03.01-plot-alpha-distributions.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "from biom import Table, load_table\n",
    "from biom.util import biom_open\n",
    "from functools import partial\n",
    "from skbio import TreeNode\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from unifrac import faith_pd\n",
    "from tempfile import NamedTemporaryFile\n",
    "import multiprocessing\n",
    "import itertools\n",
    "from itertools import repeat\n",
    "import json\n",
    "import matplotlib\n",
    "import seaborn as sns\n",
    "N_WORKERS = multiprocessing.cpu_count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "table_MG_path = 'data/finrisk/anonymized-finrisk-MG-BL_AGE-filtered_rarefied_table.biom'\n",
    "finrisk_metadata_path = 'data/finrisk/anonymized.age-only.finrisk.metadata.txt'\n",
    "wol_tree_path = 'data/wol/wol-tree.nwk'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "table_MG = load_table(table_MG_path)\n",
    "metadata = pd.read_csv(finrisk_metadata_path, sep='\\t', low_memory=False, index_col=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "class FaithPDAnalysis:\n",
    "    \n",
    "    def __init__(self, table_path, tree_path, metadata):\n",
    "        self.table_path = table_path\n",
    "        self.tree_path = tree_path\n",
    "        self.metadata = metadata\n",
    "        self.faith_pd = None\n",
    "        \n",
    "    def convert_table(self):\n",
    "        pass\n",
    "        \n",
    "    def convert_tree(self):\n",
    "        pass\n",
    "        \n",
    "    def get_faith_pd(self):\n",
    "        self.convert_table()\n",
    "        self.convert_tree()\n",
    "        table_biom_path = self.table_path\n",
    "        tree_path = self.tree_path\n",
    "        faith_series = pd.DataFrame(faith_pd(table_biom_path, tree_path))\n",
    "        faith_series.index.name = '#SampleID'\n",
    "        self.faith_pd = faith_series\n",
    "        return faith_series\n",
    "    \n",
    "    def get_observed_otus(self):\n",
    "        table = load_table(self.table_path)\n",
    "        observed_otus = pd.DataFrame((table.matrix_data > 0).sum(axis=0)).transpose()\n",
    "        observed_otus.index = table.ids('sample')\n",
    "        observed_otus.index.name = '#SampleID'\n",
    "        observed_otus.columns = ['observed_otus']\n",
    "        self.observed_otus = observed_otus\n",
    "        return observed_otus\n",
    "    \n",
    "    def merge_metadata(self):\n",
    "        faith_pd = self.get_faith_pd()\n",
    "        observed_otus = self.get_observed_otus()\n",
    "        self.metadata = faith_pd.join(observed_otus).join(self.metadata)\n",
    "        return self.metadata\n",
    "    \n",
    "    def filter_metadata(self, fn):\n",
    "        self.metadata = self.metadata.loc[fn(self.metadata)]\n",
    "        return self.metadata\n",
    "    \n",
    "    def boolean_splitter(self, fn, metadata=None):\n",
    "        if metadata is None:\n",
    "            metadata = self.metadata\n",
    "        bools = fn(metadata)\n",
    "        return metadata[bools], metadata[~bools]\n",
    "    \n",
    "    def make_splitter(self, fn):\n",
    "        def f(metadata):\n",
    "            bools = fn(metadata)\n",
    "            return metadata[bools], metadata[~bools]\n",
    "        return f"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "fpd_MG = FaithPDAnalysis(table_MG_path, wol_tree_path, metadata)\n",
    "fpd_MG.merge_metadata()\n",
    "\n",
    "TARGET = 'BL_AGE'\n",
    "\n",
    "age_filter = lambda x: (x[TARGET] >= 60) | (x[TARGET] <= 35)\n",
    "dataset_MG = fpd_MG.filter_metadata(age_filter)\n",
    "\n",
    "def age_splitter(x): return x.loc[(x[TARGET] >= 60)], x.loc[(x[TARGET] <= 35)]\n",
    "\n",
    "partial_splitter = fpd_MG.make_splitter(age_splitter)\n",
    "\n",
    "datasets = {\n",
    "    'MG': dataset_MG\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "age_cat = np.where(datasets['MG']['BL_AGE'] < 50, 'Young (<=35)', 'Old (>60)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "datasets['MG']['age_cat'] = age_cat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAnUAAAIaCAYAAABGVbvbAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAABadUlEQVR4nO3dd3wVVf7/8fclhWBIqEGKNMEkQASVDroiRWqQAGKBIAqIRoO60hTXtYCgwLILGJXFgpRVJAGJ0qUJAkmoRghNSggllEAIwSQk8/uDb+6Pawo3pNyb8fV8PHys98yZuZ/RZXzfM2fOWAzDMAQAAIBSrYyjCwAAAEDhEeoAAABMgFAHAABgAoQ6AAAAEyDUAQAAmICrowtwpKysLF29elVubm6yWCyOLgcAACBPhmEoIyNDnp6eKlMm57jcXzrUXb16VQcPHnR0GQAAAHbz9fWVl5dXjva/dKhzc3OTdOMfjru7u4OrAQAAyFt6eroOHjxozS9/9pcOddm3XN3d3VW2bFkHVwMAAHBreU0Z40EJAAAAEyDUAQAAmAChDgAAwAQIdQAAACZAqAMAADABQh0AAIAJEOoAAABMgFAHAABgAoQ6AAAAE3DKN0pkZmZqwIABio2N1QMPPKD//e9/1m0nT55Up06d8tx3+fLlatCgQUmUCQAA4DScMtR99dVX+v333/Pt4+/vr86dO+dor1SpUnGVBQAA4LScLtSdOHFCM2fO1CuvvKLJkyfn2a9Ro0YKDQ0twcoAAACcl9PNqXv77bdVv359DR482NGlAAAAlBpOFeq+++47RUVF6f3335eLi0u+fRMTE7VgwQJ99tlnWrZsmS5evFhCVQIAUHQSExM1aNAgnTt3ztGloJRzmtuv586d00cffaRBgwYpICDglv23bNmiLVu2WD97eHjotdde05AhQ4qxSgAAilZYWJhiYmIUFhamf/7zn44uB6WY04S69957T56ennrllVfy7VeuXDmFhoaqc+fOql27tlJTU7V9+3ZNmTJFkyZNUtWqVdWrV68CfXdsbGxhSgcA4LYkJSVp8eLFMgxDixcv1oMPPqiKFSs6uiyUUk4R6tasWaPVq1frk08+kaenZ759q1Spopdfftn62dPTU7169VLDhg3Vv39/hYWFFTjUBQQEqGzZsrdVOwAAt+udd96x/r1hGNq8eTOjdchTWlpavgNRDp9Tl5qaqnfffVddu3ZVx44db/s4/v7+CggI0JEjR5SSklKEFQIAUDwiIyOVkZEhScrIyNCyZcscXBFKM4eHuosXL+rcuXNatWqV/Pz8bP6SpJ07d8rPz0/BwcG3PFb2kHVqampxlgwAQJEIDAyUm5ubJMnNzU29e/d2cEUozRx++9XT01P9+/fPddvixYtVtWpVdejQQfXr18/3ONevX1dcXJzKlSunypUrF0epAAAUqZCQEEVEREiSypQpo5CQEAdXhNLM4aGuUqVKmjhxYq7bFi9erDp16thsj42Nla+vr9zd3a1tmZmZmjZtmk6fPq3evXvL1dXhpwUAwC1Vq1ZNffv21TfffKN+/frJx8fH0SWhFCt16ScsLEy7d+9Wy5YtVbNmTV27dk3R0dE6fPiw6tatq7Fjxzq6RAAA7BYSEqLDhw8zSodCK3Whrlu3bsrIyNCePXu0fv16WSwW1alTRy+88IKGDRsmLy8vR5cIAIDdqlWrpvnz5zu6DJiAU4e6AwcO5Gjr3bs3E0kBAAD+xOFPvwIAAKDwCHUAAAAmQKgDAAAwAUIdAACACRDqAAAATIBQBwAAYAKEOgAAABMg1AEAAJgAoQ4AAMAECHUAAAAmQKgDAAAwAUIdAAAOlJiYqEGDBuncuXOOLgWlHKEOAAAHCgsLU0xMjMLCwhxdCko5Qh0AAA6SmJioiIgIGYah8PBwRutQKIQ6AAAcJCwsTFlZWZKkrKwsRutQKIQ6AAAcJDIyUhkZGZKkjIwMLVu2zMEVoTQj1AEA4CCBgYGyWCySJIvFot69ezu4IpRmhDoAABxkwIABMgxDkmQYhp544gkHV4TSjFAHAICDzJ071+bzV1995ZhCYAqEOgAAHOSHH36w+RwZGemgSmAGhDoAABwk+9YrUBQIdQAAOEjt2rXz/QwUBKEOAAAHSUxMtPl89uxZB1UCMyDUAQDgII899ph1SZMyZcqoT58+ji0IpRqhDgAABwkJCZGbm5skyc3NTSEhIQ6uCKWZq6MLAAAgP0uXLlV4eLijyyg2ZcuWVXp6uipUqKC///3vji6nWPTr149RyBLASB0AAA6UlZUlFxcX1axZ09GloJRjpA4A4NT69Olj6lGe4OBgSdK8efMcXAlKO0bqAAAATIBQBwAAYALcfkWRMftk5vPnz0uSqlat6uBKig+TmQGg9CLUAXY6d+6cJHOHOgBA6UWoQ5FhMjMAAI7DnDoAAAATINQBAACYAKEOAADABAh1AAAAJkCoAwAAMAFCHQAAgAkQ6gAAAEyAUAcAAGAChDoAAAATINQBAACYAKEOAADABAh1AAAAJkCoAwAAMAFCHQAAgAkQ6gAAAEyAUAcAAGAChDoAAAATcMpQl5mZqX79+snPz09PPfVUrn1WrFih/v37q1mzZmrdurVeffVVnTx5soQrBQAAcA5OGeq++uor/f7773luX7hwoV599VUlJibqySefVKdOnbRu3ToNGDBACQkJJVgpAACAc3C6UHfixAnNnDlTI0eOzHX7xYsXNWXKFPn4+GjJkiV644039MEHH+izzz7TxYsXNXXq1BKuGAAAwPGcLtS9/fbbql+/vgYPHpzr9hUrVig1NVWDBw9WlSpVrO1t27ZVq1attGbNGiUnJ5dUuQAAAE7BqULdd999p6ioKL3//vtycXHJtU9MTIwkqU2bNjm2tWvXThkZGdqzZ0+x1gkAAOBsnCbUnTt3Th999JEGDRqkgICAPPsdP35cklSnTp0c27LbTpw4UTxFAgAAOClXRxeQ7b333pOnp6deeeWVfPtdvXpVklS+fPkc2zw9PSVJKSkpBfru2NjYAvXHX9OVK1ckSTt27HBwJQDMhGsLiopThLo1a9Zo9erV+uSTT6zBLC+GYRT59wcEBKhs2bJFflyYi5eXlySpefPmDq4EgJlwbYG90tLS8h2Icvjt19TUVL377rvq2rWrOnbseMv+2SN0uY3G5TeKBwAAYGYOH6m7ePGizp07p1WrVsnPzy/H9p07d8rPz0+tWrXSvHnzVLduXf322286ceKEKlasaNM3ey5dbvPtAAAAzMzhoc7T01P9+/fPddvixYtVtWpVdejQQfXr15cktWjRQsuXL9e2bdvUtGlTm/6//PKL3Nzc1KxZs2KvGwAAwJk4PNRVqlRJEydOzHXb4sWLVadOHZvt3bt319SpU/X111+rX79+1rXqtm7dqqioKHXr1k3e3t4lUjsAAICzcHioK6jKlStr1KhReu+99xQUFKQePXooOTlZP/74oypVqqTRo0c7ukQAAIASV+pCnSQNHDhQlSpV0ueff66FCxfKw8NDHTp00KhRo1SrVi1HlwcAAFDinDrUHThwIM9tPXr0UI8ePUqwGgAAAOfl8CVNAAAAUHiEOgAAABMg1AEAAJgAoQ4AAMAECHUAAAAmQKgDAAAwAUIdAACACRDqAAAATIBQBwAAYAKEOgAAABMg1AEAAJgAoQ4AAMAECHUAAAAmQKgDAAAwAUIdAACACRDqAAAATIBQBwAAYAKEOgAAABMg1AEAAJgAoQ4AAMAECHUAAAAmQKgDAAAwAUIdAACACRDqAAAATIBQBwAAYAKEOgAAABMg1AEAAJgAoQ4AAMAECHUAAAAmQKgDAAAwAUIdAACACRDqAAAATIBQBwAAYAKEOgAAABMg1AEAAJgAoQ4AAMAECHUAAAAmQKgDAAAwAUIdAACACRDqAAAATIBQBwAAYAKEOgAAABMg1AEAAJgAoQ4AAMAECHUAAAAmQKgDAAAwAUIdAACACRDqAAAATIBQBwAAYAKuji5Aki5duqSZM2dqz549SkhI0JUrV1StWjUFBARoxIgRatKkibXvyZMn1alTpzyPtXz5cjVo0KAkygYAAHAaThHqLly4oIiICD3wwAMKCAiQl5eXTp06pZ9++klr167VjBkz1LlzZ5t9/P39c7RJUqVKlUqqbAAAAKfhFKGubt26io6OlqurbTlHjhxRUFCQpk6dmiPANWrUSKGhoSVZJgAAgNNyijl1rq6uOQKdJDVo0EANGjTQyZMnHVAVAABA6eEUI3V5OXnypI4dO5brHLnExEQtWLBAKSkpqlGjhh588EFVrlzZAVUCAAA4nlOFurNnz2rRokXKzMzUmTNntHbtWlksFo0fPz5H3y1btmjLli3Wzx4eHnrttdc0ZMiQEqwYAADAOThVqEtMTNSsWbOsnytXrqypU6eqVatW1rZy5copNDRUnTt3Vu3atZWamqrt27drypQpmjRpkqpWrapevXoV6HtjY2OL7BxgXleuXJEk7dixw8GVADATri0oKhbDMAxHF/Fn6enpio+P1xdffKElS5boH//4h5566ql894mLi1P//v1Vp04dLV++3K7vSUtLU2xsrAICAlS2bNmiKB0mFhwcLEmaN2+egysBYCZcW2CvW+UWp3hQ4s/c3d3VoEEDTZw4Ue3bt9cHH3ygs2fP5ruPv7+/AgICdOTIEaWkpJRQpQAAAM7BKUPdzdq1a6f09HTt3bv3ln0rVqwoSUpNTS3mqgAAAJyL04e6xMRESZKLi0u+/a5fv664uDiVK1eOp2ABAMBfjlOEuri4uFxvmR44cECLFi1SuXLl1Lx5c0k3HmpIT0+36ZeZmalp06bp9OnT6tKlS65r3gEAAJiZU6Sf8PBwhYeHq02bNqpVq5ZcXFx07Ngxbdq0SYZhaMKECapQoYIkKSwsTLt371bLli1Vs2ZNXbt2TdHR0Tp8+LDq1q2rsWPHOvhsAAAASp5ThLquXbsqOTlZu3fv1tatW5WRkSEfHx91795dzzzzjJo2bWrt261bN2VkZGjPnj1av369LBaL6tSpoxdeeEHDhg2Tl5eXA88EAADAMZwi1LVo0UItWrSwq2/v3r3Vu3fvYq4IAACgdHGKOXUAAAAoHEIdAACACRDqAAAATIBQBwAAYAKEOgAAABMg1AEAAJgAoQ4AAMAECHUAAAAm4BSLDwMAbt/EiRMVFxfn6DJwm/bv3y9JCg4OdnAluF3+/v4aP368o8sg1AFAaRcXF6ed23bK+7q3o0vBbcgqkyVJOrz5sIMrwe1Idk12dAlWhDoAMAHv695qc6mNo8sA/nK2Vdzm6BKsmFMHAABgAoQ6AAAAEyDUAQAAmAChDgAAwAQIdQAAACZAqAMAADABQh0AAIAJEOoAAABMgFAHAABgAoQ6AAAAE7DrNWHp6elauXKloqKidOTIESUnJ8tiscjb21sNGjRQq1at1LVrV7m7uxd3vQAAAMjFLUPdhg0b9NZbb+nChQsyDCPH9p07d2rx4sX66KOPNGHCBD388MPFUigAAADylm+oW7t2rUaOHKmsrCzVrVtXjz76qPz9/VWhQgUZhqFLly4pLi5Oq1evVnx8vF588UXNmjVLHTt2LKn6AQAAoHxCXWpqqsaPHy9JGjdunIYMGZJrv8DAQI0ePVpffPGFpkyZorfeekvr1q2Th4dHsRRcmk2cOFFxcXGOLgO3af/+/ZKk4OBgB1eC2+Xv72+9rgGA2eQZ6iIjI3X58mUNHz48z0B3s+eee05JSUmaM2eOIiMj9fjjjxdlnaYQFxenbTu36br3dUeXgttQJuvGc0WbD292cCW4Ha7Jdk0hBoBSK8+r3JYtW+Tm5qbnn3/e7oMNHz5cX375pX7++WdCXR6ue1/XpTaXHF0G8JdTcVtFR5cAAMUqzyVN4uLi1LhxY3l5edl9MG9vbzVp0oRbjAAAACUsz1CXlJSk6tWrF/iA1atXV1JSUqGKAgAAQMHkGepSUlJ0xx13FPiAd9xxh65evVqoogAAAFAweYa63Naks1dh9gUAAEDB5fs42IkTJ7R06dICHfD48eOFqQcAAAC3Id9Qt3PnTu3cubNABzQMQxaLpVBFAQAAoGDyDHVBQUElWQcAAAAKIc9QN2nSpJKsAwAAAIWQ54MSAAAAKD1u+d6cgwcPav369Tp16pTc3d3l6+urbt26FWhRYgAAABSvfEPdlClT9MUXX0iyfQBi2rRpmjVrllq0aFH8FQIAAOCW8gx1K1as0Oeffy5Jateunfz8/HT16lX98ssvOnnypEaOHKm1a9fe1gLFAAAAKFp5hrpvv/1WFotF06ZNU48ePazt6enpeuWVV7RhwwatXr1affr0KYk6AQAAkI88H5TYv3+/AgICbAKdJLm7u+vvf/+7DMNQXFxcsRcIAACAW8sz1CUnJ6t+/fq5brv77rslSVeuXCmeqgAAAFAg+b771cXFJddt2e1ZWVnFUxUAAAAKhHXqAAAATCDfJU2WLl2qpUuX5rrNYrHkud1isWjfvn1FUR8AAADskG+oMwzjtg56u/sBAAru/PnzSnZN1raK2xxdCvCXk+yarPPnzzu6DEn5hDqebAUAACg9bvmaMACAc6tataouxV1Sm0ttHF0K8JezreI2Va1a1dFlSOJBCQAAAFMg1AEAAJiAU9x+vXTpkmbOnKk9e/YoISFBV65cUbVq1RQQEKARI0aoSZMmOfbJfjftoUOH5OHhobZt22rUqFG66667HHAGAAAAjuUUI3UXLlxQRESEKlSooK5du+rZZ5/V/fffr02bNunxxx/X2rVrbfovXLhQr776qhITE/Xkk0+qU6dOWrdunQYMGKCEhAQHnQUAAIDjOMVIXd26dRUdHS1XV9tyjhw5oqCgIE2dOlWdO3eWJF28eFFTpkyRj4+PlixZoipVqkiSAgMD9eyzz2rq1KmaPn16iZ8DAACAIznFSJ2rq2uOQCdJDRo0UIMGDXTy5Elr24oVK5SamqrBgwdbA50ktW3bVq1atdKaNWuUnJxcInUDAAA4C6cIdXk5efKkjh07pgYNGljbYmJiJElt2uR8dL9du3bKyMjQnj17SqxGAAAAZ1Akt183bdqkQ4cOqUaNGurSpYvc3Nxu6zhnz57VokWLlJmZqTNnzmjt2rWyWCwaP368tc/x48clSXXq1Mmxf3bbiRMnbuv7AQAASiu7Q938+fM1Z84cTZs2Tc2bN7e2v/LKK1q9erX1c5MmTTR//nx5eHgUuJjExETNmjXL+rly5cqaOnWqWrVqZW27evWqJKl8+fI59vf09JQkpaSkFOh7Y2NjC1zr7bhy5UqJfA+A3F25ckU7duxwdBlFjmsL4FjOcm2xO9StWbNG169f1wMPPGBt27Rpk1atWqXatWurT58+2rZtm2JiYvTNN99oyJAhBS7m3nvv1YEDB5Senq74+Hh98cUXGj58uP7xj3/oqaeeklQ875UNCAhQ2bJli/y4f+bl5SWdLfavAZAHLy8vmx+lZuHl5aWzXFwAhympa0taWlq+A1F2z6k7duyYGjZsKIvFYm1bvny5LBaLZsyYoZdeekmff/65qlSposjIyEIV7e7urgYNGmjixIlq3769PvjgA509e+OClT1Cl9toXH6jeAAAAGZmd6hLSkrK8W6zqKgo1a5dW40aNZJ0I4zdf//9Nk+rFla7du2Unp6uvXv3Srqx/ImU+7y57Lbc5tsBAACYmd2hztvbW0lJSdbP8fHxOnXqlM18N+lGsEtPTy+yAhMTEyVJLi4ukqQWLVpIkrZt25aj7y+//CI3Nzc1a9asyL4fAACgNLA71N19993asWOHzpw5I0latGiRLBaLHn74YZt+CQkJ8vHxKVARcXFxud5OPXDggBYtWqRy5cpZ71V3795dd9xxh77++mtduHDB2nfr1q2KiopS586d5e3tXaDvBwAAKO3sflBi8ODBevnllxUYGKjatWtr3759qlWrlk2oS0lJ0W+//aYOHToUqIjw8HCFh4erTZs2qlWrllxcXHTs2DFt2rRJhmFowoQJqlChgqQbT8SOGjVK7733noKCgtSjRw8lJyfrxx9/VKVKlTR69OgCfTcAAIAZ2B3qOnfurDfeeENfffWVfv/9dz3wwAN6++235e7ubu2zdOlSZWRkqG3btgUqomvXrkpOTtbu3bu1detWZWRkyMfHR927d9czzzyjpk2b2vQfOHCgKlWqpM8//1wLFy6Uh4eHOnTooFGjRqlWrVoF+u6SdP78ebkmu6ritoqOLgX4y3FNdtX58+cdXQYAFJsCLT78zDPP6Jlnnslze//+/fXYY4/pjjvuKFARLVq0sM6Vs1ePHj3Uo0ePAu0DAABgVkXyRolsHh4et7Xo8F9F1apVFXcpTpfaXHJ0KcBfTsVtFXM8wQ8AZuLU734FAACAffIcqevUqVOONovForVr1xZrQQAAACi4PENdbq/jKo5XdAEAAKDw8gx169atK8k6AAAAUAjMqQMAADABQh0AAIAJFHhJk9TUVMXGxurcuXPKyMjIs1+fPn0KUxcAAAAKwO5QZxiGpk+frnnz5umPP/7It5/FYiHUAQAAlCC7Q92nn36q2bNny9XVVY888ojq1q0rT0/P4qwNAAAAdrI71C1evFjlypXT//73P/n7+xdnTQAAACggux+USExMVKtWrQh0AAAATsjuUFejRg1dv369OGsBAADAbbI71AUFBWnPnj06e/ZscdYDAACA22B3qHv++efVvn17DR8+XNHR0bwyDAAAwInk+aBEp06dcm0/deqUBg8eLFdXV/n4+MhiseToY7FYtHbt2qKrEgAAAPnKM9QlJCTku2NGRoZOnTpV5AUBAACg4PIMdXFxcSVZBwAAAAqBd78CAACYgN2LD7/xxhtq3ry5+vfvn2+/iIgIRUdHa9KkSYUuDgBgn2TXZG2ruM3RZeA2pJVJkySVzSrr4EpwO5Jdkx1dgpXdoW7JkiWSdMtQt3PnTi1dupRQBwAlhEXhS7f9+/dLkho2aujgSnC7nOXPoN2hzl7p6elycXEp6sMCAPIwfvx4R5eAQggODpYkzZs3z8GVoLQr0jl1KSkp2rVrl6pUqVKUhwUAAMAt5DtS9+e16latWqWoqKhc+2ZmZur8+fPKzMzUgAEDiq5CAAAA3FK+oe7mteosFotSU1OVmpqa+4H+bzHiDh066PXXXy/aKgEAAJCvfEPdzWvV+fv7KygoiAcgAAAAnJDdD0pMmjRJderUKc5aAAAAcJvsDnVBQUHFWQcAAAAKIc9QFx0dLUlq2rSpypYta/1sr5YtWxauMgAAANgtz1AXHBwsi8Wi5cuXq379+tbP9speTBEAAADFL89Q16dPH1ksFnl5edl8BgAAgPPJM9RNnjw5388AAABwHkX6RgkAAAA4BqEOAADABOxe0iRbQkKC1q9fr+PHj+vq1asyDCNHH4vFog8++KBICgQAAMCtFSjU/ec//9Hs2bOVlZVlbcsOddkPURiGQagDAAAoYXaHusjISH3yySe66667NGLECK1cuVK//PKLPv/8c8XHx2vlypXatm2bhgwZoo4dOxZnzQAAAPgTu0Pd//73P7m7u2v+/PmqXr26du7cKUlq3769JOnJJ5/U/PnzNWnSJHXq1Kl4qgUAAECu7H5Q4uDBg3rggQdUvXp1m/abb8UOGjRIDRo00CeffFJ0FQIAAOCW7A51169fV5UqVayfPTw8JEnJyck2/e655x7FxsYWUXkAAACwh92hrlq1akpMTLR+rlGjhqScrwOLj48votIAAABgL7vn1DVp0kRRUVHKyspSmTJl1K5dO/3rX//SRx99pGnTpunOO+/UokWLtHfvXrVq1ao4ay7VXJNdVXFbRUeXgdtQJu3Gb6Csslm36Aln5Jpc4BWcAKBUsfsq17FjR61YsUIbNmxQx44dFRAQoG7dumnlypXq2bOntZ+Li4tee+21Yim2tPP393d0CSiE7FHpRg0bObgS3C7+DAIwM7tDXWBgoLp27SoXFxdr25QpU9SoUSP99NNPunz5surXr6/nnntO999/f7EUW9qNHz/e0SWgEIKDgyVJ8+bNc3AlAADklGeomzRpkpo2bWozCufu7m7Tx83NTSNGjNCIESOKr0IAAADcUp4PSsydO1ebN2/OdVunTp00ZcqUYisKAAAABWP30683S0hI0MWLF4u6FgAAANym2wp1AAAAcC6EOgAAABMg1AEAAJiAU6zGeebMGa1YsULr16/X0aNHlZSUpCpVqqhdu3YKCQlR7dq1rX1PnjypTp065Xms5cuXq0GDBiVRNgAAgNPIN9QtXbpUS5cuzdFusVjy3Ja9fd++fXYXMX/+fP33v/9VgwYN9Mgjj8jLy0uxsbGKiIjQ2rVrtXDhQt1zzz02+/j7+6tz5845jlWpUiW7vxcAAMAs8g11hmHc1kELul+zZs30zTff5Fi0+KuvvtKkSZP04Ycfas6cOTbbGjVqpNDQ0NuqDwAAwGzyDHVxcXElVkSXLl1ybR88eLD+/e9/a8eOHSVWCwAAQGnkFHPq8uPi4iKLxZKjPTExUQsWLFBKSopq1KihBx98UJUrV3ZAhQAAAI7n1KHup59+UkpKSq4jeVu2bNGWLVusnz08PPTaa69pyJAhJVghAACAc3DaUHf+/Hm9//77cnd318iRI63t5cqVU2hoqDp37qzatWsrNTVV27dv15QpUzRp0iRVrVpVvXr1KtB3xcbGFnX5MKErV65IEtMBABQpri0oKk4Z6lJTU/XSSy/p7NmzmjBhgnx9fa3bqlSpopdfftn62dPTU7169VLDhg3Vv39/hYWFFTjUBQQEqGzZskVWP8zJy8tLktS8eXMHVwLATLi2wF5paWn5DkQ53eLDaWlpCgkJ0e7duzVmzBg9/vjjdu3n7++vgIAAHTlyRCkpKcVcJQAAgHNxqlCXnp6u0NBQbd26VaGhoRo6dGiB9q9YsaKkGyN9AAAAfyVOE+quX7+u1157TRs3btSwYcNsbrHau39cXJzKlSvHU7AAAOAvxylCXVZWlkaPHq21a9cqODhYo0ePzrNvbGys0tPTbdoyMzM1bdo0nT59Wl26dJGrq1NOFQQAACg2TpF+Zs2apeXLl6tSpUry9vbWzJkzc/TJfntEWFiYdu/erZYtW6pmzZq6du2aoqOjdfjwYdWtW1djx44t6fIBAAAczilC3alTpyRJSUlJ+vjjj3Ptkx3qunXrpoyMDO3Zs0fr16+XxWJRnTp19MILL2jYsGHWp4gAAAD+Spwi1E2ePFmTJ0+2q2/v3r3Vu3fvYq4IAACgdHGKOXUAAAAoHEIdAACACRDqAAAATIBQBwAAYAKEOgAAABMg1AEAAJgAoQ4AAMAECHUAAAAmQKgDAAAwAUIdAACACRDqAAAATIBQBwAAYAKEOgAAABMg1AEAAJgAoQ4AAMAECHUAAAAmQKgDAAAwAUIdAACACRDqAAAATIBQBwAAYAKEOgAAABMg1AEAAJgAoQ4AAMAECHUAAAAmQKgDAAAwAUIdAACACRDqAAAATIBQBwAAYAKEOgAAABMg1AEAAJgAoQ4AAMAECHUAAAAmQKgDAAAwAUIdAACACRDqAAAATIBQBwAAYAKEOgAAABMg1AEAAJgAoQ4AAMAECHUAAAAmQKgDAAAwAUIdAACACRDqAAAATIBQBwAAYAKEOgAAABMg1AEAAJgAoQ4AAMAECHUAAAAmQKgDAAAwAVdHFyBJZ86c0YoVK7R+/XodPXpUSUlJqlKlitq1a6eQkBDVrl07xz4rVqzQ559/rkOHDsnDw0Nt27bVqFGjdNdddzngDAAAABzLKUbq5s+fr8mTJ+v8+fN65JFH9Mwzz6hevXqKiIhQ3759dejQIZv+Cxcu1KuvvqrExEQ9+eST6tSpk9atW6cBAwYoISHBQWcBAADgOE4xUtesWTN98803uv/++23av/rqK02aNEkffvih5syZI0m6ePGipkyZIh8fHy1ZskRVqlSRJAUGBurZZ5/V1KlTNX369BI/BwAAAEdyipG6Ll265Ah0kjR48GCVK1dOO3bssLatWLFCqampGjx4sDXQSVLbtm3VqlUrrVmzRsnJySVSNwAAgLNwilCXHxcXF7m4uFg/x8TESJLatGmTo2+7du2UkZGhPXv2lFh9AAAAzsCpQ91PP/2klJQUmwB3/PhxSVKdOnVy9M9uO3HiRMkUCAAA4CScYk5dbs6fP6/3339f7u7uGjlypLX96tWrkqTy5cvn2MfT01OSlJKSUqDvio2NLUSl+Ku4cuWKJNlMBwCAwuLagqLilKEuNTVVL730ks6ePasJEybI19fXus0wjCL/voCAAJUtW7bIjwtz8fLykiQ1b97cwZUAMBOuLbBXWlpavgNRTnf7NS0tTSEhIdq9e7fGjBmjxx9/3GZ79ghdbqNx+Y3iAQAAmJlThbr09HSFhoZq69atCg0N1dChQ3P0qVu3rqTc581lt+U23w4AAMDMnCbUXb9+Xa+99po2btyoYcOG6eWXX861X4sWLSRJ27Zty7Htl19+kZubm5o1a1astQIAADgbpwh1WVlZGj16tNauXavg4GCNHj06z77du3fXHXfcoa+//loXLlywtm/dulVRUVHq3LmzvL29S6JsAAAAp+EUD0rMmjVLy5cvV6VKleTt7a2ZM2fm6BMaGipJqly5skaNGqX33ntPQUFB6tGjh5KTk/Xjjz+qUqVK+QZCAAAAs3KKUHfq1ClJUlJSkj7++ONc+2SHOkkaOHCgKlWqpM8//1wLFy6Uh4eHOnTooFGjRqlWrVolUjMAAIAzcYpQN3nyZE2ePLlA+/To0UM9evQopooAAABKF6eYUwcAAIDCcYqROgAA8rJ06VKFh4c7uoxis3//fklScHCwgyspPv369VOfPn0cXYbpEeoAAHAgHx8fR5cAkyDUAQCcWp8+fRjlAezAnDoAABwoMTFRgwYN0rlz5xxdCko5Qh0AAA40bdo0RUdHa9q0aY4uBaUcoQ4AAAdJTExUZGSkJGnZsmWM1qFQCHUAADjItGnTlJmZKUnKzMxktA6FQqgDAMBBfvjhh3w/AwVBqAMAwEkYhuHoElCKEeoAAHCQXr162XwODAx0UCUwA0IdAAAO8vrrr+f7GSgIQh0AAA5UpkwZm/8Fbhf/DwIAwEHCwsLk4uIiSXJxcVFYWJiDK0JpRqgDAMBBIiMjlZGRIUnKyMjQsmXLHFwRSjNCHQAADhIYGCg3NzdJkpubm3r37u3gilCaEeoAAHCQkJAQmzl1ISEhDq4IpRmhDgAAB6lWrZr69u0ri8Wifv36ycfHx9EloRRzdXQBAAD8lYWEhOjw4cOM0qHQCHUAADhQtWrVNH/+fEeXARPg9isAAIAJEOoAAABMgFAHAIAD7du3T82bN1dcXJyjS0EpR6gDAMCBRo8erZSUFI0aNcrRpaCUI9QBAOAg+/bt0+HDhyVJhw4dYrQOhUKoAwDAQUaPHm3zmdE6FAahDgAAB8kepct26NAhB1UCMyDUAQDgIPXq1bP5XL9+fccUAlMg1AEA4CD+/v75fgYKglAHAICDbNy40ebzhg0bHFMITIFQBwCAg7i5udl8dnd3d1AlMANCHQAADpKcnGzz+fLlyw6qBGZAqAMAwEEaNmxo8/mee+5xUCUwA0IdAAAOMmXKFJvPU6dOdVAlMANXRxcA81i6dKnCw8MdXUax2b9/vyQpODjYwZUUn379+qlPnz6OLgP4y2jcuLEaNmyow4cP65577uHpVxQKI3WAnSpVqqSMjAylp6c7uhQAJvLGG2+oTJkyGj9+vKNLQSnHSB2KTJ8+fUw9yvPOO+/om2++UePGjfXPf/7T0eUAMInvvvtOWVlZ+vbbb9W2bVtHl4NSjJE6wA6JiYmKiIiQYRgKDw/XuXPnHF0SABNITEzUqlWrJEkrV67k2oJCIdQBdggLC1NWVpYkKSsrS2FhYQ6uCIAZTJw4UYZhSJIMw9DEiRMdXBFKM0IdYIfIyEhlZGRIkjIyMrRs2TIHVwTADLJH6fL6DBQEoQ6wQ2BgoHXldzc3N/Xu3dvBFQEwg+xRumzZdwSA20GoA+wQEhKiMmVu/HEpU6aMQkJCHFwRADOoV6+ezef69es7phCYAqEOsEO1atXUt29fWSwW9evXTz4+Po4uCYAJTJ8+3ebzv//9b8cUAlMg1AF2CgkJUYsWLRilA1BkGjdubB2tq1+/PosPo1AIdYCdqlWrpvnz5zNKB6BITZ8+XeXLl2eUDoXG4sMAADhQ48aNtWPHDkeXARNgpA4AAMAECHUAAAAmQKgDAAAwAUIdAACACTjFgxJff/21fv31V/366686duyYDMPQb7/9JlfXnOWdPHlSnTp1yvNYy5cvV4MGDYqzXAAAAKfjFKEu+wXGtWrVkpeXl5KTk2+5j7+/vzp37pyjvVKlSkVeHwAAxWXfvn0KDg7WggULWKcOheIUoW727Nm69957VblyZQUHBysqKuqW+zRq1EihoaElUB0AAMVn9OjRSklJ0ahRo/TDDz84uhyUYk4xp+7hhx9W5cqVHV0GAAAlat++fTp8+LAk6dChQ4qLi3NwRSjNnGKk7nYkJiZqwYIFSklJUY0aNfTggw8SDAEApcro0aNtPjNah8IotaFuy5Yt2rJli/Wzh4eHXnvtNQ0ZMsRxRQEAUADZo3TZDh065KBKYAalLtSVK1dOoaGh6ty5s2rXrq3U1FRt375dU6ZM0aRJk1S1alX16tWrQMeMjY0tpmoBAMhbrVq1lJCQYPOZV4bhdlkMwzAcXcTNsh+UyGtJk7zExcWpf//+qlOnjpYvX27XPmlpaYqNjVVAQIDKli17uyUDAHBb9u3bp6CgIOvn77//nidgkadb5RaneFCiKPj7+ysgIEBHjhxRSkqKo8sBAOCWGjdurHr16kmS6tevT6BDoZgm1ElSxYoVJUmpqamOLQQAADtlBzkCHQrLNKHu+vXriouLU7ly5XgKFgBQKiQmJmr9+vWSpHXr1uncuXMOrgilWakLdbGxsUpPT7dpy8zM1LRp03T69Gl16dKlQHPxAABwlLCwMGVlZUmSsrKyFBYW5uCKUJo5RfqZPXu2fv/9d0my/u/48eNlsVgkSWPGjLGOvoWFhWn37t1q2bKlatasqWvXrik6OlqHDx9W3bp1NXbsWMecBAAABRQZGamMjAxJUkZGhpYtW6Z//vOfDq4KpZVThLqff/45x6vBli5dav37l19+2RrqunXrpoyMDO3Zs0fr16+XxWJRnTp19MILL2jYsGHy8vIqydIBALhtgYGBWrx4sTIyMuTm5qbevXs7uiSUYk63pElJYkkTAIAjJSYmqnPnzkpLS1PZsmX1008/ycfHx9FlwUn9ZZY0AQCgtKlWrZr69u0ri8Wifv36EehQKE5x+xUAgL+qkJAQHT58WCEhIY4uBaUcI3WAnRITEzVo0CCWHABQpKpVq6b58+czSodCI9QBdgoLC1NMTAxLDgAAnBKhDrBDYmKiIiIiZBiGwsPDGa0DADgdQh1gBxYIBQA4O0IdYIfcFggFAMCZEOoAOwQGBsrNzU2SWCAUAOCUCHWAHUJCQlSmzI0/LmXKlGHpAQCA0yHUAXZggVAAgLMj1AF2CgkJUYsWLRilA1CkWAMTRYVQB9iJBUIBFAfWwERRIdQBAOAgrIGJokSoAwDAQVgDE0WJUAcAgIOwBiaKEqEOAAAHYQ1MFCVCHQAADsIamChKhDoAAByENTBRlFwdXQAAAH9lISEhOnz4MKN0KDRCHQAADpS9BiZQWNx+BQAAMAFCHQAAgAkQ6gAAAEyAUAcAAGAChDoAAAATINQBAACYAKEOAADABAh1AAAAJkCoAwAAMAFCHQAAgAkQ6gAAAEyAUAcAAGAChDoAAAATcHV0AY5kGIYkKT093cGVAAAA5C87r2Tnlz/7S4e6jIwMSdLBgwcdXAkAAIB9MjIy5OHhkaPdYuQV9/4CsrKydPXqVbm5uclisTi6HAAAgDwZhqGMjAx5enqqTJmcM+j+0qEOAADALHhQAgAAwAQIdQAAACZAqAMAADABQh0AAIAJEOoAAABMgFAHAABgAoQ6AAAAEyDUAXY6f/687r//fn311VeOLqVEXLx4Uffdd5/mzp3r6FIAAHYg1MFuzz77rPz9/RUdHZ3r9vT0dPXs2VPNmjXTsWPHSra4EvDxxx/Lw8NDTzzxhKNLsVqxYoVeeuklderUSffff79atmypoKAgffnll/rjjz9y9B83bpz8/Pxy/eu1116z6Vu5cmUNGDBAn376qa5evVpSpwQUq+w/A/YKDg5Wx44dC/Qd69atk5+fn/bt21fQ8kqlDRs2yN/fX/v373d0KX95f+l3v6Jg3n//fQUGBmr8+PFatmxZjvfOffzxxzp8+LDGjh2revXqOabIYnLu3Dl99913GjZsmMqVK+focqzWrFmj33//XQ888ICqVaumtLQ0xcTEaPLkyVq5cqXmz58vNze3HPsNHjxY3t7eNm333HNPjn7PPPOMvv76a33zzTcaOnRosZ0HcDsyMzO1dOlSff/99zpw4ICuXr2qqlWrqmXLlhoyZIiaNGlS4jVlZWVp+vTp6tChgxo3blzi338rq1at0oIFC7R//35lZGSoRo0aatWqld59990cfffv369//etf2rVrlzIzMxUQEKCXXnpJbdq0senXoUMHNWrUSNOmTdOcOXNK6lSQC0Id7HbXXXfp9ddf1/vvv6/p06frjTfesG6Li4vTnDlz1KxZMw0ZMsRxRRaTJUuWKCMjQ4GBgbd9jMzMTO3bt0/33ntvkdU1adIklS1bNkf7G2+8oYiICK1evVo9e/bMsf2ZZ57RXXfddcvj16pVS/fdd58WL15MqINTSU5OVkhIiKKjo1WtWjV17dpVFSpU0NGjR7Vy5UpFRkZqzJgxeu6550q0rjVr1ujgwYN68803S/R77fH+++9r/vz5atiwofr06SM3NzedPHlSq1atyhHqYmNjNXDgQFksFvXs2VOenp5avny5nn32WX3yySfq0KGDTf9nnnlGY8eO1a+//lqk1zgUkAEUQFZWlvH0008b/v7+xs6dOw3DMIyMjAyjT58+RpMmTYxDhw4Zly9fNiZNmmQ88sgjRpMmTYwHH3zQeOutt4yzZ8/aHCs+Pt7w9fU1ZsyYkeN7tm3bZvj6+hrh4eE5+o8dO9Y4fPiwMXToUOO+++4zWrZsaYwbN864cuVKjuNcuXLFeOedd4y2bdsaTZs2NQYMGGD88ssvxowZMwxfX18jPj7ervPu2bOn0a1bt4L8o7I6evSoMW3aNOOhhx4yevfufVvHKKi1a9cavr6+RlhYmE372LFjC3TehmEYc+bMMXx9fY1ff/21qMsEbtvw4cMNX19f4+9//7tx7do1m22HDx82OnToYPj6+hrLly+32Zb9Z8BegwYNMh555BG7+w8dOtRo3769kZmZadOekpJiHDp0yO7jFLUlS5YYvr6+xuTJk3PUlpGRkaN///79DX9/f2P79u3WtsTERKNt27bGww8/bKSlpdn0v3LlinHvvfcab7/9dvGcAOzCnDoUiMVi0cSJE+Xu7q433nhDaWlp+u9//6t9+/bppZdeUs2aNTVw4EB9+eWXql27tp599lnde++9WrRokZ544gmdO3eu0DWcPHlSTz31lAzD0JNPPqkGDRooIiJC48aNs+l3/fp1DRs2TAsXLlS9evU0ePBg1axZU88//7x2795t9/ddvHhRhw4dUrNmzeze5+rVqwoPD9fTTz+trl276rPPPlOFChU0aNAgu49RGJs3b5aU+y1VSdq4caNmz56tuXPnaseOHfkeK/u8t2/fXrRFArdp48aN2rhxo3x9fTV58uQcU0EaNGigf//735KkyZMnKyMj45bHPH78uF588UXdf//9atGihUJCQhQfH1+gui5cuKAtW7aoU6dOKlPG9j+vSUlJ6tmzpx5//HF98803unLlSoGOXRiGYWjWrFmqV6+eRo0alaM2V1fbm3aHDh3S3r171a5dO7Vq1cra7uPjo6efflqnT5/W1q1bbfYpX7682rZtq+XLlyszM7P4Tgb54vYrCqxevXoaOXKkPvroI40bN05r165V48aNNXz4cM2cOVMHDx7UsGHDNHr0aOs+8+bN04QJEzRlyhR99NFHhfr+6Ohovf322xo4cKCkGxesoUOHau3atTp9+rRq1KghSVq8eLF27dqlxx9/XBMmTLDuv3z58hwPBeRnz549kqRGjRrdsm9MTIzCw8O1cuVKpaamqmLFiho0aJCCgoIUEBCQo//Jkye1ZMkSu2tp1aqVWrdunaM9MjJSx44dU0pKivbs2aNdu3apc+fO6tSpU67Hee+992w+N23aVP/5z39Us2bNHH2zz3vXrl121wkUp6VLl0q6ccsvtzmj0o0fIy1atFBMTIy2bdumhx56KM/jnT59Wk899ZSSkpLUpUsX1a1bVzExMRo4cKAqVKhgd13R0dHKysrK9Qdg5cqVFRwcrB9++EH//Oc/NWnSJHXp0kX9+vVTmzZtZLFY7P6egtq3b5/i4+P13HPPKSMjQ6tWrVJ8fLyqVq2qv/3tb/Lx8bHpHxMTI0lq27ZtjmO1a9dOM2fOVExMjB5++GGbbU2bNtWGDRuKfJoJ7Eeow20ZMmSIVq1apeXLl8vV1VUffPCBXF1dFRkZqfLly+vFF1+06f/000/rq6++0ooVKzRhwgS5u7vf9nfXrVtXTz/9tPWzxWJR7969tWXLFu3fv98a6n744Qe5uLjo5Zdfttm/R48eCgsL06FDh+z6vjNnzkiSqlSpkuv2s2fPaunSpYqIiNCxY8fk6uqqhx56SEFBQXrkkUfyPdeEhATNmjXLrjok6eWXX8411P3www/asGGD9XNgYKDef//9HP+haNmypTp37qx7771XFSpU0PHjx/Xll19qyZIlev7557V06dIcv9o9PT3l4eFh/ecAOFr2D63c/izcrGXLloqJidGePXvyDXVTp07VhQsX9MEHH6hfv37W9jfffFPh4eGqVatWgerK7QGJO+64Q2+99ZbGjh2rDRs2KCIiQitWrFBkZKRq1aqloKAgBQUF5TnXdfv27YqKirKrDkk2x7r5KdxevXrZjEB6eHjonXfeUVBQkLXtxIkTkqQ6derkOG52W3afm2Wf965duwh1DkKow21xcXHRSy+9pOeff15dunRRo0aNdOXKFSUkJKhFixYqX758jv4PPPCAli1bpqNHjxZoSYE/8/X1zRFWqlWrJunG5OlsBw4cUPXq1VW9evUcx2jatKndoe7y5cuSlONp0WxPPfWUEhISVKtWLY0dO1a9e/dW1apV7Tp269atdeDAAbv65uezzz6TdONWcVRUlCZPnqwnn3xSX375pSpXrmztd/N/sCTJz89PkydPVnp6un788UetXr1aPXr0yHF8b29vXbp0qdB1AkXh/PnzkqQ777wz337Z27P75yY9PV2rV69WvXr11LdvX5ttI0eOtI4K2iP7h8/Nf+b+zM3NTV26dFGXLl104cIFLVu2TBEREZo1a5Y+/vhjtWnTRn379tWjjz5qc1s5KiqqQD8AW7VqZQ11SUlJkqS5c+fq/vvv18cff6xatWppy5Yt+sc//qHx48fL19fX+rRwSkqKJOW4jt/clt3nZtk/fPkB6DjMqcNty77gZP9v9lpmeY1oZQed3C4GBZHbhcbFxUXSjeUEsl29elWVKlXK9Rh51Zib7PNLT0/PdXvDhg0l3biF8/PPP2vLli26du2a3ccvSpUrV1a3bt00Y8YMxcXF2f0fgez/mGWPNPxZenp6jnlLgBn8/vvvSk9P1/3335/jx2L16tVznZKQl+wfgPbesq1SpYqeffZZRUZGKiIiQgMHDlRMTIxGjx6tf/3rXzZ9Q0NDdeDAAbv/unkUM3uOW9myZTVz5kz5+fmpfPny6tq1q0aPHq3MzEzNnz/f2t8wDEkq8C3h7B++/AB0HEbqUGQ8PT0l3ZgsnJvsX8vZoSx7su7169dz9C1s8MuuJ/sX6p/lVWNusoNhXheq2bNn6/DhwwoPD9eyZcs0ZswYvfvuu+rWrZuCgoLUokWLPC+ORTWn7s+aNm2q8uXLW+fG3Er2OeYWRg3DUHJysnx9fe2uEyhOVatWVUJCgs6ePavatWvn2S8xMdHaPy/Z15r8fgDa+4BX9vJCaWlpuS41lJfTp09r06ZN2rx5szIyMuTu7q66devavf+teHl5SbpxXfjzKOLf/vY3Sba3aLP75/YwR36jeGlpaZLED0AHItShyHh5ealmzZqKi4tTSkqKzR/6zMxM7dq1S+7u7qpfv76k//+rLvvCe7OiWJncz89PO3fu1JkzZ3Lcgt27d6/dx8l+gvT48eN59mnYsKHGjh2r119/XRs2bFB4eLi+//57hYeHq3bt2urTp48ee+yxHP8BKqo5dX+WkpKiq1ev5pgfl5fY2FhJynVU4sSJE8rKyirULXOgKDVr1kwJCQnavn17vqEu++03+T25nn2dKoofgNmB6fLly3lO18h27do1rV69WkuWLNH27duVlZWlRo0a6a233lJgYKAqVqxo078wc+qyA2J2WLtZdp03v4Emv3lz+c23y57+kt/tZxQvQh2KVGBgoD777DPNnj1bf//7363t33zzjeLj49W7d2/rgwPly5dXnTp1tG7dOiUnJ1svLvHx8Zo3b16ha+nZs6eio6M1a9asHE+/2jufTroRDj09Pa3BJz+urq7q3LmzOnfurAsXLlgfoJg5c6ZmzZqlXr16aerUqdb+hZlTl56erri4ODVt2tSm/fr16/rwww9lGIbat29vbU9KStKlS5esoTrb0aNH9Z///EcuLi569NFHc3xP9nk3b978tuoEitpjjz2m5cuXa+7cuerTp0+uP1727t2rmJgYVa9ePccbEG529913y93dXbt27ZJhGDaj6mfOnNHp06dvOXcvW/YPwGPHjuUaNg3DUExMjCIiImyekB84cKD69euX7xP2hZlTd99998nNzU2///57jn5HjhyRZPuDrkWLFpKkrVu3atiwYTb9f/nlF5s+N8t+PSSj+o5DqEORGjFihNatW6fPPvvMurL4kSNH9NNPP6l69eo2y5xIN96rOHHiRAUFBalTp066fPmy1qxZo7Zt22rt2rWFqqV///5asmSJvvvuOx09elTNmzfXyZMntWbNGj344IPavHlzjvWacuPi4qKHH35YGzZs0B9//GH3rYUqVapo6NChGjp0qPbs2aPw8HAdPXq0UOd0sz/++EOPP/64GjVqJH9/f/n4+OjixYvatm2bTp48KX9/fz3//PPW/mfOnFFQUJAeeOABNWjQQN7e3oqPj9e6deuUkZGhMWPG6O67787xPdu3b1fZsmXVrl27IqsdKIyHH37Y+mf4zTff1Pvvv29zu/Po0aN69dVXZRiGxo4dm+eyJ5Lk7u6uRx99VD/88IMiIiJsHiaaMWNGrtND8vLAAw9IuvFD6M9P2168eFGPP/64Tp48KRcXF7Vr1079+vVTp06d7FoNIDQ0VKGhoXbXcjMvLy917dpVP/zwg3744Qf16tVL0o0fgB9//LEkqUuXLtb+99xzj+6991798ssvioqKsq5Vd+7cOS1cuFA1atTIdbmTX3/9VRI/AB2JUIci5enpqQULFujjjz/WmjVrFB0drYoVK6pfv34aOXKk9SnVbIMHD1Z6eroWLlyohQsXqk6dOho/frzuuuuuQoc6Nzc3zZkzR9OmTdPq1asVGxsrf39/zZ49W2vXrtXmzZut8wBvpX///lq+fLnWrVuX69Oht9KsWTM1a9bM5kGOwipXrpxCQ0O1detW/fzzz7p06ZI8PDx0991368knn1RwcLBNAK1WrZoGDBigPXv2aNWqVbp69aoqVKigBx98UM8880yuF+nr169r1apV6tKlS4HW6wKKk8Vi0bRp0/Tiiy/q+++/17Zt29ShQwdVrFhRx44d0/r1660/VOz58zpq1Cht3bpVb731ljZt2qS6desqOjpaCQkJ8vX1tT4EditNmjRRtWrVtG3bthzLOqWmpsrV1VWvvfaagoKC7B79KypjxoxRdHS0Ro8erdWrV6tGjRravn279u/fr5YtW+rxxx+36f/OO+9o4MCBev7559WrVy95enrqxx9/VFJSksLCwnINotu3b1ezZs0K9CAaipbFyH7MBfgLGTRokA4fPqxt27bZ1d8wDPXq1UvVq1fX559/XszVOY/Vq1crNDRU//vf/6yjEICzyMzMVEREhJYtW6YDBw4oNTVVVapUUcuWLTVkyJBcF/weN26clixZkmPaw7Fjx/Thhx9q27ZtKlOmjFq3bq033nhDb775phISErRu3Tq7apo+fbpmz56tjRs32vyIzcrKsuvOQHE6e/as/v3vf2vjxo1KTk5WjRo11LNnT73wwgu53oHYt2+fpk+frh07dsgwDDVp0kQvvfRSrj8Af/vtN/Xt21fvvfeennjiiZI4HeSCUAdTO3fuXI7V0leuXKlXXnlFffv21aRJk+w+1po1a/Tyyy8rPDw81/9YmFG/fv3k4+OjTz/91NGlAKXCqVOn9Oijj2rkyJE20x/MbsKECYqMjNRPP/2U65OxKBncfoWpjRkzRikpKWrSpInKlSungwcPavPmzapYsWKON03cSpcuXTRmzBhdvHixmKp1LhcvXlSHDh2s828A3FrNmjX15JNP6quvvtLgwYP/Est7XLhwQd99951eeeUVAp2DMVIHU/vuu++0aNEiHT9+3LoYcbt27RQaGprvUggAcLsuXryoBQsWqEePHmrQoIGjyyl2v/76qzZu3Kjhw4cXaH0+FD1CHQAAgAnwmjAAAAATINQBAACYAKEOAADABHj6Ffg/J06c0JIlS/Trr7/q119/1aVLlxQYGGjzWq8/S01N1cyZM7VixQpduHBBNWrUUN++fTVs2LBcX120cuVKffXVV9Y1sho3bqwRI0ZYX6r9Z+vXr9cXX3yhI0eO6Nq1a6pZs6Y6d+6sZ599Nse7IQE4p4JeW7JfK5ibZs2aadGiRblu27Ztm7744gvt2bNH165d05133qkHHnhA//jHP2yeSg0LC9P27dt19OhRJSUlqUKFCqpfv76Cg4NzfVUgSg9CHfB/YmJiFBYWJjc3N9WtW1eXLl3Kt39GRoaGDh2qnTt3qnXr1urVq5f27Nmj6dOn68CBA5o+fbpN/9mzZ2vatGmqWLGievXqJVdXV61bt07Dhw/XRx99pMceeyzX/j4+Pnr00UdVrlw57dq1S59++qlWrlypiIgIu9+IAcBxCnptyRYUFKRatWrZtFWvXj3Xvp9//rk++ugj1apVSz169JCnp6dOnz6tzZs3KyUlxSbUffvtt7rzzjv10EMPqXLlyrp8+bLWr1+v0NBQPfvssxo3btxtnysczABgGIZhnDhxwtizZ4+RlpZmxMfHG76+vsbrr7+eZ//58+cbvr6+xptvvmnTPmbMGMPX19fYuHGjte3UqVNGkyZNjNatWxtnzpyxtl++fNl49NFHjRYtWhiXLl2ytqelpRn33Xef8dBDDxmXL1+2Of748eMNX19fIzw8vLCnDKAEFPTaMmPGDMPX19fYtm2bXcePiooy/Pz8jFdffdVIS0uz2ZaZmWlkZmbatP3xxx85jnH16lWjZ8+ehp+fn801CqULc+rg1LZv3y4/Pz/NnDlTO3fu1MCBA9WsWTO1b99ekydPVkZGRpF9V+3atdW0aVO7Xq4tSREREbJYLBo5cqRNe/bn8PBwa9vPP/+sjIwM9e/f3+adj97e3nrmmWeUnJyslStXWtsvXryo1NRU3XffffL29rY5/sMPPyxJSkpKKtgJArBy5mtLQc2YMUOenp56//33c3xHmTJlcryeLLe15O644w61b99ehmHo5MmTxVInih+hDqXC3r179eyzz6pSpUp66qmnVLVqVX355Zf5zncrTn/88Yd+++03NWjQIMeLuWvVqqV69eppx44d1rYLFy5IurHa/J9l316Jioqytvn4+MjLy0t79uzRlStXbPr//PPPkqSWLVsWzckAf2HOdm25WXR0tP773//qyy+/1C+//KLMzMwcfZKSkhQdHa327dvLw8NDP/30k2bPnq1vvvlG8fHxdn9Xenq6oqKi5Obmpnr16hXhWaAkMacOpcKmTZv02WefqUOHDpJuXICCgoK0aNEivf7663J3d1dycrLmzp1r9zEbNWqkzp0731Y9J06ckGEYeb6Vonbt2vr555+VmpqqO+64w/pQw6lTp3L0TUhIkCQdP37c2ubi4qKxY8fq7bffVs+ePdWpUyeVK1dOu3fv1r59+/Tmm2+qadOmt1U7gP/P2a4tN5s5c6bN53r16uk///mP/P39rW1xcXEyDEPe3t4aMGCAfvvtN+s2V1dXhYaG6oUXXsj1+J9++qnS09OVlJSkTZs2KSEhQWPGjFGVKlUKXTscg1CHUqFNmzbWi64kubu7q3v37po5c6aOHj0qPz8/JScn5/nEWG6CgoJu+8KbkpIiSXm+5zC7PSUlRXfccYfatWunMmXKKDw8XIMHD1a1atWs27/++mtJyjEi9/jjj8vb21vjxo3TwoULre1du3ZVx44db6tuALac7doi3QiFH330kVq2bKkqVaro1KlT+vbbbzV37lwNGzZMP/74oypUqCBJ1ndRR0RE6O6779a3336re+65R3v37tWbb76p6dOn65577lGnTp1yfM9nn32m1NRUSTcC4JgxY/Tcc8/ddt1wPEIdSoWbf5lmy77tmR2G7rrrLutSIcXN+L+361ksFrv6161bV0OGDNEXX3yh3r1769FHH7U+/erl5SVJOea9LFy4UBMmTNCIESP05JNPytPTU1FRUXrnnXf05JNPKiIiIsetXwAF42zXFkk5AmH9+vU1btw4ubm5afbs2Vq0aJGGDx8uSdZbsoZh6N///rcaNmwoSWrbtq0mTpyoZ599VnPnzs011O3atUtZWVk6e/asli9frunTp2vfvn1OcesZt4c5dSgVchsRc3FxkaRc55kUt+wg9ufRtWy5jeSNHTtW7733nqpXr64lS5boxx9/VNu2bTVjxgxJUuXKla19jxw5ogkTJqhLly565ZVXdOedd6p8+fLq2LGj3n//fZ0/f75At4MA5M7Zri356du3ryRpz5491rbsa1GtWrWsgS5b27Zt5e7urn379uV5zDJlyqhGjRoaOnSoRo4cqcjISJuHtlC6MFIH0yjJeS+1a9eWxWLJcyJyfHy8fHx8dMcdd9i0P/HEE3riiSds2mJiYqz1ZNuyZYsyMzNzfRiiefPmklSiIwfAX5mj5tT9WaVKlSRJ165ds7bVrVtX0v8PdzezWCwqX758nj8+/6xdu3aSblyTunXrVthy4QCEOphGSc57KVeunJo0aaLffvtNZ8+etbkNmpCQoGPHjtl9UVy+fLkk2fTPXk4ht0VKs9vc3Nxuq3YABVPSc+ryEhsbK8n2Kfr69evLx8dH8fHxSk9Pt1nSJCkpSRcvXrQGv1tJTEyUpFzfhoPSgX9zMI2SnvfSt29fxcbGasaMGZo4caK1Pft2avatkmx/XtVduvHk3aJFi/Tggw+qRYsW1vb77rtPkrRo0SI9/fTT1luzhmHok08+kcSSJkBJKclrS3p6ug4ePKiAgACb9sTERH3wwQeSpB49eljbLRaL+vXrp08//VRffvmlRowYYd2W/fRsly5drG3ZT+D/eXml5ORk67Wrffv2RXhGKEmEOuD/XLx4UR999JEkWZ8I2717t/WVOXfffbeef/55a/8BAwbohx9+0OLFixUfH69mzZpp9+7dioqKUvfu3a2LBGebPHmy4uLiFBAQIC8vL8XFxennn39WvXr1NHnyZJu+zZs316OPPqrVq1ere/fu6tKli8qXL6+oqCj99ttvuvvuu3PcxgXgnApybfnjjz/Ur18/NWnSRH5+fqpSpYpOnz6t9evX6+rVqxo0aJDatm1rc/zhw4dr3bp1+te//qWoqCj5+vpq7969iomJ0d13322zpMm+ffs0cuRItWzZUnXr1pW3t7dOnz6tDRs2KCUlRX379tVDDz1UEv9YUAwIdcD/SU1N1ZIlS2za4uPjrfPmWrVqZRPq3NzcNGfOHM2aNUvLly/Xzp07VaNGDb366qsaNmxYjuO3a9dOhw8f1o8//qg//vhDNWvW1IgRIzR8+PBcJ2tPnz5d8+bN0/fff6/IyEhlZmaqRo0aGjJkiEJCQvJcTgWAcynItcXDw0PBwcHavXu31q9frytXrsjT01PNmjXTgAED1L179xzHL1++vBYsWKAZM2ZozZo12r59u6pUqaLg4GC9/PLLNvPtGjdurIEDByo6OlorV67U1atX5eXlpfvuu0/9+vWzGQVE6WMxstdmAAAAQKnFkiYAAAAmQKgDAAAwAUIdAACACRDqAAAATIBQBwAAYAKEOgAAABMg1AEAAJgAoQ5AsYiMjJSfn5/8/Py0adMmR5djl7S0NM2bN0+DBw9Wu3btFBAQoNatW2vQoEH673//q4sXLzq6RADIE6EOQLH44YcfrH8fGRnpwErsc+DAAXXv3l0TJkzQ7t271bBhQ3Xt2lXNmjXT4cOHNXXqVHXp0kWHDh267e84efKk/Pz8FBwcXISVA8ANvCYMQJG7ePGitmzZIm9vb6WlpWnt2rW6du2aypUr5+jScpWQkKDg4GBdvnxZTz31lF5//XWbVytlZmZq1apV+vDDDxmtA+C0GKkDUORWrFihjIwMde3aVR06dFBqaqp++uknR5eVp7fffluXL1/WE088oXfeeccm0EmSi4uLevToocWLF6tWrVoOqhIA8keoA1Dksm+39urVS4GBgZJsb8fmZunSperTp4+aNm2qBx98UO+9955SUlIUHBwsPz8/nTx5Msc+8fHxGj9+vDp06KCAgAC1b99eo0eP1okTJ+yu9eDBg9q8ebPKlSun0aNH59vXx8dHd911l/Xz8ePH9Z///EePP/642rZtq4CAAHXs2FFvv/22zpw5Y7PvzJkz1alTJ0lSVFSUdb6hn5+fZs6cWajzunLliiZMmKAHH3xQTZs21WOPPaYlS5bke7s3PT1dc+bMUWBgoJo2barmzZsrODhYa9euzfU7/Pz81LFjR6Wlpelf//qXOnXqpCZNmmjmzJmaNGmS/Pz8FBERkeu+ly9f1r333quHH35YWVlZef8DBlAo3H4FUKROnjyp3bt3q1q1amrVqpWuX78ub29vbd68WUlJSapUqVKOfWbPnq1p06bJzc1NrVu3Vvny5bV8+XLt3r1b7u7uuX5PdHS0RowYoatXr8rf319NmzbVyZMntWzZMm3cuFFff/21/P39b1lv9kMcf/vb33KM0N3Kd999pzlz5sjPz0/NmjWTm5ub4uLi9O233+qnn35SeHi4qlevLklq1KiRunbtqlWrVqlq1ap66KGHrMdp1KjRbZ/XtWvXNHjwYO3bt08+Pj7q2LGjLl26pPHjx+upp57Kte6MjAwNHz5c27ZtU8WKFdWhQwddu3ZNW7duVVRUlEaOHKmXXnopx35ZWVl64YUXtHfvXrVu3VqNGjVS+fLl1aNHD3311VdavHix+vbtm2O/yMhIpaenKygoSGXKMJYAFBsDAIrQJ598Yvj6+hoffPCBte2NN94wfH19jQULFuTof/ToUaNx48bGfffdZ/z666/W9uTkZKN///6Gr6+v4evra8THx1u3Xb582Wjbtq1x7733Ghs2bLA53rJlywxfX18jKCjIrnpff/11w9fX1wgLCyvoqRq7du0yTp48adOWmZlphIWFGb6+vsa4ceNstsXHxxu+vr7GoEGDcj3e7ZzXjBkzDF9fX2Pw4MFGamqqtX379u1GQEBArt/36aefGr6+vkb//v2Ny5cvW9vj4uKMli1bGn5+fsbu3btt9sn+99C9e3fj/PnzOWp/6qmnDF9fX+P333/Psa1Pnz6Gn5+fceLEiVzPG0DR4CcTgCKVfZu1V69e1rbsW7C5PQUbERGh69eva8CAAQoICLC2e3l55Xk7dPHixbpw4YJGjBihhx9+2GZbYGCgunTpot9++02xsbG3rPfSpUuSlOsI4q3cd999OebYlSlTRi+++KKqV6+udevWFeh4t3NeixYtkiS98cYbNg+itGrVyubfwc0WLlwoSXrrrbfk7e1tbffz89Nzzz0nwzC0YMGCXPf9+9//ripVquRoHzBggCQpPDzcpn3//v3at2+fWrdurdq1a+d57gAKj1AHoMjs379fhw4dUr169XTvvfda21u3bi0fHx/t2rUrx9y4Xbt2SZIeffTRHMdr1apVrmHrl19+kSTrHLU/e+CBByRJv/766y1rNgxDkmSxWG7ZNzcpKSn6/vvv9eGHH2r8+PEaN26cxo0bp4yMDF26dMkaGu1R0PNKSEhQYmKi6tatm+ut5q5du+ZoS0hI0JkzZ3TXXXepWbNmObZnB/AdO3bk2GaxWPS3v/0t19q6desmb29vLV26VJmZmdb2xYsXS5L69++f634Aig5z6gAUmeyRuJ49e9q0lylTRj169NDcuXP1448/asSIEdZt586dkyTVqFEj12PWqFFDSUlJNm2nTp2SJD322GP51mNPoMoOjbezVMkvv/yi1157Ld/vuXr1qipWrGjX8Qp6Xtn/7LLn7f1Zbu3Z+9SsWTPPfVxdXa39blalSpU85zh6eHgoMDBQCxYs0MaNG9WxY0elp6crMjJS3t7euYZ2AEWLUAegSBiGoR9//FGStGrVKm3dutVme3Ywi4yMtAl1txopy95+s+wnKHv27JlnyJCke+6555Z1+/v7KzIyUvv3779l35tdvXpVr732mi5fvqzQ0FB1795dtWrVkoeHhyTpySef1K5du3KtPy8FPa/bGWW0Z5+8ai5btmy+xx4wYIAWLFig8PBwdezYUWvWrNHly5f19NNP33JfAIVHqANQJKKioqzLeBw+fDjPfocOHVJcXJz1dmG1atV07NgxnT59OtfRuj8vDSLdGE06evSoQkNDVb9+/ULV/dBDD2nKlCn6+eeflZKSovLly9u1X0xMjC5duqRu3brp5ZdfzrE9Pj6+wLUU9Lx8fHwk5f7PKK/2atWqSfr/o4K57ZOZmZnnyGl+/P39de+992rjxo26cOGCdX4dt16BksGcOgBFIvvW6+uvv64DBw7k+lf2Mhk3PzBx//33S5LWrFmT45gxMTE5br1KUrt27SSpSBY09vPzU7t27ZSamqopU6bk2/fcuXPWOYHJycmScr/FGRMTo/Pnz+dod3NzkyRdv3491+MX9Lzuuusu+fj46NixYzpw4ECO7atXr87RVqtWLVWvXl3x8fHau3dvju3Zo63Nmze3q4Y/e+KJJ5SRkaFPPvlEW7duVaNGjdSkSZPbOhaAgiHUASi09PR0a4Do3r17nv169Ogh6UZwyL7FFxQUJFdXV33zzTfat2+ftW9KSkqeIWvAgAGqVKmSZs2aZQ0hN7t8+bIWLlyoP/74w67633vvPXl5eembb77Ru+++q5SUFJvtWVlZ+vHHH9WvXz8lJCRIkurVqyfpRhi9OXgmJibq3XffzfV7KlWqJFdXV8XHx9s8TFCY88p+6nTy5Mk27TExMXm+c/fpp5+WJH3wwQe6cuWKtf3QoUP6/PPPZbFYNHDgwFz3vZUePXrojjvu0Lx585SVlcUoHVCCLEZBJnwAQC7Wrl2rl156SU2bNtV3332Xb9/AwEAdPHhQ8+bNU6tWrSTZLj7cpk0beXp6KioqStWrV5e7u7t2796tTZs26c4777QeJzo6WiEhIUpOTladOnXUsGFDlSlTRqdOndLhw4eVnp6u6OhomyU78rN//369+OKLOn36tDw8PHTfffepatWqunLlivbu3aukpCRr8GvYsKEkaciQIdq6dasqVqyoli1b6vr169q+fbt8fX1lsVi0a9cu/fTTTzZvoRg+fLg2bdokX19fNW7cWK6ururYsaP1ideCnte1a9f09NNPa9++fapWrZpatGihS5cuafv27RowYID+97//qX379vriiy+sNWRkZGjYsGHatm2bKlWqpNatW1sXH05PT1doaGiOW8p+fn6qVauWXcu0/OMf/9CiRYvk7u6uzZs3q0KFCnb9OwBQOIzUASi0ZcuWScp/lC5bdp+bR5Gef/55ffjhh2rQoIG2b9+uHTt2qEuXLpo7d64uXboki8WSIxi0bNlSy5Yt0+DBg1WmTBlt2bJF27ZtU1pamnr27KlPP/20QG+IaNSokVauXKk333xTTZs21YEDB7Ry5Urt3r1b9evX1+uvv641a9ZYA50kffLJJxo+fLi8vb21ceNGHTx4UE899ZS+/PJL663WP5s4caJ69uyp8+fPa9myZVq8eLHNCGVBz6tcuXKaO3euBg0aJMMwtHbtWp0/f17vvfeeevfuLUk5nr51c3PTf//7X40aNUo+Pj5av369oqOj1axZM82aNSvXOYIFkR3Wu3TpQqADShAjdQCc1tmzZ9WpUyfdddddWrlypaPLKXXmzJmjKVOm6PXXX9fzzz9fYt8bGhqq1atX68svv7TOEwRQ/BipA+Bwx48fzzGPLSkpSePHj1dGRkaeb0bADfv27cuxDMmOHTv06aefysXFxa4R1KJy5MgRrVu3TvXq1VPbtm1L7HsBsKQJACewbNkyzZkzR02aNNGdd96pixcv6rffftOVK1fk7++voUOHOrpEp/byyy8rMzNT99xzj8qXL68TJ05Yg96rr75aIq/nmjNnjuLi4rR582Zdv35doaGht/2WDgC3h1AHwOEefPBBHTp0SHv37tVvv/0mSapdu7aCg4M1dOhQm3eaIqfBgwdr5cqVio2N1ZUrV+Tp6ak2bdpo4MCB6tKlS4nUsHHjRkVHR6t69eoaO3Yso6uAAzCnDgAAwASYUwcAAGAChDoAAAATINQBAACYAKEOAADABAh1AAAAJkCoAwAAMIH/B7VnYcXAvgDpAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 720x595.44 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_theme(\n",
    "    font_scale=1.7,\n",
    "    rc={'figure.figsize':(10,8.27)}\n",
    ")\n",
    "sns.set_style(\"whitegrid\")\n",
    "\n",
    "ax = sns.boxplot(\n",
    "    x='age_cat',\n",
    "    y='faith_pd',\n",
    "    data=datasets['MG'].sort_values(by='age_cat', ascending=False),\n",
    "    palette=[\n",
    "        '#008000',\n",
    "        '#800080',\n",
    "\n",
    "    ],\n",
    "    hue_order=[\n",
    "        'Young',\n",
    "        'Old',\n",
    "    ]\n",
    ")\n",
    "\n",
    "ax.set_xticklabels(['Young (<=35)\\nn=1098', 'Old (>=60)\\nn=1563'])\n",
    "ax.set_ylabel('Faith\\'s PD')\n",
    "ax.set_xlabel('Age Category')\n",
    "plt.savefig('results/03.01-alpha-distributions-young-old.png')\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "faith-pd-bench",
   "language": "python",
   "name": "faith-pd-bench"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}