeuXQqgnDx5Uv/Y1KlTlSd/zb28vBQbGxvl5s2b+m1xcXFKuXLllFdffVW/bceOHQqg7NixI107AWXVqlXpztmzZ0/F19dX//0PP/ygAMrGjRvT7ffqq68qgLJo0aJsX9OTkpOTlaSkJKVTp07KM888o99+/fp1BVDq1q2rJCcn67cfPnxYAZTly5criqIoKSkpioeHh9KwYUNFq9Xq97tx44ZiaWmpeHl55diGa9euKRqNRnnjjTf025KSkhQ3NzelVatWmR6j+9ncvHlTAZS///5b/9iiRYsUQLl+/bp+2/Dhw9O1ZePGjQqgfPPNN+nO+7///U8BlKlTp2bZ3uTkZCUxMVGpUaOG8uabb+q3HzlyJMufwZO/L+fPn1cAZezYsen2O3TokAIo77//vn6b7vf10KFD6fatU6eO0q1btyzbqaPVapXq1asrlSpV0v8sde15/G/g2rVrirm5uTJ48OAszxUdHa04OjoqrVu3TvfzflK7du2Udu3aZdj+5M9B93tWrVo1JTExMdvXoftdHTVqlPLUU0/pt+/evVsBlA8++CDb49u1a6c0aNAg3bbXXntNcXR0VKKjo7M9VhQt0oMhijRnZ2c6duyYYfu1a9cYNGgQbm5umJubY2lpSbt27QC1yz4nDRo0oEqVKvrvbWxsqFmzZrpPgFnRaDQZekrq1auX7thdu3bh4OCQIWHwxRdfzPH8Oj/++CMNGzbExsYGCwsLLC0t2bZtW6avr1evXpibm6drD6Bv08WLFwkJCWHQoEHphgC8vLxo2bJlrtrj4+NDhw4dWLZsGYmJiQBs3LiRsLAwfe8FQHh4OGPGjMHT01Pfbi8vLyB3P5vH7dixA4DBgwen2z5o0KAM+yYnJ/P5559Tp04drKyssLCwwMrKisuXL+f5eZ98/hEjRqTb3rRpU2rXrs22bdvSbXdzc6Np06bptj35u5GVXbt2ceXKFYYPH67/WeqGcR4fvgsICCAlJSXbT/P79+8nKiqKsWPHGnRWTJ8+fbC0tMyw/c8//6RVq1bY29vrf+YLFixId903btwIkGMvxIQJEwgMDGTfvn2AOkT2+++/M3z4cOzt7Q32WoTxSYAhijR3d/cM2x49ekSbNm04dOgQ06dPZ+fOnRw5coQ1a9YAEBcXl+N5y5cvn2GbtbV1ro61s7PDxsYmw7Hx8fH67+/du4erq2uGYzPblpnZs2fz2muv0axZM1avXs3Bgwc5cuQI3bt3z7SNT74eXeKdbt979+4B6hvgkzLblpVRo0Zx7949/Zj5okWLsLe3Z8CAAYA6xt61a1fWrFnDO++8w7Zt2zh8+LA+HyQ31/dx9+7dw8LCIsPry6zNkyZN4qOPPqJfv378888/HDp0iCNHjlC/fv08P+/jzw+Z/x56eHjoH9cpyO+VLn/imWee4eHDhzx8+BAnJydat27N6tWrefjwIaDm5wDZJn7mZp/8yOw6rFmzhgEDBlCpUiWWLl3KgQMHOHLkCCNHjkz3N3H37l3Mzc1z/H3r27cv3t7e/PDDD4A6bBQTEyPDI8WQzCIRRVpmn762b99OSEgIO3fu1PdaAPp/wEVB+fLlOXz4cIbtYWFhuTp+6dKltG/fnvnz56fbHh0dne/2ZPX8uW0TwLPPPouzszMLFy6kXbt2/PvvvwwbNkz/yfLMmTOcPHmSxYsXM3z4cP1xV65cyXe7k5OTuXfvXro378zavHTpUoYNG8bnn3+ebntERARly5bN9/ODmgv05Jt1SEgILi4u+TrvkyIjI1m9ejWAPgn1SX/88Qdjx46lQoUKgJpE7Onpmem+j++THRsbm3R5OjpZJeRm9ve4dOlSfHx8WLlyZbrHn0x6rlChAikpKYSFhWUaqOiYmZnx+uuv8/777/P1118zb948OnXqhK+vb7avRRQ90oMhih3dP7Enp8f99NNPpmhOptq1a0d0dLS+W1hnxYoVuTpeo9FkeH2nTp3KUD8kt3x9fXF3d2f58uXpEgZv3rzJ/v37c30eGxsbBg0axJYtW/jiiy9ISkpKNzxi6J9Nhw4dAFi2bFm67U8mAeqe+8nn/e+//7h9+3a6bU/27mRHNzz3ZJLmkSNHOH/+PJ06dcrxHLnxxx9/EBcXp68H8+TNxcVFP0zStWtXzM3NMwSfj2vZsiVOTk78+OOPGRJEH+ft7c2lS5fSBQP37t3L0++ERqPBysoqXXARFhaWYRaJLjE3u3brjB49GisrKwYPHszFixcznRotij7pwRDFTsuWLXF2dmbMmDFMnToVS0tLli1blmF2gykNHz6cOXPmMGTIEKZPn0716tXZuHEjmzdvBtRPadnp3bs3n332GVOnTqVdu3ZcvHiRadOm4ePjQ3Jycp7bY2Zmxmeffcbo0aN55plnePnll3n48CGffPJJnoZIQB0m+eGHH5g9eza1atVKl8NRq1YtqlWrxnvvvYeiKJQrV45//vmHgICAPLcZ1DfTtm3b8s477xATE0Pjxo3Zt28fv//+e4Z9e/fuzeLFi6lVqxb16tXj2LFjfPnllxl6HqpVq4atrS3Lli2jdu3a2Nvb4+HhgYeHR4Zz+vr68sorr/Ddd99hZmZGjx49uHHjBh999BGenp68+eab+XpdT1qwYAHOzs5Mnjw5w/AbwLBhw5g9ezYnT56kfv36vP/++3z22WfExcXx4osv4uTkxLlz54iIiODTTz/F3t6er7/+mtGjR9O5c2defvllXF1duXLlCidPnuT7778HYOjQofz0008MGTKEl19+mXv37jFr1iwcHR1z3fbevXuzZs0axo4dS//+/bl16xafffYZ7u7uXL58Wb9fmzZtGDp0KNOnT+fOnTv07t0ba2trTpw4gZ2dHePHj9fvW7ZsWYYNG8b8+fPx8vLK9ewwUcSYOMlUCEVRsp5F4ufnl+n++/fvV1q0aKHY2dkpFSpUUEaPHq0cP348w+yArGaR9OrVK8M5n8yoz2oWyZPtzOp5goKClGeffVaxt7dXHBwclOeee07ZsGFDhtkUmUlISFAmT56sVKpUSbGxsVEaNmyorFu3Lsvs/i+//DLDOchklsWvv/6q1KhRQ7GyslJq1qypLFy4MMM5c+Opp55SAGXWrFkZHjt37pzSpUsXxcHBQXF2dlaef/55JSgoKEN7cjOLRFEU5eHDh8rIkSOVsmXLKnZ2dkqXLl2UCxcuZDjfgwcPlFGjRikVK1ZU7OzslNatWyt79uzJdKbE8uXLlVq1aimWlpbpzpPZzzElJUX54osvlJo1ayqWlpaKi4uLMmTIEOXWrVvp9svq9zWn63vy5EkFUCZOnJjlPrrXO378eP223377TWnSpIliY2Oj2NvbK0899VSGmTEbNmxQ2rVrp5QpU0axs7NT6tSpo3zxxRfp9lmyZIlSu3ZtxcbGRqlTp46ycuXKPP2eKYqizJw5U/H29lasra2V2rVrK7/88kuW13LOnDmKv7+/YmVlpTg5OSktWrRQ/vnnnwzn3LlzpwIoM2fOzPK6iKJNoyjZ9J8JIQzq888/58MPPyQoKKhIVmcUoqh46623mD9/Prdu3co0eVYUfTJEIoSR6Lqha9WqRVJSEtu3b+fbb79lyJAhElwIkYWDBw9y6dIl5s2bx6uvvirBRTEmPRhCGMnChQuZM2cON27cICEhgSpVqjBo0CA+/PBDrKysTN08IYokjUaDnZ0dPXv21E+DFsWTBBhCCCGEMDiZpiqEEEIIg5MAQwghhBAGJwGGEEIIIQyu1M0i0Wq1hISE4ODgYNBFgIQQQoiSTlEUoqOj8fDwyLFgYKkLMEJCQrKs3y+EEEKInN26dSvH6falLsBwcHAA1IuTl3K4QgghRGkXFRWFp6en/r00O6UuwNANizg6OkqAIYQQQuRDblIMJMlTCCGEEAYnAYYQQgghDE4CDCGEEEIYXKnLwcgNRVFITk4mJSXF1E0RwuDMzc2xsLCQadpCCKOSAOMJiYmJhIaGEhsba+qmCGE0dnZ2uLu7y6JrQgijkQDjMVqtluvXr2Nubo6HhwdWVlbyKU+UKIqikJiYyN27d7l+/To1atTIsViOEELkhwQYj0lMTESr1eLp6YmdnZ2pmyOEUdja2mJpacnNmzdJTEzExsbG1E0SQpRA8tElE/KJTpR08jsuhDA2+S8jhBBCCIOTAEMIIYQogcKj4pm38wqKopjk+SXAEFlq3749EydOzPX+N27cQKPREBgYaLQ2CSGEyNneyxH0/HYPszZdZPnhWyZpgwQYJYBGo8n2NmLEiHydd82aNXz22We53t/T05PQ0FD8/f3z9Xz50bVrV8zNzTl48GChPacQQhRVKVqF2VsuMnThISIeJVLLzYHmVcuZpC0yi6QECA0N1d9fuXIlH3/8MRcvXtRvs7W1Tbd/UlISlpaWOZ63XLm8/VKam5vj5uaWp2MKIigoiAMHDjBu3DgWLFhA8+bNC+25M5Pb6yqEEMYQHhXPGytOcPDafQAGNavCx73rYGNpbpL2SA9GDhRFITYx2SS33I6bubm56W9OTk5oNBr99/Hx8ZQtW5ZVq1bRvn17bGxsWLp0Kffu3ePFF1+kcuXK2NnZUbduXZYvX57uvE8OkXh7e/P5558zcuRIHBwcqFKlCj///LP+8SeHSHbu3IlGo2Hbtm00btwYOzs7WrZsmS74AZg+fToVK1bEwcGB0aNH895779GgQYMcX/eiRYvo3bs3r732GitXriQmJibd4w8fPuSVV17B1dUVGxsb/P39+ffff/WP79u3j3bt2mFnZ4ezszPdunXjwYMH+tc6d+7cdOdr0KABn3zyif57jUbDjz/+SN++fSlTpgzTp08nJSWFUaNG4ePjg62tLb6+vnzzzTcZ2r5w4UL8/PywtrbG3d2dcePGATBy5Eh69+6dbt/k5GTc3NxYuHBhjtdECGFa9x4lMGXNaQLO3SnU591z+S49v93DwWv3KWNlzjcvNODzZ+qaLLgA6cHIUVxSCnU+3myS5z43rRt2Vob5Eb377rt8/fXXLFq0CGtra+Lj42nUqBHvvvsujo6O/PfffwwdOpSqVavSrFmzLM/z9ddf89lnn/H+++/z119/8dprr9G2bVtq1aqV5TEffPABX3/9NRUqVGDMmDGMHDmSffv2AbBs2TL+97//MW/ePFq1asWKFSv4+uuv8fHxyfb1KIrCokWL+OGHH6hVqxY1a9Zk1apVvPTSS4BaNK1Hjx5ER0ezdOlSqlWrxrlz5zA3V//YAgMD6dSpEyNHjuTbb7/FwsKCHTt25Lk8/NSpU5kxYwZz5szB3NwcrVZL5cqVWbVqFS4uLuzfv59XXnkFd3d3BgwYAMD8+fOZNGkSM2fOpEePHkRGRuqvx+jRo2nbti2hoaG4u7sDsGHDBh49eqQ/XghRNIVHxTP410NcDn/EzovhdK5d0ejFGlO0Ct9svcR3O66gKFDb3ZEfBj1F1Qr2Rn3e3JAAo5SYOHEizz77bLptkydP1t8fP348mzZt4s8//8w2wOjZsydjx44F1KBlzpw57Ny5M9sA43//+x/t2rUD4L333qNXr17Ex8djY2PDd999x6hRo/SBwccff8yWLVt49OhRtq9n69atxMbG0q1bNwCGDBnCggUL9OfZunUrhw8f5vz589SsWROAqlWr6o+fNWsWjRs3Zt68efptfn5+2T5nZgYNGsTIkSPTbfv000/19318fNi/fz+rVq3SBwjTp0/nrbfeYsKECfr9mjRpAkDLli3x9fXl999/55133gHUnprnn38ee3vT/8MQQmTu9sM4Bv9ykBv31GUmQiPjuRYRQzUjvtHfiYrnjeUnOHS9aAyJPEkCjBzYWppzblo3kz23oTRu3Djd9ykpKcycOZOVK1dy+/ZtEhISSEhIoEyZMtmep169evr7uqGY8PDwXB+j+1QeHh5OlSpVuHjxoj5g0WnatCnbt2/P9pwLFixg4MCBWFiov8Ivvvgib7/9NhcvXsTX15fAwEAqV66sDy6eFBgYyPPPP5/tc+TGk9cV4Mcff+TXX3/l5s2bxMXFkZiYqB/yCQ8PJyQkhE6dOmV5ztGjR/Pzzz/zzjvvEB4ezn///ce2bdsK3FYhRPaSUrSERcZTqawtZma573m4ERHD4F8PcfthHJ7lbLG3tuR8aBR7L0cYLcA4HvSAl5cc5V5MImWszPn82br0bVDJKM+VXxJg5ECj0RhsmMKUngwcvv76a+bMmcPcuXOpW7cuZcqUYeLEiSQmJmZ7nieTGDUaDVqtNtfH6LoLHz/myS7EnHJP7t+/z7p160hKSmL+/Pn67SkpKSxcuJAvvvgiQ2Lrk3J63MzMLEM7kpKSMuz35HVdtWoVb775Jl9//TUtWrTAwcGBL7/8kkOHDuXqeQGGDRvGe++9x4EDBzhw4ADe3t60adMmx+OEEHkXk5DMrkt3CTh3h23n7xAVn4x/JUfe71mbltVccjz+8p1oBv96iPDoBKq6lGHZy81Yc/w250Oj2HM5guEtvQ3e5r2XI3j5t6PEJaUUqSGRJ0mSZym1Z88e+vbty5AhQ6hfvz5Vq1bl8uXLhd4OX19fDh8+nG7b0aNHsz1m2bJlVK5cmZMnTxIYGKi/zZ07lyVLlpCcnEy9evUIDg7m0qVLmZ6jXr162fYKVKhQId3snKioKK5fv57j69mzZw8tW7Zk7NixPPXUU1SvXp2rV6/qH3dwcMDb2zvb5y5fvjz9+vVj0aJFLFq0SD/sI4QwjPDoeJYfDuKlRYd56rMAxi47ztoTt4mKTwbgzO0oBv1yiJGLj3D5TnSW5zlzO5KBPx8kPDqBWm4OrHy1Be5OtrSpoQYmB6/dIykl+w9gebXpTBgjFx8hLimFNjVcWP1aiyIZXID0YJRa1atXZ/Xq1ezfvx9nZ2dmz55NWFgYtWvXLtR2jB8/npdffpnGjRvTsmVLVq5cyalTp9LlSzxpwYIF9O/fP0O9DS8vL959913+++8/+vbtS9u2bXnuueeYPXs21atX58KFC2g0Grp3786UKVOoW7cuY8eOZcyYMVhZWbFjxw6ef/55XFxc6NixI4sXL+bpp5/G2dmZjz76SJ8gmp3q1avz22+/sXnzZnx8fPj99985cuRIuqTVTz75hDFjxlCxYkV9Iuq+ffsYP368fp/Ro0fTu3dvUlJSGD58eD6urBDicY8SkllxOIgNp0M5ceshj3dQepW3o0ttV7r6ueHtYsf326/wx6Egtl8IZ+fFcAY28eTNzjWp6Ji2MOCJoAcMX3iYqPhk6lV2YslLTXEuYwWAn4cTTraWRMYlcSr4IY28DFOHYvWxYN5ZfYoUrUIPfzfmvtAAa4uikW+RGQkwSqmPPvqI69ev061bN+zs7HjllVfo168fkZGRhdqOwYMHc+3aNSZPnkx8fDwDBgxgxIgRGXo1dI4dO8bJkyf55ZdfMjzm4OBA165dWbBgAX379mX16tVMnjyZF198kZiYGKpXr87MmTMBqFmzJlu2bOH999+nadOm2Nra0qxZM1588UUApkyZwrVr1+jduzdOTk589tlnuerBGDNmDIGBgQwcOBCNRsOLL77I2LFj2bhxo36f4cOHEx8fz5w5c5g8eTIuLi70798/3Xk6d+6Mu7s7fn5+eHh45Pp6CiHSexibyKJ9N1i8/waRcWnDnPUrO9Gljitd6rhR09U+3VDttL7+jGjpzRebLrD57B2WH77F34EhvNK2Ki+3qcrp25GMWnyEmMQUGns5s/ClJjjapA0Fm5tpaFW9PBtOh7HncoRBAozF+67zyT/nAHi+UWVmPFsXC/OiPQihUUxVpNxEoqKicHJyIjIyEkdHx3SPxcfHc/36dXx8fGQJaxPq0qULbm5u/P7776ZuisnExsbi4eHBwoULM8z+MQT5XReFQatViI5PBg042RZuEbq70Qn8uvcaSw/cJCZRnX5etUIZRrT0pmsdN9yccvd7f+TGff7333kCbz0EoIKDNdHxScQnaWlVvTy/DGucaZ7eH4eCeH/taRp7OfPXay3z/ToUReG77VeYHaAO945s5cOHvWrnKQnVkLJ7D32S9GAIk4qNjeXHH3+kW7dumJubs3z5crZu3UpAQICpm2YSWq2WsLAwvv76a5ycnOjTp4+pmyRElm7dj+XfU6GERsbxMDaJh3FJRMYm8jAuiYexSUTFJ6EooNHAslHNaFk956TJggp5GMfPu6+x/HAQCclq/kMtNwfGdaxOD393zPP4xtzEuxxrx7Zkw+kwvth0gaD76jTUjrUqMm9wwyynhOryME7cekh0fBIONnkPsBRF4X//nefXvWrv6Zuda/JGp+pGr61hKBJgCJPSaDRs2LCB6dOnk5CQgK+vL6tXr6Zz586mbppJBAUF4ePjQ+XKlVm8eLF+Gq4QRUV8Ugqbz4ax8sgt9l+9l6tjFAVWHb1l1ADj9sM4vt16mTUngklKUTvmG3iWZXzH6nSsVbCCVxqNhl713OlcpyKrjtziQWwSY9pVw8oi6yEKz3J2eJW34+a9WA5eu0+XOq55es4UrcL7a06z8qi6UNnHveswsnX2BQiLGvnvJUzK1taWrVu3mroZRYa3t7fJllYWIjtnbkey8sgt/g5Mm22h0UDr6i485VkWJzsrnGwtKWtrSVk79eZoa8mV8EcM+uUQ2y6Ek5SixdIIeQOngyMZsegw92LUafbNq5ZjfMcatKxW3qCf9q0tzBnawjvX+7eq7sLNe0HsuxKRpwBDURTeWhXIusAQzDTwxXP1eL6xZz5abFoSYAghhMhUZGwS6wJvs/LILc6FRum3Vypry/ONK9O/UWUqO9tle47yZaxxsbci4lEih67dp3UNw/Zi7L0cwau/HyUmMYU67o5M6+tHY2/TrB76pDbVXfjjUBB7Lt/N03GngiNZFxiChZmG7wc9RXd/dyO10LgkwBBCCJHOg5hEft17jSX7b/IoQe2tsDI3o6ufKwObeNKqmkuukwzNzTR0ru3KiiO32HIuzKABxr+nQnhzZSBJKQotq5Xnp6GN8pXrYCwtq7lgpoGrd2MIjYzD3SnnQnsAyw7dBODp+h7FNrgACTCEEEKkuh+TyK97rrFk/w39zIuarva82LQK/RpU0td5yKuufqkBxtk7fNrHzyDDFr8duMHU9WdRFOhZ1405A4teTQgnO0vqVi7LyVsP2XM5ggG5GOaIjE1i/ckQAIY0r2LsJhqVBBhCCFHK3Y9J5Jc91/jtscCijrsjEzrXoEtt1wJPiWxZzYUyVuaERcVz+nYk9SqXzfe5FEVhztbLfLtNrTw8tLkXn/Txy/PskMLSproLJ289ZN+V3AUYa04EE5+kpZabAw2rOBdCC41HAgwhhChGtFqFvVciCH4QR/9GlbOdyZCTe48S+GXPdX47cIPYxwKLiZ1r0KWOq8ESJG0szWnvW5H/Toey5eydfAcYKVqFj/4+wx+HggCY2LkGEzrVKNLTNltVd+H7HVfYdyUCrVbJNlhTFIVlqa9tcLMqRfp15YYEGEIIUQzEJCSz5ngwi/ff4OrdGADWnbjND4MbUsHBOk/nUhSFxftv8OXmi/rAws/DkYmda9K5dsGmdGalq58r/50OZfPZMCZ3883z8fFJKby5MpCNZ8LQaOCzvv4Mae5l8HYaWkOvsthamhPxKJELYdHU8ci6ONWh6/e5Ev4IOytz+j1VtFZGzY+iXWdUFKr27dszceJE/ffe3t7MnTs322M0Gg3r1q0r8HMb6jxCFGWKovD7wZt8tfkiAefucDc6Icdjbt2P5X//naPFjG189PdZrt6Nwd7aAntrCw7fuE+f7/dyMrXKZG48iEnk5d+O8uk/54hNTMG/kiO/DmvMv+NbG7TX4kntfStiYabhcvgjrt19lKdjYxKSGbHoMBvPhGFlbsa8QQ2LRXAB6tTWZlXVWS17r2Q/m0TXe9G3QaUilayaX9KDUQI8/fTTxMXFZVpP4sCBA7Rs2ZJjx47RsGHDPJ33yJEjGZYjL6hPPvmEdevWERgYmG57aGgozs6FM94YFxeHh4cHGo2G27dv52oJdSEMYd7Oq3y5+WK6bZXK2tLAs6x6q1IWfw8nbCzNOHT9Pov2XSfg3B20qaVRvMvbMaKlN/0be3InKp6XfzvKtbsxPP/TAT5/pi79G1XO9vkPXbvHhBWBhEXFY2Vuxge9ajOshVehdMU72VrSolp59lyOIODcHV5tl/sVQOftvMLBa/ext7bg52GNcrWMelHSuroLOy/eZc/lCF5pWy3Tfe5GJ7DpjLqC8+BmxTu5U0cCjBJg1KhRPPvss9y8eRMvr/RR/cKFC2nQoEGegwtQlywvLG5uboX2XKtXr8bf3x9FUVizZg2DBw8utOd+kqIopKSkSMXOUmDN8WB9cNGpVkVuPYjlcvgjbj+M4/bDOP47rb65mJtpqOhgTWhkvP7YNjVceKmVN+1rVtSP4dtXsGfd662YtDKQrefDmfznSc7cjuSDXrUzFLNK0Sp8t11NjNQqUNWlDN8Nego/D6dCevWqrnVc2XM5gi3n7vBqu8zfaJ8UGZfEb/vVaZtfPV+/2AUXAG1qVADOc+TGfeKTUjItL/7nsVskpSg08CyLf6XC/bkYiwyR5ERRIDHGNLdcVnTs3bs3FStWZPHixem2x8bGsnLlSkaNGsW9e/d48cUXqVy5MnZ2dtStW5fly5dne94nh0guX75M27ZtsbGxoU6dOpmuF/Luu+9Ss2ZN7OzsqFq1Kh999BFJSeoKhosXL+bTTz/l5MmTaDQaNBqNvs1PDpGcPn2ajh07YmtrS/ny5XnllVd49CitW3XEiBH069ePr776Cnd3d8qXL8/rr7+uf67sLFiwgCFDhjBkyBAWLFiQ4fGzZ8/Sq1cvHB0dcXBwoE2bNly9elX/+MKFC/Hz88Pa2hp3d3fGjRsHwI0bN9BoNOl6Zx4+fIhGo2Hnzp0A7Ny5E41Gw+bNm2ncuDHW1tbs2bOHq1ev0rdvX1xdXbG3t6dJkyYZeqQSEhJ455138PT0xNramho1arBgwQIURaF69ep89dVX6fY/c+YMZmZm6douTGPv5Qje+esUAK+0rcqCEU3Y8mY7Tk3tyh8vN+Od7r50reNKRQdrUrQKoZHx2FiaMahZFQLebMvvo5rRsVbG2RyONpb8PLQxEzrVAGDx/hsM+fUQ9x6lDb2ERsYx6JeDzN2qBhfPNazMP+NbF3pwAdA5tZrl8aAHhEfH57C36rf9N4hOSMbX1YGueSy3XVTUdLWngoM18Ulajt98kOFxrVbRJ66WlN4LkB6MnCXFwucmWi77/RCwynmIwsLCgmHDhrF48WI+/vhjfXfnn3/+SWJiIoMHDyY2NpZGjRrx7rvv4ujoyH///cfQoUOpWrUqzZo1y/E5tFotzz77LC4uLhw8eJCoqKh0+Ro6Dg4OLF68GA8PD06fPs3LL7+Mg4MD77zzDgMHDuTMmTNs2rRJ/+bp5JTxn1xsbCzdu3enefPmHDlyhPDwcEaPHs24cePSBVE7duzA3d2dHTt2cOXKFQYOHEiDBg14+eWXs3wdV69e5cCBA6xZswZFUZg4cSLXrl2jatWqANy+fZu2bdvSvn17tm/fjqOjI/v27SM5WS02NH/+fCZNmsTMmTPp0aMHkZGR7Nu3L8fr96R33nmHr776iqpVq1K2bFmCg4Pp2bMn06dPx8bGhiVLlvD0009z8eJFqlRR/+EMGzaMAwcO8O2331K/fn2uX79OREQEGo2GkSNHsmjRIiZPnqx/joULF9KmTRuqVcvdJ0VhHOdCohiz9BjJWoWn63vwXvda+sccbCxpWc1F/6lcURTCouK5Gh6DfyVHytrlXHfCzEzDm11qUsfDkUkrAzl0/T59vt/HT0MbcScqnsl/nuRBbBJlrMyZ/ow/zzyV/TCKMbk72VK/shMngyPZdj6cF5tm/2Yak5DMwn3qQl9jO1Qz2QqiBaXRaGhd3YW1J26z50pEhjVZdl2+S/CDOBxtLOhdz0TvN0YgAUYJMXLkSL788kt27txJhw4dAPRLfTs7O+Ps7JzuzWf8+PFs2rSJP//8M1cBxtatWzl//jw3btygcmX1H9Tnn39Ojx490u334Ycf6u97e3vz1ltvsXLlSt555x1sbW2xt7fHwsIi2yGRZcuWERcXx2+//abPAfn+++95+umn+eKLL3B1VT/FODs78/3332Nubk6tWrXo1asX27ZtyzbAWLhwIT169NDne3Tv3p2FCxcyffp0AH744QecnJxYsWIFlpZqklXNmjX1x0+fPp233nqLCRMm6Lc1adIkx+v3pGnTptGlSxf99+XLl6d+/frpnmft2rWsX7+ecePGcenSJVatWkVAQIB+IThdUATw0ksv8fHHH3P48GGaNm1KUlISS5cu5csvv8xz24Th3H4Yx0uLD/MoIZnmVcvx1fP1sn2T1Gg0uDvZ5rri4+O6+bmx7vVWvPL7Ma5HxPDsvP0kpqirifpXcuS7Fxvi42LYnKr86OrnxsngSLacDcsxwFh+OIgHsUl4lbejV93iW9ES0AcYey9H8G739I8tO6j2XvRv5ImtVdEqFlYQJg8w5s2bx5dffkloaCh+fn7MnTuXNm3aZLn/smXLmDVrFpcvX8bJyYnu3bvz1VdfUb58eeM00NJO7UkwBcvsa/w/rlatWrRs2ZKFCxfSoUMHrl69yp49e9iyZQsAKSkpzJw5k5UrV3L79m0SEhJISEjIdRLn+fPnqVKlij64AGjRokWG/f766y/mzp3LlStXePToEcnJyTg6Zj0tK6vnql+/frq2tWrVCq1Wy8WLF/UBhp+fH+bmaX+M7u7unD59OsvzpqSksGTJEr755hv9tiFDhvDmm2/y6aefYm5uTmBgIG3atNEHF48LDw8nJCSETp065en1ZKZx48bpvo+JieHTTz/l33//JSQkhOTkZOLi4ggKUv/xBAYGYm5uTrt27TI9n7u7O7169WLhwoU0bdqUf//9l/j4eJ5//vkCt1XkT2RsEsMXHuZOVAK+rg78NLSx0StN1nB1YN3rrZi44gQ7LqozFka19uGd7r5Fpspl1zqufLn5Ivuu3ONRQjL21pm/DSUkp/Dz7msAvNauGhZGWCStMOlKpJ8JieRBTKK+KmrIwzi2X7gDwKASNDwCJs7BWLlyJRMnTuSDDz7gxIkTtGnThh49euj/qT5p7969DBs2jFGjRnH27Fn+/PNPjhw5wujRo43XSI1GHaYwxS2Pmd2jRo1i9erVREVFsWjRIry8vPRvhl9//TVz5szhnXfeYfv27QQGBtKtWzcSExNzde7MVvh8MvP84MGDvPDCC/To0YN///2XEydO8MEHH+T6OR5/rqyy2h/f/mQQoNFo0Gq1WZ538+bN3L59m4EDB2JhYYGFhQUvvPACwcHB+kAsuxklOc02MTMz07dfJ6uckCcDu7fffpvVq1fzv//9jz179hAYGEjdunX11y43M11Gjx7NihUriIuLY9GiRQwcOBA7u9wHqSVdZGzO+TmGEp+Uwsu/H+VK+CPcHG1Y9FITnGwLZ9qhk60lvw5vwlfP12f5y835qHedIhNcAFSvaI+PSxkSU7Tsupj1tM2/jgUTHp2Au5MNzzY03bCOobg62lDT1R5FId0y9ysOB6FVoEXV8lSvmPuZNcWBSQOM2bNnM2rUKEaPHk3t2rWZO3cunp6ezJ8/P9P9Dx48iLe3N2+88QY+Pj60bt2aV199laNHjxZyy4umAQMGYG5uzh9//MGSJUt46aWX9G/Ie/bsoW/fvgwZMoT69etTtWpVLl++nOtz16lTh6CgIEJC0npzDhw4kG6fffv24eXlxQcffEDjxo2pUaMGN2/eTLePlZUVKSkpOT5XYGAgMTEx6c5tZmaWbrgirxYsWMALL7xAYGBgutvgwYP1yZ716tVjz549mQYGDg4OeHt7s23btkzPr5t1Exoaqt/25HTcrOzZs4cRI0bwzDPPULduXdzc3Lhx44b+8bp166LVatm1a1eW5+jZsydlypRh/vz5bNy4kZEjR+bquUu6+zGJjFx8hPrTtugT6YxJq1V468+THL5+HwdrCxaPbIJH2cKdCm1upqF/o8q0qGaknt0C0Gg0+mTNLefCMt0nOUXLj7vU5ORX2lYtULXSoqR1dfV/hK4eRlKKlhVHbgEwuJivO5IZk/3UEhMTOXbsGF27dk23vWvXruzfvz/TY1q2bElwcDAbNmxAURTu3LnDX3/9Ra9evbJ8noSEBKKiotLdSip7e3sGDhzI+++/T0hICCNGjNA/Vr16dQICAti/fz/nz5/n1VdfJSws8z/uzHTu3BlfX1+GDRvGyZMn2bNnDx988EG6fapXr05QUBArVqzg6tWrfPvtt6xduzbdPt7e3ly/fp3AwEAiIiJISMhYaGjw4MHY2NgwfPhwzpw5w44dOxg/fjxDhw7VD4/k1d27d/nnn38YPnw4/v7+6W7Dhw9n/fr13L17l3HjxhEVFcULL7zA0aNHuXz5Mr///jsXL6rTCz/55BO+/vprvv32Wy5fvszx48f57rvvALWXoXnz5sycOZNz586xe/fudDkp2alevTpr1qwhMDCQkydPMmjQoHS9Md7e3gwfPpyRI0eybt06rl+/zs6dO1m1apV+H3Nzc0aMGMGUKVOoXr16pkNYpc3Ba/fo8c1utl8IB+DLzReIjjduT8bnG87z36lQLM01/DS0EbXc8jZEWBp09VP/jrdfCCcxOWOv4/qTIdy6H0f5Mla80KTkvPG2rqEGfHsuR6AoCtvO3yE8OgEXe2u61im8qfqFxWQBRkREBCkpKRneMFxdXbN842vZsiXLli1j4MCBWFlZ4ebmRtmyZfX/4DMzY8YMnJyc9DdPz5wXmynORo0axYMHD+jcubN+9gHARx99RMOGDenWrRvt27fHzc2Nfv365fq8ZmZmrF27loSEBJo2bcro0aP53//+l26fvn378uabbzJu3DgaNGjA/v37+eijj9Lt89xzz9G9e3c6dOhAhQoVMp0qa2dnx+bNm7l//z5NmjShf//+dOrUie+//z5vF+MxuoTRzPInOnTogIODA7///jvly5dn+/btPHr0iHbt2tGoUSN++eUX/XDM8OHDmTt3LvPmzcPPz4/evXun6wlauHAhSUlJNG7cmAkTJuiTR3MyZ84cnJ2dadmyJU8//TTdunXLULtk/vz59O/fn7Fjx1KrVi1efvnldL08oP78ExMTS33vRYpW4Zutlxn0y0HuRCVQrUIZvMvb8SA2iV/2XDfa8y7ad51f96rn/+r5+hlmCwhVA09nXOytiY5P5tD1e+ke02oV5u1Uey9GtfEpUUmPzXzKY2muIfhBHDfvxbI0NblzYJOCrSlTVGmUzAbXC0FISAiVKlVi//796T5p/e9//+P333/nwoULGY45d+4cnTt35s0336Rbt26Ehoby9ttv06RJk0zrGQD6ZEadqKgoPD09iYyMzJB8GB8fz/Xr1/Hx8cHGxsZAr1SIwrNv3z7at29PcHBwtr09Jfl3/U5UPBNXBHLgmvrG1b9RZab19WPXxbu8tuw4dlbm7H6nAy72eVu/Iyc7LoQzaskRtAq8270Wr7WX6cHZmbLmFMsP32Jocy8+6+ev377xdCivLTuOo40F+97rWCJKZj9uwE8HOHz9PqNb+/Dr3utoNLD77Q54lise+VJRUVE4OTll+h76JJOFTC4uLpibm2forQgPD8/yH+OMGTNo1aoVb7/9NvXq1aNbt27MmzePhQsXphv3fpy1tTWOjo7pbkKUNAkJCVy5coWPPvqIAQMG5HsoqbjbeTGcnt/s4cC1e9hZmTNnYH2+er4+dlYWdPd3o15lJ2ITU/hhxxWDPu/FsGjGLz+BVoEXmngypl3VnA8q5XRDAgHn7qBNrYWuKArfp/5sRrT0LnHBBajLtwMsSK3v0b5mhWITXOSVyQIMKysrGjVqlKEaZEBAAC1btsz0mNjYWH2mvo5umqKJOmKEKBKWL1+Or68vkZGRzJo1y9TNKXRJKVpmbDjPiEVHuBeTSB13R/4d3zpdUSmNRsPbqat4LjsYRPCDWIM8971HCYxacoRHCck08ynHtL7+xX6Z7cLQolp5yliZExYVz+nbkQDsvHSXsyFR2FmZ81IrHxO30Dh001V1b1nFZdG2/DDpoM+kSZP49ddfWbhwIefPn+fNN98kKCiIMWPGADBlyhSGDRum3//pp59mzZo1zJ8/n2vXrrFv3z7eeOMNmjZtiodHyal+JkRejRgxgpSUFI4dO0alSsV/mee8eJSQzMCfDvBTas2EYS28WDO2JVUrZJzy17q6Cy2qlicxRcs3W3M/iyorCckpjFl6jOAHcXiVt+PHIY1K5Fi6MdhYmtPetyKgziZRFIUftqu9F4OaVtHXiShp6lZywsFGrf3h4WSjvwYlkUn/EgYOHMjcuXOZNm0aDRo0YPfu3WzYsEG/YFdoaGi6mhgjRoxg9uzZfP/99/j7+/P888/j6+vLmjVrTPUShBAm9sn6sxwPeoiDjQU/DmnItL7+mS4mBam9GN3VXozVx4O5Eh6d7+dVFIX315zhyI0HONhYsGB4kxL7pmgsutkkW87e4dD1+xy9+QArczNebltyh5gszM1oV1Odrjq4uRfmxbT8eW6YLMnTVLJLUNElvnl7e8sS3qJEi4uL48aNG8U+yfOfkyGMX34CMw2seKUFTX3K5eq4l387SsC5O/Twd2P+kEb5eu4fd11l5sYLmJtpWDSiCW1rFt7qwyVFVHwSjT4LIClFoaarPZfuPGJwsyr875m6pm6aUUU8SmD3pbv0qe9R7CqUFoskz6JINxUxNtYwY7NCFFW63/HMSqIbSnKKltXHglmy/wa3H8YZ/PzBD2J5f61aGv71DtVzHVwATO7qi0YDG8+EcSr4YZ6fe/PZML7YpM50m/p0HQku8snRxpLmVdXaEJfuPMLcTMOYXC7jXpy52FvzbMPKxS64yCuTr0VSlJibm1O2bFnCw9WiPHZ2dpKsJUoURVGIjY0lPDycsmXLplvLxZAuhkXz9l8nORWsJu9NXX+W+pWd6FHXnR7+bniVL9iiWylahUkrTxIdn0wDz7K8kbpceW75ujnwTINKrDlxmy83X+T3UTkv+KdzNiSSN1cGoigwtLkXw1p457H14nFd/dzYczkCgL4NPErsjIrSSAKMJ+hW+dQFGUKURGXLls12Rdv8SkrRMn/nVb7bfpmkFAVHGwt83Rw4evMBJ4MjORkcycyNF6jj7kjPum5093fP1/oL83Zc4fCN+5SxMuebFxpgmY9Pgm92qck/p0LYczmC/Vcj9EumZyc8Op6XlxwlNjGF1tVdmPp0nTw/r0ivS21Xpv59BgUYK7VDShTJwchCSkpKlgtVCVGcWVpa5thzEZeYwuL9N3B1tKZjrYqUtcs5efHM7Uje/usU50PVcvyda7vyv2f8cXW0ITw6ni1n77DxTCgHr90nRZv2b6emqz0TOtWkV73cLcd9POgBz/94gBStwuwB9Qu0ENbHf5/htwM3aeBZlrVjW2bbYxl0L5Y3Vpwg8NZDqlYow9rXWuFkV/LqNJjCjtRS7h1qldwZFSVFXnIwJMAQQqSTnKJlzNLjbD2vLiFtbqahqXc5uvq50qWOK5Wd03dhJySn8N22K8zfdZUUrUJZO0s+7eNHn/oemb5h349JJOBcGBvPhLHvSgRJKeq/oH4NPPi0r3+2q45GxyfR69u9BN2PpU99D755oUGBhjHDo+NpN2sncUkp/DS0Ed380vfqKIrCgav3WLjvBtsu3EFR1NVK173eCh+Xgg3zCFEcSYCRDQkwhMiaoih8sO4MfxwKwsrCDJ/yZbh4J/1UzjrujnSp40pXP1eSUhTe+eskl+48AqBnXTc+7eNPBYfcleGOjEvi1z3X+GHHFbSKWhfgqwH1sxyumLQqkDXHb1OprC0bJ7bB0QCVHmdtusC8nVepUdGeTRPbYm6mIT4phXUnbrN4/w0uhKW9/rY1K/BON1/8KzkV+HmFKI4kwMiGBBhCZO377Zf5asslNBqYP7gh3f3dCboXy5ZzYWw5d4ejN+6jzeQ/hou9FdP6+tOzbu6GOZ50POgBk1YGcuOeOrtldGsfJnfzTVfPYv3JEN5InZK66tUWNPbO/ayR7ETGJtFm1nai4pN5v2ctIuOS+ONQEA9i1SFSW0tznmtUiREtvale0cEgzylEcSUBRjYkwBAic38evcXbf50C4NM+fgxv6Z1hn/sxiWw7f4ct5+6w5/Jd4pO09GvgwcdP+1GugEWmYhKS+d+G8/xxSC2uV9PVnjkDG+Dn4UTwg1h6fLOH6Phk3uhUg0ldahbouZ40b+cVZm26mG5bpbK2DG/pxcDGVSTXQohUEmBkQwIMITLaeTGcUUuOkqJVGNOuGu/1qJXjMXGJKUQnJFHRwbCFurZfuMM7f50i4lEiluYaJnXxZfuFOxy58YCGVcqy6tUWBq8fEJeYQufZu7j9MI6mPuUY2cqbzrVdS3ydAiHySgKMbEiAIUR6p4MjGfjzAWITU3jmqUp8/Xx9zExcvvjeowSmrDnNlnN39NscrC3YMKGN0eok3I9JJDo+qcA1OoQoyaSSpxClWHhUPOtO3OZ8aFSOqwwH3YvlpcWH9XUdvniunsmDC4Dy9tb8NLQRs/rXo4yVmofxWT9/oxZhKlfGSoILIQxICm0JUUJotQrLDgcxa+MFohOSATWPoFPtinSq7UrzquWwtkhLmrz3KIHhiw4T8Uhd3nz+kIZFaiVQjUbDgMaetK9ZgbCoeOpVLmvqJgkh8kCGSIQoAS6ERTFlzWlOBD0EwKu8HWGR8SQka/X72FmZ06aGC51qu9KiannGL1eLRlUqa8uasS1xdSy+i54JIQpHXt5DpQdDiGIsLjGFb7Zd5tc910jWKthbW/B2N1+GNPciMVnLvisRbLsQzvYLd7gTlcDms3fYfDYtr8HJ1pIlI5tIcCGEMDgJMIQopnZdusuH605z6766Uml3Pzc+6eOHm5MaLNhamdO5jiud67iiKP6cuR3F1vN32HbhDmduR2FjacaC4Y2ltoMQwihkiESIYiY8Op7p/55n/ckQANydbJjW158udVxzf46oeAAqSs+FECIPZIhEiBLq2M37jFx8lMi4JMw08FIrH97sUhN767z9KUtgIYQwNgkwhCgm9l6O4OXfjhKXlEIdd0e+eK4edSvLmhhCiKJJAgwhioGAc3d4fdlxElO0tKnhwk9DG2FnJX++QoiiS/5DCVHE/R14m0mrTpKiVejm58q3Lz6Vrp6FEEIURRJgCFGELT8cxPtrT6Mo8MxTlfiyfz1ZH0MIUSxIgCFEEfXrnmtM/+88AIObVeGzvv5Fooy3EELkhgQYQhQxiqLwzbbLzN16GYBX2lZlSo9aaDQSXAghig8JMIQoQhRF4fMN5/llz3UAJnWpyfiO1SW4EEIUOxJgCGEAsYnJBD+Iw8HGAncn2zwffz8mka3n77A+MIS9VyIA+Kh3HUa19jF0U4UQolBIgCFELiiKQtD9WILux3Lrfhy3HsRy634stx7EEXw/lnsxifp9Kzvb0tSnHE29y9HUpxw+LmUy7YEIfhDLlrN32Hw2jCM37qNNralrpoHPn6nLC02rFNbLE0IIg5MAQ4gcXAiL4v01pzmeulJpVhysLYhJ7ckIfnCbNcdvA+Bib01TH2eaepejlrsjR67fZ/O5MM7cjkp3fB13R7r6udK7nrusDyKEKPYkwBAiC0+uVGplbkaV8nZ4OtviWc4OT2c7PMvZUtlZve9kZ8mjhGSO33zA4ev3OXzjPoG3HhLxKIENp8PYcDos3fk1GmjiVY6ufq5083PDs5ydiV6pEEIYngQYQmRi58VwPvr7jH6l0m5+rnzSxy/H/Ap7awva1qxA25oVAIhPSuFUcCRHbtzn0PX7XAiNwr+SE938XOlU2xUXe2ujvxYhhDAFWU1ViMeER8fz2b/n+eexlUo/7eNHVz83E7dMCCFMT1ZTFSKPtFqF5UeCmLnxAtHxyZhpYERLHyZ1zftKpUIIISTAEILQyDjG/3GCozcfAOBfyZEZz8hKpUIIURASYIhS7Ur4I4YtOERIZDx2Vua81dWX4S28ZL0PIYQoIAkwRKl1IugBIxcf4UFsElVdyrBkZFOZySGEEAYiAYYolXZeDOe1pceJS0qhfmUnFo5oQnmZ0SGEEAYjAYYoddaduM3kP0+SrFVoU8OFH4c0oowkcgohhEHJf1VRqizYe53P/j0HQJ/6Hnz1fH2sLCTfQgghDE0CDFEqKIrCrM0Xmb/zKgAjWnrzce86mJnJKqVCCGEMEmCIEi85Rcv7a0+z6mgwAG9382Vs+2qyBLoQQhiRBBiiRItJSGbCihNsPR8uq5QKIUQhkgBDlFghD+MYteQo50OjsLIw47sXn6KblPwWQohCIQGGKJECbz3k5d+Ocjc6ARd7K34a2phGXs6mbpYQQpQaEmCIEuefkyFM/vMkCclafF0dWDCiMZWdpYCWEEIUJgkwRImhKArfbrvCnK2XAOhYqyLfvviULFYmhBAmIP95RYkQn5TCO3+dYn3qMuujW/swpWdtzGUaqhBCmITJKwzNmzcPHx8fbGxsaNSoEXv27Mly3xEjRqDRaDLc/Pz8CrHFoqgJj47nhZ8Psv5kCBZmGmY8W5cPe9eR4EIIIUzIpD0YK1euZOLEicybN49WrVrx008/0aNHD86dO0eVKhmnEn7zzTfMnDlT/31ycjL169fn+eefL8xmCxNQFIWo+GQiHiUQEZ1AxKNE7kbHE/EokbUnbnP7YRxOtpbMH9KQltVcTN1cIYQo9TSKoiimevJmzZrRsGFD5s+fr99Wu3Zt+vXrx4wZM3I8ft26dTz77LNcv34dLy+vXD1nVFQUTk5OREZG4ujomO+2C+NSFIU9lyP4Zc81rt2N4e6jBBKTtVnuX9WlDL8Ob0zVCvaF2EohhChd8vIearIejMTERI4dO8Z7772XbnvXrl3Zv39/rs6xYMECOnfunG1wkZCQQEJCgv77qKio/DVYFJr9VyKYHXCJozcfZHjMwdoCFwdrXOytqOBgjYu9NZWdbRnYuApOdpYmaK0QQojMmCzAiIiIICUlBVdX13TbXV1dCQsLy/H40NBQNm7cyB9//JHtfjNmzODTTz8tUFtF4Th8/T6zAy5y8Np9AKwszBjSzIte9dyp6GBNBQdrbCzNTdxKIYQQuWHyWSRPrgehKEqu1ohYvHgxZcuWpV+/ftnuN2XKFCZNmqT/PioqCk9Pz3y1VRjHsZsPmBNwib1XIgCwMjfjxaaevNa+Om5ONiZunRBCiPwwWYDh4uKCubl5ht6K8PDwDL0aT1IUhYULFzJ06FCsrKyy3dfa2hpra+sCt1cY3vnQKL7YdIGdF+8CYGGmYUATT17vUJ1KZW1N3DohhBAFYbIAw8rKikaNGhEQEMAzzzyj3x4QEEDfvn2zPXbXrl1cuXKFUaNGGbuZwkjWHA/mvTWnSUzWYm6moX/DyozrWB3PclJxUwghSgKTDpFMmjSJoUOH0rhxY1q0aMHPP/9MUFAQY8aMAdThjdu3b/Pbb7+lO27BggU0a9YMf39/UzRbFEByipaZGy/w697rAHTwrcAnffzwKl/GxC0TQghhSCYNMAYOHMi9e/eYNm0aoaGh+Pv7s2HDBv2skNDQUIKCgtIdExkZyerVq/nmm29M0WRRAA9iEhm//IQ+12J8x+q82bkmZlIQSwghShyT1sEwBamDYRoXwqJ4+bej3Lofh52VOV8/X58edd1N3SwhhBB5UCzqYIjSY+PpUN768ySxiSl4lrPll2GNqeUmwZ0QQpRkEmAIo9FqFeZsvcR3268A0Kp6eb5/sSHOZbKf+SOEEKL4kwBDGEV4dDzvrznN1vPhAIxq7cOUHrWwMDf5+npCCCEKgQQYwqCuR8Tw8+5rrD4WTGKKFisLM2Y8U5fnGlU2ddOEEKLkeHgLDs6DegPBo4GpW5MpCTCEQZwKfsiPu66y8UwYurThhlXK8kkfP+pVLmvStgkhRIkSfgF+fwaiQ+DsOhh3BKyL3kKPEmCIfFMUhb1XIvhx11X2Xbmn396xVkXGtKtGE2/nXJV9F0IIkUvBR2FZf4hLXQwyOgR2z4Iu00zbrkxIgCHyZceFcL4OuMiZ2+rqtOZmGvrU9+DVdlVlhogQpUViLJxbB57NoHw1U7em5LuyDVYOhaQYqNQYmoyCda/BgR+gwWCo4GvqFqYjAYbIs9PBkYxccgRFAVtLcwY28WR0Gx8qO0uZbyFKBW0KBC6DHZ9DdChUaQkjN5q6VSXbmdWw5lXQJkG1jjDgd3VY5Nx6uLQRNrwNw/6GItRrLAGGyLNZmy+gKOpQyFfP16ecTDsVonRQFLi8BQKmwt3zadtDTqhBh5m56dpWkh3+RQ0gUMDvWXjmJ7BI/b/bYyZc2wHXd8HZteD/rEmb+jiZMyjyZP+VCPZcjsDSXMMnT/tJcCFEaXH7OCx5Gv4YoAYXNmWh6//A0g6S4+DeVVO30LgUBa7ugOUvworBkJxYOM+58wvYMBlQoMloeO7XtOACwNkbWr+p3t/8PiREG79duSQ9GCLXFEXhi00XABjUtApVysuQiBAl3v3rsP0ztYsewNwamo9R39RsneHsGrh9DO6chgo1TdtWY0hOhDN/qXkOd86kbQ8+DN6tjfe8Wi1sehcO/6x+3+49aP9e5kMgrSbAyeXw4AbsmgVdPzNeu/JAejBErm0+G8bJ4EjsrMwZ17GGqZsjhDCmpDjY8hF83yQ1uNBA/Rdh/DF1xoKts7qfW131a9iZLE9VLMXeh91fwdy6aiLlnTNqb429q/r43QvGff7149OCix5fQocpWedXWNpCj1nq/YPz1GmsRYD0YIhcSU7R8uXmiwCMbu1DBQdrE7dICGE0t47AujFwTy3zT7WO0PlTcK+XcV9Xf/XrnRIQYCgK3L8GB+erSaxJsep2B3do9io0GgF7ZsP+b437Jh56EgKXgsZczbeo93zOx9TsBr494eIG2Pg2DFtv8oRPCTBErqw+HszVuzE421kyum1VUzdHCGEMSfGw83PY/x0oWrB3g6fngm+PrI/R92CcNl67tClwfTdUbpL/glJardojEHIcEh5BYjQkxqTef5T2VUlJO8atLrQYD37PpOU9VKytfjVmD8axJerXOn1zF1zodJ8BV7er1+rsGvB/zjjtyyUJMESO4pNSmLv1MgCvd6iOo42liVskhDC44GPqUECE2lNJ/RfVNyzdUEhWXP3Ur9GhEHMPypQ3bLsUBf5+Xc0xcKsHIzeDVT7yv3Z9Abtm5m7fGt2gxevg0zZjL4Cu1sTdi3lvQ24kxsDpP9X7jYbn7Vhnb2g9SQ0SN38ANbqCtYPBm5hbEmCIHP1+4CahkfG4O9kwpLmXqZsjhOmEn4f4SDC3VJMdza3UT7bmj90sbdXHi4vkBNg5E/bNTe21cIXec6FWz9wdb+0Azj7w4Lqa6Fm1vWHbt3OGGlwAhJ2CfybAsz/nrfv/4qa04KLFOLUomJWD2htiVQas7NXXYWUPNo7qtqy4pAYYMeFqnoZdufy9rqycXQcJUWqw4N0278e3mgAn/0hN+PwCuk43bPvyQAIMka2o+CR+2KmOw77ZuSY2ljLPXZRSZ9fBn7n5RKkBJ09wqQ7ldbdq6lcnz8KrFREfpfYqPB786IMhazAzU+tXrH0traZF3efVZMG8vmm6+asBRtgZwwYYx39X3yQBmrwMRxfC6VXgXh9ajsvdOe5dhTWvpJ2j2/8K1iZre3CqApFB6jCJV8uCne9Jx1OHRxoOU39GeWVpo/4M/xig5pI0GAIVaxm2jbkkAYbI1s+7rvEwNonqFe15tmElUzdHiPxLTgCLfCYnPwqHf1NrDTh4gMYMUhLT37TJqTsr6ptPZJA6Hv44cysoV1Ut81z/BfBqlb83kZwEH1MXw0qIzHofjXlavkGZCtB7DtR+On/P51oXzv9j2DyMK1vV3gqANpOh00fgUgM2vgMBH6lDM9U6ZH+OhEewcoh6HTybQbfPDdO2Cr7GCTDCz8OtQ+rPpsHg/J+nZjfw7QUX/1NraAz/xyQJn3kOMLy9vRk5ciQjRoygSpUqxmiTKCLCo+NZsPc6AJO7+mJhLrOaRTF09xJs+UB9w+o2Q63hkBeKogYXcffVN9KXt6cvdKSj1aqBRnykOhPh3pX0t/vX1MfvXlBvgUuhbBU116H+C2rgYQiP7sKqoeqbqmUZ9Y1FFwSle12pwYXfs9Dzq4LlTugSPQ01kyT0FKwarrax3kDo+KG6vekrqTMslsFfL8HLO6CcT+bnUBR1qmf4OXXY5/klmf/c8qNiLbgSYPiZJMd/U7/69gAHt4Kdq/sMuLoNbuxRpxnX7V/w9uVRngOMt956i8WLFzNt2jQ6dOjAqFGjeOaZZ7C2lmmLJc33268Ql5RCfc+ydPNzNXVzhMib2Ptq9/rhX9LeTDdPUYtBVeuY+/OcWQ0X/gUzC+g3L+s3KTMzMLNRu6gdXMGrRfrHtSkQeUsNeC78q5Z1fhiUmnz4BVRpAQ0GQZ1+ah5AfqQkq2+8UbfVIZmXd6SdS1EgJQlSEtSvyQlqrkgZl/w91+PcUqeq3r2oFqYqyBt5ZLDavZ/4CLzbQJ/v0z59azTQa7b6ST/kuFpRc3RA5jkTB35QZ1KYWcCA38DRPf9telKF1CEHQ84kSYpPyzVpmMfkzsw4e0Gbt9RpqyZaiC7PH0nHjx/PsWPHOHbsGHXq1OGNN97A3d2dcePGcfz4cWO0UZhA0L1Y/jgUBMC73X1l2XVRfKQkwaGf4buGcOhHNbjw7al+Ule08OdLanXK3Ii+k1qmGWj7TuZ1IHLLzFxN3KvZFfp8C5MvwXMLUoMdDQQdUD9xf1UTVr8MEVfy/hxbp6qfWK3sYeCy9IGKRqO+8Vs7qDkWju6GCS5AzS2xcVIX4ooowOyK+EhY9ryaO1KhFgxcmjFYsbRRt5epCOFnYd1YNXh63PXdEPCxer/7TKjSPP9tyowxAowL/6pLsDtWhuqdDHPO1m/C6G3g8ZRhzpdH+e7zrl+/Pt988w23b99m6tSp/PrrrzRp0oT69euzcOFClCd/4KJYmR1wkWStQpsaLrSsZqB/QkIY25WtML+VWmgo7gFUrAND18KLy6HffDX3If6h+sk34VH259IPjTxQp0e2mWTYtlraqt3WQ9fCm2eh01Rwqamu63F6FfzSMWMOR3bOrIYD36v3+80r3MQ+jSat4FZ+8zCSE9WlyMPPqfU3Bv8FtmUz39epEgz8Hcws1eXi985JeywyWA0ilRR1+KnJ6Py1Jzu6qaqP7qg9ZYZwbLH69akhhksENrc06QJ0+Q4wkpKSWLVqFX369OGtt96icePG/PrrrwwYMIAPPviAwYMLkKAiTOpcSBR/nwwB4N3upsk+FiJP7l5SP/kufU79BG1XXu1Kf3VP2nCIpY36pmTvqn7y/TuTT76PO7VKTZIzs4RnfjTu1FOnSmoA8/ph9ROnZ3M1h2JpfziyIOfj75yDv1NnVbSaqBZoKmwFKRmuKPDPG+qKoJZlYPAqKOuZ/TFVmkPP1PLY26bB5QB1mGHlUIiNUNvTe45xkhutHdSeBoCISwU/372ras8TGjXAKCHynINx/PhxFi1axPLlyzE3N2fo0KHMmTOHWrXS3oi6du1K27b5mL8rTC4mIZkpa0+jKNC7njv+lZxM3SQhshcZrH7aT4xWx9ubjYG2b2f+6dfRAwb8Dot7wbm/Ye9sdZz6SVGhai8IQPt304pJGZtGA5Ubw/D1sP4NOLUC/pukJol2nZ75p9G4h7BysFrWump76PhR4bT1SfqS4fnowdg5U80/0JjDgCXqNNTcaDxSTQg9tgj+GgVV26q5GbbO6jCKpW3e25JbFWtBVLCaD1LQIRhdcmf1zjkHVsVIngOMJk2a0KVLF+bPn0+/fv2wtMwY1depU4cXXnjBIA0UhSc+KYVXfj/KyVsPcbK15J1u0nshHqPVQuw9eBQG0ak3M3M1y9+E3bCcXasGFxVqq28qLtWz379KM+j1lToFcttn6syQml3THlcU+Heimg/g3gBavWnM1mfOwlrtNXGpoa5kenCe+in3uV/T51VotbD2VXWGipMnPLcQzE1UfcDtsSESRcl9z0H0Hdid2hPRezbU6JK35+0xK3V650F1qiwaNbfF2Ttv58mrCrXUIbmCVvRMSYLAP9T7ea3cWcTl+Tfx2rVreHllX82xTJkyLFq0KN+NEoUvOUXLG8tPsO/KPeyszFn8UhNZjr00S4pTP1VGXE4LKB7deazWw2PMrUwyBU7vwn/q18Yjcw4udBqNUKc7Hl0Iq0erU091x55cDpc2qa/rmR9N94at0UDbyepskLWvwuXNsLAbDFqpTm8F2P1lalut1eEfQ5fpzosKtdUeiLgHEBWiDvvkxpWtavKtewP155JXFlbqLJGf26nJoZ0+MlySZHb0JcPPF+w8FzeqVUHLVISa3QveriIkzzkY4eHhHDp0KMP2Q4cOcfToUYM0ShQurVbhnb9OseXcHawszPh1eGOeqpLD+gOiZNs7Vy0dffE/uH1MnfaoCy7sXNRP/bpPiEEHTdRI1AJYuufPbWlrne5fpOU6rBikVr6MvA0b31Mfbz8lbWErU/LrBy9tSM0dOacOB906Apc2q2W0Qc01MNFMAT1LGzVJFfJWD+PyFvVrzW75f24HV3h1Nwz7W12LozBU0C16VsAeDF3lzqcGF68S87mQ5wDj9ddf59atWxm23759m9dff90gjRKFR1EUPv3nLGtO3MbcTMMPgxrKrJHSLuZe2myEVhPhhT9g9HZ48xx8FAHvXIXX9qaN9YecMFlTubgRUNQ3V6fKeTtW98nXwUNNDF37qppomBAJlRpByzeM0uR8qdRI7WVxrQsxd9UcktWjAQUaj1LfnIoCtzzOJElJgqs71Ps1uma/b07sK6o5KIU1pb5CajAVHarmweTHwyC4sk2933CYQZpVlOQ5wDh37hwNGzbMsP2pp57i3LlzBmmUKDyzAy6x5MBNNBr4+vn6dKkjBbVKvX1z1CJH7vXVqZO1ekHlRmqX9+OfsHSfmO+cUQs8mcKFf9WvtXrl73gHV3hhqTrEcHGD2l1vbq1OaTXV0EhWnCrDyE1Qs4daLCshCio3Ves8FBV5nap665Aa0NmVN30PTF7ZOIFj6jBQfnsxTiwFFHXVVkNVci1C8hxgWFtbc+fOnQzbQ0NDsbAoYn+QIls/777Kd9vVYj7T+vjR7ylZa8TokhNM3YLsRYWolS9B7aHIbp0MZx+wdoTkeMMWHMqthGi4tlO9Xyufa2iA2jvQ+7E6Ch0/TBtfL2qs7eGFZerwTc3uag+MocpfG0JeS4brhkeqdzZtonB+6fMw8vH7r01JDTAwTOXOIijPAUaXLl2YMmUKkZFpi+g8fPiQ999/ny5d8pj9K0xm+eEgPt+g/lG83c2XoS28Tdugou7aTvhvsjqzIL/O/wPTXWGLiaYR5sbuL9WAoUoL9Z9+dszM0qYThgYavWkZXA5Q19coV63gAcFTg6HHl+r4fYsiPtRrZg7t31OTPQ1Z/toQdAHGvauQGJPz/pcD1K8FHR4xlYJU9LyyVc1tsi2X/0Xmirg8Bxhff/01t27dwsvLiw4dOtChQwd8fHwICwvj66+/NkYbhYH9czKE99eqXZhj2lXj9Q65zLwvra5sVYs4HfkF9uTzd1yrVadEosD+b9VS1oamKGqJ5D8GwueV1UJReXH/Wtp8/E4f524sWxdgmCIPQzd7pHZvw4y7N3sFOk8tnp+kiwr7iupsCBR16mh2Ht5Sk1Y1ZnlbG6YoKUiAcSw1ubPBoPyv8lvE5TnAqFSpEqdOnWLWrFnUqVOHRo0a8c0333D69Gk8PUtOgZCSSFEUfj9wgzdXBqIoMKhZFd7tXkS7gouK63vUstK6lSiPLFCn4eXVpY1qIqEm9U9u07vqLABDSE6Ekyvhp7aw5Gl12mJitFrZMS9v/DtnqjNFqnfO/RLUunHzkMA8N7tAkhPTutdr9S7c5xbZ0yd6nsp+vyupvReVm6proxRH+gAjjzkY0WHq3ymUyOROnXwlTZQpU4ZXXnnF0G0RRnQ/JpF3/jrF1vNq/ky/Bh581tdfFjHLzq0jam9Acrw63v3wllpi+tDPanXH3FKUtLUSWk2AmAg48Tv8NVJN2tN1K+dV3AM4uggO/6xmsgNY2kGDwfDgutrzsnIovLIr5/oId86l9XjolsbOjXSJnkmFN83uxm41ydHeVV1fRBQdbnXVNVRyKhmuHx4pxkPruqG5qNvq8KlNLisfBy5T10qp0qLo5vsYQL6zMs+dO0dQUBCJiYnptvfp06fAjRKGtf9qBG+uDOROVAJW5ma816MWL7XyluAiO6En1XUtkmLUqW/PL1FnLKweBYfmq+P01va5O9fN/RB8RJ2d0HysWsb44c204YzR2/I2ln7/GhycryaIJcWq2+xdoekrarEpu3LqtLlfOqj7/vUSDFmT/ayIHf8DFHUNi7xk8+sSPROi1G7i/AZLeXU+dfaIb8/sE1FF4XPNRaJnckJagm5xzb8AtRy9g7sa4N+9BJ5NcnfcmbXq1xK07khm8lXJ85lnnuH06dNoNBr9qqm6N6uUlBTDtlDkTeTt1MQ3H5JStMwJuMT8XVdRFKhWoQzfvvgUfh6yvki2ws/D78+o0+eqtFDrQFjagN8zsONzuH9VXfug5fjcnU/Xe/HUYHWMGtT1MBZ0VYdN/hgAL23MOWCJj4JdX6hLkOuKXrn6q8GO/3Ppx3Fty6rLdf/aSV1Aavtn0OXTzM97+5gaPGnMoMMHuXtNOrpEzxt71GGSwggwtFp1Simo+ReiaNENkdw5q/6sMgsAb+5Tg2N7t8ILSo2lgm9qgHEhdwHGwyB1vRaNmRogl2B5Dv0nTJiAj48Pd+7cwc7OjrNnz7J7924aN27Mzp07jdBEkWspSeobyvyWhFw9y/M/HmDeTjW4eLGpJ/+Mb114wUV8pJowGHG5cJ7PUO5dhd/6qmtueDwFg1aBVRn1MTNzaJ26LsX+79SVG3MSdloda9aYpQ9IbMuqK0bauahj1atHq9PWMqMoao7F943VAljaZDUpbug6GLM36yQx1zrQN7Vg1r656uJemdn2mfq13gv56671aKB+LayZJLePqmXLrR3BWxZVLHLK11B76xIfqUN1mbmUmj9To0vhFcYyFn1Fz1wmel7cqH6t0qL45p7kUp4DjAMHDjBt2jQqVKiAmZkZZmZmtG7dmhkzZvDGG0Wo8l1pFHZKjaSTYjn320QCbz3E0caCeYMbMuPZethZFVKdktvH4Mc2sH68+qb4+zPqH1VWb6BFxcMgWNJHffOq6KcOKzy+sBSoC3s5Vlb3CVya8zn3zlW/+j2TsZCOsze8uFz9Z3xpI2zJJPch7Aws6glrX1Gfs1xVGLwahq6Fah1y/ufs/xy0SF3Ge91YCH/in+D13XBth7okeV7ySh7n3kD9WliJnuf/Ub/W6Fq0akAIlblFWon1rIZJDFEevKjIay0M3eynEt57AfkIMFJSUrC3V7tyXVxcCAkJAcDLy4uLFwtYk10UTFDaGjGdNYcZ4X6LjRPb0rNuIc2VVxQ4MA8WdFNzDGzLARo14Wv5C/BtA9j3DcTeL5z25EVUqBpcRAWrn8CGrcv804WFFbRKDaT3faP2GmXl/nU4u0a932pC5vt4NlUX1AJ1xUxdkau4h7DxXXVmSNB+NXmz08cw9iDUyKE+xZM6fwrebdRPlCsHp9XyUJS03otGI/K/+qQuZyPsdPbXwxAUpeDVO4Xx6WeSZBJg3LuqDjOaWYJPu8JtlzHogqkng/fMxD1Uh4cAfHsYrUlFRZ4DDH9/f06dUqcfNWvWjFmzZrFv3z6mTZtG1aolr9RpcRJ9eQ8ADxQ1AJxqvYxKjvn4hPfwFpxYBo/u5v6Y2PvqdM7NU0CbBLX7wBsnYEKguqaDTVm1hyDgY5hdW51CGXoy720zlPhIuH0cTv0JO2aoazs8uA5lvdQFk3S5EplpOAzKVFBfz+m/st7vwPfqKpHVOqXVi8iM/7Nq8ACw8R0ImKr2/Bz6Uc00r9MXXj8Mbd7K33x5cwvov0gta3zvCqx9TR0bv7QZgg+Dha26amd+6RI9UxKMX9Hz7gU1cdXcqnjPPijpdImemZUM180e8WqRsYewONIt8BYVrOZJZefKVnWIs0ItKF/N+G0zsTz3mX/44YfExKgV2qZPn07v3r1p06YN5cuXZ+XKlQZvoMglRdH3YKx0m8yYh3PQhJ1Sl57OS6Zy7H21Sz4ySP2EUbu3+unWu23W2fq3jqgzFSJvqf/4u30OTUar3fe2ZaHrZ2pp4zOr4fBP6j+dE7+rtxpd1YRHS5sCX4JMKYq63sGtQ+qba8QV9WtMeMZ9HTxg+Pqcl5m2tFVng2z7FPbOVodNnrw2j8LTygDr8jay03qS+sZ5YqmaLwHqP64es9ShkIKyr6Be50Xd1RVS93ydlpPR7BVwcMv/uQsz0VPXe1G1PVg7GO95RMFkVzJcNzxSnGePPM6unDqL69EdNeescqOs9y1FwyOQjwCjW7e0MbOqVaty7tw57t+/j7Ozs0x7NKGEu1dxSL5HomJO7TbPQZSFOqa/bRrU6Ze7KZVaLax7TQ0uLGzU+g9n16o3Zx9oNFytsaD7dK/VwoHv1OfQJqv7PL84LenvcVZ20HCoGuzcOqTWbjj3t/rP5sK/ULe/Aa8GaqB0cjkcWwwRlzLfx94VyldXP0mUrwF1n8/9dNEmo9VAIOISXPhH7WV43KEf1etXqTF4t875fBoN9JqjrmR6cz+0fQuavWbYHIPKjaDnV+qKoTumq9usHdUVUwvKo4EaYIQGAkMLfr6s6KanSnGtos3VT/0aeUut12LrrH6fGAM39qr3S0qAAWqPxKM7cPd81gFGcqLagwESYGQmOTkZGxsbAgMD8ff3128vV65kZ8IWB+cObeEp4KJZdVrX8QTtK2rVyQfX1TfC3BRPOvCdWl3O3BpGpX7KOLZELcD04Dps/QS2T1f/OOq/qL55X06tRun3LDz9Tc5dnhoNVGmu3gKmqm27tNkwAYaiqP+8ji2G8+vTqm9a2qnd6RVqPxZQVC9Y96yNIzR9FXbPgt1fqUNCugA7PgoO/6reb/1m7rPkLazUpE9FMV5th0bD1STc46lliluON0wmuz7R04glwyODUwMYTakYvy7WbMuCUxX1w8qds2lB9vU96lBa2SppQwslQYVa6nTw7IYIb+59rDhcNr0cJUieAgwLCwu8vLyk1kUR9ODCbgDiPZpibqYBM2t1aGLlEHVKZcPhUDabUu4398PW1DoJPb5IyxnoPVs9z9m16ht38BH1zfv8evVxc2t1/0Yj8j7drGY3NcC4slWdYZLfNSBiIiDwD/VN896VtO1u9aDxS+Df3zhjvc1fgwM/qLN3rmxLS748tkitoeFSM++fVDQa40/b6/mlWnkwJkJ9DYagT/Q0YkVPXfdylebZ58iIosHNXw0wwk6nBRj64ZFuxX966uMqppYMzy7R80Jq7Zaa3UtNcbg8v8oPP/yQKVOmcP9+EZwJUEoF3YulcrSaeOvd4LFFg2r1Bq/Walf91k+yPsGju2rZaiUF6g5Qg4XHWZVRhzZGb4Ux+9SKkdZO4OILL29T38Tz88+iclM1+TPuPgQfzfvxoP7Rfl0LAj5Sgwsre7X9r+yEMXvUypbGSiSzK6e+doA9X6lfkxPUmTSgzhwpiv9ILKxhyGp4dZfh8hicfdTfCWMmesrskeJFl4ehm0miKCUv/0InpzVJFCWt/kUpGR6BfAQY3377LXv27MHDwwNfX18aNmyY7pZX8+bNw8fHBxsbGxo1asSePXuy3T8hIYEPPvgALy8vrK2tqVatGgsXLszz85Yk6w+coabZbQAq1Hms8JBGA90/BzRw5i81GfNJ2hRYM1qtn+HiC73nZB8suPmrn4DfvaFOmSxIQp+5BVTvpN7XLfyTVztnqLNW3OrC09/CWxfVoZq8lLsuiBbj1MTWoANwYx+cXAGPwtSE0boDCqcNRYGZGbjXU+8box5G7H31+oIEGMWFq66iZ+pMkrsX1JwMC5vc5SUVJ7oAIzIIEh5lfDzslDrLxNIOqpaAqbm5lOckz379+hnsyVeuXMnEiROZN28erVq14qeffqJHjx6cO3eOKlWqZHrMgAEDuHPnDgsWLKB69eqEh4eTnJxssDYVN8kpWq4Fbgfgkb0P9mVc0u/gXl9NzAxcqk4hHRWQPoDY/aW6JoClHQz4Lffraxjqk3nN7urskstb1KWy8+LeVfUPV2MOQ//OeUEvY3B0V3t3ji5U8zEe3lK3txxX+opA6RI9Q06oCb2GdGmz2sNW0S9jwTJRNOlqYYRfUIfNdL0X3m3UpO+SxK6cukx9TLha/v/JHAvd8Ei1juostFIizwHG1Kl5fBPIxuzZsxk1ahSjR48GYO7cuWzevJn58+czY8aMDPtv2rSJXbt2ce3aNX1iqbe3t8HaUxztunSXGvFnwAJsq2fxqaDTR2oORfAR9c1cl1B5dYe6RDeoPRe6ccTCVL2zWkb7zhk1ic+pcu6P1RWxqtrONMGFTqsJajKsbvEmm7Jqzktpo+s1MkbJcN3wiKw9UnyU9VaHLBMfqdM39aunlrDhEZ0KvmqAcTeTAEO3dk4pGh6BfAyRGEpiYiLHjh2ja9f0v2xdu3Zl//79mR6zfv16GjduzKxZs6hUqRI1a9Zk8uTJxMXFZfk8CQkJREVFpbuVJMsP36KRmToN09yreeY7ObhBm9RaDAFTISkOokLU9S9Q1DfD+i8UToOfZFcOKqcuEKT7hJNbuhUJ/Z41bJvyytlbneKq0/SV3PcElSS6mSS6RE9DSYxVk2hBhkeKEzOztGGSoP3qMCKU3AJp+oqe59Nvf3grtafVrGSURs+DPAcYZmZmmJubZ3nLrYiICFJSUnB1dU233dXVlbCwsEyPuXbtGnv37uXMmTOsXbuWuXPn8tdff/H6669n+TwzZszAyclJf/P0zGYmRTFzJyqefReDqa+5qm6o0iLrnVuMAydPdRxw3zdqUmdshJq70OOLwmlwVnSfaC7lIcC4exHCz4KZRdF402kzSR2qsSwDzV41dWtMo1zVtETPJ//JFsTV7ZAcp057dKtnuPMK49MNkxz4Qa2VU74GlPMxbZuMRb8myROJnrrkTs9m8OQQdgmX5yGStWvXpvs+KSmJEydOsGTJEj79NIvloLPxZHEuRVGyLNil1WrRaDQsW7YMJyd1VdDZs2fTv39/fvjhB2xtM45tTZkyhUmTJum/j4qKKjFBxl/HgqmjXMNak6yWrs5ubNrSFjp/AqtHqYmRAFYO8PwS048J1uymLid+bafau5Kb9pxN/T2s1rForEhYwRdGblbbXsr+iehpNGqip67glruBgoHHZ4+UpKmNpYGuB+P+NfVrSf4Er59J8sQsqlI6PAL5CDD69u2bYVv//v3x8/Nj5cqVjBo1KlfncXFxwdzcPENvRXh4eIZeDR13d3cqVaqkDy4AateujaIoBAcHU6NGjQzHWFtbY22dj/UbijitVmHV0Vt0Tx0ewbNZ7lbWPPSjmosB6lLeRaEevqu/uk5G1G21UFZOXaiKAmdS8y9MPTzyOM8mpm6B6Xk8lVYyvOGwgp8v+g6cXafer/10wc8nCteTPU4ldXgE0pZtf3hTrVhqVUZd80hXubQo9LQWMoPlYDRr1oytW7fmen8rKysaNWpEQEBAuu0BAQG0bNky02NatWpFSEgIjx6lTQO6dOkSZmZmVK6ch+TAEuDg9XvcvBdLc4vUACO74REdjUYtFe3gDm3fAb9+Rm1jrmk0jw2TbM55//Dzaqa2uRXUKn2fCoo0XZl4QyV67vtGHR6p1Bi8Mv+/IIqwirXV3ANQEz5z83+quCpTHuxSey91yxNcDlCn0bvULBof5gqZQQKMuLg4vvvuuzy/yU+aNIlff/2VhQsXcv78ed58802CgoIYM2YMoA5vDBuW9ilo0KBBlC9fnpdeeolz586xe/du3n77bUaOHJnp8EhJtuLwLTRoaWqRWrmyShYJnk/yaABvXYCOHxitbfmi6zq9vFntociObvZI9c5g45T9vqJwGTLRMzoMji5Q77efIsMjxZGVHZRLfWOt2j5/qwEXJ08u3V4Ki2s9Ls9DJE8uaqYoCtHR0djZ2bF06dI8nWvgwIHcu3ePadOmERoair+/Pxs2bMDLywuA0NBQgoKC9Pvb29sTEBDA+PHjady4MeXLl2fAgAFMnz49ry+jWHsQk8imM2FU1YRSJiVSXW67uCe/+bRVy44/DFKTpLKaMqsoafkXfs8UXvtE7ugSPRMi1Z6mguRh7J2rVqGt3CStIJsofnzawL3LpePvtYKvOkR4V1f7I7WHXgKM3JkzZ066AMPMzIwKFSrQrFkznJ2d89yAsWPHMnbs2EwfW7x4cYZttWrVyjCsUtqsC7xNYoqWPuWCIBZ1znVxL+pkVUb9R3Rlq1rVM6sAI+y0WhLc3FoWvCqKNBrwqA/Xdxcs0TMqVC1eBtJ7Udx1mQb1XgDPpqZuifE9XjL8xl410C5TASo3Nm27TCTPAcaIESOM0AyRW4qisOKwWi2yd9mbaoCR2+GRoq5mdzXAuLwFWk/MfB9d70WNLoZbR0MYlnsDNcAoSKLnvrnqdNfKTdWZQqL4snaAKs1M3YrCoQ8wzqcNj9Tsnv+FHIu5POdgLFq0iD///DPD9j///JMlS5YYpFEia4G3HnLxTjTWFmZ4x6XW+C8pAYYu0TPoIMQ9yPi4oqTlX/gXodkjIj1domd+l26PCoGji9T7HaT3QhQjugDjwc20FadL6fAI5CPAmDlzJi4uGef5V6xYkc8//9wgjSqVokJgbj1YPz7b3VYeUXsvBta2xvzBdUCTVgmzuHP2Uv9AlRS1uNKTQk7AgxtqzknN7oXePJFLukTPO2fzl+i5d47ae+HZHKp2MGjThDAq+wpgVx5Q1AUkLWzV5NZSKs8Bxs2bN/HxyViJzcvLK11Cpsij47+p86eP/6YOE2TiUUIy60+GADDEI1Td6OoHtmULqZGFILuqnrrhkZrd1JwNUTQVpKJn5G04tli9L70Xojiq8Fj+WLWOJW9htzzIc4BRsWJFTp06lWH7yZMnKV/ehAtOFWeKAieXp32/aUqmn/z+OxVCbGIKVV3KUCPhjLrRs4SNbeqnq25Rl5LXUZS0gksyPFK06RI9Ie/DJHvnQEoiVGkJPqVnWWtRguhKhkOpT0TPc4Dxwgsv8MYbb7Bjxw5SUlJISUlh+/btTJgwgRdeMNGCWcVd0EG169/KXi3UEnEJDv+cbhdFUfjtwE0ABjbxRHPrkPpASStc49lMrW0Rdx9uH0vbfvsYRAapa31UL8HVAEsK3TBJXgpuRQbD8dQ8Lum9EMWVrqInmlI/lJvnAGP69Ok0a9aMTp06YWtri62tLV27dqVjx46Sg5Ffut6LOv2g08fq/Z0z4dFd/S5Hbz7gbEgUNpZmDKhXDkJPqg+UtOxsc0uollrz4PGqnrrS4L49SnWXY7GhT/QMzP0xe2arvRdercC7jTFaJYTxebVUq5fW6KLmZJRieQ4wrKysWLlyJRcvXmTZsmWsWbOGq1evsnDhQqysinktBlNIikvLLaj/Ajw1BNzrQ0IUbEtbPG7RvusAPPNUJZwfnlZXJnSspK6QWtI8XtUTQKtNu0YyPFI8eDylfr1zFpITc97/4S01/wik7oUo3tz8YdxR6L/Q1C0xuTzXwdCpUaNGpouLiTy68J8aTDhVUT+5mZlBj1mwsBucWApNRnHbrhabz94BYHhLb7j4k3psbhY4K46qdwY0alGtqBC1umd0CFg7pvVuiKLN2Ucd6oqPVGsCuNfPfv89X6trNni3UQuuCVGclcJ1RzKT5x6M/v37M3PmzAzbv/zyS55//nmDNKpUOblC/Vp/oBpcgFrXou4AQIEN77D0wA1StAotqpanlpsjBB1I3a+E5V/olHFJq3x3afNjwyM9wdLGdO0SuafRpAUVOQ2TPAxSg2lQey+EECVCngOMXbt20atXxmVnu3fvzu7duw3SqFIjOgyublPv138x/WNdPlUTGoMPE3l4GQAvtfJWZ1bollsvKQW2MqMbJrm0Cc79rd4vDWsZlCS5TfTU9V74tAXvVsZulRCikOR5iOTRo0eZ5lpYWloSFRVlkEaVGqf/BEWrlkN+skvN0QPavgXbpvGGdilHnVvQqbYrhJ9Rh1SsHNQaGCVVjW6wfboaYIBaV0FKRhcvujyMa7vg8C/q1OuUhNSviZCcoH6V3gshSqQ8Bxj+/v6sXLmSjz/+ON32FStWUKdOHYM1rMRTFAhMnT3S4MXMd2k+ltAdP+OhDWNWxa2Ym/VSp7QCeDYp2fXt3eqCg4eaewFQu3fxX9CttNEFGPevwobJ2e/r007NvhdClBh5DjA++ugjnnvuOa5evUrHjuonym3btvHHH3/w119/GbyBJVbYaQg/q64KmkXX/4GgGBbFD+IXq9nUD14K98Y9FmCU4OERUMfwa3RJq4vgJ7NHip1yPtDhQ7WGibklmFuBhXXafd3Nyg4aDDZ1a4UQBpbnAKNPnz6sW7eOzz//nL/++gtbW1vq16/P9u3bcXR0NEYbSyZd7QvfHmCb+TL3i/bdIEDbiMsOTakRfRi2fKgGJlCy8y90anZXAwxbZ6gqVR2LpXZvm7oFQggTydc01V69eukTPR8+fMiyZcuYOHEiJ0+eJCUlJYejBSlJav4FZEzuTHXrfixbz98BNFj1+gJWdYGLG9QHNeZpsyxKsprdodNUtavd3NLUrRFCCJEHeZ5ForN9+3aGDBmCh4cH33//PT179uTo0aOGbFvJdWUbxNyFMhWgeuZ1HZbsv4GiQJsaLnjVaghNX0l70L1e6Vjsy8wM2kyCarKiphBCFDd56sEIDg5m8eLFLFy4kJiYGAYMGEBSUhKrV6+WBM+8OPmH+rXugEw/mcckJLPyqLos+8hWqSvXtnsXTq2C2IiSn38hhBCi2Mt1D0bPnj2pU6cO586d47vvviMkJITvvvvOmG0rmeIewMWN6v36mS8Ot+bEbaLjk/FxKUO7mqm17G3LwjM/qsW1mowqnLYKIYQQ+ZTrHowtW7bwxhtv8Nprr0mJ8II4s0ad++/qrw51PEGrVVicuu7I8BZemJk9Vgq8Rhf1JoQQQhRxue7B2LNnD9HR0TRu3JhmzZrx/fffc/fu3ZwPFOnpS4Nn3nux90oEV+/GYG9twXONKhdiw4QQQgjDyXWA0aJFC3755RdCQ0N59dVXWbFiBZUqVUKr1RIQEEB0dLQx21kyRFyB4MPqUr51B2S6i27V1P6NKuNgIzMnhBBCFE95nkViZ2fHyJEj2bt3L6dPn+att95i5syZVKxYkT59+hijjSXHqdTei2qdwME1w8PXI2LYcfEuGg2MaOlduG0TQgghDCjf01QBfH19mTVrFsHBwSxfvtxQbSqZtNq04ZEsSoMv2X8DgA6+FfF2KQXTUIUQQpRYBQowdMzNzenXrx/r1683xOlKppv7IPKWumiXb88MD0fHJ/Fn6tTUl1p5F3LjhBBCCMMySIAhckFXGtyvH1jaZnh47+UIYhJT8HEpQ+vqLoXbNiGEEMLAJMAoDFotXPhPvV9vYKa7HLnxAFArd2o0mkz3EUIIIYoLCTAKQ/g5iH8IlmXAs1mmuxy5cR+Axt7lCrFhQgghhHFIgFEYbu5Xv1ZpBuYZa5s9SkjmbEgkAE28M19ZVQghhChOJMAoDDf3qV+9Wmb68ImgB2gVqOxsi7tTxvwMIYQQoriRAMPYFCWtB8OrVaa76PIvmsjwiBBCiBJCAgxju3cVYsLB3Bo8Gma6y9HU/AsJMIQQQpQUEmAYm254pHJjsLTJ8HBSipYTQQ8Byb8QQghRckiAYWz64ZHM8y/OhkQRl5RCWTtLqlWwL8SGCSGEEMYjAYax5RBgHLmeOj3Vq1z6pdmFEEKIYkwCDGN6GASRQaAxh8pNM93liD7/QoZHhBBClBwSYBjTzQPqV48GYJ1x+ENRFI7eTJ1B4iMJnkIIIUoOCTCMKYf6F1fvxnA/JhFrCzP8PZwKsWFCCCGEcUmAYUw51L/QTU9t4FkWKwv5UQghhCg55F3NWB6Fw73LgAaqNM90F12BraYyPCKEEKKEkQDDWHS9F67+YJt5AqcscCaEEKKkkgDDWHKYnnonKp6g+7GYaaBhlbKF1y4hhBCiEEiAYSw5JHjqei9quzviYGNZWK0SQgghCoUEGMYQex/unFXvZxFgHJUFzoQQQpRgEmAYw61DgALla4B9xUx3Scu/kAJbQgghSh4JMIwhh+GR6PgkzodGAdKDIYQQomQyeYAxb948fHx8sLGxoVGjRuzZsyfLfXfu3IlGo8lwu3DhQiG2OBdyqH9xPOghWgWqlLPD1THjCqtCCCFEcWfSAGPlypVMnDiRDz74gBMnTtCmTRt69OhBUFBQtsddvHiR0NBQ/a1GjRqF1OJcSHgEIYHq/SzzL3Trj0jvhRBCiJLJpAHG7NmzGTVqFKNHj6Z27drMnTsXT09P5s+fn+1xFStWxM3NTX8zNzcvpBbnQvBhUFLAqQqU9cx0l8PXZYEzIYQQJZvJAozExESOHTtG165d023v2rUr+/fvz/bYp556Cnd3dzp16sSOHTuy3TchIYGoqKh0N6PKof5FYrKWwFsPASmwJYQQouQyWYARERFBSkoKrq6u6ba7uroSFhaW6THu7u78/PPPrF69mjVr1uDr60unTp3YvXt3ls8zY8YMnJyc9DdPz8x7FQwmhwDj9O1IEpK1lCtjRbUKZYzbFiGEEMJELEzdAI1Gk+57RVEybNPx9fXF19dX/32LFi24desWX331FW3bts30mClTpjBp0iT991FRUcYLMpLiIfioej+HBc4aezln+TqFEEKI4s5kPRguLi6Ym5tn6K0IDw/P0KuRnebNm3P58uUsH7e2tsbR0THdzWhCjkNKApSpCOWrZbrLESmwJYQQohQwWYBhZWVFo0aNCAgISLc9ICCAli0zH17IzIkTJ3B3dzd08/Ln8foXmfROaLUKR2+mJnjKCqpCCCFKMJMOkUyaNImhQ4fSuHFjWrRowc8//0xQUBBjxowB1OGN27dv89tvvwEwd+5cvL298fPzIzExkaVLl7J69WpWr15typeRRpd/4d0604ev3n3Ew9gkbCzN8PMwYk+KEEIIYWImDTAGDhzIvXv3mDZtGqGhofj7+7Nhwwa8vLwACA0NTVcTIzExkcmTJ3P79m1sbW3x8/Pjv//+o2fPnqZ6CWlSkiHokHo/ywXO1OGRpzydsTQ3eY0zIYQQwmg0iqIopm5EYYqKisLJyYnIyEjD5mPcPga/dASbsvDOdTDLGEC8uTKQtSdu80anGkzqUtNwzy2EEEIUgry8h8rHaEN5fHpqJsEFpC1wJgW2hBBClHQSYBjKjewXOAuNjCP4QRzmZhqeqiIBhhBCiJJNAgxD0GohKPsCW7r8izrujthbm7z8iBBCCGFUEmAYQvg5iI8EyzLgVj/TXfQFtmR4RAghRCkgAYYh6PIvqjQD88x7J46m9mA0lQJbQgghSgEJMAwh6jZozLMcHlEUhWsRjwCo7S71L4QQQpR8kgxgCF0+hbaTQZuc6cP3YhKJT9Ki0YBHWdtCbpwQQghR+CTAMBRrhywfunU/FgA3RxusLKTTSAghRMkn73aFIPhBHACVnaX3QgghROkgAUYhSAsw7EzcEiGEEKJwSIBRCIIfqEMkntKDIYQQopSQAKMQSA+GEEKI0kYCjEKg68GQHAwhhBClhQQYRqYoivRgCCGEKHUkwDCyiEeJJCRrMdOAm5ONqZsjhBBCFAoJMIxMNzwiNTCEEEKUJvKOZ2QyPCKEEKI0kgDDyKTIlhBCiNJIAgwjuyUzSIQQQpRCEmAYmQyRCCGEKI0kwDAyqYEhhBCiNJIAw4gUReG29GAIIYQohSTAMKK7jxL0NTDcy0oNDCGEEKWHBBhGpMu/cHeyxdJcLrUQQojSQ971jEgXYFSS/AshhBCljAQYRiQJnkIIIUorCTCMSKaoCiGEKK0kwDAiqeIphBCitJIAw4hkiEQIIURpJQGGkWi1ir4Hw1OGSIQQQpQyEmAYScSjBBJTa2C4OUkNDCGEEKWLBBhGcktqYAghhCjF5J3PSHT5F1IDQwghRGkkAYaRSP6FEEKI0kwCDCORKapCCCFKMwkwjESmqAohhCjNJMAwElmmXQghRGkmAYYRaLUKwQ9liEQIIUTpJQGGEehqYJibaXCXGhhCCCFKIQkwjEBXA8PN0QYLqYEhhBCiFJJ3PyOQBE8hhBClnQQYRiDLtAshhCjtJMAwAunBEEIIUdpJgGEEUmRLCCFEaScBhhHIEIkQQojSzuQBxrx58/Dx8cHGxoZGjRqxZ8+eXB23b98+LCwsaNCggXEbmEdaraIvsuVZTnowhBBClE4mDTBWrlzJxIkT+eCDDzhx4gRt2rShR48eBAUFZXtcZGQkw4YNo1OnToXU0ty7+yiBxBS1Boabo9TAEEIIUTqZNMCYPXs2o0aNYvTo0dSuXZu5c+fi6enJ/Pnzsz3u1VdfZdCgQbRo0aKQWpp7ugRPdyepgSGEEKL0Mtk7YGJiIseOHaNr167ptnft2pX9+/dnedyiRYu4evUqU6dOzdXzJCQkEBUVle5mTJLgKYQQQpgwwIiIiCAlJQVXV9d0211dXQkLC8v0mMuXL/Pee++xbNkyLCwscvU8M2bMwMnJSX/z9PQscNuzIwmeQgghRBFI8tRoNOm+VxQlwzaAlJQUBg0axKeffkrNmjVzff4pU6YQGRmpv926davAbc6O1MAQQgghIHfdAEbg4uKCubl5ht6K8PDwDL0aANHR0Rw9epQTJ04wbtw4ALRaLYqiYGFhwZYtW+jYsWOG46ytrbG2tjbOi8iE9GAIIYQQJuzBsLKyolGjRgQEBKTbHhAQQMuWLTPs7+joyOnTpwkMDNTfxowZg6+vL4GBgTRr1qywmp6tW/elB0MIIYQwWQ8GwKRJkxg6dCiNGzemRYsW/PzzzwQFBTFmzBhAHd64ffs2v/32G2ZmZvj7+6c7vmLFitjY2GTYbiparcLth5LkKYQQQpg0wBg4cCD37t1j2rRphIaG4u/vz4YNG/Dy8gIgNDQ0x5oYRUl4dAJJKYrUwBBCCFHqaRRFUUzdiMIUFRWFk5MTkZGRODo6GvTcR2/cp/+PB6jsbMvedzPmgwghhBDFWV7eQ00+i6Qk0SV4ekqCpxBCiFJOAgwDkimqQgghhEoCDAOSKapCCCGESgIMA5Iy4UIIIYRKAgwDkiESIYQQQiUBhoGkq4FRToZIhBBClG4SYBiIrgaGhZkGV4fCK00uhBBCFEUSYBjIrdThEfeyNliYy2UVQghRusk7oYHo8y/KyvCIEEIIIQGGgQTflxkkQgghhI4EGAYiNTCEEEKINBJgGEjwQ5miKoQQQuhIgGEg+nVIZIqqEEIIIQGGIaRoFUIeSg6GEEIIoSMBhgGER8en1cBwtDF1c4QQQgiTkwDDAHTDIx5lbTE305i4NUIIIYTpSYBhALIGiRBCCJGeBBgGIDUwhBBCiPQkwDCACg7WNPZypra7o6mbIoQQQhQJFqZuQEnwQtMqvNC0iqmbIYQQQhQZ0oMhhBBCCIOTAEMIIYQQBicBhhBCCCEMTgIMIYQQQhicBBhCCCGEMDgJMIQQQghhcBJgCCGEEMLgJMAQQgghhMFJgCGEEEIIg5MAQwghhBAGJwGGEEIIIQyu1K1FoigKAFFRUSZuiRBCCFG86N47de+l2Sl1AUZ0dDQAnp6eJm6JEEIIUTxFR0fj5OSU7T4aJTdhSAmi1WoJCQnBwcEBjUZjsPNGRUXh6enJrVu3cHSUZdsNRa6rcch1NQ65rsYj19Y48npdFUUhOjoaDw8PzMyyz7IodT0YZmZmVK5c2Wjnd3R0lF9+I5DrahxyXY1DrqvxyLU1jrxc15x6LnQkyVMIIYQQBicBhhBCCCEMTgIMA7G2tmbq1KlYW1ubuiklilxX45DrahxyXY1Hrq1xGPO6lrokTyGEEEIYn/RgCCGEEMLgJMAQQgghhMFJgCGEEEIIg5MAQwghhBAGJwGGAcybNw8fHx9sbGxo1KgRe/bsMXWTipXdu3fz9NNP4+HhgUajYd26dekeVxSFTz75BA8PD2xtbWnfvj1nz541TWOLkRkzZtCkSRMcHByoWLEi/fr14+LFi+n2kWubP/Pnz6devXr64kQtWrRg48aN+sfluhbcjBkz0Gg0TJw4Ub9Nrmv+fPLJJ2g0mnQ3Nzc3/ePGuq4SYBTQypUrmThxIh988AEnTpygTZs29OjRg6CgIFM3rdiIiYmhfv36fP/995k+PmvWLGbPns3333/PkSNHcHNzo0uXLvp1ZUTmdu3axeuvv87BgwcJCAggOTmZrl27EhMTo99Hrm3+VK5cmZkzZ3L06FGOHj1Kx44d6du3r/6fslzXgjly5Ag///wz9erVS7ddrmv++fn5ERoaqr+dPn1a/5jRrqsiCqRp06bKmDFj0m2rVauW8t5775moRcUboKxdu1b/vVarVdzc3JSZM2fqt8XHxytOTk7Kjz/+aIIWFl/h4eEKoOzatUtRFLm2hubs7Kz8+uuvcl0LKDo6WqlRo4YSEBCgtGvXTpkwYYKiKPL7WhBTp05V6tevn+ljxryu0oNRAImJiRw7doyuXbum2961a1f2799volaVLNevXycsLCzdNba2tqZdu3ZyjfMoMjISgHLlygFybQ0lJSWFFStWEBMTQ4sWLeS6FtDrr79Or1696Ny5c7rtcl0L5vLly3h4eODj48MLL7zAtWvXAONe11K32JkhRUREkJKSgqura7rtrq6uhIWFmahVJYvuOmZ2jW/evGmKJhVLiqIwadIkWrdujb+/PyDXtqBOnz5NixYtiI+Px97enrVr11KnTh39P2W5rnm3YsUKjh8/zpEjRzI8Jr+v+desWTN+++03atasyZ07d5g+fTotW7bk7NmzRr2uEmAYwJPLviuKYtCl4IVc44IaN24cp06dYu/evRkek2ubP76+vgQGBvLw4UNWr17N8OHD2bVrl/5xua55c+vWLSZMmMCWLVuwsbHJcj+5rnnXo0cP/f26devSokULqlWrxpIlS2jevDlgnOsqQyQF4OLigrm5eYbeivDw8AzRoMgfXaazXOP8Gz9+POvXr2fHjh1UrlxZv12ubcFYWVlRvXp1GjduzIwZM6hfvz7ffPONXNd8OnbsGOHh4TRq1AgLCwssLCzYtWsX3377LRYWFvprJ9e14MqUKUPdunW5fPmyUX9fJcAoACsrKxo1akRAQEC67QEBAbRs2dJErSpZfHx8cHNzS3eNExMT2bVrl1zjHCiKwrhx41izZg3bt2/Hx8cn3eNybQ1LURQSEhLkuuZTp06dOH36NIGBgfpb48aNGTx4MIGBgVStWlWuq4EkJCRw/vx53N3djfv7WqAUUaGsWLFCsbS0VBYsWKCcO3dOmThxolKmTBnlxo0bpm5asREdHa2cOHFCOXHihAIos2fPVk6cOKHcvHlTURRFmTlzpuLk5KSsWbNGOX36tPLiiy8q7u7uSlRUlIlbXrS99tpripOTk7Jz504lNDRUf4uNjdXvI9c2f6ZMmaLs3r1buX79unLq1Cnl/fffV8zMzJQtW7YoiiLX1VAen0WiKHJd8+utt95Sdu7cqVy7dk05ePCg0rt3b8XBwUH/PmWs6yoBhgH88MMPipeXl2JlZaU0bNhQPw1Q5M6OHTsUIMNt+PDhiqKo06imTp2quLm5KdbW1krbtm2V06dPm7bRxUBm1xRQFi1apN9Hrm3+jBw5Uv83X6FCBaVTp0764EJR5LoaypMBhlzX/Bk4cKDi7u6uWFpaKh4eHsqzzz6rnD17Vv+4sa6rLNcuhBBCCIOTHAwhhBBCGJwEGEIIIYQwOAkwhBBCCGFwEmAIIYQQwuAkwBBCCCGEwUmAIYQQQgiDkwBDCCGEEAYnAYYQQgghDE4CDCFEiaDRaFi3bp2pmyGESCUBhhCiwEaMGIFGo8lw6969u6mbJoQwEQtTN0AIUTJ0796dRYsWpdtmbW1totYIIUxNejCEEAZhbW2Nm5tbupuzszOgDl/Mnz+fHj16YGtri4+PD3/++We640+fPk3Hjh2xtbWlfPnyvPLKKzx69CjdPgsXLsTPzw9ra2vc3d0ZN25cuscjIiJ45plnsLOzo0aNGqxfv964L1oIkSUJMIQQheKjjz7iueee4+TJkwwZMoQXX3yR8+fPAxAbG0v37t1xdnbmyJEj/Pnnn2zdujVdADF//nxef/11XnnlFU6fPs369eupXr16uuf49NNPGTBgAKdOnaJnz54MHjyY+/fvF+rrFEKkKvB6rEKIUm/48OGKubm5UqZMmXS3adOmKYqiLh0/ZsyYdMc0a9ZMee211xRFUZSff/5ZcXZ2Vh49eqR//L///lPMzMyUsLAwRVEUxcPDQ/nggw+ybAOgfPjhh/rvHz16pGg0GmXjxo0Ge51CiNyTHAwhhEF06NCB+fPnp9tWrlw5/f0WLVqke6xFixYEBgYCcP78eerXr0+ZMmX0j7dq1QqtVsvFixfRaDSEhITQqVOnbNtQr149/f0yZcrg4OBAeHh4fl+SEKIAJMAQQhhEmTJlMgxZ5ESj0QCgKIr+fmb72Nra5up8lpaWGY7VarV5apMQwjAkB0MIUSgOHjyY4ftatWoBUKdOHQIDA4mJidE/vm/fPszMzKhZsyYODg54e3uzbdu2Qm2zECL/pAdDCGEQCQkJhIWFpdtmYWGBi4sLAH/++SeNGzemdevWLFu2jMOHD7NgwQIABg8ezNSpUxk+fDiffPIJd+/eZfz48QwdOhRXV1cAPvnkE8aMGUPFihXp0aMH0dHR7Nu3j/HjxxfuCxVC5IoEGEIIg9i0aRPu7u7ptvn6+nLhwgVAneGxYsUKxo4di5ubG8uWLaNOnToA2NnZsXnzZiZMmECTJk2ws7PjueeeY/bs2fpzDR8+nPj4eObMmcPkyZNxcXGhf//+hfcChRB5olEURTF1I4QQJZtGo2Ht2rX069fP1E0RQhQSycEQQgghhMFJgCGEEEIIg5McDCGE0clIrBClj/RgCCGEEMLgJMAQQgghhMFJgCGEEEIIg5MAQwghhBAGJwGGEEIIIQxOAgwhhBBCGJwEGEIIIYQwOAkwhBBCCGFw/we7TFBwR+/EQQAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(6, 4))\n",
    "plt.plot(history.history['accuracy'], label='Training Accuracy')\n",
    "plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n",
    "plt.title('Training and Validation Accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.legend()\n",
    "plt.show()\n",
    "\n",
    "plt.figure(figsize=(6, 4))\n",
    "plt.plot(history.history['loss'], label='Training Loss')\n",
    "plt.plot(history.history['val_loss'], label='Validation Loss')\n",
    "plt.title('Training and Validation Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Loss')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}