reproduce_glorys_ose_results.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### CUDA SETTINGS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os \n",
    "os.environ['CUDA_VISIBLE_DEVICES'] = '1'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### IMPORTS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from src.reproduce.reproduce_utils import call_cfg_key, reconstruct_from_config, eval_ose, plot_results\n",
    "from omegaconf import OmegaConf\n",
    "import datetime"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### IMPORTANT FILE PATHS"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# downloaded cmems data\n",
    "concatenated_6_nadirs_filepath = 'data/gridded/input_test_6sat_2023.nc'\n",
    "test_1_nadir_filepath = 'data/concat/concatenated_ref_2023_reprocessed.nc'\n",
    "\n",
    "# model hydra config\n",
    "your_config = 'models/4dvarnet_glorys_10years/config.yaml'\n",
    "# best model ckpt path\n",
    "best_ckpt_path = 'models/4dvarnet_glorys_10years/val_mse=26.93737-epoch=018.ckpt'\n",
    "# name of the dir under which the reconstructed files will be stored\n",
    "your_rec_dir = 'rec_dir'\n",
    "\n",
    "# reconstructed files will be stored in <your_rec_dir>/reproduce/version_0/\n",
    "# careful, it will be version_1, 2, 3... instead of version_0 and so on if you use this notebook multiple times\n",
    "REC_PATHS = your_rec_dir+'/reproduce/version_0/test_data_{}.nc'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### CONFIG OVERRIDES"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# min and max time of REFERENCE DATA\n",
    "min_time=\"2023-01-01\"\n",
    "max_time=\"2023-12-31\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "min/max time: 2023-01-01 / 2023-12-31\n",
      "min/max offseted time: 2023-01-16 / 2023-12-16\n"
     ]
    }
   ],
   "source": [
    "min_time_date = datetime.datetime.strptime(min_time, '%Y-%m-%d')\n",
    "max_time_date = datetime.datetime.strptime(max_time, '%Y-%m-%d')\n",
    "\n",
    "time_offset = datetime.timedelta(days=15)\n",
    "\n",
    "min_time_offset = min_time_date + time_offset\n",
    "max_time_offset = max_time_date - time_offset\n",
    "\n",
    "min_time_offset_str = min_time_offset.strftime('%Y-%m-%d')\n",
    "max_time_offset_str = max_time_offset.strftime('%Y-%m-%d')\n",
    "\n",
    "print('min/max time: {} / {}\\nmin/max offseted time: {} / {}'.format(min_time, max_time, min_time_offset_str, max_time_offset_str))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### LOADING CONFIG (and overriding with your information)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "config = OmegaConf.load(your_config)\n",
    "OmegaConf.set_struct(config, True)\n",
    "\n",
    "# override path to cmems input data\n",
    "OmegaConf.update(config, key='paths.ose_6_sats', value=concatenated_6_nadirs_filepath)\n",
    "\n",
    "# override output path of reconstructed paths\n",
    "OmegaConf.update(config, key='trainer.logger.save_dir', value=your_rec_dir)\n",
    "\n",
    "# override time necessary time values in config\n",
    "OmegaConf.update(config, key='datamodule.domains.train.time._args_', value=[min_time, min_time_offset_str])\n",
    "OmegaConf.update(config, key='datamodule.domains.val.time._args_', value=[min_time, min_time_offset_str])\n",
    "\n",
    "#full_time allocated to the test dataset\n",
    "OmegaConf.update(config, key='datamodule.domains.test.time._args_', value=[min_time, max_time])\n",
    "\n",
    "# pre metric fn is applied to a a window cropped on each side by 15 days\n",
    "OmegaConf.update(config, key='model.pre_metric_fn.time._args_', value=[min_time_offset_str, max_time_offset_str])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### PREDICT WITH MODEL:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/DATASET/envs/p24hasle/miniconda/envs/repro_env/lib/python3.11/site-packages/lightning_fabric/plugins/environments/slurm.py:168: PossibleUserWarning: The `srun` command is available on your system but is not used. HINT: If your intention is to run Lightning on SLURM, prepend your python command with `srun` like so: srun python /DATASET/envs/p24hasle/miniconda/envs/repro_env/lib/ ...\n",
      "  rank_zero_warn(\n",
      "GPU available: True (cuda), used: True\n",
      "TPU available: False, using: 0 TPU cores\n",
      "IPU available: False, using: 0 IPUs\n",
      "HPU available: False, using: 0 HPUs\n",
      "Global seed set to 333\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "test_regrid: None\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Restoring states from the checkpoint path at models/4dvarnet_glorys_10years/val_mse=26.93737-epoch=018.ckpt\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Norm stats (0.48749464750289917, 0.4605884253978729)\n",
      "dataset sizes\n",
      "dataset length\n",
      "dataset sizes\n",
      "dataset length\n",
      "dataset sizes\n",
      "dataset length\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/DATASET/envs/p24hasle/miniconda/envs/repro_env/lib/python3.11/site-packages/pytorch_lightning/utilities/migration/utils.py:55: PossibleUserWarning: The loaded checkpoint was produced with Lightning v2.2.4, which is newer than your current Lightning version: v2.0.9\n",
      "  rank_zero_warn(\n",
      "LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [1]\n",
      "Loaded model weights from the checkpoint at models/4dvarnet_glorys_10years/val_mse=26.93737-epoch=018.ckpt\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Testing DataLoader 0: 100%|██████████| 85/85 [00:18<00:00,  4.53it/s]rec_dir/test_data_0.nc\n",
      "rec_dir/test_data_1.nc\n",
      "rec_dir/test_data_2.nc\n",
      "rec_dir/test_data_3.nc\n",
      "rec_dir/test_data_4.nc\n",
      "rec_dir/test_data_5.nc\n",
      "rec_dir/test_data_6.nc\n",
      "rec_dir/test_data_7.nc\n",
      "rec_dir/test_data_8.nc\n",
      "rec_dir/test_data_9.nc\n",
      "rec_dir/test_data_10.nc\n",
      "rec_dir/test_data_11.nc\n",
      "rec_dir/test_data_12.nc\n",
      "rec_dir/test_data_13.nc\n",
      "rec_dir/test_data_14.nc\n",
      "rec_dir/test_data_15.nc\n",
      "rec_dir/test_data_16.nc\n",
      "rec_dir/test_data_17.nc\n",
      "rec_dir/test_data_18.nc\n",
      "rec_dir/test_data_19.nc\n",
      "|     |       -13 |       -12 |       -11 |      -10 |        -9 |        -8 |        -7 |        -6 |        -5 |        -4 |        -3 |        -2 |      -1 |         0 |         1 |         2 |         3 |         4 |         5 |         6 |\n",
      "|:----|----------:|----------:|----------:|---------:|----------:|----------:|----------:|----------:|----------:|----------:|----------:|----------:|--------:|----------:|----------:|----------:|----------:|----------:|----------:|----------:|\n",
      "| mu  |   0.14276 |   0.08941 |   0.065   |   0.0722 |   0.04953 |   0.08661 |   0.08818 |   0.084   |   0.08862 |   0.11977 |   0.10634 |   0.14439 | 0.12252 |   0.08611 |   0.09307 |   0.14514 |   0.1409  |   0.16067 |   0.14811 |   0.17371 |\n",
      "| sig |   0.03569 |   0.03725 |   0.03781 |   0.0366 |   0.03879 |   0.03598 |   0.0355  |   0.03692 |   0.03792 |   0.03743 |   0.03898 |   0.03749 | 0.03975 |   0.04061 |   0.04114 |   0.03893 |   0.04022 |   0.04196 |   0.03998 |   0.0395  |\n",
      "| lx  | nan       | nan       | nan       | nan      | nan       | nan       | nan       | nan       | nan       | nan       | nan       | nan       | 1.976   | nan       | nan       | nan       | nan       | nan       | nan       | nan       |\n",
      "| lt  | nan       | nan       | nan       | nan      | nan       | nan       | nan       | nan       | nan       | nan       | nan       | nan       | 3.039   | nan       | nan       | nan       | nan       | nan       | nan       | nan       |\n",
      "Testing DataLoader 0: 100%|██████████| 85/85 [11:39<00:00,  8.23s/it]\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "reconstruct_from_config(config, your_rec_dir, best_ckpt_path)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### OSE RESULTS:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "leadtime 0 - RMSE: 0.7953 | PSD: 18.0000\n",
      "leadtime 1 - RMSE: 0.7716 | PSD: 17.0000\n",
      "leadtime 2 - RMSE: 0.7250 | PSD: 17.0000\n",
      "leadtime 3 - RMSE: 0.7020 | PSD: 17.0000\n",
      "leadtime 4 - RMSE: 0.6600 | PSD: 16.0000\n",
      "leadtime 5 - RMSE: 0.6693 | PSD: 17.0000\n",
      "leadtime 6 - RMSE: 0.5968 | PSD: 13.0000\n"
     ]
    }
   ],
   "source": [
    "lead_times = range(13,20)\n",
    "\n",
    "RMSE_array = []\n",
    "\n",
    "for lead_time in lead_times:\n",
    "    a,b = eval_ose(\n",
    "        path_alongtrack = test_1_nadir_filepath,\n",
    "        path_rec = REC_PATHS.format(lead_time),\n",
    "        time_min = min_time_offset_str,\n",
    "        time_max= max_time_offset_str\n",
    "    )\n",
    "\n",
    "    print('leadtime {} - RMSE: {:.4f} | PSD: {:.4f}'.format(lead_time-lead_times[0], a, b))\n",
    "    RMSE_array.append(a)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### PLOT RESULTS (<html><span style=\"color:red\">RED</span></html> on graph)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4AAAAGwCAYAAAD40QvTAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/TGe4hAAAACXBIWXMAAA9hAAAPYQGoP6dpAAC6aklEQVR4nOzdd1zU9R/A8dcdGwRcgLgQt4gTt5kjcWvaULMcaZppKlmWTbUcaWmuzExxZW6t/GUqlVtLRTFzDwwHCC5AERXu8/vjGwcnwzsED+H9fDw+D7nvfb7f+9zb47j3fZZOKaUQQgghhBBCCJHv6a3dACGEEEIIIYQQj4ckgEIIIYQQQghRQEgCKIQQQgghhBAFhCSAQgghhBBCCFFASAIohBBCCCGEEAWEJIBCCCGEEEIIUUBIAiiEEEIIIYQQBYSttRuQFxkMBi5fvoyrqys6nc7azRFCCCGEGZRSxMfHU7JkSfR6+Y5bCCEyIglgBi5fvkyZMmWs3QwhhBBCZMOFCxcoXbq0tZshhBB5kiSAGXB1dQW0PyBubm45em2DwUBMTAweHh7y7eRDSKzMJ7Eyn8TKfBIry0i8zJdbsYqLi6NMmTLGv+NCCCHSkwQwAynDPt3c3HIlAUxMTMTNzU0+IDyExMp8EivzSazMJ7GyjMTLfLkdK5m+IYQQmZO/UEIIIYQQQghRQEgCKIQQQgghhBAFhCSAQgghhBBCCFFASAIohBBCCCGEEAWEJIBCCCGEEEIIUUBIAiiEEEIIIYQQBYQkgEIIIYQQQghRQEgCKIQQQgghhBAFhCSAQgghhBBCCFFASAIohBBCCCGEEAWEJIBCCCGEEEIIUUBIAiiEEEIIIYQQBYSttRtQEOmSEyDpNugzyL91NmDjmHo76XYWV9KDrVM26yYAKrMWgq1zNuveAQyZN8PWxfy6+jTtTU4ElWzedR9W18YZdLr/6t4FlZRDdZ1A99//afI9UPdzpq7eEfQ2Wdc1GLTXlSE59XVluA+Ge1lc1wH0ttmomwSGu1nUtQe9XTbqJoMhMfO6Ojuwsbe8rjJA8p3U+1JilfI7mFXddNe1BRuH/+oqSE7IoboW/N4/zvcIQ3Im71d55D3Ckt/7x/QekeX7uzXfIzKqa833CGwyv18IIUSukgTQCry2V8j8zpIdoMUvqbfXemb+wdGzObTelnr7p3Jw92rGdYvWg3b7U2//4ge3/824rrsfdDyaentzfYg9lnFdFx949nzq7d+ehusHMq7rUByej0m9va09RG/PuK6NM7wYn3p75/NweWPGdQF6pfnwuac3XFiTed3ut1I/DO57HcIXZ173uWhw9NB+PjgSTs/JvG6XcChUTvv57w/h+JeZ1+3wDxSurv18dCL8My7zum33QbH62s8nZ0DYu+mq6AEvwNDyd/BupR08Mw8OvJn5dZv/D0p11H4+vwz+fDXzuk+tgrIvaj9fXA+7umdet9FCKN9P+zlyM2zvlHnderOh8lDt55id8HvLzOvWngJ+o7SfbxyEzQ0yr+s/BmqO1X6OPQ4b/Y13pcTKqNo7UOcL7efbEfCzb+bXrTQE6n+t/Xz3KqzzzLyub19ovEj7OTkBVhXKvG6ZF6DZ6tTbWdV9jO8R+tv/msYqRV55j+iRJqHNA+8RukNv43Xmm8zrWvE9wuiZreDVQvvZmu8R5fpkfr8QQohcJUNAhRBCCCGEEKKA0CmlMhu3U2DFxcXh7u5ObGwsbm5uOXptg8FATOR5PDw80MsQ0CzrGvROREdH4+npiV7dyxPDu/LqEFCDwUBMTAweXmXQ26YMqZQhoEC6YZ3GWKX8DsoQ0EzqJmAwJJvGKrUReeI9Iq8NATXcv0NMdGTm7+8yBPS/uvYYsEl9f88oVtmUm3+/hRAiv5AhoFagbJy1DyPm/NFL+6ElR+s6P7xOtuo6PbyOuXUNaT74pf3A+zAW1XUAHHKhrj1g//jqGgwom9upH+xAS6xSkquHsaiubeoHvRytawN6M1/DltTV6U1/N1JildHv4IN1s7yuLnfqQh6p65x1rB6sa/Z1c/A9Iq088h5h9vv7436PyIg13yMMWST2QgghcpUMARVCCCGEEEKIAkISQCGEEEIIIYQoICQBFEIIIYQQQogCQhJAIYQQQgghhCggJAEUQgghhBBCiAJCEkAhhBBCCCGEKCAkARRCCCGEEEKIAkISQCGEEEIIIYQoICQBFEIIIYQQQogCQhJAIYQQQgghhCggJAEUQgghhBBCiAJCEkAhhBBCCCGEKCAkARRCCCGEEEKIAkISQCGEEEIIIYQoICQBFEIIIYQQQogCwuoJ4Jw5c/D19cXR0ZGAgAB27tyZZf1ly5ZRq1YtnJ2d8fb25tVXX+XatWvG+xctWoROp0tXEhMTc/upCCGEEEIIIUSeZtUEcOXKlQQFBfHhhx9y6NAhmjVrRvv27YmIiMiw/q5du+jTpw8DBgzg6NGjrF69mv379/Paa6+Z1HNzcyMyMtKkODo6Po6nJIQQQgghhBB5llUTwGnTpjFgwABee+01qlWrxvTp0ylTpgzffPNNhvX//PNPypUrx/Dhw/H19eWpp57i9ddf58CBAyb1dDodJUqUMClCCCGEEEIIUdDZWuuB7927R2hoKKNHjzY53qZNG/bs2ZPhOU2aNOHDDz9k48aNtG/fnujoaNasWUPHjh1N6t26dQsfHx+Sk5OpXbs2n332GXXq1Mm0LXfv3uXu3bvG23FxcQAYDAYMBkN2n2KGDAYDSqkcv25+JLEyn8TKfBIr80msLCPxMl9uxUpiL4QQD2e1BPDq1askJyfj5eVlctzLy4uoqKgMz2nSpAnLli2jR48eJCYmkpSURJcuXZg1a5axTtWqVVm0aBE1atQgLi6OGTNm0LRpUw4fPkylSpUyvO6kSZMYN25cuuMxMTE5PnfQYDAQGxuLUgq93upTMPM0iZX5JFbmk1iZT2JlGYmX+XIrVvHx8Tl2LSGEyK+slgCm0Ol0JreVUumOpTh27BjDhw/nk08+oW3btkRGRjJq1CgGDx7MggULAGjUqBGNGjUyntO0aVPq1q3LrFmzmDlzZobXff/99xk5cqTxdlxcHGXKlMHDwwM3N7dHfYomDAYDOp0ODw8P+YDwEBIr80mszCexMp/EyjISL/PlVqxkvr8QQjyc1RLA4sWLY2Njk663Lzo6Ol2vYIpJkybRtGlTRo0aBUDNmjVxcXGhWbNmjB8/Hm9v73Tn6PV66tevz+nTpzNti4ODAw4ODhmemxt/xHU6Xa5dO7+RWJlPYmU+iZX5JFaWkXiZLzdiJXEXQoiHs9o7pb29PQEBAYSEhJgcDwkJoUmTJhmek5CQkO7N3cbGBtB6DjOilCIsLCzD5FAIIYQQQgghChKrDgEdOXIkvXv3pl69ejRu3Jh58+YRERHB4MGDAW1o5qVLl1iyZAkAnTt3ZuDAgXzzzTfGIaBBQUE0aNCAkiVLAjBu3DgaNWpEpUqViIuLY+bMmYSFhfH1119b7XkKIYQQQgghRF5g1QSwR48eXLt2jU8//ZTIyEj8/f3ZuHEjPj4+AERGRprsCdivXz/i4+OZPXs2b7/9NoULF6ZVq1ZMnjzZWOfmzZsMGjSIqKgo3N3dqVOnDjt27KBBgwaP/fkJIYQQQgghRF6iU5mNnSzA4uLicHd3JzY2NlcWgYmOjsbT01PmKjyExMp8EivzSazMJ7GyjMTLfLkVq9z8+y2EEPmF/IUSQgghhBBCiAJCEkAhhBBCCCGEKCAkARRCCCGEEEKIAkISQCGEEEIIIYQoICQBFEIIIYQQQogCQhJAIYQQQgghhCggJAEUQgghhBBCiAJCEkAhhBBCCCGEKCAkARRCCCGEEEKIAkISwMdIKa0IIYQQQgghhDXYWrsBBcmFC1C3ro7KlYtSt66OGjXA318r7u7Wbp0QQgghhBAiv5ME8DE6cgSuXdOxd689e/ea3lemDMaEMOXfqlXB0dE6bRVCCCGEEELkP5IAPkbPPAMHDhjYsyeOiAh3jh7VceQIXLyo9Q5euAAbN6bWt7GBSpVMk8IaNaB8ee0+IYQQQgghhLCEJICPkaMj1KkDpUol4unphl6vA+DmTfjnH60cOZL6740bcOKEVtasSb2OkxP4+aUOH01JDkuWBJ3OOs9NCCGEEEIIkfdJApgHFC4MTz2llRRKQWRk+qTw2DG4cwdCQ7WSVpEi6XsLq1fXjgshhBBCCCGEJIB5lE6n9eiVLAlt2qQeT06Gc+fS9xieOqX1GO7cqZW0SpVKP7+wWjWtJ1EIIYQQQghRcEgC+IRJmRdYqRJ065Z6PDERTp407S385x+IiIBLl7SyaVNqfb0eKlbEZCXSGjWgQgWwlVeFEEIIIYQQ+ZJ81M8nHB2hVi2tpBUbC0ePph9Keu2a1mt46hSsXZta38EhdX5h2l7DUqVkfqEQQgghhBBPOkkA8zl3d2jSRCsplIIrV9InhUePQkICHDqklQev82BS6O8PRYs+3ucjhBBCCCGEyD5JAAsgnQ5KlNBK69apxw0GCA83TQz/+UcbWhobC7t3ayWtkiVNE0J/f60H0dn58T4nIYQQQgghxMNJAiiM9HptDmCFCvDss6nH797Vhoo+OL/w/Hm4fFkrW7ak1tfptGs82FtYqZLMLxRCCCGEEMKa5OO4eCgHBy2Jq1HD9HhcnLYtxYNDSWNi4MwZraxfn1rf3l5bffTBoaRlysj8QiGEEEIIIR4HSQBFtrm5QaNGWkkrOjp9b+E//8Dt23D4sFYevM6Dm9rL/EIhhBBCCCFyniSAIsd5esIzz2glhcEA//6bvrfwxAmtJ3HPHq2kVaKEjipVitC4sY569aBuXShXTnoLhRBCCCGEyC5JAMVjodeDr69WOndOPX7vHpw+nb7H8Nw5iIrSERXlwPbtqfWLFtUSwYCA1H/Ll5ekUAghhBBCCHNIAiisyt4eqlfXSlq3bsGRIwZ27ozn9Gk3Dh7UceQIXL8Ov/2mlRSFC2vJYEpCGBCgLUKj1z/WpyKEEEIIIUSeJwmgyJMKFYKGDcHX9w6enq7o9Tru3dN6B0NDU8vff8PNm/DHH1pJ4eYGdeqkJoQBAdoqpJIUCiGEEEKIgkwSQPHEsLdP7ekbOFA7dv++toF9aCgcPKj9e/iwNq9w+3ZMho8WKpSaFKb0FlapAjY21nk+QgghhBBCPG6SAIonmp0d1K6tlQEDtGP378Px46kJYWgohIVpw0p37tRKChcX7dy0w0erVpX9CoUQQgghRP4kH3NFvmNnBzVraqVfP+1YUhKcPGk6fDQsTNuaYvduraRwcoJatVITwrp1wc9Pu64QQgghhBBPMqvPiJozZw6+vr44OjoSEBDAzrTdMxlYtmwZtWrVwtnZGW9vb1599VWuXbtmUmft2rX4+fnh4OCAn58f69PuRi4KJFtbbaGZPn1gxgzYtQtiY7WN7JcuhaAgePppbZjonTvw55/w9dfQv7/WQ+jqqs1JHDIE5s+HQ4e0FUyFEEIIIYR4kli1B3DlypUEBQUxZ84cmjZtyrfffkv79u05duwYZcuWTVd/165d9OnTh6+++orOnTtz6dIlBg8ezGuvvWZM8vbu3UuPHj347LPP6NatG+vXr6d79+7s2rWLhg0bPu6nKPIwGxuoVk0rr7yiHTMYtG0p0s4pPHhQm1O4b59WUtjbaxvXp11oxt8fHBys83yEEEIIIYR4GJ1SSlnrwRs2bEjdunX55ptvjMeqVatG165dmTRpUrr6X375Jd988w1nz541Hps1axZTpkzhwoULAPTo0YO4uDh+/fVXY5127dpRpEgRli9fbla74uLicHd3JzY2Fjc3t+w+vQwZDAaio6Px9PREL0tSZimvxMpggLNnTecUHjyorT76IDs7LQlMO3y0Zk1wdMztNuaNWD0JJFbmk1hZRuJlvtyKVW7+/RZCiPzCaj2A9+7dIzQ0lNGjR5scb9OmDXv27MnwnCZNmvDhhx+yceNG2rdvT3R0NGvWrKFjx47GOnv37uWtt94yOa9t27ZMnz4907bcvXuXu3fvGm/HxcUB2h8og8Fg6VPLksFgQCmV49fNj/JSrCpU0MqLL2q3lYLw8JRkUGdMCm/c0HHokDZEdP58ra6traJ69ZQVTBV162pzDJ2ccq59eSlWeZ3EynwSK8tIvMyXW7GS2AshxMNZLQG8evUqycnJeHl5mRz38vIiKioqw3OaNGnCsmXL6NGjB4mJiSQlJdGlSxdmzZplrBMVFWXRNQEmTZrEuHHj0h2PiYkhMTHRkqf1UAaDgdjYWJRS8g3xQ+T1WBUqBM2bawW0pPDiRRsOH7bl77/tOHLEjr//tuP6dT2HD2vbUyxcqAPAxkZRuXISNWokUbPmfWrWvE/16kk4O2evQz6vxyovkViZT2JlGYmX+XIrVvHx8Tl2LSGEyK+svgqoTqczua2USncsxbFjxxg+fDiffPIJbdu2JTIyklGjRjF48GAWLFiQrWsCvP/++4wcOdJ4Oy4ujjJlyuDh4ZErQ0B1Oh0eHh7yAeEhnsRYeXlpQz9TKAUXLhgIDYVDh1J7CqOjdRw/bsfx43asWqV1Ber1iqpVU7akUNSpo+1bWKjQwx/3SYyVtUiszCexsozEy3y5FSvH3B5vL4QQ+YDVEsDixYtjY2OTrmcuOjo6XQ9eikmTJtG0aVNGjRoFQM2aNXFxcaFZs2aMHz8eb29vSpQoYdE1ARwcHHDIYOUOvV6fs3/E791D9/bbONaqhb5rV/TFi+fctfMpnU6X8/8Pj1m5clp5/nnttlJw6ZLpnMLQUIiK0nHsmLYy6fffa19Y6HTaZvVpN6+vUwcy+l4iP8TqcZFYmU9iZRmJl/lyI1YSdyGEeDirJYD29vYEBAQQEhJCt27djMdDQkJ49tlnMzwnISEB2wd26LaxsQG0Xj6Axo0bExISYjIPcMuWLTRp0iSnn4Ll/voL3ezZFAHUoEHa/gLPPKOVZs20XclFvqfTQenSWunSJfV4ZKRpQnjwoJYonjihlWXLUutWrmy6eX2tWo//eQghhBBCiCePVYeAjhw5kt69e1OvXj0aN27MvHnziIiIYPDgwYA2NPPSpUssWbIEgM6dOzNw4EC++eYb4xDQoKAgGjRoQMmSJQEYMWIETz/9NJMnT+bZZ5/lp59+4rfffmPXrl1We55GxYuj3nyTpC1bsDt1CuNqIV9+qS0f2bBhakLYsKG2z4AoMLy9oVMnraS4csU0IQwNhQsX4NQpraxYkVJTT8mSHvj46ChdGkqVwvhvys8lS8oWFUIIIYQQBZ1Vt4EAbSP4KVOmEBkZib+/P1999RVPP/00AP369eP8+fNs27bNWH/WrFnMnTuX8PBwChcuTKtWrZg8eTKlSpUy1lmzZg0fffQR586do0KFCkyYMIHnnnvO7DY9lm0gkpPRb9sGf/wBv/8O//5rWtHZWesVTEkIa9eGAja0RZZUz1h0tJYMph1C+uDLJzMeHqZJ4YM/ly6tDS/NYsrsE09eV+aTWFlG4mU+2QZCCCGsx+oJYF702PcBVArOndMSwT/+0EpMjOmJRYtCixapCWHlyvn7UzryYcoSMTEG9u27QUJCES5f1nPpEly8qA0hTfk5zU4nWXJxSZ8UPpgoenrCf6OvnzjyujKfxMoyEi/zSQIohBDWY/VVQAVaIpey0dygQdrO4//8k9o7uH07XL8O69ZpBbRP4a1apSaEpUtb9zkIqypWDAIC7uPpmXFHsVLaS+jBpPDBRPHGDbh9O3WIaWZsbbUhq1n1JJYsCbIgnxBCCCFE3iIJYF6k10PNmloJCoL79+HAgdSEcPdu7dP60qVaAahUKTUZbNlSywiE+I9Op70kihXLesGYhIT0SeGDP0dFQVKSNhfxwoWsH7dYscwTxJRjhQvn+85sIYQQQog8QxLAJ4GdHTRurJUPP4Q7d7QkMCUhPHAATp/Wyty52qfpWrVMVxg1ZzM5UeA5O2vfJVSqlHmdpCQtCcyqJ/HiRUhMhGvXtHL4cNaPmVlPYsrPXl5P7pBTIYQQQoi8RBLAJ5GTE7RurRWAmzdhxw4tGfz9dzh6FMLCtDJ1qjZe78EVRmU5SJFNtrap21hkRiltOOnDhpxev671OqZ8f5EZG5vUIaeZ9SSWKqX9agghhBBCiMxJApgfFC6sbSiXsqlcVBRs3ZqaEJ4/r/UY7t4Nn36qfUp+cIVR6V4ROUin09YtKlpUG8mcmYQEuHw5657EyEhITtZ+vngx68ctWvThC9i4u+fscxVCCCGEeJJIApgflSgBL72kFYDw8NRk8I8/tH0EtmzRCkCRIqYrjFapIpOyxGPh7AwVK2olM0lJ2n6IDxtyeueO1qN4/TocOZL59ZycdHh7F6diRR0VKoCvL5Qvn1okQRRCCCFEfiYJYEHg6wuvvaYVpbQhoikJ4fbt2li99eu1AtryjSkrjLZqBWXLWrf9okCztU3txcuMUtpI6KwWr7l0SZuPeOeOjnPnbDl3LuNrFS2amgw+mByWKaNNyRVCCCGEeFJJAljQ6HTg76+VESO07pXQ0NSEcPdubUze999rBbTumbQrjBYvbt3nIMQDdDqtI7tIEahRI/N6d+7AxYsG/v77JjduFOb8eT3nzmEsMTGpvYgHDqQ/38ZG+z4kbVKYNkksWlQ6z4UQQgiRt0kCWNClLBDTsCF88IH2CXnv3tSEcP9+OHNGK99+q53z4Aqjrq7WfQ5CmMnJSdtu09X1XoZ7Jt66pY2YTpsUppTwcLh7V/s3ZVT1g9zcTJPDtAmij4+svSSEEEII69MppZS1G5HXxMXF4e7uTmxsLG5ubjl6bYPBQHR0NJ6enugz2rE7r4mNNV1h9J9/TO+3tYUGDVKHizZunGOfcp+4WFmRxMp82Y2VwaAtSJNZghgZmfX5Op22EE1mvYeennmv91BeV5aReJkvt2KVm3+/hRAiv5AeQJE1d3fo3FkroK3GkXaF0fBw2LNHK599pnWxPPVUakJYt66sMCryBb0+dS7iU0+lvz8hQVtwN7MEMSEBLlzQyvbt6c93ds6899DXV7a4EEIIIUTOkARQWMbLC3r21Apon3ZTNqT/4w8tQQwJ0QpoW1S0aJG6qEy1anmvm0OIHODsDH5+WnmQUtriu5klhxcvagniP/+k72RP4e2deYLo7Z1+OKsQQgghREYkARSPxtcXBgzQilJw7Fhq7+C2bdrSjD/+qBXQPqm2apWaEPr4WK/tQjwmOp323YmXFzRqlP7+u3chIiLj5PDcOYiL04aYRkZq6zQ9yMEh/YqlKQmir69M0xVCCCFEKkkARc7R6aB6da0MH66tMHrwoOkKo5GRsGyZVkBbkSNluGirVuDhYd3nIIQVODhApUpaeZBS2k4tmSWHERFaAnnihFYy4uGRPjlMSRBLl5ZR2kIIIURBIgmgyD0pC8Q0aADvvw+JielXGD17Vivz5mnn1KyZmhBmNNFKiAJGp9O2lyhaFOrVS39/UpI2rzCzBPH6dW17i5gY+Ouv9Ofb2Wkd8Rklh+XK5frTE0IIIcRjJquAZiC3VwE9f+k8Hh4eGa58ZqO3wdHW0Xj79r3bmV5Lr9PjZOeUrboJ9xPI7L9ep9PhbOecrbp37t/BoAyZtsPF3iW17vUrGHbu1IaKbt+ebvKTs8GGe82aYffZZ9xrVI9kQ7JZ101MSsyyrrOdM7r/5iHeTbpLkiEpR+o62Tmh12n/p/eS73E/+X6O1HW0dcRGb5NlXYPBQExMDGW8y2Bnq+1Ufj/5PveS72V6XQdbB2z1thbXTTIkcTfpbqZ17W3ssbOxs7husiGZxKTETOva2dhhb2NvcV2DMnDn/h3jfSmxSvkdzKrug2z1tjjYaqvcKqVIuJ+QI3Ut+b239D3iXoKTce7hibO3tW0szsP5cPg3ApJSXk5KD0lpVpqxSwAUhQsbCKino0ljHY0aQ70AKFToMb1HWFD3Yb/3j+M94s69O0Reicz0/d2a7xEZ1bXme4SNzkZWARVCCCuRBDADuZ0A2nyW+XirDpU68EuvX4y3XSa6ZPrBsblPc7b122a87fGFB1cTrmZYt17JeuwfuN94u9z0cvwb+2+Gdf08/Dg65KjxdvU51TkWcyzDuj7uPpwPOm+8Xf+7+hy4nMEO2kBx5+LEjIox3m6xqAXb/81gOUTAOUnH7fGpL82OIzzYWCQmw7oAakxq3RdXv8iaY2syrXvr/VvGD4P9fuzH4sOLM60b/U40Hi7asNShvwxlzoE5mdYNHxFOucLlABi1ZRRf7v0y07r/vPEP1T2rAzB221jGbR+Xad19r+2jfqn6AHyx+wve/e3dTOv+3vt3WpVvBcDX+77mzV/fzLTu/176Hx0rdwRgUdgiXv3p1UzrrnphFS9WfxGA1UdX031N90zrLnx2If1q9wPgl1O/0Gl5p0zrzm4/m6ENhgKw7fw2Wi5umWndKa2nMKrpKAD2X9pPg/kNMq07pvkYxrYYC8DR6KP4f+Ofad13Gr/DF22+AOD8zfP4zvDNtO6QekP4uuPXAMTcjsHzS89M6/at1ZdFXRcBWpJWaFKhTOu+4PcCq19cbbytG5f5Qkm59R5Rzr4eryTsN/Yc7mtSDoNbxu8RjvF+vJ50lKZNoWlTCPzRCu8Rds7c/iA1+e34Q0c2nt6YYV14PO8RQ34ZwjcHvsm0bl54j9jadystyrUArPse0admH0kAhRDCSmTdOJE3OTlhOHGChF69UDY22vg1IUSuKV5c28ll2TJtpHaZMpnXTbwDM2ZA9+7athinTz2+dgohhBDi0UgPYAZkCGjeGN7lZOukfUMcH8+9CeNIXvEDGP5rx7PPwkcfQjW/dNeVIaAyBFSGgGoe9T0iOTnZJFZKafMN//xTx4G9zuzeDUeOgMFGGy6awtUVGjSERg2hcWMdzZs4U+i/DlAZAipDQGUIqBBCWJckgBnI7QQwN/7o5UfpYnXyJIwbBytWaEsj6nTQoweMHQtVqli7uVYlryvzSazMZ06s4uLgzz+1RX737NF+vnXLtI6NDdSqhXHIaNOm2uqj+Y28tsyXW7GSBFAIIR5O/kKJJ0eVKvDDD/D33/D881oSuGKFtvN2377aaqJCiMfKzQ3atNG+mwkJ0basOHgQZs2Cnj21oaTJyemP+fhAr17w9dcQFqbVEUIIIUTukwRQPHn8/WHNGjh0CLp0AYMBlizREsTXXoN/M164QgiR+2xtoU4dePNNWL5c26cwIkL7+c03tfv0+vTHihRJTSR/+w3i4639TIQQQoj8SRJA8eSqXRt++gn27YP27bUuhAULtN2033gDLl60dguFEGg9fj17aj2ABw/CzZtab+HYsRAYqM0ZjI83PVa4cPpEUgghhBCPThJA8eSrXx82btQmIbVuDffvw9y5ULEijBgBkZHWbqEQIg1XV+1XdcwY2LJFGzYaFqYNB+3VSxseajCkP/ZgIpmU+dosQgghhMhEthLApKQkfvvtN7799lvi/xunc/nyZW49OPNfiMepSROtC2HbNnj6abh7F2bOhAoV4J13ZCsJIfKolEVihgzRtqE4f17rwF+5EoYPh4AArc6Dx4oUMU0k4+Ks/UyEEEKIvM/iVUD//fdf2rVrR0REBHfv3uXUqVOUL1+eoKAgEhMTmTt3bm619bGRVUDzhkeKlVLw++/w8cfasoQALi7aJ8e334ZixXK+wVYkryvzSazMl5dideuWNto7ZbXRvXshNta0jl4PNWpoq4w2aaL96+OjLRj8OOSleOV1sgqoEEJYj8XvuiNGjKBevXrcuHEDJ6fUPaO6devG77//nqONEyLbdDqta2DPHm14aL16cPs2TJoEvr5al8HNm9ZupRDCTIUKQatW2nc6v/4K165pCwJ/8w288or2a20wwOHDMGdO6rHSpbUN62fMgAMHZNioEEIIYWvpCbt27WL37t3Y29ubHPfx8eHSpUs51jAhcoROpy0Q064dbNgAn3yifUL89FNteOjbb2vzBF1drd1SIYQFbGy03r4aNWDwYO1YZKTWQ5hSDh2Cy5dh9WqtADg7Q8OGqfsRNm4M7u7Wex5CCCHE42ZxD6DBYCA5gw2bLl68iKt8iBZ5lU6nbRlx8KD2SdDPT+sB/PhjrZtgyhSth1AI8cTy9oYXXoCvvtKGi8bGwtatMH689j1Q4cKQkGB6rEgRqFlTWzj4++8hPFwbQS6EEELkVxYngIGBgUyfPt14W6fTcevWLcaMGUOHDh1ysm1C5Dy9XvuE+Pff2qbylStrY8neew/Kl9c+Od65Y+1WCiFygLMztGgBH36ojQS/dg3++Qe+/Rb69NHWh1IKjhzRFg7u3Vt7GyhVyjSRvH/f2s9ECCGEyDkWJ4DTpk1j+/bt+Pn5kZiYSK9evShXrhyXLl1i8uTJFjdgzpw5+Pr64ujoSEBAADt37sy0br9+/dDpdOlK9erVjXUWLVqUYZ3ExESL2ybyMRsbeOklOHoUFi/WPvVFR8PIkdqnwtmztVVEhRD5hl4P1avDoEHar/2ZM9qw0bVrtV/9hg3Bzi79MXd300Tyxg1rPxMhhBAi+yyeA1iqVCnCwsJYsWIFoaGhGAwGBgwYwMsvv2yyKIw5Vq5cSVBQEHPmzKFp06Z8++23tG/fnmPHjlG2bNl09WfMmMHnn39uvJ2UlEStWrV48cUXTeq5ublx8uRJk2OOjo4WtU0UELa2WlfASy9pnwg/+0zbcXrYMG1Y6EcfQb9+8MCcVyFE/lCiBDz3nFZAGwCwf3/qPMI9e7SEb/t2raSoXj11HmHTptp3SEIIIcSTwKJtIO7fv0+VKlX43//+h5+f3yM/eMOGDalbty7ffPON8Vi1atXo2rUrkyZNeuj5P/74I8899xzh4eH4+PgAWg9gUFAQNx9hhUfZBiJvsEqs7t2DBQtgwgRIWdSoXDlt8ZjevbWEMQ+S15X5JFbmk1hpK4uePGm6uMzp0+nreXlBkyaKli1j6dvXDTe3ghkvc8k2EEIIYT0WfZq1s7Pj7t276HJgU6V79+4RGhrK6NGjTY63adOGPXv2mHWNBQsW0Lp1a2Pyl+LWrVv4+PiQnJxM7dq1+eyzz6hTp06m17l79y530wz3i/tvN2GDwYDBYDD3KZnFYDCglMrx6+ZHVomVrS28/jr07QvffYdu0iR0589D//6oiRNRH3+s9Rba2Dy+NplBXlfmk1iZT2KlqVJFK/37a7ejo7WewT17dOzZA6GhcOWKjvXrdaxfX5jRoxXduil691a0apXn3i7yhNx6bRX016oQQpjD4u6MYcOGMXnyZObPn4/tI/SGXL16leTkZLy8vEyOe3l5ERUV9dDzIyMj+fXXX/nhhx9MjletWpVFixZRo0YN4uLimDFjBk2bNuXw4cNUqlQpw2tNmjSJcePGpTseExOT43MHDQYDsbGxKKUK7Dfq5rJ6rHr0gM6dcV68mEKzZ6M/cwZd374kjR/PrXfeIbFTJ21SUR5g9Vg9QSRW5pNYZa5JE60AJCbC4cN27Nhhz5o1DkRE2LNsGSxbpqNEiWS6dUvkhRfu4OcnmxCmyK3XVnx8fI5dSwgh8iuLhoBC6obvhQoVokaNGri4uJjcv27dOrOuc/nyZUqVKsWePXto3Lix8fiECRNYunQpJ06cyPL8SZMmMXXqVC5fvpxuT8K0DAYDdevW5emnn2bmzJkZ1smoB7BMmTLcuHEjV4aAxsTE4OHhIR+oHiJPxerWLZg9G92XX6L7bwUIVaMGaswY6NpV22bCivJUrPI4iZX5JFaW0YY1xnD2rAc//GDDqlVw/Xrqe0OtWopXXlG89JK2ZUVBlluvrbi4OIoUKSJDQIUQIgsWd+EVLlyY559//pEfuHjx4tjY2KTr7YuOjk7XK/ggpRTBwcH07t07y+QPQK/XU79+fU5nNGnjPw4ODjg4OGR4bm586NHpdLl27fwmz8TKzQ0++ACGDoUZM2DqVHRHjqB74QWoU0fbWL5jR6smgnkmVk8AiZX5JFaW0et1NG2qp1kzHTNmaKuGLlkC//sfHD6s4/BhHe+9B61ba9OKu3WDB75HLTBy47Ulr1MhhHg4ixPAhQsX5sgD29vbExAQQEhICN26dTMeDwkJ4dlnn83y3O3bt3PmzBkGDBjw0MdRShEWFkaNGjUeuc1C4O6uLQgzbBhMnaolg4cOQefO2nrxn34KgYFW7xEUQlifvb02QKBrV7h+HVatgqVLtfmDW7ZoxcUFnn9eSwZbtpT5gkIIIXJftr8qi4mJYdeuXezevZuYmJhsXWPkyJHMnz+f4OBgjh8/zltvvUVERASDBw8G4P3336dPnz7pzluwYAENGzbE398/3X3jxo1j8+bNnDt3jrCwMAYMGEBYWJjxmkLkiCJFYPx4CA+Hd9/Vdpz+6y9o2xaefhq2brV2C4UQeUjRojB4cOoqomPGaFtH3L6t9RAGBoKPD7z3nrZZvRBCCJFbLE4Ab9++Tf/+/fH29ubpp5+mWbNmlCxZkgEDBpCQkGDRtXr06MH06dP59NNPqV27Njt27GDjxo3GVT0jIyOJiIgwOSc2Npa1a9dm2vt38+ZNBg0aRLVq1WjTpg2XLl1ix44dNGjQwNKnKsTDFS8OkyfDuXPw1lvg4AC7dkGrVlrZvdvaLRRC5DEVK8LYsdpG9Lt3a4lhkSLazjNTpkCNGtrI8q++AjPWRBNCCCEsYvEiMK+//jq//fYbs2fPpmnTpgDs2rWL4cOHExgYaLKn35NK9gHMG57IWF2+DBMnwnffaXsKArRpo20wn4tfQjyRsbISiZX5JFaWeZR43b0Lv/yiDRH95Re4f187rtdrbyG9e2tDSZ2dc77d1iD7AAohhPVY/K67du1aFixYQPv27XFzc8PNzY0OHTrw3XffsWbNmtxooxBPjpIlYfZsbYzXoEHavoJbtmjzAzt3hoMHrd1CIUQe5OAAzz0H69dDZCR8/TU0aqRtRL9pE7z8srbZ/Kuvwh9/aMeFEEKI7LB4EZiEhIQMV+n09PS0eAioEPlW2bLw7bfahJ7x41OXAfzf/7Rl/8aN08Z5CSHEA4oVgyFDtHL6tNYr+P332pTjRYu0Urq0lhT27g3Vq1u7xQVDcnIy91O6ZoUQIo+xs7PDxsyVxCweAvrMM89QrFgxlixZgqOjIwB37tyhb9++XL9+nd9++83yFucxMgQ0b8hXsTp9Wkv6fvgBUn7lunfXJgJVq/bIl89XscplEivzSawsk5vxUkqbL7h0qbaa6M2bqffVraslgi+9pPUSPgmepCGgSimioqK4mTboQgiRBxUuXJgSJUqge8hq9BYngP/88w/t2rUjMTGRWrVqodPpCAsLw9HRkc2bN1M9H3wVKQlg3pAvY3XsmJYIrlql3dbroVcvbWuJSpWyfdl8GatcIrEyn8TKMo8rXomJ2mCCpUu1fQaTkrTjNjbaQsS9e8Ozz4KTU6414ZE9SQlgZGQkN2/exNPTE2dn54d+sBJCiMdNKUVCQgLR0dEULlwYb2/vLOtbnACC1uP3/fffc+LECZRS+Pn58fLLL+OUl//aWEASwLwhX8fq77+1deB//FG7bWMDffrAxx+Dr6/Fl8vXscphEivzSawsY414Xb0KK1dqo8z37Us97uoKL76oJYNPP61915SXPCkJYHJyMqdOncLT05NixYrlQAuFECL3XLt2jejoaCpXrpzlcNBsJYD5nSSAeUOBiFVoqJYI/vKLdtvWFvr3h48+gjJlzL5MgYhVDpFYmU9iZRlrx+vkSW2u4Pffw/nzqcfLlk2dL5gDI85zxJOSACYmJhIeHk65cuXyzZfcQoj8686dO5w/fx5fX1/jVL2MWPyuO2nSJIKDg9MdDw4OZvLkyZZeToiCLSBAG8v155/aWu9JSTBvnrZR2LBh2rYSQghhhipVtB1nzp6FHTvgtdfA3R0iImDSJPDzg/r1YeZMiI62dmufLDLsUwjxJDD3vcriBPDbb7+latWq6Y5Xr16duXPnWno5IQRo20Rs3gw7d0KLFtoegrNnQ4UKMHKkfFoTQphNr4dmzbTtSKOitCnHnTppAwwOHIARI7Qdazp10u67c8faLRZCCPE4WZwARkVFZTix0MPDg8jIyBxplBAF1lNPwdat2kZfTZtqqz189ZU2L3D0aG3CjxBCmMnRUZsLuGGDNqBg5kytFzA5WRt53qMHlCih9Rbu2CH7CwohREFgcQJYpkwZdu/ene747t27KVmyZI40SogCr2VLrTdw82Zo0AASEmDyZC0R/PhjuHHD2i0UQjxhPDy0keX79sHx4/DBB9r8wLg4WLAAmjeH8uW1KcgnT1q7tUIIIXKLxQnga6+9RlBQEAsXLuTff//l33//JTg4mLfeeouBAwfmRhuFKJh0Om1e4J9/al/f16kDt25pG8v7+sKnn2qf3IQQwkJVq8KECdrm8tu2wYAB4OYG//6rHa9aVfvuafZsGXiQH0yaNAmdTkdQUJDJ8RYtWqDT6dDpdDg4OFCqVCk6d+7MunXrjHXWrl2LjY0NERERGV67atWqDB8+PNttGzt2LDqdjsGDB5scDwsLQ6fTcT7tikYP0aJFi3TPMbu+++47mjVrRpEiRShSpAitW7dmX9qldv8zZ84c44IbAQEB7Ny50+T+devW0bZtW4oXL27cOi2t8+fPG/8PHiyrV6/OkecixIMsTgDfffddBgwYwJAhQyhfvjzly5dn2LBhDB8+nPfffz832ihEwabTaZN1QkNh3Trw94fYWG31UF9f+PxzLTEUQggL6fVaz9/8+dp8wRUroGNHbWea/fu1HkNvb+jSBVav1kaliyfL/v37mTdvHjVr1szw/oEDBxIZGcmZM2dYu3Ytfn5+9OzZk0GDBgHQpUsXihUrxuLFi9Odu3v3bk6ePMmAAQMsbldycjKG/8YcOzo6smDBAk6dOmXxdXLLtm3beOmll9i6dSt79+6lbNmytGnThkuXLhnrrFy5kqCgID788EMOHTpEs2bNaN++vUmyfPv2bZo2bcrnn3+e4eOUKVOGyMhIkzJu3DhcXFxo3759rj9PUUCpbIqPj1f79u1TR44cUYmJidm9TJ4UGxurABUbG5vj105OTlaRkZEqOTk5x6+d30isMpGcrNTKlUpVraoUKAXK4OGh4t59VyXv26dUUpK1W5inyevKfBIry+SneF25otT06UoFBBjfZhQo5e6u1MCBSu3YoZTBkP3r51ascvrv9507d9SxY8fUnTt3jMcMBqVu3Xr8JTvxjo+PV5UqVVIhISGqefPmasSIESb3Z3RMKaWCg4MVoEJCQpRSSo0cOVKVL19eGR5oRP/+/VVAQIBSSqmpU6cqf39/5ezsrEqXLq3eeOMNFR8fb6y7cOFC5e7urjZs2KCqVaumbGxs1Llz59SYMWNUrVq1VGBgoHrxxReN9Q8dOqQAFR4ebjx29OhR1b59e+Xi4qI8PT3VK6+8omJiYpRSSvXt21cBJiXtuY8qKSlJubq6qsWLFxuPNWjQQA0ePNikXtWqVdXo0aPTnR8eHq4AdejQoYc+Vu3atVX//v0fuc2i4MnoPSsj2d58p1ChQtSvX5+yZcvy66+/cvz48RxIR4UQD6XXQ/fu8M8/sHQpVKyILiYG1ylT0DdoAMWKaV/Xf/UVhIXJqg5CCIt5emqrhR44AMeOwfvva1uTxsZqq4s+/bS2SPEnn8Dp09Zu7eOVkACFCj3+kpBgeVuHDh1Kx44dad26tUXn9e3blyJFihiHgg4YMIBz586xfft2Y53bt2+zatUqY++fXq9n5syZ/PPPPyxevJg//viDd99994HYJTBp0iTmz5/P0aNH8fT0NN73+eefs3btWvbv359hmyIjI2nevDm1a9fmwIEDbNq0iStXrtC9e3cAZsyYQePGjY09mpGRkZSxYD/dh0lISOD+/fsULVoUgHv37hEaGkqbNm1M6rVp04Y9e/Zk+3FCQ0MJCwvLVq+qEOayOAHs3r07s2fPBrTNBuvVq0f37t2pWbMma9euzfEGCiEyYWMDr7wCx49jCA4mMTAQ5eamfULbsEHbPqJOHSheHLp105b/+/tvSQiFEBapVg0mTtQ2l//jD3j1VS0hCQ/X9h2sXBkaNYI5c+DaNWu3VqRYsWIFBw8eZNKkSRafq9frqVy5snH+nZ+fHw0bNmThwoXGOqtWrSI5OZmXXnoJgKCgIFq2bImvry+tWrXis88+Y9WqVSbXvX//PnPmzKFJkyZUqVIFFxcX431169ale/fujB49OsM2ffPNN9StW5eJEydStWpV6tSpQ3BwMFu3buXUqVO4u7tjb2+Ps7MzJUqUoESJEtjY2Fj83DMzevRoSpUqZUymr169SnJyMl5eXib1vLy8iIqKyvbjLFiwgGrVqtGkSZNHaq8QWbE4AdyxYwfNmjUDYP369SiluHnzJjNnzmT8+PE53kAhxEPY2kLfvtxcsgR19ao2ceeLL6BDB3B11VYM/fFH7ev8WrW0r/ZfeEFb3eHoUW1klxBCPIRery1QHBwMV67ADz9A+/bad1F//QVDh2rzBbt2hbVr4e5da7c4dzg7a9OuH3dxdja/jRcuXGDEiBF8//33ODo6Zut5KqVMNpUeMGAAa9asIT4+HoDg4GCee+45ChcuDMDWrVsJDAykVKlSuLq60qdPH65du8bt27eN17C3t890LiLA+PHj2blzJ1u2bEl3X2hoKFu3bqVQoULGkrIv9dmzZ81+XhMnTjS5RmaL26Q1ZcoUli9fzrp169LF88GNtx+MmyXu3LnDDz/8IL1/ItdZnADGxsYau783bdrE888/j7OzMx07duR0QRsHIkReY2MD9erBO+9om3xdv659Mvv8c2jbFlxctK/o167VVnfw9wcvL21I6TffaGvDS0IohHgIZ2d46SXYuBEuXdJGnNepA/fvw08/ad8xlSgBgwfD7t35621Fp9PeSh93sSSnCA0NJTo6moCAAGxtbbG1tWX79u3MnDkTW1tbkpOTszw/OTmZ06dP4+vrazzWs2dPdDodK1eu5MyZM+zatcuYqPz777906NABf39/1q5dS2hoKF9//TWg9fqlcHJyyjI5qlChAgMHDmT06NGoB140BoOBzp07ExYWZlJOnz7N008/bXZsBg8ebHL+w7Yw+/LLL5k4cSJbtmwxSV6LFy+OjY1Nut6+6OjodL2C5lqzZg0JCQn06dMnW+cLYS5bS08oU6YMe/fupWjRomzatIkVK1YAcOPGjWx/yySEyCW2ttpa7g0awHvvaZ/ODhzQNpvftg127YKYGG15v5Tlpr28oEUL7av+li2hUiXLPnkIIQoULy8ICtLK0aPa1ORly+DiRfj2W62ULw+9e2uj1itWtHaL879nnnmGI0eOmBx79dVXqVq1Ku+9995Dh0YuXryYGzdu8PzzzxuPubq68uKLL7Jw4ULOnTtH+fLladGiBQAHDhwgKSmJqVOnotdrfQsPDv801yeffEKFChWMny9T1K1bl7Vr11KuXDlsbTP++Gpvb//Q5LZo0aLGjoyH+eKLLxg/fjybN2+mXr166R4rICCAkJAQunXrZjweEhLCs88+a9b1H7RgwQK6dOmCh4dHts4XwlwWJ4BBQUG8/PLLFCpUCB8fH+Mv/44dO6hRo0ZOt08IkZPs7KBxY6188AHcu6cNGd26VSt79mhju1au1ApAyZJaQpiSFFaoIAmhECJD1atrAw4mTNC+Y1q6VBtwcO4cjBunlcaN4eWXoVUrHWnWABE5yNXVFX9/f5NjLi4uFCtWLN3xhIQEoqKiSEpK4tKlS6xbt46vvvqKN954g5YtW5rUHTBgAM2aNePYsWO88847xt68ChUqkJSUxKxZs+jcuTO7d+9m7ty52Wq7l5cXI0eO5IsvvjA5PnToUL777jteeuklRo0aRfHixTlz5gwrVqzgu+++w8bGhnLlyvHXX39x/vx5ChUqRNGiRY0JqaWmTJnCxx9/zA8//EC5cuWMPX0pQ0cBRo4cSe/evalXrx6NGzdm3rx5REREmOxpeP36dSIiIrh8+TIAJ0+eBDDOU0xx5swZduzYwcaNG7PVXiEskp0lRg8cOKDWrVtnsrzv//73P7Vr167sXC7PkW0g8gaJlflyLFaJiUpt367U2LFKtWihlIOD6RrwoFTp0kq98opSCxYode7co60FbwXyujKfxMoyEq+M3bql1LJlSrVrp5Ren/pW4uWVpO7ff/K2gXhSZbYNBP9tmWBvb6+8vb1Vp06d1Lp16zK9TpUqVZRer1cXLlwwOT5t2jTl7e2tnJycVNu2bdWSJUsUoG7cuKGUSt0G4kEp20CkFRcXp4oXL55uK4dTp06pbt26qcKFCysnJydVtWpVFRQUZNye4uTJk6pRo0bKycnpkbeB8PHxSbetBKDGjBljUu/rr79WPj4+yt7eXtWtW1dt377d5P6FCxeadZ33339flS5dWt4/xCMx9z1Lp1R+GpmfM+Li4nB3dyc2NhY3N7ccvbbBYCA6OhpPT89sfytVUEiszJdrsUpMhD//TO0h/PNPbRhpWmXLpg4XbdECfHxy7vFzgbyuzCexsozE6+EiI2H5cli6VFG7dgILFjjlaKxy+u93YmIi4eHh+Pr6yjQXIUSeZ+57lsVDQIUQBYijY+rwz3HjtI2o9u5NnUP4118QEQGLF2sFwNfXdA5h6dLWa78QIk/x9tZ2qAkKUly8GA84WbtJQghR4EgCKIQwn7MzPPOMVgBu39bmDab0EO7fr20OFh4OKftFVahg2kP4kBXXhBAFg729tVsghBAFkySAQojsc3GBwECtAMTHa2u+p/QQHjgAZ89qZf58rU7lyqk9hC1aaGvFCyGEEEKIx0ISQCFEznF1hXbttAIQFwc7d2rJ4NatcOgQnDqllXnztDpVq6Ymgy1aIMsCCiGEEELkHrNnXk+ZMoU7d+4Yb+/YsYO7d+8ab8fHxzNkyJCcbZ0Q4snm5gYdO8IXX2i9gdeuwc8/w1tvabtG63Rw4oS2CX2PHtqGYv7+8Oab2trxV69a+xkIIYQQQuQrZieA77//PvHx8cbbnTp14tKlS8bbCQkJfPvttznbOiFE/lK4MHTuDNOmwcGDWoK3fj2MGAE1a2p1jh6Fr7+GF14ADw/t+IgRWr3r163afCGEEEKIJ53ZQ0Af3C1Cdo8QQjyyokWha1etgNZDuH176hzCf/6BI0e0MnOm1mNYq1bqHMKnn9aSSiGEEEIIYRaZAyiEyDuKFYPnntMKQHS0lhCmzCE8fhzCwrQyfbqWENapkzqHsFkzcHe3WvOFEEIIIfI6SQCFEHmXpye8+KJWAKKiTHsIT57UhpIePAhTp4JeDwEBqT2ETz2lLUwjhBBCCCEACxPA+fPnU6hQIQCSkpJYtGgRxYsXBzCZHyiEELmiRAltsZgePbTbly9riWBKD+GZM9pehPv3awvP2NhAvXqp+xA2bQpOsvG0EEKInKPT6Vi/fj1dU6Yz5JJy5coRFBREUFBQrj6OyP/MXgSmbNmyfPfdd3z11Vd89dVXlChRgqVLlxpvz58/n7Jly1rcgDlz5uDr64ujoyMBAQHs3Lkz07r9+vVDp9OlK9WrVzept3btWvz8/HBwcMDPz4/169db3C4hxBOgZEno1UvbUuL0abhwAZYuhf79oXx5SE6Gv/6Czz+Htm2hcGF0zZpR6Isv4OJFa7deCCEei0mTJqHT6dIlDi1atDB+lnJwcKBUqVJ07tyZdevWGeusXbsWGxsbIiIiMrx21apVGT58eLbbNnbsWHQ6HYMHDzY5HhYWhk6n4/z582Zfq0WLFmYlR+XKlWP69OmWNTQLkZGRtG/fPseulxcdOnSIHj164O3tjYODAz4+PnTq1IkNGzYY1wU5f/48Op2OsLCwTK+zZ88eOnToQJEiRXB0dKRGjRpMnTqV5ORkk3ppP+cXKlSIWrVqsWjRIgC2b9+OnZ0du3btMjnn9u3blC9fnrfeeguAc+fO8dJLL1GyZEkcHR0pXbo0zz77LKdOnXqkWKxbt47AwEA8PDxwc3OjcePGbN68OV29h+UjkyZNon79+ri6uuLp6UnXrl05efKkSZ2xY8dStWpVXFxcKFKkCK1bt+avv/56pPaDBQng+fPnCQ8Pf2ixxMqVKwkKCuLDDz/k0KFDNGvWjPbt22f6JjNjxgwiIyON5cKFCxQtWpQXU4aHAXv37qVHjx707t2bw4cP07t3b7p3754jwRJC5HGlS8Mrr8CCBdrm8+fPw6JF0K8f+PhAUhK6PXsoNG0augoVtONHj1q3zUIIkYv279/PvHnzqJmy0vIDBg4cSGRkJGfOnDF+YO3ZsyeDBg0CoEuXLhQrVozFixenO3f37t2cPHmSAQMGWNyu5ORkDAYDAI6OjixYsOCRP5jnpLTte5gSJUrg4OCQyy2ynp9++olGjRpx69YtFi9ezLFjx1i9ejVdu3blo48+IjY21qzrrF+/nubNm1O6dGm2bt3KiRMnGDFiBBMmTKBnz57pFphcuHAhkZGRHD58mB49evDqq6+yefNmmjdvzrBhw+jXrx+3b9821n/33XdxcHBg0qRJ3Lt3j8DAQOLi4li3bh0nT55k5cqV+Pv7m93ezOzYsYPAwEA2btxIaGgoLVu2pHPnzhw6dMhYx5x8ZPv27QwdOpQ///yTkJAQkpKSaNOmjclzqly5MrNnz+bIkSPs2rWLcuXK0aZNG2JiYh7pOaCsqEGDBmrw4MEmx6pWrapGjx5t1vnr169XOp1OnT9/3nise/fuql27dib12rZtq3r27JnpdRITE1VsbKyxXLhwQQHqxo0bKjk5OUfL/fv31eXLl9X9+/dz/Nr5rUisJFY5Xs6cUUnz5qnExo2VAmMxdOigkrduVclJSdZvYx4q8rqSeD1psbpx44YCVGxsrAWfRjJ3584ddezYMXXnzp3UgwaDUrduPf5iMFjc/vj4eFWpUiUVEhKimjdvrkaMGGFyf0bHlFIqODhYASokJEQppdTIkSNV+fLlleGBNvTv318FBAQopZSaOnWq8vf3V87Ozqp06dLqjTfeUPHx8ca6CxcuVO7u7mrDhg2qWrVqysbGRp07d06NGTNG1apVSwUGBqoXX3zRWP/QoUMKUOHh4cZjR48eVe3bt1cuLi7K09NTvfLKKyomJkYppVTfvn0VYFLSnpv2OT9YL6v27du3T7Vu3VoVK1ZMubm5qaefflqFhoaaXBNQ69evV0opFR4ergC1du1a1aJFC+Xk5KRq1qyp9uzZY3LO7t27VbNmzZSjo6MqXbq0GjZsmLp165bx/itXrqhOnTopR0dHVa5cOfX9998rHx8f9dVXX6V7TimSk5PVuHHjVKlSpZS9vb2qVauW+vXXX433m9u2tG7duqWKFSumunXrlmmdlNdFyvUPHTqU6XWee+65dPf9/PPPClArVqwwHksb0xRFixZVI0eOVEppv5vVqlVTQ4cOVUop9ccffyg7Ozu1f/9+pVTq6ydtjpCb/Pz81Lhx44y3s5OPREdHK0Bt37490zqxsbEKUL/99luG92f4npUBs+cA/vXXX1y/ft2ki3vJkiWMGTOG27dv07VrV2bNmmX2NyD37t0jNDSU0aNHmxxv06YNe/bsMesaCxYsoHXr1vj4+BiP7d2719j1m6Jt27ZZdvVPmjSJcePGpTseExNDYmKiWW0xl8FgIDY2FqUUer3ZHbAFksTKfBIrM7m4YOjYkdinnqL42bO4zp2Lw8aN6P4r9+rW5faQIdxt106bP1jAyevKMhIv8+VWrB7LegQJCfDfegiP1a1b4OJi0SlDhw6lY8eOtG7dmvHjx5t9Xt++fXn77bdZt24drVu3ZsCAAUybNo3t27fTokULQBtut2rVKqZMmQKAXq9n5syZlCtXjvDwcIYMGcK7777LnDlzjNdNSEhg0qRJzJ8/n2LFiuHp6Wm87/PPP6d+/frs37+f+vXrp2tTZGQkzZs3Z+DAgUybNo07d+7w3nvv0b17d/744w9mzJjBqVOn8Pf359NPPwXAw8Mj3XXWrVtHrVq1GDRoEAMHDjS5L6P2hYeH07dvX2bOnAnA1KlT6dChA6dPn8Y1i0XGPvzwQ7788ksqVarEhx9+yEsvvcSZM2ewtbXlyJEjtG3bls8++4wFCxYQExPDm2++yZtvvsnChQsBbdrThQsX+OOPP7C3t2f48OFER0dn+f82Y8YMpk6dyrfffkudOnUIDg6mS5cuHD16lEqVKpnVtgdt2bKFa9eu8e6772b6uDqdLst2pb3OO++8k+6+zp07U7lyZZYvX06PlDn+aSQnJ7N27VquX7+OnZ0doPUaL1myhCZNmtC6dWveeustPvjgA+rVqwdo//d6vZ41a9YQFBSETS7+TTcYDMTHx1O0aFHjsezkIyk9k2mvk9a9e/eYN28e7u7u1KpV69EanWV6mEa7du3U559/brz9999/K1tbW/Xaa6+pqVOnqhIlSqgxY8aYezl16dIlBajdu3ebHJ8wYYKqXLnyQ8+/fPmysrGxUStXrjQ5bmdnp5YtW2ZybNmyZcre3j7Ta0kPYN4sEiuJ1WOJ1YkTyjBokDI4OKT2CFaqpJK/+UYl375t9fbmqVhJkXjl8Vg9lh7AW7dMRhA8tpKmd8gcy5cvV/7+/sa2W9IDqJRSDRs2VO3btze53adPH+Pt4OBg5eTkpG7cuJHh+atWrVLFihUz3l64cKECVFhYmEm9Mf/1ACqlVM+ePVWrVq2UUul7AD/++GPVpk0bk3NTPq+dPHnyoc8nrYx60jJr34OSkpKUq6ur2rBhg/EYGfQAzp8/33j/0aNHFaCOHz+ulFKqd+/eatCgQSbX3blzp9Lr9erOnTvq5MmTClB//vmn8f7jx48rIMsewJIlS6oJEyaYHKtfv74aMmSI2W170Oeff64Adf36deOxffv2KRcXF2NJiUVWPYAp18ns9dKlSxdVrVo1421AOTo6KhcXF2VjY6MAVbRoUXX69GmT8z755BOl1+tVQECAun//vsl9s2fPVs7OzsrV1VW1bNlSffrpp+rs2bMZPv6jmDJliipatKi6cuWK8Zil+YjBYFCdO3dWTz31VLr7NmzYoFxcXJROp1MlS5ZU+/bty7QtOd4DGBYWxmeffWa8vWLFCho2bMh3330HQJkyZRgzZgxjx461KAF98FsDpZRZ3yQsWrSIwoULZ7jikqXXdHBwyLDnUq/X58q3uDqdLteund9IrMwnsTKfSayqVIFvv4VPP4VZs+Drr9GdPo3ujTdg7FgYPhzeeAOKFLF2s61CXleWkXiZLzdi9Vji7uys9cY9bs7OZle9cOECI0aMYMuWLTg6Ombr4R787DRgwACCgoKYPXs2rq6uBAcH89xzz1G4cGEAtm7dysSJEzl27BhxcXEkJSWRmJjI7du3cfmv59Le3j7TuYgA48ePp1q1amzZssWkdxAgNDSUrVu3GlejT+vs2bNUrlw5W88zrYzaFx0dzSeffMIff/zBlStXSE5OJiEhIdP1KlKkvY63t7fxWlWrViU0NJQzZ86wbNkyYx2lFAaDgfDwcE6dOoWtra2xNwu0xXZSYp2RuLg4Ll++TNOmTU2ON23alMOHD5vdNnPUrFnTuNBLpUqVSEpKMus8IN08v7THH/ys/tVXX9G6dWsuXLjAyJEjeeutt6hYsaJJnY8++ohPP/2U0aNHp+vBHDp0KH369GHr1q389ddfrF69mokTJ/Lzzz8TGBiYrg3Lli3j9ddfN97+9ddfadasWZbPZ/ny5YwdO5affvop3WvWknzkzTff5O+//063sA1Ay5YtCQsL4+rVq3z33XfGuYQPPp4lzH6nvHHjBl5eXsbb27dvp127dsbb9evX58KFC2Y/cPHixbGxsSEqKsrkeHR0tMnjZEQpRXBwML1798be3t7kvhIlSmTrmkKIAs7LC8aPh4gI+OorKFMGrlyBDz/Ufh45UrtPCFGw6XTaUMzHXcz4cjxFaGgo0dHRBAQEYGtri62tLdu3b2fmzJnY2tqmW3HxQcnJyZw+fRpfX1/jsZ49e6LT6Vi5ciVnzpxh165dxsVf/v33Xzp06IC/vz9r164lNDSUr7/+GoD79+8br+Hk5JTlF/IVKlRg4MCBjB49Ol2iYDAY6Ny5M2FhYSbl9OnTPP3002bHJisZta9fv36EhoYyffp09uzZQ1hYGMWKFePevXtZXitlqCKkJgIpi8oYDAZef/11k+dx+PBhTp8+TYUKFYzP3ZwOkQeZk3Rk1bYHpQwdTbs6pYODAxUrVkyXjGUlJUE/fvx4hvefOHHCZJgqaJ/pK1asSMuWLVm9ejVDhw7l2LFjGT6XjIavAri6utKlSxcmTJjA4cOHadasWabDobt06WLyf5I2Ac/IypUrGTBgAKtWraJ169bp2m5uPjJs2DB+/vlntm7dSunSpdPd7+LiQsWKFWnUqBELFizA1taWBQsWZNm2hzE7AfTy8jKu8nnv3j0OHjxI48aNjffHx8ebvKAext7enoCAAEJCQkyOh4SE0KRJkyzP3b59O2fOnMlw1anGjRunu+aWLVseek0hhAC0jeODgrRVRJcuhRo14PZtLSmsUAH69IEjR6zdSiGEyNQzzzzDkSNH0n2YffnllwkLC3vofKjFixdz48YNnn/+eeMxV1dXXnzxRRYuXEhwcDDly5c3zgc8cOAASUlJTJ06lUaNGlG5cmUuX76crbZ/8sknnDp1ihUrVpgcr1u3LkePHqVcuXLG5COlpO1hfFhya0k9gJ07dzJ8+HA6dOhA9erVcXBw4OrVq5Y/sQyey4PPo2LFitjb21OtWjWSkpI4cOCA8ZyTJ09y8+bNTK/p5uZGyZIl0/Ug7dmzh2rVqmW7rW3atKFo0aJMnjw529dIe52pU6emu+/nn3/m9OnTvPTSS5meX7FiRZ5//nnef//9bLdBp9NRtWpVk1U203J1dTX5v3DKYt/g5cuX069fP3744Qc6duyY7n5z8hGlFG+++Sbr1q3jjz/+MPnCJStKKe7evWtW3cyYPQS0Xbt2jB49msmTJ/Pjjz/i7Oxs0i36999/U6FCBYsefOTIkfTu3Zt69erRuHFj5s2bR0REhHEvmPfff59Lly6xZMkSk/MWLFhAw4YN8ff3T3fNESNG8PTTTzN58mSeffZZfvrpJ3777bcMu1SFECJTdnbalhIvvwybN8OUKdpm80uXaqV9e3j3XWje3KJv5oUQIre5urqm+4zk4uJCsWLF0h1PSEggKiqKpKQkLl26xLp16/jqq6944403aNmypUndAQMG0KxZM44dO8Y777xj7D2qUKECSUlJzJo1i86dO7N7927mzp2brbZ7eXkxcuRIvvjiC5PjQ4cO5bvvvuOll15i1KhRFC9enDNnzrBixQq+++47bGxsKFeuHH/99Rfnz5+nUKFCFC1aNMNhweXKlWPHjh307NkTBwcHihcvnml7KlasyNKlS6lXrx5xcXGMGjUqy8TAHO+99x6NGjVi6NChDBw4EBcXF44fP05ISAizZs2iSpUqtGvXjoEDBzJv3jxsbW0JCgp66OOOGjWKMWPGUKFCBWrXrs3ChQsJCwszGWpqqUKFCjF//nx69OhBx44dGT58OJUqVeLWrVts2rQJIN0XCg/uZQfg5+fHt99+a9xi5M0338TNzY3ff/+dUaNG8cILL9C9e/cs2/L2229Tq1YtDhw48NDeubCwMMaMGUPv3r3x8/PD3t6e7du3ExwczHvvvWdhFEwtX76cPn36MGPGDBo1amTs6XNycsLd3R0wLx8ZOnQoP/zwAz/99BOurq7G67i7u+Pk5MTt27eZMGECXbp0wdvbm2vXrjFnzhwuXrxosgVetmQ5QzCN6Oho9dRTTymdTqdcXV3VunXrTO5v1aqV+uCDD8y9nNHXX3+tfHx8lL29vapbt67J0qd9+/ZVzZs3N6l/8+ZN5eTkpObNm5fpNVevXq2qVKmi7OzsVNWqVdXatWstalPKEqs5NYk8reTkZBUZGamSk5Nz/Nr5jcTKfBIr8z1SrPbvV6p7d6X0+tSFGerXV2r1aqWSknK+sVYmryvLSLzMl1uxyum/3+YuqPAkyGwRGP7bCsHe3l55e3urTp06pfuMl1aVKlWUXq9XFy5cMDk+bdo05e3trZycnFTbtm3VkiVLTBb9SNlm4UFpF4FJERcXp4oXL55uK4dTp06pbt26qcKFCysnJydVtWpVFRQUZNyG4OTJk6pRo0bKyckp020glFJq7969qmbNmsrBwSHdNhAPOnjwoKpXr55ycHBQlSpVUqtXr063iAwZLAKTdiGUlMWJtm7dajy2b98+FRgYqAoVKqRcXFxUzZo1TRZwiYyMVB07dlQODg6qbNmyasmSJRZtA2FnZ5fpNhAPa1tG9u/fr1544QXl6empbG1tVbFixVTbtm3VihUr0m0DkVFJ+b/YsWOHateunXJ3d1f29vbKz89Pffnllyrpgb+haWOaVmBgoMniRJnVjYmJUcOHD1f+/v6qUKFCytXVVdWoUUN9+eWXj/y+k9FWIoDq27evSb2H5SOZxWrhwoVKKe39p1u3bqpkyZLG388uXbrkyCIwuv8aYLbY2FgKFSqULtu/fv06hQoVSjcn70kUFxeHu7s7sbGxuLm55ei1DQYD0dHReHp6yiIBDyGxMp/Eynw5EquzZ2HaNAgOhpStYipUgHfegb594RG/Hc4r5HVlGYmX+XIrVjn99zsxMZHw8HB8fX2zvZiKEEI8Lua+Z1n8ruvu7p7h2PGiRYvmi+RPCCEeqkIF+Ppr+Pdf+OQTKFpUSwrfeAN8fLTFZK5ft3YrhRBCCCHSMXsOYP/+/c2qFxwcnO3GCCHEE8XTE8aN0+YCBgfD1KlaUvjxxzBpEgwcCG+9pSWFQgghhBB5gNk9gIsWLWLr1q3cvHmTGzduZFqEEKLAcXGBYcPgzBn44QeoXRsSEmDGDK238JVX4IF9mIQQQgghrMHsHsDBgwezYsUKzp07R//+/XnllVcoWrRobrZNCCGeLLa28NJL0LMn/PabtnLob7/BsmVaadNG6y1s1UpWDhVCCCGEVZjdAzhnzhwiIyN577332LBhA2XKlKF79+5s3rw53WadQghRoOl0EBgIISEQGqolhHo9bNkCrVtDvXqwciUkJVm7pUIIIYQoYCxaBMbBwYGXXnqJkJAQjh07RvXq1RkyZAg+Pj7cunUrt9oohBBPrrp1YflybXjom29qK4QePKglhZUra4vJJCRYu5VCCCGEKCCyvfayTqdDp9OhlMJgMORkm4QQIv/x9YVZsyAiAsaOhWLFIDxcSwp9fLTFZK5etXYrhRBCCJHPWZQA3r17l+XLlxMYGEiVKlU4cuQIs2fPJiIigkKFCuVWG4UQIv8oXhzGjNESwdmztcTw6lUtKSxbVltMJjzc2q0UQgghRD5ldgI4ZMgQvL29mTx5Mp06deLixYusXr2aDh06yIa3QghhKWdnGDoUTp2CFSu0oaJ37mhJYcWK2mIyBw9au5VCCCGEyGfMXgV07ty5lC1bFl9fX7Zv38727dszrLdu3boca5wQQuR7trbQowd07w5//KGtHLpli5YUrlihLRrz7rvav7JyqBAiF+h0OtavX0/Xrl2t3RQhxGNgdtddnz59aNmyJYULF8bd3T3TIoQQIht0OnjmGdi8GQ4dgl69wMZG20aiTZvUxWRk5VAhhJmioqIYMWIEFStWxNHRES8vL5566inmzp1LwmNefOr69esMGzaMKlWq4OzsTNmyZRk+fDixsbEm9W7cuEHv3r2Nnyt79+7NzZs3TeqMGDGCgIAAHBwcqF27drrHGjt2rHGtirTFxcUlF5+hEE8Os3sAFy1alIvNEEIIYVS7trZv4IQJ8NVXMH8+hIVpSeEHH8DIkdC/v7YBvRBCZODcuXM0bdqUwoULM3HiRGrUqEFSUhKnTp0iODiYkiVL0qVLl8fWnsuXL3P58mW+/PJL/Pz8+Pfffxk8eDCXL19mzZo1xnq9evXi4sWLbNq0CYBBgwbRu3dvNmzYYKyjlKJ///789ddf/P333+ke65133mHw4MEmx5555hnq16+fS89OiCeL2QmgOdasWcMLL7yQk5cUQoiCq1w5mDEDPvkE5syBmTPh/HkYPlxbNObNN7Xi4WHlhgpRcCilSLj/+LducbZzRmfBMPAhQ4Zga2vLgQMHTHq+atSowfPPP5/lHs5HjhxhxIgR7N27F2dnZ55//nmmTZv2SAv++fv7s3btWuPtChUqMGHCBF555RWSkpKwtbXl+PHjbNq0iT///JOGDRsC8N1339G4cWNOnjxJlSpVAJg5cyYAMTExGSaAhQoVMmnr4cOHOXbsGHPnzs12+4XITyxKAJOSkjh58iR2dnZUrlzZePynn37ik08+4cSJE5IACiFETitWDD7+GN55BxYtgqlT4exZ+PRTbc5g//5ar2CFCtZuqRD5XsL9BApNevwrn996/xYu9ub1+l+7do0tW7YwceLETIc9ZpZMJiQk0K5dOxo1asT+/fuJjo7mtdde480338zx0WCxsbG4ublha6t9HN27dy/u7u7G5A+gUaNGuLu7s2fPHmMCaKn58+dTuXJlmjVrliPtFuJJZ/YcwGPHjlG5cmVq1qxJtWrVeO6557hy5QrNmzenb9++BAYGcubMmdxsqxBCFGxOTvDGG3DyJKxeDfXqQWKi1jtYubK2mMyBA9ZupRDCys6cOYNSKl3CVLx4cWPv2HvvvZfhucuWLePOnTssWbIEf39/WrVqxezZs1m6dClXrlzJsTZeu3aNzz77jNdff914LCoqCk9Pz3R1PT09iYqKytbj3L17l2XLljFgwIBst1WI/MbsHsDRo0fj6+vLzJkzWbZsGStXruSff/7hlVde4X//+x+urq652U4hhBApbGzghRfg+edh+3atF/DXX2HVKq20aqWtHNqmjawcKkQOc7Zz5tb7t6zyuJZ6sJdv3759GAwGXn75Ze7evZvhOcePH6dWrVomPYdNmzbFYDBw8uRJvLy80p1TvXp1/v33XwCaNWvGr7/+mmW74uLi6NixI35+fowZMybLNoM27NaS4a9prVu3jvj4ePr06ZOt84XIj8xOAPft28fGjRupW7cuTz31FCtXrmTUqFEMHDgwN9snhBAiMzodtGihlb//hi+/1FYK/eMPrdSsqSWC3buDnZ21W5u/3L8PUVEQGQmXL2slPh6bli0hgx4MkX/odDqzh2JaS8WKFdHpdJw4ccLkePny5QFwcnLK9Nyskq3Mjm/cuJH79+8/9NoA8fHxtGvXjkKFCrF+/Xrs0rw3lShRIsNexpiYmAwTT3PMnz+fTp06UaJEiWydL0R+ZHYCGB0dTalSpQAoXLgwzs7ONG/ePNcaJoQQwgI1a8KSJTB+PEyfDvPmaUnhK6+krhw6YAA8wiIOBUJyMkRHpyZ1mZWYGHhgEQ09UMzNDb7/Hjp3tk77hQCKFStGYGAgs2fPZtiwYRZtf+Dn58fixYu5ffu28bzdu3ej1+tN1n9Iy8fHx6xrx8XF0bZtWxwcHPj5559xdHQ0ub9x48bExsayb98+GjRoAMBff/1FbGwsTZo0Mfs5pAgPD2fr1q38/PPPFp8rRH5mdgKo0+nQ61OnDOr1epNvbYQQQuQBZcvCtGnaojHffKOtIhoRAUFBMG4cDB0Kw4YVvF4qgwGuXn14YnflilbXHHZ24O0NJUtCyZKo8+fRHzyIevZZLdYffgh6s6faC5Gj5syZQ9OmTalXrx5jx46lZs2a6PV69u/fz4kTJwgICMjwvJdffpkxY8bQt29fxo4dS0xMDMOGDaN3797Z7oUDreevTZs2JCQk8P333xMXF0dcXBwAHh4e2NjYUK1aNdq1a8fAgQP59ttvAW0biE6dOpnMZzxz5gy3bt0iKiqKO3fuEBYWBmjJq729vbFecHAw3t7etG/fPtvtFiI/MjsBVEpRuXJlY/f/rVu3qFOnjklSCNpGn0IIIaysSJHUnr8lS7ThoadPaz2EX3wB/frB229DpUrWbumjUQquX886qYuM1EpSknnXtLGBEiW0xC5NgpeuFCtmkuCpxEQS3ngDl0WLtK07QkO12Lu55c5zFyILFSpU4NChQ0ycOJH333+fixcv4uDggJ+fH++88w5DhgzJ8DxnZ2c2b97MiBEjqF+/vsk2EI8iNDSUv/76C9CGqKYVHh5OuXLlAG0RmuHDh9OmTRsAunTpwuzZs03qv/baa2zfvt14u06dOumuYzAYWLRoEf369cPGxuaR2i5EfqNTWW0Ek8bixYvNumDfvn0fqUF5QVxcHO7u7sbliXOSwWAgOjoaT0/PdMmzMCWxMp/EynwFNlbJyfDTTzB5Muzbpx3T6eC557R5gv8Nt0rLqrFSCmJjH57YXb4M9+6Zd02dTuv5zCyhSykeHloSaCFjvH75Bf3QoXD3LlSpAj/+CFWrWny9/Cy3Xls5/fc7MTGR8PBwfH190w1XFEKIvMbc9yyzewDzQ2InhBAFlo2Nlux16wY7d2orh/7yC6xdq5XmzbVEsH373F85ND4+64Qupdy5Y/41ixd/eGLn6fl4FsN59VVtTuZzz2lbdjRooPUEdu2a+48thBBCPIRFG8Gnde/ePaKjozE8MFeibNmyj9woIYQQuUSng6ef1so//2hDQ5ct07aT2L4d/P1h1Cjo2RNsLfwTkZDw8KTu8mW4ZcES+kWKZJ7QpQzPLFECHBwsa2tuq19fGwLavbsW127d4KOPYOzYbPUuCiGEEDnF4gTw1KlTDBgwgD179pgcT1k2ODk5OccaJ4QQIhf5+8OiRfDZZ9piMd9+qyWFfftqC5gEBaF79lltDtuVK1kndZcva0M2zeXmlnlCl/b2Q5aUz9M8PSEkROtZnT5dm38ZGqol3EWKWLt1QgghCiiLE8BXX30VW1tb/ve//+Ht7Z3tjTmFEELkEWXKaD2BH30Ec+dqyeDFi+jfeQfP0aPRmbt4CoCzM5QqlfXiKd7eBWc7Cjs7+OorqFcPBg6EX3/Vegd//FFLwIUQQojHzOIEMCwsjNDQUKrKhHYhhMhfCheG0aO1LSO+/x71xRfoTp3S7nNwePgcu5IlwdU19+cQPolefhmqV9eGgp49Cw0bwsKF2hBRIYQQ4jGyOAH08/Pj6tWrudEWIYQQeYGjI7z2GqpfP66GhlKsYkX0RYtKYveoateGAwe0+ZW//QY9emhDQidMsHy+pRBCCJFNFq+9PHnyZN599122bdvGtWvXjBt5pt3QUwghRD6g15Ps46PNV5PkL2cUKwabNsF772m3p0zRVl69ds267RJCCFFgWPyVY+vWrQF45plnTI7LIjBCCCGEGWxs4PPPoW5d6N9f6w2sVw/WrYP/NrQWQgghcovFCeDWrVtzox1CCCFEwdK9O/j5afsDnj0LTZrAd9/BK69Yu2VCCCHyMYuHgDZv3jzLYqk5c+YYd6sPCAhg586dWda/e/cuH374IT4+Pjg4OFChQgWCg4ON9y9atAidTpeuJCYmWtw2IYQQIlf5+8P+/dChAyQmQu/e2iI89+9bu2WiANHpdPz444/WboYQ4jGxOAHMSStXriQoKIgPP/yQQ4cO0axZM9q3b09ERESm53Tv3p3ff/+dBQsWcPLkSZYvX55uRVI3NzciIyNNiqOjY24/HSGEEMJyRYrAhg3w8cfa7RkzIDAQoqOt2y7xxIuKimLEiBFUrFgRR0dHvLy8eOqpp5g7dy4JCQnWbt4ju3//Pu+99x41atTAxcWFkiVL0qdPHy5fvmxS7+7duwwbNozixYvj4uJCly5duHjxokmdCRMm0KRJE5ydnSlcuHC6x8qsg0Gn0xEtv6viCWPVZcemTZvGgAEDeO211wCYPn06mzdv5ptvvmHSpEnp6m/atInt27dz7tw5ihYtCkC5cuXS1dPpdJQoUcLsdty9e5e7d+8ab6csZmMwGDAYDJY8pYcyGAwopXL8uvmRxMp8EivzSazMJ7GyzCPHa+xYqF0bXb9+6LZvRwUEoNas0fYNzGdy67Ulr9VU586do2nTphQuXJiJEydSo0YNkpKSOHXqFMHBwZQsWZIuXbrkyGPdu3cPe3v7HLmWJRISEjh48CAff/wxtWrV4saNGwQFBdGlSxcOHDhgrBcUFMSGDRtYsWIFxYoV4+2336ZTp06EhoZiY2NjfA4vvvgijRs3ZsGCBekeq0ePHrRr187kWL9+/UhMTMTT0zN3n6gQOcxqCeC9e/cIDQ1l9OjRJsfbtGnDnj17Mjzn559/pl69ekyZMoWlS5cav8X57LPPcHJyMta7desWPj4+JCcnU7t2bT777DPqZDGxftKkSYwbNy7d8ZiYmBwfOmowGIiNjUUphV5v1Q7YPE9iZT6JlfkkVuaTWFkmR+LVpAk2v/xCkVdfxfbsWWjenLhJk7jz0ks521gry63XVnx8fI5d62Fu37ud6X02ehscbR3NqqvX6XGyc8qyrou9i8XtGzJkCLa2thw4cAAXl9Tza9SowfPPP49SKtNzjxw5wogRI9i7dy/Ozs48//zzTJs2jUKFCgFa4nPz5k0aNmzIrFmzsLe3p3///qxevZojR46YXCsgIICOHTvy6aefsm3bNt59912OHj2KnZ0d1atX54cffsDHx8fi5wfg7u5OSEiIybFZs2bRoEEDIiIiKFu2LLGxsSxYsIClS5caFzL8/vvvKVOmDL/99htt27YFMH4OXLRoUYaP5eTkZPJZMyYmhj/++CPDZFGIvM5qCeDVq1dJTk7Gy8vL5LiXlxdRUVEZnnPu3Dl27dqFo6Mj69ev5+rVqwwZMoTr168b5wFWrVqVRYsWUaNGDeLi4pgxYwZNmzbl8OHDVKpUKcPrvv/++4wcOdJ4Oy4ujjJlyuDh4YGbm1sOPWONwWBAp9Ph4eEhH6geQmJlPomV+SRW5pNYWSbH4uXpCQcOoPr1Q/fTT7iPHInbqVOor74CK/Sy5Ibcem09zukehSYVyvS+DpU68EuvX4y3Pb/0JOF+xkMum/s0Z1u/bcbb5WaU42qC6X7LakzmyVpGrl27xpYtW5g4caJJ8peWLpOtXRISEmjXrh2NGjVi//79REdH89prr/Hmm2+aJEe///47bm5uhISEoJSicOHCjBs3jv3791P/v17rv//+m0OHDrF69WqSkpLo2rUrAwcOZPny5dy7d499+/Zl2o7sio2NRafTGYdxhoaGcv/+fdq0aWOsU7JkSfz9/dmzZ48xAbTUkiVLcHZ25oUXXsiJZgvxWGUrAUxKSmLbtm2cPXuWXr164erqyuXLl3FzczN+O2SuB3/xU7aTyEjKH4xly5bh7u4OaMNIX3jhBb7++mucnJxo1KgRjRo1Mp7TtGlT6taty6xZs5g5c2aG13VwcMDBwSHdcb1enysfenQ6Xa5dO7+RWJlPYmU+iZX5JFaWybF4FS6sbQsxcSJ88gm6uXPR/f03rFkD3t450lZry43XlrxONWfOnEEpRZUqVUyOFy9e3DiyaejQoUyePDnducuWLePOnTssWbLEmDzOnj2bzp07M3nyZOMX9y4uLsyfP99k6Gfbtm1ZuHChMQFcuHAhzZs3p3z58ly/fp3Y2Fg6depEhQoVAKhWrVqOPu/ExERGjx5Nr169jF/gR0VFYW9vT5EiRUzqZtXhYI7g4GB69epl0isoxJPC4gTw33//pV27dkRERHD37l0CAwNxdXVlypQpJCYmMnfuXLOuU7x4cWxsbNL98kVHR6frFUzh7e1NqVKljMkfaG8eSikuXryYYQ+fXq+nfv36nD592oJnKYQQQliZXg8ffaTtF9irF+zZAwEBWhLYpIm1W1fg3Xr/Vqb32ehtTG5Hv5P5IiF6nWnSen7E+UdqV1oPfqG+b98+DAYDL7/8ssnaB2kdP36cWrVqmfQcNm3aFIPBwMmTJ42f0WrUqJFu3t/AgQPp378/06ZNw8bGhmXLljF16lQAihYtSr9+/Wjbti2BgYG0bt2a7t27453JFxrt27c3rgzv4+PD0aNHs3yu9+/fp2fPnhgMBubMmZNlXci6w+Fh9u7dy7Fjx1iyZEm2zhfC2iz+qmzEiBHUq1ePGzdumHzr0a1bN37//Xezr2Nvb09AQEC6sdshISE0yeQPW9OmTbl8+TK3bqW+6Z46dQq9Xk/p0qUzPEcpRVhYWKZvMEIIIUSe1qEDHDigbRkRGQktWsDcuZDFHC6R+1zsXTItaef/Paxu2vl/mdW1VMWKFdHpdJw4ccLkePny5alYsWKWvVZZJUZpj2c0tLRz5844ODiwfv16NmzYwN27d3n++eeN9y9cuJC9e/fSpEkTVq5cSeXKlfnzzz8zfKz58+cTFhZGWFgYGzduzPL53r9/n+7duxMeHk5ISIjJ9J0SJUpw7949bty4YXJOVh0ODzN//nxq165NQEBAts4XwtosTgB37drFRx99lO5bHx8fHy5dumTRtUaOHMn8+fMJDg7m+PHjvPXWW0RERDB48GBAm5vXp08fY/1evXpRrFgxXn31VY4dO8aOHTsYNWoU/fv3N76ZjRs3js2bN3Pu3DnCwsIYMGAAYWFhxmsKIYQQT5yKFWHvXnjxRW2PwDfegNde0/YOFOIBxYoVIzAwkNmzZ3P7duYL0GTEz8+PsLAwk/N2796NXq+ncuXKWZ5ra2tL3759WbhwIQsXLqRnz544Ozub1KlTpw7vv/8+e/bswd/fnx9++CHDa5UqVYqKFStSsWLFLBeJSUn+Tp8+zW+//UaxYsVM7g8ICMDOzs6kwyEyMpJ//vkn0w6HrNy6dYtVq1YxYMAAi88VIq+weAiowWAgOTk53fGLFy/i6upq0bV69OjBtWvX+PTTT4mMjMTf35+NGzcaf9EjIyNN9gQsVKgQISEhDBs2jHr16lGsWDG6d+/O+PHjjXVu3rzJoEGDiIqKwt3dnTp16rBjxw4aNGhg6VMVQggh8o5ChWDlSm1biNGjITgYjhyBtWuhTBlrt07kMXPmzKFp06bUq1ePsWPHUrNmTfR6Pfv37+fEiROZ9l69/PLLjBkzhr59+zJ27FhiYmIYNmwYvXv3NqvH7LXXXjPO7du9e7fxeHh4OPPmzaNLly6ULFmSkydPcurUKZMv+i2VlJTECy+8wMGDB/nf//5HcnKycWpR0aJFsbe3x93dnQEDBvD2229TrFgxihYtyjvvvEONGjWMq4ICREREcP36dSIiIkhOTiYsLAzQelPTrm+xcuVKkpKSePnll7PdbiGsTlmoe/fuauDAgUoppQoVKqTOnTun4uPjVatWrVS/fv0svVyeFBsbqwAVGxub49dOTk5WkZGRKjk5Ocevnd9IrMwnsTKfxMp8EivLPNZ4bdmiVNGiSoFSHh5KbduW+4+Zg3IrVjn99/vOnTvq2LFj6s6dOzlyvcft8uXL6s0331S+vr7Kzs5OFSpUSDVo0EB98cUX6vbt28Z6gFq/fr3x9t9//61atmypHB0dVdGiRdXAgQNVfHy88f6+ffuqZ599NtPHbdasmfLz8zM5FhUVpbp27aq8vb2Vvb298vHxUZ988skjvQbCw8MVkGHZunWrsd6dO3fUm2++qYoWLaqcnJxUp06dVEREhMm1+vbt+9DrKKVU48aNVa9evbLdZiFyk7nvWTqlLJtEcPnyZVq2bImNjQ2nT5+mXr16nD59muLFi7Njx458sRlmXFwc7u7uxMbG5so2ENHR0Xh6espqZQ8hsTKfxMp8EivzSaws89jjFR4Ozz0HYWFgYwPTpsGwYZDDy+rnhtyKVU7//U5MTCQ8PBxfX9/HusXEk0wpRdWqVXn99ddNttgSQuQ+c9+zLB4CWrJkScLCwli+fDkHDx7EYDAwYMAAXn75ZVkKVwghhHhcfH1h924YNAiWLYMRI2D/fvj2W3hg3pUQj0N0dDRLly7l0qVLvPrqq9ZujhAiE9naB9DJyYn+/fvTv3//nG6PEEIIIczl7AxLl0K9evDOO/D993D0qLaHYLly1m6dKGC8vLwoXrw48+bNS7fvnhAi77A4Afz5558zPK7T6XB0dKRixYr4+vo+csOEEEIIYQadDoKCoHZt6N4dDh3SEsIVKyDNIhdC5DYLZxUJIazE4gSwa9eu6HS6dL/kKcd0Oh1PPfUUP/74o3z7I4QQQjwuLVpAaKg2L/DAAWjbFj7/XOsZfALmBQohhHg8LJ55HRISQv369QkJCSE2NpbY2FhCQkJo0KAB//vf/9ixYwfXrl3jnXfeyY32CiGEECIzZcrAzp3w6qtgMMC770LPnmDhXnBCCCHyL4t7AEeMGMG8efNMNs985plncHR0ZNCgQRw9epTp06fL/EAhhBDCGhwdYcECbb/A4cNh1So4dgzWr9c2lBdCCFGgWdwDePbs2QyXVnZzc+PcuXMAVKpUiatXrz5664QQQghhOZ0O3ngDtm6FEiXgn3+0hPDXX63dMiGEEFZmcQIYEBDAqFGjiImJMR6LiYnh3XffpX79+gCcPn2a0qVL51wrhRBCCGG5p57S5gU2bgw3b0LHjjBhgjY8VAghRIFkcQK4YMECwsPDKV26NBUrVqRSpUqULl2a8+fPM3/+fABu3brFxx9/nOONFUIIIYSFSpbUegJffx2Ugo8+guefh7g4a7dMCCGEFVg8B7BKlSocP36czZs3c+rUKZRSVK1alcDAQPR6LZ/s2rVrTrdTCCGEENnl4ABz52rbQwwdCj/+CA0bav9WqWLt1gnxROnXrx83b97kxx9/tHZThMgWi3sAQdvyoV27dgwfPpwRI0bQtm1bY/InhBBCiDzqtddgxw4oVQpOnNDmBf70k7VbJXJBv3790Ol06HQ67Ozs8PLyIjAwkODgYAwyBPiRzJgxg0WLFlm7GQCsW7eOwMBAPDw8cHNzo3HjxmzevDldvbVr1+Ln54eDgwN+fn6sX7/e5P4dO3bQuXNnSpYsiU6nyzC5TXk9PVi++OKL3Hp6IpdkK2u7ffs2GzduZO7cucycOdOkCCGEECIPa9hQmxfYrBnEx0PXrvDJJzIvMB9q164dkZGRnD9/nl9//ZWWLVsyYsQIOnXqRFJSkrWb98Ryd3encOHC1m4GoCVugYGBbNy4kdDQUFq2bEnnzp05dOiQsc7evXvp0aMHvXv35vDhw/Tu3Zvu3bvz119/Gevcvn2bWrVqMXv27EwfKzIy0qQEBwej0+l4/vnnc/U5ilygLHTw4EFVokQJ5ebmpmxsbJSHh4fS6XTKxcVF+fr6Wnq5PCk2NlYBKjY2NsevnZycrCIjI1VycnKOXzu/kViZT2JlPomV+SRWlnni4nXvnlLDhyulzQxUqmNHpW7ceCwPnVuxyum/33fu3FHHjh1Td+7cSX/n/VuZl6Q75te9n/DwutnQt29f9eyzz6Y7/vvvvytAfffdd0oppcLDwxWgDh06ZKxz48YNBaitW7cqpZRKSkpS/fv3V+XKlVOOjo6qcuXKavr06emuvWDBAuXn56fs7e1ViRIl1NChQ433jRkzRpUpU0bZ29srb29vNWzYsEzbPmbMGFWrVi21YMECVaZMGeXi4qIGDx6skpKS1OTJk5WXl5fy8PBQ48ePNzlv6tSpyt/fXzk7O6vSpUurN954Q8XHxxvvX7hwoXJ3d1fr169XlSpVUg4ODqp169YqIiIi3WPPnTtXlS5dWjk5OakXXnhB3Ujzu/FgbJs3b66GDRumRo0apYoUKaK8vLzUmDFjTNp2/Phx1bRpU+Xg4KCqVaumQkJCFKDWr1+faRyyy8/PT40bN854u3v37qpdu3Ymddq2bat69uyZ4fnmtuvZZ59VrVq1eqS2ipyV5XtWGhbPAXzrrbfo3Lkz33zzDYULF+bPP//Ezs6OV155hREjRuRociqEEEKIXGJnBzNmaPMCBw2CX37RhoT++CNUr27t1uV9qwplfl/JDtDil9Tbaz0hOSHjup7NofW21Ns/lYO7D2yl1Utlt5XptGrVilq1arFu3Tpee+01s84xGAyULl2aVatWUbx4cfbs2cOgQYPw9vame/fuAHzzzTeMHDmSzz//nPbt2xMbG8vu3bsBWLNmDV999RUrVqygevXqREVFcfjw4Swf8+zZs/z6669s2rSJs2fP8sILLxAeHk7lypXZvn07e/bsoX///jzzzDM0atQIAL1ez8yZMylXrhzh4eEMGTKEd999lzlz5hivm5CQwIQJE1i8eDH29vYMGTKEnj17GtsKcObMGVatWsWGDRuIi4tjwIABDB06lGXLlmXa3sWLFzNy5Ej++usv9u7dS79+/WjatCmBgYEYDAa6du1K2bJl+euvv4iPj+ftt982K/aWMhgMxMfHU7RoUeOxvXv38tZbb5nUa9u2LdOnT8/241y5coVffvmFxYsXZ/sawnosTgDDwsL49ttvsbGxwcbGhrt371K+fHmmTJlC3759ee6553KjnUIIIYTIDb17awlft25w5ow2RHTRInjhBWu3TOSSqlWr8vfff5td387OjnHjxhlv+/r6smfPHlatWmVMAMePH8/bb79t0hmQsj1YREQEJUqUoHXr1tjZ2VG2bFkaNGiQ5WMaDAaCg4NxdXXFz8+Pli1bcvLkSTZu3Iher6dKlSpMnjyZbdu2GRPAoKAgkzZ+9tlnvPHGGyYJ4P3795k9ezYNGzYEtMStWrVq7Nu3z9imxMREFi9ebNzSbNasWXTs2JGpU6dSokSJDNtbs2ZNxowZA2j7Yc+ePZvff/+dwMBAtmzZwtmzZ9m2bZvx/AkTJhAYGPiQyFtu6tSp3L592/j/AhAVFYWXl5dJPS8vL6KiorL9OIsXL8bV1VU+9z+hLE4A7ezs0Ol0gPbiiYiIoFq1ari7uxMREZHjDRRCCCFELqtbV5sX2LMn/P47vPgivPeetmegjY21W5c3db+V+X26B2L2fHQWF3pgOYZnz2e3RWZTShk/y5lr7ty5zJ8/n3///Zc7d+5w7949ateuDUB0dDSXL1/mmWeeyfDcF198kenTp1O+fHnatWtHhw4d6Ny5M7a2mX8MLVeuHK6ursbbXl5e2NjYmCw66OXlRXR0amy3bt3KxIkTOXbsGHFxcSQlJZGYmMjt27dxcXEBwNbWlnr16hnPqVq1KoULF+b48ePGBLBs2bIm+1k3btwYg8HAyZMns0wA0/L29ja27eTJk5QpU8bk3IclwMuWLeP111833v71119p1qxZlucsX76csWPH8tNPP+Hp6Wly34P/39l5DaQVHBzMyy+/jKOjY7avIazH4kVg6tSpw4EDBwBo2bIln3zyCcuWLSMoKIgaNWrkeAOFEEII8RgULw6bNsE772i3J0+GDh3g2jXrtiuvsnXJvNg4ml/X1unhdXPY8ePH8fX1BTAmVEqlDjO9f/++Sf1Vq1bx1ltv0b9/f7Zs2UJYWBivvvoq9+7dA8DJ6YHn8IAyZcpw8uRJvv76a5ycnBgyZAhPP/10usdJy87OzuR2ymqmDx5LWdH033//pUOHDvj7+7N27VpCQ0P5+uuvM3w+GSU+WSVDKfdlVSertmUn2erSpQthYWHGkjZpzcjKlSsZMGAAq1atonXr1ib3lShRIl1vX3R0dLpeQXPt3LmTkydPmj2EWOQ9FieAEydOxNvbG4DPPvuMYsWK8cYbbxAdHc28efNyvIFCCCGEeExsbeGLL2D5cnB2hi1btDmCYWHWbpnIIX/88QdHjhwxrtzo4eEBaCs8pgh74P97586dNGnShCFDhlCnTh0qVqzI2bNnjfe7urpSrlw5fv/990wf18nJiS5dujBz5ky2bdvG3r17OXLkSI49rwMHDpCUlMTUqVNp1KgRlStX5vLly+nqJSUlGTsyQOudu3nzJlWrVjUei4iIMDl379696PV6KleunK22Va1alYiICK5cuWI8tn///izPcXV1pWLFisaSVZK9fPly+vXrxw8//EDHjh3T3d+4cWNCQkJMjm3ZsoUmTZpY+Ew0CxYsICAggFq1amXrfGF9Fg0BVUrh4eFB9f8mh3t4eLBx48ZcaZgQQgghrKRnT/Dz0+YFnjsHTZrA/PnQq5e1WyYscPfuXaKiokhOTubKlSts2rSJSZMm0alTJ/r06QNoiVmjRo34/PPPKVeuHFevXuWjjz4yuU7FihVZsmQJmzdvxtfXl6VLl7J//35jLyLA2LFjGTx4MJ6enrRv3574+Hh2797NsGHDWLRoEcnJyTRs2BBnZ2eWLl2Kk5MTPj4+OfZcK1SoQFJSErNmzaJz587s3r2buXPnpqtnZ2fHsGHDmDlzJnZ2drz55ps0atTIZEimo6Mjffv25csvvyQuLo7hw4fTvXv3TId/PkxgYCAVKlSgb9++TJkyhfj4eD788EMg615Fcyxfvpw+ffowY8YMGjVqZOzpc3Jywt3dHYARI0bw9NNPM3nyZJ599ll++uknfvvtN3bt2mW8zq1btzhz5ozxdnh4OGFhYRQtWpSyZcsaj8fFxbF69WqmTp36SO0W1mVRD6BSikqVKnHx4sXcao8QQggh8oKaNWH/fmjXDu7cgZdfhpEjQfaPe2Js2rQJb29vypUrR7t27di6dSszZ87kp59+wibN3M7g4GDu379PvXr1GDFiBOPHjze5zuDBg3nuuefo0aMHDRs25Nq1awwZMsSkTt++fZk+fTpz5syhevXqdOrUidOnTwNQuHBhvvvuO5o2bUrNmjX5/fff2bBhA8WKFcux51q7dm2mTZvG5MmT8ff3Z9myZUyaNCldPWdnZ9577z169epF48aNcXJyYsWKFSZ1KlasyHPPPUeHDh1o06YN/v7+JgvJWMrGxoYff/yRW7duUb9+fV577TVjkv2oc+i+/fZbkpKSGDp0KN7e3saSdjGeJk2asGLFChYuXEjNmjVZtGgRK1euNC6EA1oPap06dahTpw4AI0eOpE6dOnzyyScmj7dixQqUUrz00kuP1G5hXTqVdtC3GapXr86CBQuMKy7lR3Fxcbi7uxMbG4ubm1uOXttgMBAdHY2np6fJRGaRnsTKfBIr80mszCexsky+jVdysrZR/MSJ2u0WLWDlSnhgkQlL5Fascvrvd2JiIuHh4fj6+spiF/nAokWLCAoK4ubNm5nWGTt2LD/++GO6YbA5bffu3Tz11FOcOXOGChUq5OpjiYLD3Pcsi991p0yZwqhRo/jnn38eqYFCCCGEeALY2Girga5dC4UKwbZtEBCg9Q4KIcyyfv16QkJCOH/+PL/99huDBg2iadOmkvwJq7B4G4hXXnmFhIQEatWqhb29fbpJqdevX8+xxgkhhBAij3juOahaVZsXeOoUNGsG33wDr75q7ZYJkefFx8fz7rvvcuHCBYoXL07r1q1lHp2wGouHgC5evDjL+/v27ftIDcoLZAho3iCxMp/EynwSK/NJrCxTYOIVG6ttHr9hg3Z7yBD46iuwtzf7EjIEVAghcp6571kW9wDmhwRPCCGEENnk7g4//gjjx8OYMTBnDhw+DKtXw3/bRAkhhMi7svW129mzZ/noo4946aWXiI6OBrSVpo4ePZqjjRNCCCFEHqTXawvDbNgAbm6we7c2L3DvXmu3LFekbOgthBB5mbnvVRb3AG7fvp327dvTtGlTduzYwYQJE/D09OTvv/9m/vz5rFmzxuLGCiGEEOIJ1KmTthhMt25w7Bg0bw6zZsGgQfCI+5vlBfb29uj1ei5fvoyHhwf29vaPvG+bEELkNKUU9+7dIyYmBr1ej/1DhuRbnACOHj2a8ePHM3LkSFxdXY3HW7ZsyYwZMyxvsRBCCCGeXJUrw59/aovBrF0LgwdrSeHs2fCEz5vT6/X4+voSGRnJ5cuXrd0cIYTIkrOzM2XLln3o3GqLE8AjR47www8/pDvu4eHBtWvXLL2cEEIIIZ50rq7aHMDJk+GDD2DBAjhyREsIS5e2duseib29PWXLliUpKYnk5GRrN0cIITJkY2ODra2tWaMULE4ACxcuTGRkJL6+vibHDx06RKlSpSy9HHPmzOGLL74gMjKS6tWrM336dJo1a5Zp/bt37/Lpp5/y/fffExUVRenSpfnwww/p37+/sc7atWv5+OOPOXv2LBUqVGDChAl069bN4rYJIYQQwkw6HYweDXXqwEsvwb592rzA1avh6aet3bpHotPpsLOzw87OztpNEUKIR2bxIjC9evXivffeIyoqCp1Oh8FgYPfu3bzzzjv06dPHomutXLmSoKAgPvzwQw4dOkSzZs1o3749ERERmZ7TvXt3fv/9dxYsWMDJkydZvnw5VatWNd6/d+9eevToQe/evTl8+DC9e/eme/fu/PXXX5Y+VSGEEEJYqm1bOHAAataE6Gh45hmYORMs23VKCCFELrF4H8D79+/Tr18/VqxYgVIKW1tbkpOT6dWrF4sWLcLGxsbsazVs2JC6devyzTffGI9Vq1aNrl27MmnSpHT1N23aRM+ePTl37hxFixbN8Jo9evQgLi6OX3/91XisXbt2FClShOXLl5vVLtkHMG+QWJlPYmU+iZX5JFaWkXg94PZtGDgQUv729u4N334LTk5PzD6AQgiRH1k8BNTOzo5ly5bx6aefcujQIQwGA3Xq1KFSpUoWXefevXuEhoYyevRok+Nt2rRhz549GZ7z888/U69ePaZMmcLSpUtxcXGhS5cufPbZZzg5OQFaD+Bbb71lcl7btm2ZPn16pm25e/cud+/eNd6Oi4sDtD/mOb30s8FgQCklS0qbQWJlPomV+SRW5pNYWUbi9QAnJ1i6FOrVQ/fuu+iWLkX98w9qzRoMZcvmSqwk9kII8XDZ2gaiefPmVKhQgQoVKmT7ga9evUpycjJeXl4mx728vIiKisrwnHPnzrFr1y4cHR1Zv349V69eZciQIVy/fp3g4GAAoqKiLLomwKRJkxg3bly64zExMSQmJlr61LJkMBiIjY1FKSXfED+ExMp8EivzSazMJ7GyjMQrE716Ye/jQ+FBg9AfOoSqV4+b33zDzZo1czxW8fHxOXYtIYTIryxOAAMDAylRogS9evXilVdewd/f/5Ea8OBKNUqpTFevMRgM6HQ6li1bhru7OwDTpk3jhRde4Ouvvzb2AlpyTYD333+fkSNHGm/HxcVRpkwZPDw8cmUIqE6nw8PDQz4gPITEynwSK/NJrMwnsbKMxCsL3bpBQADqhRfQh4ZSrFcv7D/8EOePP0ZvwdSRh3F8wredEEKIx8HiBPDy5cusWLGC5cuXM2XKFPz9/XnllVfo1asXpS1Y6rl48eLY2Nik65mLjo5O14OXwtvbm1KlShmTP9DmDCqluHjxIpUqVaJEiRIWXRPAwcEBBweHdMf1en2u/BHX6XS5du38RmJlPomV+SRW5pNYWUbilYVy5WDnTnjjDXSLF+O0ciX6d99FX6hQjj2ExF0IIR7O4nfK4sWL8+abb7J7927Onj1Ljx49WLJkCeXKlaNVq1ZmX8fe3p6AgABCQkJMjoeEhNCkSZMMz2natCmXL1/m1q1bxmOnTp1Cr9cbk8/GjRunu+aWLVsyvaYQQgghHhMnJ1i4EMPs2dwMDgZnZ2u3SAghCpxH+qrM19eX0aNH8/nnn1OjRg22b99u0fkjR45k/vz5BAcHc/z4cd566y0iIiIYPHgwoA3NTLu1RK9evShWrBivvvoqx44dY8eOHYwaNYr+/fsbh3+OGDGCLVu2MHnyZE6cOMHkyZP57bffCAoKepSnKoQQQoicoNPBG2+Q/AjrCAghhMg+i4eApti9ezfLli1jzZo1JCYm0qVLFyZOnGjRNXr06MG1a9f49NNPiYyMxN/fn40bN+Lj4wNAZGSkyZ6AhQoVIiQkhGHDhlGvXj2KFStG9+7dGT9+vLFOkyZNWLFiBR999BEff/wxFSpUYOXKlTRs2DC7T1UIIYQQQggh8gWL9wH84IMPWL58OZcvX6Z169a8/PLLdO3aFed8NIwjN/cRioyLRJegk32izCB7aplPYmU+iZX5JFaWkXiZT/YBFEII67G4B3Dbtm2888479OjRg+LFi5vcFxYWRu3atXOqbfnOpbhLlP6qNFWLVqVjlY50qNSBpmWa4mCbfgEaIYQQQgghhMhpFieAD27SHhsby7Jly5g/fz6HDx8mOTk5xxqX3xy4fAAdOk5cP8GJvSeYuncqLnYutPJtRbuK7WhXsR3li5S3djOFEEIIIYQQ+VS25wD+8ccfBAcHs27dOnx8fHj++edZsGBBTrYt33m26rNcefsKa8PWsjdmL5vPbubK7StsOLWBDac2AFCpaCVjMtiiXAuc7fLP0FohhBBCCCGEdVmUAF68eJFFixYRHBzM7du36d69O/fv32ft2rX4+fnlVhvzlWLOxehasSuDmgwCHfx95W82ndnEpjOb2H1hN6evn+b0vtPM2jcLBxsHnvZ52pgQViteLcsN7YUQQgghhBAiK2YngB06dGDXrl106tSJWbNm0a5dO2xsbJg7d25uti9f0+v01C5Rm9olajP6qdHE3Y3jj/A/2HRmE7+e+ZWI2AhCzoUQci6Et7e8TRm3MsZk8BnfZ3B3dLf2UxBCCCGEEEI8QcxOALds2cLw4cN54403qFSpUm62qcByc3Cja9WudK3aFaUUJ66e0HoHz25i+/ntXIi7wHcHv+O7g99hq7elSZkmtKugJYS1StRCr5NV54QQQgghhBCZMzsB3LlzJ8HBwdSrV4+qVavSu3dvevTokZttK9B0Oh3VPKpRzaMabzV+i4T7CWw/v92YEJ66dood/+5gx787+OCPD/By8aJtxba0q9COwAqBFHcu/vAHEUIIIYQQQhQoFu8DmJCQwIoVKwgODmbfvn0kJyczbdo0+vfvj6ura26187HKzX2Ecmrvo3M3zrH5zGY2nd3E7+d+5/b928b7dOioX6q+sXewQakG2OhtcqL5j5XsqWU+iZX5JFbmk1hZRuJlPtkHUAghrMfiBDCtkydPsmDBApYuXcrNmzcJDAzk559/zsn2WcWTkACmdS/5Hrsjdht7B/++8rfJ/UUcixBYIZB2FdrRtmJbSrqWzJHHzW3yYcp8EivzSazMJ7GyjMTLfJIACiGE9TxSApgiOTmZDRs2EBwcLAngQzyODwiX4i6x5ewWNp3dxJazW7iZeNPk/ppeNY29g03LNsXexj5X2vGo5MOU+SRW5pNYmU9iZRmJl/kkARRCCOvJkQQwv3nSE8C0kgxJ7Lu0z7jVxIHLB1Ck/pcXsi9EK99WtK/YnrYV2uJbxDfX22Qu+TBlPomV+SRW5pNYWUbiZT5JAIUQwnqyvRG8eDKkrBbapEwTPm35KTG3Ywg5F8KmM5vYfHYz0bej+fnkz/x8Uuu5rVKsinGrieY+zXGyc7LyMxBCCCGEEELkFEkACxgPFw961ehFrxq9MCgDYVFhxt7BPRf2cPLaSU5eO8mMv2bgaOtIc5/mxoSwSrEqshG9EEIIIYQQTzBJAAswvU5PXe+61PWuywfNPuD/7d15XFV1/j/w17mXfUf2fRFkEZTlgiAuhebW+KBlJpsmtbImp5oypmm01a1o6ltj9lXLfpNrM/ptTG1mVCINRXEB5BIqICqyXmQRQUBBuef3B3nqCurVwHPhvp6Px33kOfeez/183g8fwcvPOZ9P8+Vm7C7bLW1EX9VShfTT6Ug/nY6X01+Gn72fFAaTA5JhZ87ba4iIiIiIBhIGQJLYW9jjobCH8FDYQxBFEUUNRVIY3Fe+D+XN5fgs7zN8lvcZTBQmSPJJkgLhSLeRnB0kIiIiIjJwDIDUK0EQEO4SjnCXcKQmpqKtsw2ZZzOlrSZOnT+FveV7sbd8LxbsXgB3G3dMHjoZU4Km4L7A++Bk5ST3EIiIiIiI6DoMgKQXazNr3D/sftw/7H4AwKnzp6SN6PeU7UFtay3WFazDuoJ1UAgKxHvFS1tNqDxVA3IjeiIiIiKiwYYBkO5I0JAgBMUH4fn459FxtQP7K/ZLs4PH6o7hUNUhHKo6hIV7F2KI5RBMGjpJ2oje3cZd7u4TERERERklBkD6xcxNzDEhcAImBE7AB/ige/GYH2cHM05n4Pyl89h0bBM2HdsEAIhyj5JmB0f7jIap0lTmERARERERGQcGQOpz3nbemBMzB3Ni5uCq9ioOVR2StprI0+RBXauGulaN9w68B1szW0wInCAFQj8HP7m7T0REREQ0aDEAUr8yUZhgjO8YjPEdg6XJS1HXVodvT38rbUTf0N6AbcXbsK14GwAg1DlUCoNjfMbI23kiIiIiokGGAZDuKldrVzw+4nE8PuJxaEUtjmqOSrODB6sOorihGMUNxVh2eBksTCyQ4J6AaaHTMDFwIka6j4RCUMg9BCIiIiKiAYsBkGSjEBRQeaqg8lThjXFvoOlSk7QR/a5Tu1B9sRqZVZnIrMoEADhZOiE5IBkTArqfNxzqOJR7DxIRERER3QYGQDIYjpaO+HX4r/Hr8F9DFEX8UPsDthVuw5GGI9hXvg+Nlxrx1Ymv8NWJrwAAfvZ+mBAwARMDJyI5IBluNm4yj4CIiIiIyLAxAJJBEgQBkW6RcBPc8Kbrm+gSu3Ck+gh2l+3Gd2e+w6GqQyhvLscX6i/whfoLAECEawQmBkzEhMAJGO83HrbmtjKPgoiIiIjIsDAA0oBgqjRFkm8SknyT8Nb4t9DW2Yasiix8d+Y77C7bDXWtGsfqjuFY3TEsO7wMSkGJUd6jpBnCBO8EmCnN5B4GEREREZGsGABpQLI2s8aUoO7VQgGgvq0e35/9HrvP7MZ3Zd/hTNMZZFdmI7syG0v2LYGVqRXG+o7FxMCJmBAwgQvKEBEREZFRYgCkQcHF2gWPDH8Ejwx/BABw9sJZKQzuKduDurY6pJ9OR/rpdABcUIaIiIiIjBMDIA1K/g7+0mb0oijiWN0x6XbRveV7uaAMERERERklBkAa9K4tKBPpFomXE1/Gla4rXFCGiIiIiIyS7A9BrVy5EgEBAbCwsEBsbCyysrJu+NnMzEwIgtDjVVxcLH1m7dq1vX7m8uXLd2M4NABcW1DmrfFvYd+T+9D0lybs/N1O/CnxT4hyjwIAaTGZ6f+cjiHvD0HSF0l46/u3sK98Hzq7OuUdABERERHRHZJ1BnDz5s2YN28eVq5ciaSkJHz22WeYOnUqTpw4AV9f3xteV1JSAjs7O+nYxcVF5307OzuUlJTonLOwsOjbztOgwQVliIiIiMhYyBoAP/roI8yZMwdPP/00AGDZsmVIT0/HqlWrkJaWdsPrXF1d4eDgcMP3BUGAu7t7X3eXjMQvWVBmYuBEBDoGckEZIiIiIjJIsgXAzs5O5OXlYf78+TrnJ02ahOzs7JteGx0djcuXLyM8PBxvvPEG7r33Xp33W1tb4efnh66uLkRFRWHJkiWIjo6+YXsdHR3o6OiQjltaWgAAWq0WWq32dod2U1qtFqIo9nm7g5Gh1MrXzhdPRj2JJ6OelBaU2V22G7vP7sa+8n29LihzLRAm+9+dBWUMpVYDAWulP9bq9rBe+uuvWrH2RES3JlsAbGhoQFdXF9zcdH85dnNzQ21tba/XeHh4YPXq1YiNjUVHRwc2bNiACRMmIDMzE+PGjQMAhIaGYu3atYiMjERLSws+/vhjJCUloaCgAMHBwb22m5aWhkWLFvU4X19f3+fPDmq1WjQ3N0MURSgUvG3wZgy1Vm6CGx4LfAyPBT6GK11XkF+fj/3V+7Gvah+O1h1FeXM51qjXYI16DQAgdEgoxnqNxVivsUj0SISNmU2f98lQa2WIWCv9sVa3h/XSX3/V6uLFi33WFhHRYCWIoijK8cU1NTXw8vJCdnY2EhMTpfPvvPMONmzYoLOwy81Mnz4dgiDgm2++6fV9rVaLmJgYjBs3DsuXL+/1M73NAPr4+KCpqUnnWcO+oNVqUV9fDxcXF/6CcAsDsVZtnW3IqsjC7rLd2FO2B+pzap33TRQmiPeMl2YIE7wTYKY0+8XfOxBrJRfWSn+s1e1hvfTXX7VqaWmBo6Mjmpub+/znNxHRYCHbDKCzszOUSmWP2b66uroes4I3k5CQgI0bN97wfYVCgbi4OJSWlt7wM+bm5jA3N+/12v74IS4IQr+1PdgMtFrZWthi2rBpmDZsGoAbLChTlY3sqmwszVoKK1MrjPMb170h/S9cUGag1UpOrJX+WKvbw3rprz9qxboTEd2abAHQzMwMsbGxyMjIwIMPPiidz8jIQEpKit7t5Ofnw8PD44bvi6IItVqNyMjIX9Rfojuhz4Iyu07twq5TuwBwQRkiIiIi6l+yrgKampqKmTNnQqVSITExEatXr0ZFRQXmzp0LAFiwYAGqq6uxfv16AN2rhPr7+2P48OHo7OzExo0bsWXLFmzZskVqc9GiRUhISEBwcDBaWlqwfPlyqNVqrFixQpYxEv2cv4M/5sTMwZyYOdKCMt+d+Q67y3Zjb/neXheUubbdRHLA3VlQhoiIiIgGL1kD4IwZM9DY2IjFixdDo9EgIiICO3bsgJ+fHwBAo9GgoqJC+nxnZydeeeUVVFdXw9LSEsOHD8d///tfTJs2TfrMhQsX8Pvf/x61tbWwt7dHdHQ09u3bh/j4+Ls+PqKbEQQBkW6RiHSLxMuJL+NK1xUcqT7SvcJo2W4crDyI8uZy/D3/7/h7/t8BAJGukd23iwZOwHi/8bA1t5V5FEREREQ0kMi2CIwha2lpgb29fb88RK7ValFXVwdXV1c+q3ALxl6rawvKXJshVNeqdd43UZgg3ite2m4i0CwQ3h7eRlmr22Hsf69uB2t1e1gv/fVXrfrz5zcR0WAh6wwgEd2YtZk1pgRNwZSgKQB0F5TZXbYbp5tOI7syG9mV2ViybwnMleaI8YjBKK9RiPeKxyjvUQhwCOAzhEREREQkYQAkGiD0WVDmYNVBHKw6KF3jbOWMeK94xHt2B8I4zzg4WTnJNQQiIiIikhkDINEA9fMFZbq6unD41GGcunQKuZpcHK4+DHWtGg3tDdhRugM7SndI1wUNCeqeIfxxpjDKPQoWJhYyjoSIiIiI7hYGQKJBQBAEBNoHIiE4AbOiZgEAOq52oOBcAY5UH8Hh6sM4Un0EJxtP4tT5Uzh1/hT+UfgPAICpwhQj3Uf+dOuo1ygEOwXf8X6ERERERGS4GACJBilzE/Pu2z+94vECXgAAnL90Hrk1uThcdRhHao7gcNVh1LfXI7cmF7k1uViR071dioOFA+I843RmCrkFBREREdHAxwBIZESGWA7BpKGTMGnoJACAKIooby7vDoQ/zhTmafJw4fIFZJzJQMaZDOlaP3s/nUAY6xkLK1MruYZCRERERHeAAZDIiAmCAH8Hf/g7+GNGxAwAwJWuKzhWd0wKhIerD6OovgjlzeUoby6XNqlXCkpEuEborDoa5hwGpUIp55CIiIiI6CYYAIlIh6nSFNEe0Yj2iMazqmcBAC0dLcityf0pFFYdhqZVg4JzBSg4V4DVR1cDAGzMbKDyVEmrjsZ7xcPbzlvO4RARERHRzzAAEtEt2ZnbITkgGckBydK5qpaq7kD44/OEuTW5aO1sRebZTGSezZQ+52nrqXPrqMpTBTtzbtBMREREJAcGQCK6I9523vC288ZDYQ8BALq0XShqKNJ5nvBY3THUXKzBtuJt2Fa8DQAgQEC4S7i0QM0or1GIcI2AqdJUxtEQERERGQcGQCLqE0pF9zOBEa4RmBMzBwDQ1tmGo5qjOltRlDeX43j9cRyvP4416jUAAEsTS8R4xEiBcJT3KPjZ+0EQBDmHRERERDToMAASUb+xNrPGWL+xGOs3VjpX21qLnOocKRAeqT6C5o5mHKg8gAOVB6TPuVi5dD9H+OPzhHGecXC0dJRjGERERESDBgMgEd1V7jbumB4yHdNDpgMAtKIWpY2lUiA8XH0YBbUFqG+vx39O/gf/Ofkf6dphTsN0nicc6TYS5ibmcg2FiIiIaMBhACQiWSkEBUKcQxDiHIJZI2cBAC5fvQx1rVrn1tFT50/hZONJnGw8iY0/bAQAmCnNEOUe9dNWFF6jEDQkiLeOEhEREd0AAyARGRwLEwskeCcgwTtBOtfY3oicmhxp1dHDVYfReKlRuo30GkcLR50FZuK94uFi7SLHMIiIiIgMDgMgEQ0ITlZOmBI0BVOCpgAARFFE2YUynVVHj2qOoulyE9JPpyP9dLp0bYBDgBQIVZ4qeCu5NyEREREZJwZAIhqQBEFAoGMgAh0D8dvI3wIAOrs6UXiuUOfW0aKGIpRdKEPZhTJsPr4ZAKAUulcsjfOMg8pThTivOES4RsBMaSbnkIiIiIj6HQMgEQ0aZkozxHrGItYzFn+I+wMAoPlyM3JrcnG4+nD3q+owzrWdQ8G5AhScK8D/y/9/AABzpTlGuo+EyqM7EKo8VQhzDoNSoZRzSERERER9igGQiAY1ewt7TAicgAmBEwAAXV1dUJepcbbzLPI0ecityUVuTS6aLjf99Dxhbve1VqZWiPGI+Wmm0DMOQ4cMhUJQyDgiIiIiojvHAEhERkUQBHjZeCHaNRoPhz8MoPt5wjNNZ5BTk4Pcmlzk1OQgryYPbVfasL9iP/ZX7Jeutze3h8pTJQXCOK84+Nj5cOVRIiIiGhAYAInI6AmCgKFDhmLokKF4NOJRAECXtgsljSXdgbA6B7maXORr8tHc0YzdZbuxu2y3dL2LlUv3baM/u33U3cZdruEQERER3RADIBFRL5QKJcJdwhHuEi7tT3il6wqO1x/vDoQ/zhQW1hWivr0eO0p3YEfpDul6bztvaZbw2ozhEMshcg2HiIiICAADIBGR3kyVpohyj0KUexSeiX0GQPem9QW1BVIgzK3JxYn6E6hqqUJVSxW2FW+Trg90DNR5njDGIwa25rYyjYaIiIiMEQMgEdEvYGFigVHeozDKe5R0rrWzFUc1R3VC4anzp3Cm6QzONJ2RtqMQICDUOVRnpjDKPQqWppZyDYeIiIgGOQZAIqI+ZmNmg3F+4zDOb5x0rulSE/I0edLzhDnVOahsqURRQxGKGoqw4YcNALhHIREREfUvBkAiorvA0dIREwMnYmLgROncudZz0jYUOTU5yKnJQV1bHfcoJCIion7DAEhEJBM3GzfcP+x+3D/sfgDd21FUtVTp3DrKPQqJiIioLzEAEhEZCEEQ4GPvAx97HzwY9iAA7lFIREREfYsBkIjIgN1sj8Kfb0ehrlVzj0IiIiK6JQZAIqIB5ud7FM6Omg2ge4/CY3XHdG4f5R6FREREdD3ZA+DKlSvxwQcfQKPRYPjw4Vi2bBnGjh3b62czMzNx77339jhfVFSE0NBQ6XjLli148803cfr0aQwdOhTvvPMOHnzwwX4bAxGR3EyVpoj2iEa0R7S0R+GlK5fww7kfdG4fLaov0nuPQmtTa5lGQ0RERP1F1gC4efNmzJs3DytXrkRSUhI+++wzTJ06FSdOnICvr+8NryspKYGdnZ107OLiIv354MGDmDFjBpYsWYIHH3wQW7duxSOPPIL9+/dj1KhRvTVHRDQoWZpa/uI9CsMdwxHrE4twl3CEuYQh0DEQJgrZ/+2QiIiI7pAgiqIo15ePGjUKMTExWLVqlXQuLCwMDzzwANLS0np8/toMYFNTExwcHHptc8aMGWhpacHOnTulc1OmTIGjoyP++c9/6tWvlpYW2Nvbo7m5WSdo9gWtVou6ujq4urpCoeBqfTfDWumPtdIfa9XTjfYo7I2Z0gzBQ4IR5hKGMOfuV6hzKEKcQ2BlanWXe25Y+HdLf/1Vq/78+U1ENFjI9s+4nZ2dyMvLw/z583XOT5o0CdnZ2Te9Njo6GpcvX0Z4eDjeeOMNndtCDx48iJdfflnn85MnT8ayZctu2F5HRwc6Ojqk45aWFgDdP6C0Wq2+Q9KLVquFKIp93u5gxFrpj7XSH2vVk725PZL9k5HsnyydO9d6DjnVOdh/Zj8qL1WipLEExQ3FuHT1Eo7XH8fx+uM6bQgQ4Ofgh1DnUCkUXguIxvJ8If9u6a+/asXaExHdmmwBsKGhAV1dXXBzc9M57+bmhtra2l6v8fDwwOrVqxEbG4uOjg5s2LABEyZMQGZmJsaNGwcAqK2tva02ASAtLQ2LFi3qcb6+vh6XL1++3aHdlFarRXNzM0RR5L8Q3wJrpT/WSn+slX4ECFDZqxAcGAx7e3soFApoRS2qW6txsukkSi+UorSpFKUXSnGq6RSaOppw9sJZnL1wFrtO7dJpy9nSGcEOwQh2DEaQQ5D0Z09rz0G1RQX/bumvv2p18eLFPmuLiGiwkv1Bjut/+IuieMNfCEJCQhASEiIdJyYmorKyEv/zP/8jBcDbbRMAFixYgNTUVOm4paUFPj4+cHFx6ZdbQAVBgIuLC39BuAXWSn+slf5YK/31Vit3N3fEIlbnc6Ioor69HkUNRShuKEZxQ7H058qWSjRcakDDpQYc1BzUuc7GzAahTqE9Zg2HDhk6IJ8z5N8t/fVXrSwsLPqsLSKiwUq2n7DOzs5QKpU9Zubq6up6zODdTEJCAjZu3Cgdu7u733ab5ubmMDc373FeoVD0yw9xQRD6re3BhrXSH2ulP9ZKf/rWyt3WHe627rg3QHel5tbO1u5AWF+EooYfX/VFOHX+FFo7W5GryUWuJlfnGlOFKYKGBOk8ZxjmEoYQpxBYmxn2yqT8u6W//qgV605EdGuyBUAzMzPExsYiIyNDZ4uGjIwMpKSk6N1Ofn4+PDw8pOPExERkZGToPAf47bffYvTo0X3TcSIi0puNmY201+DPdXZ14vT501IgvBYOixuK0X6lXTq+np+9X48FaMJcwuBs5Xy3hkRERDSgyXqPTWpqKmbOnAmVSoXExESsXr0aFRUVmDt3LoDuWzOrq6uxfv16AMCyZcvg7++P4cOHo7OzExs3bsSWLVuwZcsWqc2XXnoJ48aNw1//+lekpKRg+/bt+O6777B//35ZxkhERD2ZKc26g5xLGBD203mtqEVlc6VOMLx2S2lDewPKm8tR3lze8zlDK2ed2cJr//Wx8xlUzxkSERH9UrIGwBkzZqCxsRGLFy+GRqNBREQEduzYAT8/PwCARqNBRUWF9PnOzk688sorqK6uhqWlJYYPH47//ve/mDZtmvSZ0aNHY9OmTXjjjTfw5ptvYujQodi8eTP3ACQiGgAUggJ+Dn7wc/DDlKApOu81tDf8NFv4s1nDiuYKNLQ3IKsiC1kVWTrXWJtaS7OEoU6hUjgMGhIEU6Xp3RwaERGRQZB1H0BDxX0ADQNrpT/WSn+slf4GSq1aO1tR0lAizRReC4il50txVXu112tMFCbdzxleN2sY6hx6x88ZDpR6GQLuA0hEJJ+Bt8waERHRz9iY2SDWMxaxnrqrk17puoLTTad7LEBT3FCMtitt0oqlW7FV5zpfe1+dfQyvhUMXa5e7OSwiIqJ+wQBIRESDkqnSFKHO3dtMPIifFhsTRRFVLVU9biUtqi9CfXs9KporUNFcgW9Pf6vTnpOlU68L0Pja+0IhcMaPiIgGBgZAIiIyKoIgwMfeBz72Ppg0dJLOe43tjb0uQHP2wlk0XmrE/or92F+hu6iYlakVQpxCEOYchmCbYNwz7B6ovFSwMbO5m8MiIiLSCwMgERHRj5ysnDDGdwzG+I7ROd9+pR0lDSU9Zg1LG0vRfqUd+bX5yK/NBwAsOrQIAgSEuYQhzjOu++UVhxFuI2Bhwo3KiYhIXgyAREREt2BlaoVoj2hEe0TrnL+qvYozTWdQVF+EwnOFyD6bjcLzhahqqcKJ+hM4UX8C6wrWAeje4D7SLVInFIa7hMNEwR/FRER09/CnDhER0R0yUZhgmNMwDHMahunDpqMupHtly7r2OuRU5yC3Jhc5NTnIqclBQ3sDjmqO4qjmKD7L+wwAYGliiWiPaCkUqjxVCHYK5jOFRETUbxgAiYiI+pi7jTumh0zH9JDpALoXnilvLtcJhbk1ubjYeRHZldnIrsyWrrU3t0esZ6xOKPS19+WG9kRE1CcYAImIiPqZIAjwd/CHv4M/fjP8NwAArajFycaT3YGwunuWML82H80dzdhTtgd7yvZI17tYuSDOK04nFLrZuMk1HCIiGsAYAImIiGSgEBTSNhWPj3gcQPfehcfrj+uEwsK6QtS312NH6Q7sKN0hXe9j5yOFQpWnCipPFRwsHGQaDRERDRQMgERERAbCVGmKKPcoRLlH4emYpwEAl65cQsG5gp+eJ6zOQXFDMSpbKlHZUomvi76Wrg8eEqwTCqPdo2FtZi3XcIiIyAAxABIRERkwS1NLJHgnIME7QTrX0tGCfE2+tMBMTnUOyi6UofR8KUrPl+Ifhf8A0D3LONxluLTqqMpThRFuI2CmNJNrOEREJDMGQCIiogHGztwO4/3HY7z/eOlcY3ujzqqjOdU50LRqUFhXiMK6Qnyh/gIAYKY0w0i3kdIsYZxXHMKcw6BUKOUaDhER3UUMgERERIOAk5UTJgdNxuSgydK5mos10rOE11YePX/pvHR8jbWpNWI8YnRC4VDHoVx5lIhoEGIAJCIiGqQ8bT2REpqClNAUAN3bUZRdKNMJhXk1eWi70oasiixkVWRJ1zpaOHaHwZ+FQi9bL4ZCIqIBjgGQiIjISAiCgEDHQAQ6BmJGxAwAQJe2CyWNJVIozK3JhbpWjabLTcg4k4GMMxnS9e427j8Fwh+fK3S2cpZrOEREdAcYAImIiIyYUqFEuEs4wl3CMTtqNgCgs6sTx+qO6YTCY3XHUNtai3+f/Df+ffLf0vX+Dv46oTDWMxZ25nZyDYeIiG6BAZCIiIh0mCnNEOMRgxiPGDyLZwEA7Vfaoa5V64TCksYSnL1wFmcvnMVXJ74CAAgQEOIcohMKo9yjYGlqKeeQiIjoRwyAREREdEtWplYY7TMao31GS+eaLzcjT5OHnOoc5Gq6N68vby5HcUMxihuKseGHDQAAE4UJIlwjpFAY6x4LV7jKNRQiIqPGAEhERER3xN7CHskByUgOSJbO1bXVIbcmV2fj+nNt56CuVUNdq8bnRz8HAATaB6L0xVK5uk5EZLQYAImIiKjPuFq7YlrwNEwLngage+XRqpYq6bbRa/8NHRIqc0+JiIwTAyARERH1G0EQ4GPvAx97HzwU9hAAoKurC2eqz8jcMyIi46SQuwNERERkXARBgK2ZrdzdICIySgyARERERERERoIBkIiIiIiIyEgwABIRERERERkJBkAiIiIiIiIjwQBIRERERERkJBgAiYiIiIiIjAQDIBERERERkZFgACQiIiIiIjISDIBERERERERGQvYAuHLlSgQEBMDCwgKxsbHIysrS67oDBw7AxMQEUVFROufXrl0LQRB6vC5fvtwPvSciIiIiIho4ZA2Amzdvxrx58/D6668jPz8fY8eOxdSpU1FRUXHT65qbmzFr1ixMmDCh1/ft7Oyg0Wh0XhYWFv0xBCIiIiIiogHDRM4v/+ijjzBnzhw8/fTTAIBly5YhPT0dq1atQlpa2g2ve/bZZ/HYY49BqVRi27ZtPd4XBAHu7u5696OjowMdHR3ScUtLCwBAq9VCq9Xq3Y4+tFotRFHs83YHI9ZKf6yV/lgr/bFWt4f10l9/1Yq1JyK6NdkCYGdnJ/Ly8jB//nyd85MmTUJ2dvYNr1uzZg1Onz6NjRs3YunSpb1+prW1FX5+fujq6kJUVBSWLFmC6OjoG7aZlpaGRYsW9ThfX1/f57eOarVaNDc3QxRFKBSy34Fr0Fgr/bFW+mOt9Mda3R7WS3/9VauLFy/2WVtERIOVbAGwoaEBXV1dcHNz0znv5uaG2traXq8pLS3F/PnzkZWVBROT3rseGhqKtWvXIjIyEi0tLfj444+RlJSEgoICBAcH93rNggULkJqaKh03NzfD19cX5ubmfX7rqFarRWtrKywsLPgLwi2wVvpjrfTHWumPtbo9rJf++qtWnZ2dAABRFPusTSKiwUbWW0CB7ts1f04UxR7nAKCrqwuPPfYYFi1ahGHDht2wvYSEBCQkJEjHSUlJiImJwSeffILly5f3eo25uTnMzc2l42u3gPr5+d3WWIiIiEh+Fy9ehL29vdzdICIySLIFQGdnZyiVyh6zfXV1dT1mBYHu/5nn5uYiPz8fL7zwAoCfniEwMTHBt99+i+Tk5B7XKRQKxMXFobS0VO++eXp6orKyEra2tr2G0V+ipaUFPj4+qKyshJ2dXZ+2PdiwVvpjrfTHWumPtbo9rJf++qtWoiji4sWL8PT07LM2iYgGG9kCoJmZGWJjY5GRkYEHH3xQOp+RkYGUlJQen7ezs0NhYaHOuZUrV2LPnj3417/+hYCAgF6/RxRFqNVqREZG6t03hUIBb29vvT9/J+zs7PgLgp5YK/2xVvpjrfTHWt0e1kt//VErzvwREd2crLeApqamYubMmVCpVEhMTMTq1atRUVGBuXPnAuh+Nq+6uhrr16+HQqFARESEzvWurq6wsLDQOb9o0SIkJCQgODgYLS0tWL58OdRqNVasWHFXx0ZERERERGRoZA2AM2bMQGNjIxYvXgyNRoOIiAjs2LFDevZOo9Hcck/A6124cAG///3vUVtbC3t7e0RHR2Pfvn2Ij4/vjyEQERERERENGLIvAvPcc8/hueee6/W9tWvX3vTahQsXYuHChTrn/va3v+Fvf/tbH/Wu75mbm+Ptt9/WWXSGesda6Y+10h9rpT/W6vawXvpjrYiI5COIXCuZiIiIiIjIKHCjIiIiIiIiIiPBAEhERERERGQkGACJiIiIiIiMBAMgERERERGRkWAAvItWrlyJgIAAWFhYIDY2FllZWXJ3ySDt27cP06dPh6enJwRBwLZt2+TuksFKS0tDXFwcbG1t4erqigceeAAlJSVyd8sgrVq1CiNGjJA2nk5MTMTOnTvl7taAkJaWBkEQMG/ePLm7YnAWLlwIQRB0Xu7u7nJ3y2BVV1fj8ccfh5OTE6ysrBAVFYW8vDy5u0VEZFQYAO+SzZs3Y968eXj99deRn5+PsWPHYurUqbe9z6ExaGtrw8iRI/G///u/cnfF4O3duxfPP/88Dh06hIyMDFy9ehWTJk1CW1ub3F0zON7e3njvvfeQm5uL3NxcJCcnIyUlBcePH5e7awYtJycHq1evxogRI+TuisEaPnw4NBqN9CosLJS7SwapqakJSUlJMDU1xc6dO3HixAl8+OGHcHBwkLtrRERGhdtA3CWjRo1CTEwMVq1aJZ0LCwvDAw88gLS0NBl7ZtgEQcDWrVvxwAMPyN2VAaG+vh6urq7Yu3cvxo0bJ3d3DN6QIUPwwQcfYM6cOXJ3xSC1trYiJiYGK1euxNKlSxEVFYVly5bJ3S2DsnDhQmzbtg1qtVrurhi8+fPn48CBA7z7hYhIZpwBvAs6OzuRl5eHSZMm6ZyfNGkSsrOzZeoVDUbNzc0AuoMN3VhXVxc2bdqEtrY2JCYmyt0dg/X888/j/vvvx8SJE+XuikErLS2Fp6cnAgIC8Oijj+LMmTNyd8kgffPNN1CpVPjNb34DV1dXREdH4/PPP5e7W0RERocB8C5oaGhAV1cX3NzcdM67ubmhtrZWpl7RYCOKIlJTUzFmzBhERETI3R2DVFhYCBsbG5ibm2Pu3LnYunUrwsPD5e6WQdq0aROOHj3KOxRuYdSoUVi/fj3S09Px+eefo7a2FqNHj0ZjY6PcXTM4Z86cwapVqxAcHIz09HTMnTsXL774ItavXy9314iIjIqJ3B0wJoIg6ByLotjjHNGdeuGFF/DDDz9g//79cnfFYIWEhECtVuPChQvYsmULZs+ejb179zIEXqeyshIvvfQSvv32W1hYWMjdHYM2depU6c+RkZFITEzE0KFDsW7dOqSmpsrYM8Oj1WqhUqnw7rvvAgCio6Nx/PhxrFq1CrNmzZK5d0RExoMzgHeBs7MzlEplj9m+urq6HrOCRHfij3/8I7755ht8//338Pb2lrs7BsvMzAxBQUFQqVRIS0vDyJEj8fHHH8vdLYOTl5eHuro6xMbGwsTEBCYmJti7dy+WL18OExMTdHV1yd1Fg2VtbY3IyEiUlpbK3RWD4+Hh0eMfW8LCwrgYGhHRXcYAeBeYmZkhNjYWGRkZOuczMjIwevRomXpFg4EoinjhhRfw9ddfY8+ePQgICJC7SwOKKIro6OiQuxsGZ8KECSgsLIRarZZeKpUKv/vd76BWq6FUKuXuosHq6OhAUVERPDw85O6KwUlKSuqxTc3Jkyfh5+cnU4+IiIwTbwG9S1JTUzFz5kyoVCokJiZi9erVqKiowNy5c+XumsFpbW3FqVOnpOOysjKo1WoMGTIEvr6+MvbM8Dz//PP4xz/+ge3bt8PW1laaZba3t4elpaXMvTMsr732GqZOnQofHx9cvHgRmzZtQmZmJnbt2iV31wyOra1tj+dIra2t4eTkxOdLr/PKK69g+vTp8PX1RV1dHZYuXYqWlhbMnj1b7q4ZnJdffhmjR4/Gu+++i0ceeQRHjhzB6tWrsXr1arm7RkRkVBgA75IZM2agsbERixcvhkajQUREBHbs2MF/+exFbm4u7r33Xun42nM0s2fPxtq1a2XqlWG6tq3IPffco3N+zZo1eOKJJ+5+hwzYuXPnMHPmTGg0Gtjb22PEiBHYtWsX7rvvPrm7RgNYVVUVfvvb36KhoQEuLi5ISEjAoUOH+P/2XsTFxWHr1q1YsGABFi9ejICAACxbtgy/+93v5O4aEZFR4T6ARERERERERoLPABIRERERERkJBkAiIiIiIiIjwQBIRERERERkJBgAiYiIiIiIjAQDIBERERERkZFgACQiIiIiIjISDIBERERERERGggGQiIiIiIjISDAAEtGAceDAAURGRsLU1BQPPPCA3N0BAAiCgG3btgEAzp49C0EQoFar77i9vmiDiIiI6EYYAIkIAPDEE09AEAQIggBTU1O4ubnhvvvuwxdffAGtVit39wAAqampiIqKQllZGdauXSt3d3rw8fGBRqNBRESEXp9/4oknegTZ222DiIiI6HYwABKRZMqUKdBoNDh79ix27tyJe++9Fy+99BJ+9atf4erVq3J3D6dPn0ZycjK8vb3h4ODQZ+1euXKlT9pRKpVwd3eHiYmJrG0QERER3QgDIBFJzM3N4e7uDi8vL8TExOC1117D9u3bsXPnTp0Zt48++giRkZGwtraGj48PnnvuObS2tgIA2traYGdnh3/96186bf/73/+GtbU1Ll682Ot3d3R04MUXX4SrqyssLCwwZswY5OTkAPjptsjGxkY89dRTEAThhjOA/v7+WLJkCR577DHY2NjA09MTn3zyic5nBEHAp59+ipSUFFhbW2Pp0qVSH2NjY2FhYYHAwEAsWrRIJ/iWlpZi3LhxsLCwQHh4ODIyMnTa7e32zePHj+P++++HnZ0dbG1tMXbsWJw+fRoLFy7EunXrsH37dmnmNTMzs9c29u7di/j4eJibm8PDwwPz58/X6dc999yDF198Ea+++iqGDBkCd3d3LFy4sNf6EBERkXFjACSim0pOTsbIkSPx9ddfS+cUCgWWL1+OY8eOYd26ddizZw9effVVAIC1tTUeffRRrFmzRqedNWvW4Ne//jVsbW17/Z5XX30VW7Zswbp163D06FEEBQVh8uTJOH/+vHRbpJ2dHZYtWwaNRoMZM2bcsM8ffPABRowYgaNHj2LBggV4+eWXe4S1t99+GykpKSgsLMRTTz2F9PR0PP7443jxxRdx4sQJfPbZZ1i7di3eeecdAIBWq8VDDz0EpVKJQ4cO4dNPP8Vf/vKXm9auurpaCox79uxBXl4ennrqKVy9ehWvvPIKHnnkEWnWVaPRYPTo0b22MW3aNMTFxaGgoACrVq3C3//+dym0XrNu3TpYW1vj8OHDeP/997F48eIeYyYiIiKCSEQkiuLs2bPFlJSUXt+bMWOGGBYWdsNr/+///k90cnKSjg8fPiwqlUqxurpaFEVRrK+vF01NTcXMzMxer29tbRVNTU3FL7/8UjrX2dkpenp6iu+//750zt7eXlyzZs1Nx+Hn5ydOmTKlR/+nTp0qHQMQ582bp/OZsWPHiu+++67OuQ0bNogeHh6iKIpienq6qFQqxcrKSun9nTt3igDErVu3iqIoimVlZSIAMT8/XxRFUVywYIEYEBAgdnZ29trX3mp+fRuvvfaaGBISImq1WukzK1asEG1sbMSuri5RFEVx/Pjx4pgxY3TaiYuLE//yl7/0+r1ERERkvDgDSES3JIoiBEGQjr///nvcd9998PLygq2tLWbNmoXGxka0tbUBAOLj4zF8+HCsX78eALBhwwb4+vpi3LhxvbZ/+vRpXLlyBUlJSdI5U1NTxMfHo6io6Lb7m5iY2OP4+nZUKpXOcV5eHhYvXgwbGxvp9cwzz0Cj0aC9vR1FRUXw9fWFt7f3Db/nemq1GmPHjoWpqeltj+GaoqIiJCYm6tQ/KSkJra2tqKqqks6NGDFC5zoPDw/U1dXd8fcSERHR4MQASES3VFRUhICAAABAeXk5pk2bhoiICGzZsgV5eXlYsWIFAN3FVJ5++mnpNtA1a9bgySef1AkxPyeKIgD0eP/64PlLXN+OtbW1zrFWq8WiRYugVqulV2FhIUpLS2FhYSH18WZtXs/S0vIX97u3GvRWr+tDpiAIBrN6KxERERkOBkAiuqk9e/agsLAQDz/8MAAgNzcXV69exYcffoiEhAQMGzYMNTU1Pa57/PHHUVFRgeXLl+P48eOYPXv2Db8jKCgIZmZm2L9/v3TuypUryM3NRVhY2G33+dChQz2OQ0NDb3pNTEwMSkpKEBQU1OOlUCgQHh6OiooKnbEePHjwpm2OGDECWVlZN1xl1MzMDF1dXTdtIzw8HNnZ2ToBNDs7G7a2tvDy8rrptURERETXYwAkIklHRwdqa2tRXV2No0eP4t1330VKSgp+9atfYdasWQCAoUOH4urVq/jkk09w5swZbNiwAZ9++mmPthwdHfHQQw/hz3/+MyZNmqRz6+T1rK2t8Yc//AF//vOfsWvXLpw4cQLPPPMM2tvbMWfOnNsex4EDB/D+++/j5MmTWLFiBb766iu89NJLN73mrbfewvr167Fw4UIcP34cRUVF2Lx5M9544w0AwMSJExESEoJZs2ahoKAAWVlZeP3112/a5gsvvICWlhY8+uijyM3NRWlpKTZs2ICSkhIA3SuW/vDDDygpKUFDQ0OvQfG5555DZWUl/vjHP6K4uBjbt2/H22+/jdTUVCgU/F84ERER3R7+9kBEkl27dsHDwwP+/v6YMmUKvv/+eyxfvhzbt2+HUqkEAERFReGjjz7CX//6V0RERODLL79EWlpar+3NmTMHnZ2deOqpp2753e+99x4efvhhzJw5EzExMTh16hTS09Ph6Oh42+P405/+hLy8PERHR2PJkiX48MMPMXny5JteM3nyZPznP/9BRkYG4uLikJCQgI8++gh+fn4Aulc+3bp1Kzo6OhAfH4+nn35aWiH0RpycnLBnzx60trZi/PjxiI2Nxeeffy7drvnMM88gJCQEKpUKLi4uOHDgQI82vLy8sGPHDhw5cgQjR47E3LlzMWfOHCmYEhEREd0OQeztwRYioj7w5Zdf4qWXXkJNTQ3MzMzuynf6+/tj3rx5mDdv3l35PiIiIqKBxETuDhDR4NPe3o6ysjKkpaXh2WefvWvhj4iIiIhujreAElGfe//99xEVFQU3NzcsWLBA7u4QERER0Y94CygREREREZGR4AwgERERERGRkWAAJCIiIiIiMhIMgEREREREREaCAZCIiIiIiMhIMAASEREREREZCQZAIiIiIiIiI8EASEREREREZCQYAImIiIiIiIzE/wfWcK2/E+oG+AAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# save_name: named of saved graph if not None\n",
    "plot_results(RMSE_array, save_name=None, your_year=min_time_date.strftime('%Y'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "4dvarnet-starter",
   "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": "3.11.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}