{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "b6cc3c6b-85c7-4a04-9aef-8ffc055aa93c",
   "metadata": {},
   "source": "# Names Analysis & Modeling"
  },
  {
   "cell_type": "code",
   "id": "initial_id",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T19:54:31.700090Z",
     "start_time": "2025-09-25T19:54:31.689969Z"
    }
   },
   "source": [
    "import pandas as pd \n",
    "import unicodedata \n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import numpy as np \n",
    "import re \n",
    "import sys \n",
    "import os\n",
    "from collections import Counter\n",
    "\n",
    "from rich.jupyter import display\n",
    "\n",
    "sys.path.append(os.path.abspath(\"..\")) \n",
    "from collections import Counter \n",
    "from core.utils.data_loader import DataLoader\n",
    "from core.utils.region_mapper import RegionMapper\n",
    "from core.config.pipeline_config import PipelineConfig"
   ],
   "outputs": [],
   "execution_count": 116
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T18:26:50.162866Z",
     "start_time": "2025-09-25T18:26:50.159601Z"
    }
   },
   "cell_type": "code",
   "source": "LETTERS = 'abcdefghijklmnopqrstuvwxyz'",
   "id": "16647cc71aea7594",
   "outputs": [],
   "execution_count": 47
  },
  {
   "cell_type": "code",
   "id": "f1a69290-a9c0-40d0-9fe8-a06d8a466671",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T17:39:49.310278Z",
     "start_time": "2025-09-25T17:39:49.304876Z"
    }
   },
   "source": [
    "config = PipelineConfig(\n",
    "    paths={\n",
    "        \"root_dir\": \"../data\",\n",
    "        \"data_dir\": \"../data/dataset\",\n",
    "        \"models_dir\": \"../models\",\n",
    "        \"outputs_dir\": \"../data/processed\",\n",
    "        \"logs_dir\": \"../logs\",\n",
    "        \"configs_dir\": \"../configs\",\n",
    "        \"checkpoints_dir\": \"../checkpoints\"\n",
    "    }\n",
    ")\n",
    "\n",
    "loader = DataLoader(config)\n",
    "\n",
    "def normalize_letters(s):\n",
    "    \"\"\"Normalize accents -> ascii, lowercase, keep only a-z.\"\"\"\n",
    "    s = str(s)\n",
    "    s = unicodedata.normalize(\"NFKD\", s)\n",
    "    s = s.encode(\"ascii\", errors=\"ignore\").decode(\"utf-8\")\n",
    "    s = s.lower()\n",
    "    s = re.sub(r\"[^a-z]\", \"\", s)\n",
    "    return s"
   ],
   "outputs": [],
   "execution_count": 3
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T17:59:34.598256Z",
     "start_time": "2025-09-25T17:58:54.210735Z"
    }
   },
   "cell_type": "code",
   "source": [
    "df = loader.load_csv_complete(config.paths.data_dir / \"names_featured.csv\")\n",
    "df['province'] = RegionMapper.clean_province(df['province'])"
   ],
   "id": "e48c6fd9a213bcd2",
   "outputs": [],
   "execution_count": 23
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T17:59:38.255948Z",
     "start_time": "2025-09-25T17:59:38.249016Z"
    }
   },
   "cell_type": "code",
   "source": "df.columns",
   "id": "2715f291947f5158",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['name', 'sex', 'region', 'words', 'length', 'probable_native',\n",
       "       'probable_surname', 'identified_name', 'identified_surname',\n",
       "       'ner_entities', 'ner_tagged', 'annotated', 'identified_category',\n",
       "       'province'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 24
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T18:01:22.242283Z",
     "start_time": "2025-09-25T18:01:21.580597Z"
    }
   },
   "cell_type": "code",
   "source": "df.describe().T",
   "id": "93f8859e3e9c4350",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "                count       mean       std  min   25%   50%   75%   max\n",
       "words       6467942.0   2.869578   0.46841  1.0   3.0   3.0   3.0  11.0\n",
       "length      6467942.0  20.141236  3.796574  0.0  18.0  21.0  23.0  60.0\n",
       "ner_tagged  5018124.0   0.997939  0.045348  0.0   1.0   1.0   1.0   1.0\n",
       "annotated   5018124.0   0.997939  0.045348  0.0   1.0   1.0   1.0   1.0"
      ],
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>words</th>\n",
       "      <td>6467942.0</td>\n",
       "      <td>2.869578</td>\n",
       "      <td>0.46841</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>11.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>length</th>\n",
       "      <td>6467942.0</td>\n",
       "      <td>20.141236</td>\n",
       "      <td>3.796574</td>\n",
       "      <td>0.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>23.0</td>\n",
       "      <td>60.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ner_tagged</th>\n",
       "      <td>5018124.0</td>\n",
       "      <td>0.997939</td>\n",
       "      <td>0.045348</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>annotated</th>\n",
       "      <td>5018124.0</td>\n",
       "      <td>0.997939</td>\n",
       "      <td>0.045348</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 28
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T18:02:53.214472Z",
     "start_time": "2025-09-25T18:02:52.427559Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# Group by province and compute counts\n",
    "word_stats = (\n",
    "    df.groupby(\"province\")[\"identified_category\"]\n",
    "    .value_counts(normalize=True)  # get proportions\n",
    "    .unstack(fill_value=0)          # reshape into columns per word count\n",
    ")\n",
    "\n",
    "display(word_stats.head(12))"
   ],
   "id": "378147d2abc9ab24",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "identified_category   compose    simple\n",
       "province                               \n",
       "AUTRES               0.206217  0.793783\n",
       "BANDUNDU             0.626906  0.373094\n",
       "BAS-CONGO            0.090813  0.909187\n",
       "EQUATEUR             0.124238  0.875762\n",
       "KASAI-OCCIDENTAL     0.261266  0.738734\n",
       "KASAI-ORIENTAL       0.076224  0.923776\n",
       "KATANGA              0.180624  0.819376\n",
       "KINSHASA             0.076792  0.923208\n",
       "MANIEMA              0.461150  0.538850\n",
       "NORD-KIVU            0.119626  0.880374\n",
       "ORIENTALE            0.160905  0.839095\n",
       "SUD-KIVU             0.409647  0.590353"
      ],
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>identified_category</th>\n",
       "      <th>compose</th>\n",
       "      <th>simple</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>province</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>AUTRES</th>\n",
       "      <td>0.206217</td>\n",
       "      <td>0.793783</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BANDUNDU</th>\n",
       "      <td>0.626906</td>\n",
       "      <td>0.373094</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BAS-CONGO</th>\n",
       "      <td>0.090813</td>\n",
       "      <td>0.909187</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EQUATEUR</th>\n",
       "      <td>0.124238</td>\n",
       "      <td>0.875762</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KASAI-OCCIDENTAL</th>\n",
       "      <td>0.261266</td>\n",
       "      <td>0.738734</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KASAI-ORIENTAL</th>\n",
       "      <td>0.076224</td>\n",
       "      <td>0.923776</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KATANGA</th>\n",
       "      <td>0.180624</td>\n",
       "      <td>0.819376</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>KINSHASA</th>\n",
       "      <td>0.076792</td>\n",
       "      <td>0.923208</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MANIEMA</th>\n",
       "      <td>0.461150</td>\n",
       "      <td>0.538850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NORD-KIVU</th>\n",
       "      <td>0.119626</td>\n",
       "      <td>0.880374</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>ORIENTALE</th>\n",
       "      <td>0.160905</td>\n",
       "      <td>0.839095</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUD-KIVU</th>\n",
       "      <td>0.409647</td>\n",
       "      <td>0.590353</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data",
     "jetTransient": {
      "display_id": null
     }
    }
   ],
   "execution_count": 30
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T18:05:12.951347Z",
     "start_time": "2025-09-25T18:05:12.736698Z"
    }
   },
   "cell_type": "code",
   "source": [
    "ax = word_stats.plot(\n",
    "    kind=\"bar\",\n",
    "    stacked=True,\n",
    "    figsize=(12, 6),\n",
    "    colormap=\"tab20c\"\n",
    ")\n",
    "\n",
    "plt.xlabel(\"Province\")\n",
    "plt.ylabel(\"Proportion (%)\")\n",
    "plt.legend(title=\"Name Category\", bbox_to_anchor=(1.05, 1), loc=\"upper left\")\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ],
   "id": "6d5c1abb55b7076a",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ],
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABKcAAAJOCAYAAABiG2QNAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjMsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvZiW1igAAAAlwSFlzAAAPYQAAD2EBqD+naQAAmKxJREFUeJzs3XdclXX/x/H3YYmKE9NwkStRXChmJpaC+suZu7QcmSPLVVZqmuLeM3dqORumgHty33Y7cuJMvBNnKu6Fogic3x8+PHfEEDzAdY6+no/H7/G7uc7FOW8/knDefK/vZTKbzWYBAAAAAAAABnAwOgAAAAAAAABeXJRTAAAAAAAAMAzlFAAAAAAAAAxDOQUAAAAAAADDUE4BAAAAAADAMJRTAAAAAAAAMAzlFAAAAAAAAAxDOQUAAAAAAADDOBkdILPEx8crNjZWDg4OMplMRscBAAAAACDdmc1mxcfHy8nJSQ4OrEeBfXhhyqnY2FgdOXLE6BgAAAAAAGS48uXLy8XFxegYQKq8MOXUk8a4fPnycnR0NDhNYnFxcTpy5IjN5rN1zM86zO/ZMTvrMD/rMD/rMD/rMD/rMD/rMD/rMD/r2Pr8nuRj1RTsyQtTTj25lM/R0dEm/wF5wtbz2TrmZx3m9+yYnXWYn3WYn3WYn3WYn3WYn3WYn3WYn3VsfX5sZwN7QpUKAAAAAAAAw1BOAQAAAAAAwDCUUwAAAAAAADDMC7PnFAAAAAAAeD7ExcXp0aNHRsdAClxcXFK9MT/lFAAAAAAAsAtms1mRkZG6deuW0VHwFA4ODipWrJhcXFyeei7lFAAAAAAAsAtPiqn8+fMrW7Zs3JXQRsXHx+vixYu6dOmSihYt+tS/J8opAAAAAABg8+Li4izFlLu7u9Fx8BQvvfSSLl68qNjYWDk7O6d4LhuiAwAAAAAAm/dkj6ls2bIZnASp8eRyvri4uKeeSzkFAAAAAADsBpfy2Ye0/D1RTgEAAAAAALtWunRp9e3bN9HxlStXyt/f34BESbt06ZIGDRqkN998U5UqVVLTpk0VHBycpudYv369rl+/njEBDUI5BQAAAAAA7N6aNWu0a9cuo2Mk68yZM2rRooVu3bqlqVOnatWqVWrTpo2GDBmiBQsWpOo5Lly4oD59+ig6OjqD02YumyinYmJi1KhRI+3evTvZc/744w+1atVKFStWVIsWLXT06NFMTAgAAAAAAGxZoUKFNGzYMMXExBgdJUlDhw6Vl5eXvv32W/n4+Kho0aJ699139cUXX+jbb7/VnTt3nvocZrM5E5JmPsPLqYcPH+rzzz/Xn3/+mew59+/fV9euXeXr66uVK1fKx8dH3bp10/379zMxKQAAAAAAsFV9+vTR5cuXNX/+/GTP2b9/v9q0aaOKFSuqUqVK6tKli65cuSLp8SWA7dq106xZs1S1alXVqFFDwcHB2rBhg2rXri1fX1+NHz/e8lwxMTEaMWKEqlWrpmrVqumLL77QrVu3knzdyMhI7dq1Sx07dky0F1PLli313XffWTZ6TyljQECA5f+vXLlSkrR582Y1aNBAFStWVMuWLbVnzx7Lc8fHx2vChAmWjDNnzlTdunUti4Nu376tb775Rm+88YaqVKmiL7/8Urdv35Yk7d69W/7+/hoyZIiqVKmi6dOny8vLS8eOHbM8//Xr11W2bFmdPXv26X9BKTC0nDp58qRat26tc+fOpXjeunXrlCVLFn311VcqUaKEBg4cqOzZs2vDhg2ZlBQAAAAAANiyAgUKqFevXpo9e7bOnz+f6PG7d++qW7duqlGjhtasWaP58+fr3Llzmjt3ruWcsLAwnT9/Xr/++qsaNmyowMBALVq0SLNmzVL//v01b948/fHHH5KkSZMm6ejRo/ruu++0aNEiRUVFqXfv3klmO3HihMxms8qXL5/osaxZs8rX11dOTk5Pzbh8+XLL/2/QoIHCw8PVr18/de/eXatWrVKTJk3UpUsXS1k0Z84cBQcHa+LEifr+++/173//O8FsevTooePHj2v27Nn6/vvvFRERof79+1sev3DhgmJiYrRy5Uo1b95cVapU0caNGy2Pb9y4UWXKlJGnp2eq/56SYmg5tWfPHlWrVk0///xziucdOnRIVapUsbSLJpNJlStX1sGDBzMhJQAAAAAAsAft2rWTp6enRo4cmeixBw8e6JNPPtGnn36qIkWKqEqVKqpXr16CK7nMZrMGDRokT09Pvfvuu4qOjlbPnj3l5eWlli1byt3dXadOnVJ0dLSWLFmioUOHqkKFCipdurTGjRunPXv26MSJE4le+8klezly5Egx/9My5s2b1/L/XV1dNX/+fLVu3VqNGzeWp6en2rdvrzfffFM//vijJGnZsmXq06eP/Pz8VLZsWY0ZM8ZyaWB4eLj27Nmj8ePHq0KFCqpQoYLGjx+v0NBQnTp1ypKpc+fO8vT0VMGCBdWwYcMEC4XWr1+vhg0bpurvJiVOVj+DFdq2bZuq865evaqSJUsmOObu7p7ipYDJiYuLS/PnZIb4+HhlzZpV8fHxRkexS8zPOszv2TE76zA/6zA/6zA/6zA/6zA/6zA/6zA/69j6/Gz1PW9mcXR0VGBgoNq2bastW7YkeOyll15S06ZN9cMPP+j48eM6efKkTpw4ocqVK1vOcXd3t1xelyVLFklS4cKFLY+7uroqJiZG58+f16NHj/Tee+8leI34+HidOXNGpUuXTnA8d+7ckh6XVE8KpqSkJuPfRUREaP369QkW/Tx69Eh+fn66ceOGrly5kmC1VvHixZUrVy5J0qlTp5QzZ04VK1bM8niJEiWUK1cunTp1ylKk/f3P//bbb2vkyJE6fvy4XnrpJR04cCDBpY7PytByKrWio6Pl4uKS4JiLi8szbXJ25MiRdMnk7Owsb29vOTo6psvzOTo6qmzZsunyXE/ExcXp2LFjevToUbo+b3pgftax9fkxO+swP+swP+swP+swP+swP+swP+swP+swP+vY8vzsUeXKldWiRQuNHDlSnTt3thy/fPmyWrRoIW9vb73xxhtq3bq1/v3vf+vQoUOWc5ycEtck/9wjSvpfCbhs2TJLmfWEu7t7ovO9vb1lMpl09OhRvfnmmwkeu3//vj799FP169dPefLkeWrGf+bo0qWLmjZtmuC4q6ur5c/yz03Un3z8z57l78/595LzSUknPV6xVb16dW3cuFH58+dXxYoV9fLLLyf5PGlhF+VUlixZEhVRMTExcnV1TfNzlS9fPl3/Ubp06ZJN3gnAxcVFHh4e8vb2NjpKspifdWx1fszOOszPOszPOszPOszPOszPOszPOszPOszPOuk9v7i4uHRblGHPvvjiC7399tsJNkffvHmzcuXKpTlz5liOLV68+JnugFekSBE5Ojrq1q1bKlOmjKTHm4MPHDhQAwYMkJubW4Lz8+bNqxo1amjhwoWqWbNmgsJrxYoV2rdvnzw8PLR69eoUM/6zKCtWrJj++uuvBHs+jRs3TsWKFVOrVq2UP39+HTt2TF5eXpKk8+fPWy4xLFasmO7cuaNTp06pePHikh7vDR4VFaVixYrp5s2bSf7ZGzVqpO+//14vv/xyulzSJ9lJOVWgQAFdu3YtwbFr164pf/78aX4uR0fHdCunpMcl2cOHD9Pt+dJbev5ZMwLzs44tz4/ZWYf5WYf5WYf5WYf5WYf5WYf5WYf5WYf5WcfW52dv8uTJoy+++EKDBg1SoUKFJD2+tO7ixYvatWuXChcurPXr12vTpk1JblL+NG5ubmrVqpUCAwM1bNgwubu7a/To0bp48WKCy+D+bsCAAWrTpo169+6tzp07K0eOHPrXv/6lKVOmqG/fvsqVK9dTM2bNmlXS4/2i8uTJo44dO+r9999X+fLlVatWLYWGhuqHH37QwoULJT3eg2vatGkqWLCg8uTJoxEjRkh6XHKVKFFCb775pvr166dvvvlGkjR06FBVrVpVr776quWOfv9Up04dDRkyROfOndOoUaPSPLukGLohempVrFhRYWFhlqbQbDbrwIEDqlixosHJAAAAAACALWrZsqV8fHwsH9evX19NmjRRr1691KJFC+3evVv9+vVTRETEM62q69+/v6pXr65evXqpdevWcnJy0ty5c5MtGkuWLKlly5ZJkrp3765mzZppzZo1GjlypDp27JiqjHnz5lWTJk3Up08fLV++XJUqVdK4ceO0bNkyNWjQQL/88osmTpyoqlWrSpI6deqkunXrqmfPnurQoYNq164tk8kkZ2dnSdLYsWNVpEgRdezYUR999JFKlSqlGTNmpPjndnNz05tvvqlKlSoleQnjszCZn2X9WgYoXbq0Fi1apGrVqkl6vAl6jhw55OrqqqioKNWtW1cNGzbUe++9p59++kkbNmzQpk2bEl3bmZy4uDgdPHhQlSpVStdG+uzZszbZvmfJksXqWzlmBuZnHVucH7OzDvOzDvOzDvOzDvOzDvOzDvOzDvOzDvOzTnrPL6Pe+9qCBw8e6PTp0ypWrNgzbfPzIvrtt99Urlw5yybsN27cUPXq1bV169ZkV3ilxnvvvadWrVqpRYsWyZ6Tlr8vm1055efnp3Xr1kl63MrNmTNH+/fvV/PmzXXo0CHNnTs31cUUAAAAAADAi+bnn3/W119/rZMnTyoiIkKBgYEqX778MxdTv//+u2bOnKmIiAi9/fbb6ZbTZvacOnHiRIofV6hQQUFBQZkZCQAAAAAAwG4NHjxYQ4cO1XvvvSez2azq1as/9bK9lISEhGjr1q0aNmyYsmfPnm45baacAgAAAAAAQPopUKCAZs6cmW7PN3r06HR7rr+z2cv6AAAAAAAA8PyjnAIAAAAAAIBhKKcAAAAAAABgGMopAAAAAAAAGIZyCgAAAAAAAIahnAIAAAAAAIBhKKcAAAAAAABgGMopAAAAAAAAGIZyCgAAAAAA2L24ePNz+VovAiejAwAAAAAAAFjL0cGkgUHHdPravQx9nWL5smtkM+80f97Zs2c1bNgwHThwQLly5VKnTp3Uvn17RUREaNSoUQoLC1P27Nn17rvv6pNPPpGDg4O+/fZbnT9/Xjly5NDKlSuVJ08eDRs2TGfOnNHMmTMVHx+vTz75RO3bt5cklS5dWiNGjNCcOXN0/fp1+fv7a9iwYcqePbskKSwsTOPGjdPx48eVN29edenSRW3atJEkXbx4UYMGDVJYWJhcXV3VoEED9e/fX87OzjKbzZo5c6Z+/PFHPXjwQL6+vho8eLAKFiyYLjOlnAIAAAAAAM+F09fuKTwyyugYiTx8+FCdOnWSt7e3fvnlF50/f159+/ZVrly5NGrUKPn7+2v58uU6ffq0Bg0aJDc3N3Xs2FGStG7dOnXu3FkhISGaNGmS+vTpI19fXy1evFgbNmzQ2LFj1ahRI+XNm1eSNHXqVI0YMULu7u76+uuvNXjwYE2cOFERERHq0KGDOnbsqJEjR+rQoUMaOnSo8uXLp7p162r48OHKli2bgoODdf36dfXq1UvFixfX+++/ryVLlmj16tWaOHGi8uXLpwULFqhTp05avXq1nJ2drZ4Pl/UBAAAAAABkoO3bt+vGjRsaNWqUSpUqJX9/fw0aNEi3bt1S1qxZNXz4cJUoUUJ16tRR7969NW/ePMvn5smTR71791bRokXVrFkz3b17VwMHDlSJEiX00UcfKTY2VmfPnrWc36VLF9WqVUvly5fXwIEDtX79et29e1e//PKLypYtq88//1zFixdXs2bN9MEHH1he68KFC8qRI4cKFiyoypUra+7cuXrrrbckSfPmzdNXX32latWqqUSJEho2bJhu376t//znP+kyH8opAAAAAACADHT69GkVK1ZMbm5ulmMtWrTQqVOn5O3tLSen/13Y5uPjo6tXr+rOnTuSpMKFC8tkMkmSXF1dJUmFChVK8HFMTIzl8ytXrmz53+XKlVNcXJxOnz6tiIgIVahQIUEuHx8fRURESJI6d+6s1atXq3r16vr888918eJFFS5cWPfu3VNkZKQ+++wz+fj4yMfHR76+vrp165bOnDmTLvPhsj4AAAAAAIAM9Pfy6e+yZMmS6Fh8fLwkKS4uLtnPdXBIfq3R3y+ze/JcDg4Oyb7Wk9dp0qSJqlevri1btujf//63evXqpS5duuijjz6S9PhywWLFiiX4/Fy5ciWbIy1YOQUAAAAAAJCBXnnlFZ09e1bR0dGWY2PHjtWyZct07NgxPXr0yHI8LCxMefPmVe7cuZ/ptY4fP27530ePHpWzs7OKFSumYsWK6dChQwnODQsLsxROkydP1vXr19WmTRvNmTNHffr00aZNm5QzZ065u7vr6tWr8vT0lKenpzw8PDR+/HidPn36mTL+E+UUAAAAAABABvLz81O+fPk0ePBgRUREaOvWrfrpp580ZcoUxcTEWI5v2bJF3377rdq0aWO5lC+tpk2bpj179ujQoUMaMWKEmjVrpuzZs6tt27Y6fvy4Jk2apNOnTysoKEjLli3T+++/L0k6deqUhg0bpvDwcP3555/atm2bypYtK0nq2LGjpkyZotDQUJ05c0aDBg3SgQMHVLx48XSZD5f1AQAAAACA50KxfNlt8jWcnJw0c+ZMDRs2TM2aNVO+fPn01VdfqU6dOipYsKBGjhyppk2bKm/evOrQoYO6dev2zPmaNm2q/v37686dO2rYsKEGDhwoSSpYsKDmzJmjcePGacGCBSpYsKD69++vFi1aSJICAwM1dOhQtWvXTrGxsapVq5blcz/66CPdu3dPgwcPVlRUlMqVK6f58+en22V9lFMAAAAAAMDuxcWbNbKZd6a9lqND2lY2lShRQgsXLkx0vGzZslq6dGmSn9OzZ88EH1erVk0nTpxIcOyfH7/++uvq3bt3ks9XvXp1BQUFJfmYu7u7pk2bluRjjo6O+uyzz/TZZ58l+bi1uKwPAAAAAADYvbSWRfbyWi8CyikAAAAAAAAYhsv6AAAAAAAAngP/vMTPXrByCgAAAAAAAIahnAIAAAAAAIBhKKcAAAAAAABgGMopAAAAAAAAGIZyCgAAAAAAAIahnAIAAAAAAIBhKKcAAAAAAAAMsHv3bpUuXTpDnrt06dLavXt3hjx3eqOcAgAAAAAAMICPj4+2b99udAzDUU4BAAAAAAC7Zzab7e61XFxc9NJLL6XLc9kzJ6MDAAAAAAAAWMtkMunSpUuKiYnJ0NdxcXGRh4dHmj9v0aJF+v7773Xt2jWVKlVKX3/9teLi4tS+fXudOHFCf/31lwICAjRnzhwNGzZMN2/eVIsWLdS6dWv1799fp06dUrVq1TRx4kS5ubmpf//+cnNz019//aWdO3eqePHiGjx4sCpXrpzotWNiYjRu3DitXr1aklSzZk0NGjRIuXPntnYc6YKVUwAAAAAA4LkQExOjhw8fZuj/PUv59ccff2jcuHEaMmSI1q9fL19fX/Xp00fx8fGJzp07d65mzpyp4cOHa/HixerRo4f69u2r+fPn6+DBg/r1118t5/70008qWbKkgoKCVLVqVXXt2lU3btxI9JyTJk3S0aNH9d1332nRokWKiopS79690/znyCiUUwAAAAAAABnowoULMplMKliwoAoXLqw+ffpo/PjxSV4e+Mknn8jLy0uNGjWSu7u7GjZsqBo1aqhKlSqqXr26Tp06ZTm3ZMmS+uKLL1SiRAkNGDBAuXLl0rp16xI8X3R0tJYsWaKhQ4eqQoUKKl26tMaNG6c9e/boxIkTGf5nTw0u6wMAAAAAAMhAfn5+evXVV9W4cWOVLVtWAQEBatWqlc6cOZPo3CJFilj+t6urqwoVKpTg47+v3Pr7JXwODg4qW7asIiIiEjzf+fPn9ejRI7333nsJjsfHx+vMmTMZdrfAtKCcAgAAAAAAyEBZs2bV8uXLtWfPHv3rX//SypUr9eOPP6pfv36JznV0dEzwsYND8he9OTklrHXi4uISnR8XFydJWrZsmbJly5bgMXd39zT9OTIKl/UBAAAAAABkoLCwMM2ZM0evv/66BgwYoA0bNujhw4eJyqW0On78uOV/x8XFKTw8PNFKqCJFisjR0VG3bt2Sp6enPD095ebmptGjR+v69etWvX56YeUUAAAAAABABnJ1ddWMGTOUL18+Va9eXXv37tX9+/d169Ytq553z549WrBggWrVqqUlS5YoOjpab7/9doJz3Nzc1KpVKwUGBmrYsGFyd3fX6NGjdfHiRRUuXNiq108vlFMAAAAAAOC54OLiYpOvUaZMGY0cOVIzZ87UsGHDVLBgQY0fP1758uWzKou/v79+//13TZkyRWXLltX333+vnDlzJjqvf//+Gjt2rHr16qVHjx6patWqmjt3bqJLCI1COQUAAAAAAOye2WyWh4dHpr2WyWRK0+e88847eueddxIdf3LHvMKFCye6e15oaGiCj8eMGZPg41y5ciU69s/nlR7veRUYGKjAwMA0Zc4s7DkFAAAAAADsXlrLInt5rRcB5RQAAAAAAAAMw2V9AAAAAAAAdia5y/nsESunAAAAAAAAYBjKKQAAAAAAABiGcgoAAAAAANgNs9lsdASkQlr+niinAAAAAACAzXN2dpYk3b9/3+AkSI2YmBhJkqOj41PPZUN0AAAAAABg8xwdHZU7d25duXJFkpQtWzaZTCaDUyEp8fHxunr1qrJlyyYnp6dXT5RTAAAAAADALrz88suSZCmoYLscHBxUtGjRVBWIlFMAAAAAAMAumEwmeXh4KH/+/Hr06JHRcZACFxcXOTikbjcpyikAAAAAAGBXHB0dU7WXEewDG6IDAAAAAADAMJRTAAAAAAAAMAzlFAAAAAAAAAxDOQUAAAAAAADDUE4BAAAAAADAMJRTAAAAAAAAMAzlFAAAAAAAAAxDOQUAAAAAAADDUE4BAAAAAADAMJRTAAAAAAAAMAzlFAAAAAAAAAxDOQUAAAAAAADDUE4BAAAAAADAMJRTAAAAAAAAMIyh5dTDhw/19ddfy9fXV35+flqwYEGy527evFn169eXj4+P2rRpo2PHjmViUgAAAAAAAGQEQ8upcePG6ejRo1q4cKGGDBmi6dOna8OGDYnO+/PPP9W3b19169ZNISEhKlOmjLp166bo6GgDUgMAAAAAACC9GFZO3b9/X8uXL9fAgQPl7e2tunXrqnPnzlq6dGmic3fs2KGSJUuqadOmKlq0qD7//HNdvXpVJ0+eNCA5AAAAAAAA0oth5VR4eLhiY2Pl4+NjOValShUdOnRI8fHxCc7NnTu3Tp48qf379ys+Pl4rV66Um5ubihYtmtmxAQAAAAAAkI6cjHrhq1evKk+ePHJxcbEcy5cvnx4+fKhbt24pb968luMNGjRQaGio2rZtK0dHRzk4OGjOnDnKlSuXEdEBAAAAAACQTgwrp6KjoxMUU5IsH8fExCQ4fvPmTV29elWDBw9WxYoV9eOPP2rAgAEKCgqSu7t7ml43Li7OuuB/4+jomG7PlVHS88+b3pifdWx9fszOOszPOszPOszPOszPOszPOszPOszPOszPOuk1P1v+ewCSY1g5lSVLlkQl1JOPXV1dExyfMGGCXn31Vb3//vuSpOHDh6t+/fpasWKFunbtmqbXPXLkiBWp/ydr1qwqW7ZsujxXRjpx4oRNbhzP/KxjD/NjdtZhftZhftZhftZhftZhftZhftZhftZhftax1fkBmcGwcqpAgQK6efOmYmNj5eT0OMbVq1fl6uqqnDlzJjj32LFjateuneVjBwcHeXl56eLFi2l+3fLly9tFa55eSpcubXQEu8b8nh2zsw7zsw7zsw7zsw7zsw7zsw7zsw7zsw7zs056zS8uLi7dFmUAmcWwcqpMmTJycnLSwYMH5evrK0nav3+/ypcvLweHhPu058+fXxEREQmOnT59WuXLl0/z6zo6Or5Q5dSL9GfNCMzv2TE76zA/6zA/6zA/6zA/6zA/6zA/6zA/6zA/6zA/vMgMu1tf1qxZ1bRpUwUGBurw4cPasmWLFixYoPbt20t6vIrqwYMHkqTWrVvrl19+UXBwsM6ePasJEybo4sWLatasmVHxAQAAAAAAkA4MWzklSQMGDFBgYKA6dOggNzc39ezZU/Xq1ZMk+fn5afTo0WrevLkaNGige/fuac6cOYqMjFSZMmW0cOHCNG+GDgAAAAAAANtiaDmVNWtWjR07VmPHjk302IkTJxJ83KpVK7Vq1SqzogEAAAAAACATGHZZHwAAAAAAAEA5BQAAAAAAAMNQTgEAAAAAAMAwlFMAAAAAAAAwDOUUAAAAAAAADEM5BQAAAAAAAMNQTgEAAAAAAMAwlFMAAAAAAAAwDOUUAAAAAAAADEM5BQAAAAAAAMNQTgEAAAAAAMAwlFMAAAAAAAAwDOUUAAAAAAAADEM5BQAAAAAAAMNQTgEAAAAAAMAwlFMAAAAAAAAwDOUUAAAAAAAADEM5BQAAAAAAAMNQTgEAAAAAAMAwlFMAAAAAAAAwDOUUAAAAAAAADEM5BQAAAAAAAMNQTgEAAAAAAMAwlFMAAAAAAAAwDOUUAAAAAAAADEM5BQAAAAAAAMNQTgEAAAAAAMAwlFMAAAAAAAAwDOUUAAAAAAAADEM5BQAAAAAAAMNQTgEAAAAAAMAwlFMAAAAAAAAwDOUUAAAAAAAADEM5BQAAAAAAAMNQTgEAAAAAAMAwlFMAAAAAAAAwDOUUAAAAAAAADEM5BQAAAAAAAMNQTgEAAAAAAMAwlFMAAAAAAAAwDOUUAAAAAAAADEM5BQAAAAAAAMNQTgEAAAAAAMAwlFMAAAAAAAAwDOUUAAAAAAAADEM5BQAAAAAAAMNQTgEAAAAAAMAwlFMAAAAAAAAwDOUUAAAAAAAADEM5BQAAAAAAAMNQTgEAAAAAAMAwlFMAAAAAAAAwDOUUAAAAAAAADEM5BQAAAAAAAMNQTgEAAAAAAMAwlFMAAAAAAAAwjNOzfuLFixd1/fp1OTg4KF++fCpQoEB65gIAAAAAAMALIE3l1P79+/Xjjz9q+/btunXrluW4yWSSu7u7atasqVatWqly5crpnRMAAAAAAADPoVSVUxEREQoMDNSNGzdUu3ZtTZ48WSVKlFDu3LllNpt18+ZNnThxQvv379dXX32lAgUKaOjQoSpZsmRG5wcAAAAAAIAdS1U5NXToUPXo0UPVqlVL8vGXX35ZL7/8st566y19/vnn2r59u4YOHarFixena1gAAAAAAAA8X1JVTi1atChNT+rn5yc/P79nCoQXi4uLi9ERkmSruQAAAAAAeN4884bokhQTE6PTp0/LbDarWLFiypIlS3rlwgvAbDbLw8PD6BjJMpvNMplMRscAAAAAAOC59szl1M6dO/Xll19KkmJjY2UymTRmzBjVqlUrvbLhOWcymTQjNEIXbkUbHSWRQrmz6lP/EkbHAAAAAADguffM5dTIkSP13XffqWzZspKkdevWKTAwUP/+97/TKxteADsiris8MsroGIl4vexGOQUAAAAAQCZwSM1JzZs3V2hoaMJPdHBQZGSkYmJi9ODBA129elWOjo4ZEhIAAAAAAADPp1SVU2PGjNGqVavUokULbd26VZI0YsQITZgwQRUqVJCPj4+WLVum0aNHZ2hYAAAAAAAAPF9SVU69+uqrmjJlisaMGaO1a9eqRYsWunr1qtatW6fdu3drz5492rhxo1577bU0vfjDhw/19ddfy9fXV35+flqwYEGy5544cUJt2rRRhQoV1LhxY/3+++9pei0AAAAAAADYnlSVU0+UKlVKkyZN0rhx47R+/Xo1b95ce/fuVY4cOZ7pxceNG6ejR49q4cKFGjJkiKZPn64NGzYkOu/u3bvq1KmTSpYsqdWrV6tu3brq0aOHrl+//kyvCwAAAAAAANuQqnLqxo0bGjt2rLp166bAwEBlyZJFEydO1Pjx47Vx40Y1b95cmzdvTtML379/X8uXL9fAgQPl7e2tunXrqnPnzlq6dGmic4OCgpQtWzYFBgbK09NTvXr1kqenp44ePZqm1wQAAAAAAIBtSVU59dlnn+ncuXPy9/dXfHy8OnbsKEkqUaKExo8fr0mTJmnLli1q3rx5ql84PDxcsbGx8vHxsRyrUqWKDh06pPj4+ATn7tmzRwEBAQk2XF+xYoXeeuutVL8eAAAAAAAAbI9Tak46evSoli9fruLFiysmJkaVKlXSjRs3lDdvXknSK6+8orFjx+rcuXOpfuGrV68qT548cnFxsRzLly+fHj58qFu3blmeW5LOnz+vChUq6JtvvlFoaKgKFSqkfv36qUqVKql+vSfi4uLS/DnJsYe7E6bnnze9MT/r2Pr8mJ11mJ91mJ91mJ91mJ91mJ91mJ91mJ91mJ910mt+tvz3ACQnVeVUvXr11K1bN1WsWFERERGqXLlygvLoiaJFi6b6haOjoxMUU5IsH8fExCQ4fv/+fc2dO1ft27fXd999p7Vr1+qjjz7S+vXr5eHhkerXlKQjR46k6fzkZM2aVWXLlk2X58pIJ06cUHR0tNExEmF+1rGH+TE76zA/6zA/6zA/6zA/6zA/6zA/6zA/6zA/69jq/IDMkKpyatSoUdq6datOnz6t2rVrq169ela/cJYsWRKVUE8+dnV1TXDc0dFRZcqUUa9evSRJZcuW1Y4dOxQSEqKPP/44Ta9bvnx5u2jN00vp0qWNjmDXmN+zY3bWYX7WYX7WYX7WYX7WYX7WYX7WYX7WYX7WSa/5xcXFpduiDCCzpKqcunPnjurUqZOmJ759+7Zy5cqV7OMFChTQzZs3FRsbKyenxzGuXr0qV1dX5cyZM8G5L730kooXL57g2CuvvKJLly6lKZP0uOh6kcqpF+nPmhGY37NjdtZhftZhftZhftZhftZhftZhftZhftZhftZhfniRpWpD9A8//FDTp0/XtWvXnnrupUuXNHnyZHXo0CHF88qUKSMnJycdPHjQcmz//v0qX768HBwSxqpUqZJOnDiR4NipU6dUqFCh1MQHAAAAAACAjUrVyqmffvpJc+fOVYMGDVS8eHG98cYbKlGihPLkyaO4uDjdunVLJ06c0P79+xUREaG2bdvqp59+SvE5s2bNqqZNmyowMFCjRo3SlStXtGDBAo0ePVrS41VUOXLkkKurq9577z0tWbJE3377rZo0aaLg4GCdP39e77zzjvUTAAAAAAAAgGFSVU65uLioR48e6tSpk1avXq3//Oc/Cg4O1o0bN2QymeTu7q6yZcuqefPmql+/vtzc3FL14gMGDFBgYKA6dOggNzc39ezZ07KflZ+fn0aPHq3mzZurUKFCmjdvnkaOHKm5c+eqRIkSmjt3rgoUKPDsf3IAAAAAAAAYLlXl1BPZsmXTu+++q3fffTddXjxr1qwaO3asxo4dm+ixf17GV6VKFa1cuTJdXhcAAAAAAAC2IVV7TgEAAAAAAAAZgXIKAAAAAAAAhqGcAgAAAAAAgGEopwAAAAAAAGCYNG2I/kRUVJROnjyp2NhYmc3mBI9VrVo1XYIBAAAAAADg+ZfmciokJESBgYGKjo5O9JjJZNLx48fTJRgAAAAAAACef2kupyZPnqxWrVqpV69ecnNzy4hMAAAAAAAAeEGkec+pW7duqX379hRTAAAAAAAAsFqay6natWtr06ZNGZEFAAAAAAAAL5g0X9ZXoEABTZ48WevXr5enp6ecnZ0TPD569Oh0CwcAAAAAAIDnW5rLqdu3b6tRo0YZkQUAAAAAAAAvmDSXU6yMAgAAAAAAQHpJczklSVu2bNG8efN06tQpxcXFqVixYvrggw/UtGnTdI4HAAAAAACA51may6mffvpJY8eO1QcffKCuXbsqPj5eBw4c0NChQ/Xo0SO1atUqI3ICAAAAAADgOZTmcmrevHkaMmRIglVSderUUalSpTR79mzKKQAAAAAAAKSaQ1o/4fr166pUqVKi4z4+Prp06VJ6ZAIAAAAAAMALIs3lVJkyZRQcHJzoeFBQkEqWLJkemQAAAAAAAPCCSPNlfV9++aU6duyo3bt3q2LFipKkgwcPKjw8XLNnz073gAAAAAAAAHh+pXnllI+Pj1auXKkKFSooIiJCf/31l6pWrar169fr9ddfz4iMAAAAAAAAeE6leeWUJJUoUUIDBgxI7ywAAAAAAAB4waSqnGrfvr2mT5+unDlzql27djKZTMmeu2jRonQLBwAAAAAAgOdbqsqp1157Tc7OzpKkatWqZWggAAAAAAAAvDhSVU716NHD8r8LFy6sBg0ayMXFJcE59+/f16+//pq+6QAAAAAAAPBcS1U5dePGDT148ECSNGDAAJUqVUp58uRJcE54eLgmTJig9u3bp39KAAAAAAAAPJdSVU7t2bNHffr0kclkktlsVosWLRLsO2U2myVJTZo0yZiUAAAAAAAAeC6lqpx6++23FRoaqvj4eNWpU0fLly9X3rx5LY+bTCZlzZo10WoqAAAAAAAAICWpKqckqWDBgpKk2rVrK1u2bCpUqFCGhQIAAAAAAMCLwSGtnxAWFiYnp1R3WgAAAAAAAECy0twytW3bVp999pnee+89FSxYUFmyZEnweNWqVdMtHAAAAAAAAJ5vaS6nZs6cKUkaPHhwosdMJpOOHz9ufSoAAAAAAAC8ENJcToWHh2dEDgAAAAAAALyAnmnzqAcPHmjVqlWKiIhQXFycihcvrgYNGih37tzpHA8AAAAAAADPszRviP7f//5X9erV06xZs3Tx4kVdvHhRc+bMUf369XXy5MmMyAgAAAAAAIDnVJpXTo0cOVI1atTQ8OHDLXfti42N1aBBgzRq1CgtWLAg3UMCAAAAAADg+ZTmlVMHDx5Uly5dLMWUJDk5OalLly4KCwtL13AAAAAAAAB4vqW5nHrppZd07ty5RMfPnTun7Nmzp0soAAAAAAAAvBjSfFnfe++9p0GDBql3796qUKGCJOnQoUOaNm2aWrVqle4BAQAAAAAA8PxKczn10UcfKTo6WhMmTNDt27clSfny5VPHjh3VqVOndA8IAAAAAACA51eayymTyaSePXuqZ8+eun79urJkySI3N7eMyAYAAAAAAIDnXJrLKUnauXOnfv75Z506dUomk0mlS5fW+++/r0qVKqVzPAAAAAAAADzP0rwh+vLly9W1a1dlzZpV7777rlq0aCFJat++vTZt2pTuAQEAAAAAAPD8SvPKqVmzZmno0KGWUuqJqlWrauLEiapXr166hQMAAAAAAMDzLc0rp27duqWKFSsmOu7r66srV66kSygAAAAAAAC8GNJcTr3//vsaO3asbt68aTkWHR2t2bNnq23btukaDgAAAAAAAM+3NF/Wt3//fh0+fFi1atVS0aJF5ezsrLNnz+revXsqWLCgNmzYYDl369at6RoWAAAAAAAAz5c0l1OtWrVSq1atMiILAAAAAAAAXjBpLqeaNWsm6fGlfGfPnlV8fLyKFi0qNze3dA8HAAAA2BIXFxejIyTJVnMBAJAaaS6nHj16pPHjx2vZsmWKi4uT2WyWk5OTGjdurKFDh/KNEQAAAM8ls9ksDw8Po2Mky2w2y2QyGR0DAIA0S3M5NXbsWG3btk2zZs2Sj4+P4uPjFRYWphEjRmjy5Mnq169fRuQEAAAADGUymTQjNEIXbkUbHSWRQrmz6lP/EkbHAADgmaS5nFqzZo2mTp2qatWqWY699dZbypIli7744gvKKQAAADy3dkRcV3hklNExEvF62Y1yCgBgtxzS+glms1nu7u6JjufNm1f37t1Ll1AAAAAAAAB4MaS5nHr99dc1YcIERUX97zdGd+7c0aRJkxKspgIAAAAAAACeJs2X9X399ddq3769atasqWLFikmSTp8+rSJFimjWrFnpHhAAAAAAAADPrzSXUzly5NCaNWv022+/6dSpU8qSJYuKFSumGjVqyMEhzQuxAAAAAAAA8AJLcznVqFEjTZ8+XQEBAQoICMiITAAAAAAAAHhBpHmpk4ODgx49epQRWQAAAAAAAPCCSfPKqVq1aunDDz9U7dq1VahQIbm4uCR4vEePHukWDgAAAAAAAM+3NJdTJ06ckLe3t65cuaIrV64keMxkMqVbMAAAAAAAADz/0lxOLV68OCNyAAAAAAAA4AWU6nIqJCREmzdvlrOzs+rUqaOGDRtmZC4AAAAAAAC8AFK1IfrChQv19ddf68GDB4qOjla/fv00adKkjM4GAAAAAACA51yqVk799NNPGjlypJo2bSpJ2rRpkwYMGKDPPvuMfaYAAAAAAADwzFK1cur8+fOqXr265WN/f39FR0cn2hAdAAAAAAAASItUlVOxsbFycvrfIisnJydlyZJFMTExGRYMAAAAAAAAz79UlVMAAAAAAABARkj13frWr18vNzc3y8fx8fHavHmz8ubNm+C8J/tSpcbDhw81dOhQbdq0Sa6ururUqZM6deqU4uf89ddfaty4sWbPnq1q1aql+rUAAAAAAABge1JVThUsWFALFixIcMzd3V1LlixJcMxkMqWpnBo3bpyOHj2qhQsX6uLFi+rXr58KFiyot99+O9nPCQwM1P3791P9GgAAAAAAuLi4GB0hSbaaC8hMqSqnQkND0/2F79+/r+XLl+u7776Tt7e3vL299eeff2rp0qXJllOrVq3SvXv30j0LAAAAAOD5ZTab5eHhYXSMZJnNZplMJqNjAIZJ9WV96S08PFyxsbHy8fGxHKtSpYpmz56t+Ph4OTgk3A7r5s2bGj9+vBYsWKBGjRpldlwAAAAAgJ0ymUyaERqhC7eijY6SSKHcWfWpfwmjYwCGMqycunr1qvLkyZNgCWO+fPn08OFD3bp1K9FeVmPGjFGzZs1UqlSpzI4KAAAAALBzOyKuKzwyyugYiXi97EY5hReeYeVUdHR0omtrn3wcExOT4PjOnTu1f/9+rVmzxurXjYuLs/o5nnB0dEy358oo6fnnTW/Mzzq2Pj9mZx3mZx3mZx3mZx3mZx3mZx3mZx3mZx3mZ530mp8t/z0AyTGsnMqSJUuiEurJx66urpZjDx480ODBgzVkyJAEx5/VkSNHrH4OScqaNavKli2bLs+VkU6cOKHoaNtbusr8rGMP82N21mF+1mF+1mF+1mF+1mF+1mF+1mF+1mF+1rHV+QGZwbByqkCBArp586ZiY2Pl5PQ4xtWrV+Xq6qqcOXNazjt8+LDOnz+vXr16Jfj8Ll26qGnTpho2bFiaXrd8+fJ20Zqnl9KlSxsdwa4xv2fH7KzD/KzD/KzD/KzD/KzD/KzD/KzD/KzD/KyTXvOLi4tLt0UZQGYxrJwqU6aMnJycdPDgQfn6+kqS9u/fr/LlyyfYDL1ChQratGlTgs+tV6+eRowYoRo1aqT5dR0dHV+ocupF+rNmBOb37JiddZifdZifdZifdZifdZifdZifdZifdZifdZgfXmSGlVNZs2ZV06ZNFRgYqFGjRunKlStasGCBRo8eLenxKqocOXLI1dVVnp6eiT6/QIECcnd3z+zYAAAAAAAASEcOTz8l4wwYMEDe3t7q0KGDhg4dqp49e6pevXqSJD8/P61bt87IeAAAAAAAAMhghq2ckh6vnho7dqzGjh2b6LETJ04k+3kpPQYAAAAAAAD7YejKKQAAAAAAALzYKKcAAAAAAABgGMopAAAAAAAAGIZyCgAAAAAAAIahnAIAAAAAAIBhKKcAAAAAAABgGMopAAAAAAAAGIZyCgAAAAAAAIahnAIAAAAAAIBhKKcAAAAAAABgGMopAAAAAAAAGIZyCgAAAAAAAIahnAIAAAAAAIBhKKcAAAAAAABgGMopAAAAAAAAGIZyCgAAAAAAAIahnAIAAAAAAIBhKKcAAAAAAABgGMopAAAAAAAAGIZyCgAAAAAAAIahnAIAAAAAAIBhKKcAAAAAAABgGMopAAAAAAAAGIZyCgAAAAAAAIahnAIAAAAAAIBhKKcAAAAAAABgGMopAAAAAAAAGIZyCgAAAAAAAIahnAIAAAAAAIBhKKcAAAAAAABgGMopAAAAAAAAGIZyCgAAAAAAAIahnAIAAAAAAIBhKKcAAAAAAABgGMopAAAAAAAAGIZyCgAAAAAAAIahnAIAAAAAAIBhKKcAAAAAAABgGMopAAAAAAAAGIZyCgAAAAAAAIahnAIAAAAAAIBhKKcAAAAAAABgGMopAAAAAAAAGIZyCgAAAAAAAIahnAIAAAAAAIBhKKcAAAAAAABgGMopAAAAAAAAGMbJ6AAAAPvj4uJidIQk2WouAAAAAMmjnAIApInZbJaHh4fRMZJlNptlMpmMjgEAAAAglSinAABpYjKZNCM0QhduRRsdJZFCubPqU/8SRscAAAAAkAaUUwCANNsRcV3hkVFGx0jE62U3yikAAADAzrAhOgAAAAAAAAxDOQUAAAAAAADDUE4BAAAAAADAMJRTAAAAAAAAMAzlFAAAAAAAAAxDOQUAAAAAAADDUE4BAAAAAADAMJRTAAAAAAAAMAzlFAAAAAAAAAxDOQUAAAAAAADDUE4BAAAAAADAMJRTAAAAAAAAMAzlFAAAAAAAAAxDOQUAAAAAAADDUE4BAAAAAADAMIaWUw8fPtTXX38tX19f+fn5acGCBcme++9//1vvvPOOfHx81LhxY23dujUTkwIAAAAAACAjGFpOjRs3TkePHtXChQs1ZMgQTZ8+XRs2bEh0Xnh4uHr06KEWLVooODhY7733nnr37q3w8HADUgMAAAAAACC9OBn1wvfv39fy5cv13XffydvbW97e3vrzzz+1dOlSvf322wnOXbNmjV5//XW1b99ekuTp6anQ0FCtX79eXl5eRsS3cHFxMfT1k2OruQAAAAAAAP7OsHIqPDxcsbGx8vHxsRyrUqWKZs+erfj4eDk4/G9RV7NmzfTo0aNEz3H37t1MyZocs9ksDw8PQzOkxGw2y2QyGR0DAAAAAAAgWYaVU1evXlWePHkSrPDJly+fHj58qFu3bilv3ryW4yVKlEjwuX/++ad27dql9957L82vGxcX9+yh/8HR0VEzQiN04VZ0uj1neimUO6s+9S+Rrn/e9Obo6Gh0hKdifs+O2VmH+VmH+VmH+VmH+VmH+VmH+VmH+VmH+VknveZny38PQHIMK6eio6MTXXr25OOYmJhkP+/GjRvq2bOnKleurICAgDS/7pEjR9L8OUnJmjWrypYtqx0R1xUeGZUuz5mevF5206f+JXTixAlFR9teefZkfraO+T07Zmcd5mcd5mcd5mcd5mcd5mcd5mcd5mcd5mcdW50fkBkMK6eyZMmSqIR68rGrq2uSn3Pt2jV9+OGHMpvNmjZtWoJL/1KrfPnydtGap5fSpUsbHcGuMb9nx+ysw/ysw/ysw/ysw/ysw/ysw/ysw/ysw/ysk17zi4uLS7dFGUBmMaycKlCggG7evKnY2Fg5OT2OcfXqVbm6uipnzpyJzr98+bJlQ/RFixYluOwvLRwdHV+ocupF+rNmBOb37JiddZifdZifdZifdZifdZifdZifdZifdZifdZgfXmRpX3qUTsqUKSMnJycdPHjQcmz//v0qX758ohVR9+/fV+fOneXg4KAlS5aoQIECmZwWAAAAAAAAGcGwcipr1qxq2rSpAgMDdfjwYW3ZskULFiywrI66evWqHjx4IEmaM2eOzp07p7Fjx1oeu3r1quF36wMAAAAAAIB1DLusT5IGDBigwMBAdejQQW5uburZs6fq1asnSfLz89Po0aPVvHlzbdy4UQ8ePFCrVq0SfH6zZs00ZswYI6IDAAAAAAAgHRhaTmXNmlVjx461rIj6uxMnTlj+94YNGzIzFgAAAAAAADKJoeUUAABAWrm4uBgdIUm2mgsAAMDWUU4BAAC7YTab5eHhYXSMZJnNZplMJqNjAAAA2BXKKQAAYDdMJpNmhEbowq1oo6MkUih3Vn3qX8LoGAAAAHaHcgoAANiVHRHXFR4ZZXSMRLxedqOcAgAAeAYORgcAAAAAAADAi4tyCgAAAAAAAIahnAIAAAAAAIBhKKcAAAAAAABgGDZEBwAAAAA74OLiYnSEJNlqLgD2g3IKAAAAAGyc2WyWh4eH0TGSZTabZTKZjI4BwE5RTgEAAACAjTOZTJoRGqELt6KNjpJIodxZ9al/CaNjALBjlFMAAAAAYAd2RFxXeGSU0TES8XrZjXIKgFXYEB0AAAAAAACGoZwCAAAAAACAYSinAAAAAAAAYBjKKQAAAAAAABiGcgoAAAAAAACGoZwCAAAAAACAYSinAAAAAAAAYBjKKQAAAAAAABiGcgoAAAAAAACGoZwCAAAAAACAYZyMDgAARnBxcTE6QpJsNRcAAAAAZBTKKQAvHLPZLA8PD6NjJMtsNstkMhkdAwAAAAAyBeUUgBeOyWTSjNAIXbgVbXSURArlzqpP/UsYHQMAgAxhqyuEbTUXALwoKKcAvJB2RFxXeGSU0TES8XrZjXIKAPBcYuUyACA5lFMAAAAAMhwrlwEAyaGcAgAAAJApWLkMAEiKg9EBAAAAAAAA8OKinAIAAAAAAIBhKKcAAAAAAABgGMopAAAAAAAAGIZyCgAAAAAAAIahnAIAAAAAAIBhKKcAAAAAAABgGCejAwB4di4uLkZHSMQWMwEAAAAAbBflFGCnzGazPDw8jI6RJLPZLJPJZHQMAAAAAIAdoJwC7JTJZNKM0AhduBVtdJQECuXOqk/9SxgdAwAAAABgJyinADu2I+K6wiOjjI6RgNfLbpRTAAAAAIBUY0N0AAAAAAAAGIaVUwAAZDJbvXGAreZC+rLVv2dbzQUAADIe5RQAAJnIlm9mIHFDg+cdX38AAMAWUU4BAJCJbPVmBhI3NHgR8PUHAABsEeUUAACZzBZvZiBxQ4MXBV9/AADA1rAhOgAAAAAAAAxDOQUAAAAAAADDUE4BAAAAAADAMJRTAAAAAAAAMAzlFAAAAAAAAAxDOQUAAAAAAADDUE4BAAAAAADAMJRTAAAAAAAAMAzlFAAAAAAAAAxDOQUAAAAAAADDUE4BAAAAAADAMJRTAAAAAAAAMAzlFAAAAAAAAAxDOQUAAAAAAADDUE4BAAAAAADAMJRTAAAAAAAAMAzlFAAAAAAAAAxDOQUAAAAAAADDUE4BAAAAAADAMJRTAAAAAAAAMAzlFAAAAAAAAAxjaDn18OFDff311/L19ZWfn58WLFiQ7Ll//PGHWrVqpYoVK6pFixY6evRoJiYFAAAAAABARjC0nBo3bpyOHj2qhQsXasiQIZo+fbo2bNiQ6Lz79++ra9eu8vX11cqVK+Xj46Nu3brp/v37BqQGAAAAAABAejGsnLp//76WL1+ugQMHytvbW3Xr1lXnzp21dOnSROeuW7dOWbJk0VdffaUSJUpo4MCByp49e5JFFgAAAAAAAOyHYeVUeHi4YmNj5ePjYzlWpUoVHTp0SPHx8QnOPXTokKpUqSKTySRJMplMqly5sg4ePJiZkQEAAAAAAJDODCunrl69qjx58sjFxcVyLF++fHr48KFu3bqV6Nz8+fMnOObu7q7IyMjMiAoAAAAAAIAM4mTUC0dHRycopiRZPo6JiUnVuf88LyVms9ny3I6Ojs8SORFHR0eVeimbXGzwnoee7tkUFxenuLg4o6Mki/lZx1bnx+ysw/ysw/ysw/ysw/ysw/ysw/ysw/ysw/ysk97ze/I8T94DA/bAZDboK3b9+vUaMWKEduzYYTkWERGhBg0aaPfu3cqdO7fleNeuXfXqq6/qiy++sBwbP368IiIiNHv27FS9XkxMjI4cOZJu+QEAAAAAsFXly5dPtMgDsFWGrZwqUKCAbt68qdjYWDk5PY5x9epVubq6KmfOnInOvXbtWoJj165dS3SpX0qcnJxUvnx5OTg4WPauAgAAAADgeWI2mxUfH295nw3YA8O+WsuUKSMnJycdPHhQvr6+kqT9+/dbCqS/q1ixor777juZzWaZTCaZzWYdOHBAH3/8capfz8HBgdYYAAAAAADAxhh2xW3WrFnVtGlTBQYG6vDhw9qyZYsWLFig9u3bS3q8iurBgweSpLffflt37tzRyJEjdfLkSY0cOVLR0dGqX7++UfEBAAAAAACQDgzbc0p6vNF5YGCgNm3aJDc3N3300Ufq2LGjJKl06dIaPXq0mjdvLkk6fPiwhgwZooiICJUuXVpDhw5V2bJljYoOAAAAAACAdGBoOQUAAAAAAIAXmw3eSBMAAAAAAAAvCsopAAAAAAAAGIZyCgAAAAAAAIahnAIAAAAAAIBhKKcAAAAAAABgGMopAxw+fFgxMTGWj7ds2aLhw4dr+vTpioyMNDAZXjSPHj3StWvX9OjRI6OjALp165Z27dpldAy7Fh8fb3QEu/b3781AZoqJidG6deuMjgEAgGFMZrPZbHSIF8W1a9fUuXNnnThxQmvXrlXx4sU1e/ZsTZ06VRUrVpSbm5uOHDmipUuXqmTJkkbHtUkDBgxI8rizs7Ny5MihsmXLqm7dunJxccnkZPZl2bJlWr58ucLDwy3HSpcurdatW6tt27YGJsOL7D//+Y+6du2q48ePGx3FLjG/Z3fw4EEFBQVpw4YN2r17t9Fx7E5MTIy2bNmioKAgfffdd0bHsSthYWGWr727d+/y328S+NnPOtOnT9dHH32krFmzWo5FRkYqf/78cnB4vE7hzp076tOnjxYsWGBUTACQk9EBXiSTJ09W9uzZ9e9//1sFChTQ7du3NXPmTNWsWVNz586VJE2dOlUTJ07UrFmzDE5rXx4+fKirV69q9erVmjJlihYvXqyXX37Z6Fg2Jy4uTt27d9e+ffvUvHlzdenSRbly5dKVK1d05MgRjR07Vtu2bdOsWbMsP7AgsRMnTsjJyUnFixeXyWRK8Fh4eLhGjBihJUuWGJQOQGpcunRJISEhCg4O1tmzZ5UzZ041b97c6Fh25cCBAwoODtb69et19+5dlStXzuhIduHSpUsKDg5WcHCwzp07p+zZs6tJkyZq06aN0dHsCj/7pc6MGTPUpk2bBOVUgwYNFBISoiJFikh6XDCzcjlpXl5eiX7WkyRHR0flzJlTZcqU0Ycffig/Pz8D0gHPF8qpTLRt2zZNnz5dBQoUsHz86NEjvfvuu5Zz6taty5vaFIwePTrFx+Pi4vT5559r3LhxmjRpUialsh8LFy7UyZMntXbtWnl4eCR4rFmzZurSpYs6dOigRYsWqWPHjsaEtGF//vmnevTooXPnzkmSSpUqpTlz5sjDw0NRUVGaNGmSfvrpJ8sPewBsS3R0tDZu3Kjg4GDt2bNHkmQ2m9WvXz+1bduWlRepcPHiRQUHByskJERnz56VyWRSgwYN1LFjR5UvX97oeDbryddeUFCQ9u7dK2dnZ73xxhs6f/68lixZIi8vL6Mj2ix+9rNOUhfJcOFM6i1atCjJ4/Hx8bp7964OHjyoXr16afz48QoICMjkdMDzhXIqE92+fVv58+e3fLxr1y45OTmpevXqlmM5cuRQbGysEfGeC46OjurUqZO6d+9udBSbFBQUpC+//DJRMfWEh4eHvvzyS02bNo1yKgkjRoyQm5ubli5dKmdnZ02dOlUjRoxQnz591LVrV926dUu9e/fWhx9+aHRUAH/z+++/KyQkRBs2bFBsbKyqV6+uoUOHKiAgQDVr1pSfnx/FVAru37+vjRs3auXKldq3b5/c3NxUq1Yt9e3bV5999pm6d+/OdgQp6NevnzZv3ixnZ2fVrFlTEydO1FtvvaVs2bLJ29tbTk78OG4NfvZDRnrttddSfLxu3boqUKCAZs+eTTkFWInvhpmoaNGi+vPPP1WwYEE9evRI27Zt02uvvaZs2bJZztmxYwerLqyUL18+3b9/3+gYNuncuXOqUKFCiueUK1dO58+fz6RE9uXo0aOaN2+efHx8JEmjRo1S3bp1FR4erpIlS2rYsGHJFn8vuuDg4Keec+LEiYwPYqcuXrz41HOuX7+eCUnsU8eOHeXp6akhQ4aoXr16Cb7v4ulq1Kghd3d3+fv7q3v37nrttdcoVNIgJCREnp6e+uCDD1StWjW9+uqrRkd67vCzH4zk5+enyZMnGx0DsHv8ZJGJ3n33XQUGBurDDz/Uvn37dOPGDcvqlEePHum3337T5MmT9fHHHxsb1M4dOnRIhQsXNjqGTcqRI4cuX76sQoUKJXvOxYsXlTdv3kxMZT/u3buXoDx+shKyRo0aGjZsmFGx7MK0adNSdR7lXtL8/f2T3PPi78xm81PPeVF9+umnWrt2rQYOHKhFixbJ399fderU4VKqVCpXrpzCwsJ04MABOTo6ytnZWVWrVjU6lt3YsmWL1q1bp+XLl2vUqFEqWLCg6tSpo4CAAP6bTSf87Jc8k8nE11kGM5vNrL4F0gHlVCZq3769pMcrCEwmk8aMGaOaNWtKkkaOHKnly5frvffeU4cOHYyMadOSWz0QHx+vqKgohYWFacqUKerVq1cmJ7MPtWvX1owZMzRv3rwkf1Axm82aOXOm/P39DUhnH/45NwcHB/6bTYXQ0NBUnRcTE5PBSezT1q1bjY5g13r27KmePXvq2LFjWrNmjVasWKEZM2aoYMGCMpvNOnPmDJelpWDx4sW6fPmy1q9frzVr1uj7779X7ty5Vbt2bUnsX/M0hQsXVteuXdW1a1edPHlSa9as0fr167Vw4UJJ0vz589WuXTuVLVvW4KS2iZ/9rGM2mzVixAhlyZLFcuzRo0caP368smfPLunx5vJ4dj///PNTr0wA8HQmMz9R2ITIyEi5uLiwYuUpkrtjxpMv4yJFiuiDDz6gLEjG1atX1apVKxUpUkRdu3ZVuXLllCtXLl29elXHjh3TzJkzdfv2bf388898LSbBy8tLO3bskLu7u+WYj4+PVq1axeW4Vjp48KDlduq7d+82Oo7diYmJ0ZYtWxQUFKTvvvvO6Dh2Yd++fVq7dq02btyomzdvqmTJkmrVqpXlF0lI3rlz57RmzRqtW7dOJ0+eVK5cudS4cWO1bNmS1WhpcPjwYa1du1YbNmzQ5cuXVaZMGQUFBRkdy+bws591BgwYkOpzn7b5/Ito+vTpSR43m82WDdEjIiK0ePFiCmbASpRTmejixYvy8PBIcWltTEyMNm7cqMaNG2diMvtx4cKFJI87OTkpR44c7COSCpGRkRo2bJj+9a9/JTju4OCgOnXqaODAgQk27sf/eHl5qVOnTgm+zmbPnq02bdooV65cCc7t0aNHZsezO5cuXVJISIiCg4N19uxZ5cyZU82bN1e/fv2MjmY3Dhw4oODgYK1fv153795VuXLl9Ouvvxody67Ex8dr586dWrt2rbZu3Wq5ix9SJzw8XOvWrdO6det04cIFHT9+3OhIdsdsNmvv3r1au3athg4danQcm8PPfjBSu3btkjzu7OysHDlyqHTp0mrRooXlbuwAnh3lVCYqU6aMtm/fnmDVRdeuXTVixAhLGXDt2jXVrFmTH+6Q4a5fv65jx47p9u3bypUrl8qVK8dqqadI7geUfzKZTMneevhF9+R26sHBwZYSwGw2q1+/fmrbti17NqTCxYsXFRwcrJCQEJ09e1Ymk0kNGjRQx44dVb58eaPj2bWYmBi+Bq1w6NAhVaxY0egYNmfv3r2pPpe9vBKbP3++GjVqxJv/DHT48GGNHTtWS5cuNToKgBcYe05loqR6wL1793Kddxo9ePBAP/30kzZv3qyTJ0/q3r17cnNzU6lSpVS/fn21bNmSNxep4O7urjfffFORkZGKj4/XgwcPdPHiReXKlcuyBwESWrx4sdER7Nbvv/+ukJAQbdiwQbGxsapevbqGDh2qgIAA1axZU35+fvx3m4L79+9r48aNWrlypfbt2yc3NzfVqlVLffv21Weffabu3buzZ9JTHDp0SLNnz9aECROUPXt2+fj46MGDB5bHfX19KZVTcPHiRa1du1bvvvuucubMqYcPH2rixInatWuX8uTJo06dOqlWrVpGx7RJ7dq1S7BqPrnfC5tMJn45mYRffvlFEydOVJUqVdS4cWP93//9X6LVyrDO7du3deDAAaNj2KRGjRqpcePGatSoUYo3FAJgPcop2JXr16+rQ4cOunr1qurUqaM6derIzc1N9+7d04kTJzR58mT9/PPPWrhwoXLnzm10XJu0adMmTZ8+XQsXLlSePHlUv379BG/QSpYsqRUrVlAUIF117NhRnp6eGjJkiOrVq8dlGGlUo0YNubu7y9/fX927d9drr70mJye+hafWgQMH1LFjR73zzjt69OiR5fi4ceNUoEABRUZGauDAgVq1apXeeecdA5PapmPHjql9+/Z66aWX1KhRI+XMmVP9+vXTli1b9OGHHypHjhzq16+fRo8ezQ01klC/fn1t375dJUqUUL169VSnTh0VLVrU6Fh2Y+PGjfrjjz+0YcMGzZs3T8OHD5efn58aN26sgICABBt9A+mtYcOGWrdunaZMmaKKFSuqUaNGatCgAVcbABmAn2xhV8aNG6esWbNq/fr1SX5T6Nevnz766CPNnj1b/fv3NyChbfvXv/6lL774Qt27d0/ww9yiRYtUsGBBRUZGqmvXrlq+fLnef/99A5PapuQ2ZXV0dFTOnDlVpkwZffjhh/Lz8zMgnW379NNPtXbtWg0cOFCLFi2Sv7+/6tSpw+bJqVSuXDmFhYXpwIEDcnR0lLOzM5f/pMGsWbPUrl07ffnll5ZjJpNJlSpVstzM4MSJE/r1118pp5IwZcoUNWrUyLIf0vnz57Vhwwa1bdtWffv2lSTlzZtXc+bMoZxKwuTJk/Xo0SPt2rVLW7ZsUdu2bZUnTx7VrVtX9erV49/BVChbtqzKli2rzz//XEePHtWGDRs0adIkDRo0SAEBAWrUqJH8/Pzk6OhodFQ8Z7p3767u3bvr1KlT2rBhg37++WeNGTNGr7/+uho3bqy6devyCzcgnbDnVCZKzZ2+2HMqZTVq1NDUqVPl6+ub7Dm7du3SkCFDtGnTpkxMZh/atWunmjVrqmvXrpZjlStXVkhIiOVrcM6cOdq6dat++eUXo2LarOQ2So6Pj7fcseXHH3/U+PHjFRAQkMnp7MOxY8e0Zs0abdiwQZGRkSpYsKAuXbqkadOmqU6dOkbHs2mXL1/W+vXrtWbNGh09elS5c+dW7dq1tWrVKgUHB6tUqVJGR7RZVatW1bJlyxLM6J/ff8PDw/XBBx9o3759RsW0Wb6+vvrpp58sl44uWbJEI0eO1OLFiy3fj0+fPq1mzZrp4MGDBia1D2azWWFhYdq6dau2bNmi2NhYy2pwSue0OXLkiLZs2aLNmzfr1q1b2rlzp9GR7NJ//vMfde3alfcfqRQREaH169drw4YN+uuvv1SrVi3LSj4Az46VU5ksLCwswXXyZrNZhw8fVmRkpKTH13wjebdu3bK8kUhOsWLFLPNEQseOHdPw4cMTHPtnP12nTh3Nnj07M2PZjddeey3Fx+vWrasCBQpo9uzZ/ICSDG9vb3l7e6tfv37at2+f1q5dq40bN6pnz54qWbKkWrVqpfbt2xsd0yYVKFBAHTt2VMeOHXXu3DmtWbNG69atU1xcnD744AM1btxYLVu2ZBVGEmJjYxP9ZjsoKEgFCxa0fJwtW7Zk9wJ60cXGxiZYbbtz507lyJFDlStXthx79OiRnJ2djYhnd0wmkypXrqzKlSurZ8+eWrp0qWbOnKlFixZRDqTBjRs3FB4ervDwcF26dIl/+5Lh7++f4p3CJSXY3gFPV6JECfXo0UM9evTQvn37NHLkSPXo0YP/fgErUU5lsqRuL/9kSTyeLi4u7qn7rDg5OSXYUwT/YzKZEi153717d4L9pRwcHNhvygp+fn6aPHmy0TFszoABAzRw4EC5ublZjvn6+srX11fffPONdu7cqbVr12r69OmUU0kIDg5WgwYNLP9tFi1aVJ988ok++eQTnThxQmvXrtW6deu0dOlSfjhOQtGiRRUWFpZgM9tXXnklwTkHDhxgU/lklClTRjt27NB7772nGzduaMeOHfq///s/OTg4WM5ZvXo15UAq3bhxQ6GhoQoNDbVsKN+sWTN+qZEKly9f1qZNm7Rp0yYdOHBApUqVUoMGDTRkyJAEZTP+p2fPnkZHeO5cuHDB8nV45MgRValSRcOGDTM6FmD3KKcyUXh4uNER7J7JZHrqb3+QvBIlSmj79u1q06aN5dg/i6gdO3aodOnSmR3tuWE2myn3khAcHKwvvvgiQTn1hIODg/z8/OTn56eYmBgD0tm+AQMGqGbNmgkuC3+idOnSKl26tD7//HMdOnTIgHS2r0mTJhozZowqV66c5BvYy5cva+rUqfrkk08MSGf7evTooU8//VTbt2/XiRMn5ODgoG7dukl6vFfXypUrtWTJEk2fPt3gpLbr5MmTCg0N1datW3XkyBGVLl1aAQEB6tmzp8qUKWN0PJt27tw5SxFw9OhRFSpUSA0bNtSQIUMolFOhWbNmTz3nypUrCgkJyYQ09isiIkKbN2/Wpk2bdPz4cXl7e6tRo0aaOnWq8ufPb3Q84LlAOWWDfv31V7Vs2dLoGDbJbDarRo0aTz2HAitprVq10pgxY+Tt7a0KFSokevz48eP69ttv+e2PFX7++eckZ/uiS+3lUhR7SUvt/CpWrJjBSexTx44dtXPnTjVs2FAtWrRQlSpVlDt3bt25c0dhYWFauXKlqlWrplatWhkd1SbVqFFDS5Ys0Zo1a1SkSBG1bNlSJUqUkPS4eN61a5cmTpyo2rVrG5zUNtWtW1eRkZGqWrWqmjRpoilTpsjDw8PoWHajXr16ypcvn+rXr69BgwbxPTadPHz4UJs3b1ZQUJB+//13OTk5qUuXLkbHsjlTpkzR5s2bderUKRUrVkwNGzbUlClTuOMmkAHYED0TxcbGau7cudqyZYscHR319ttvq1OnTpYi5fDhwxo2bJiOHTvGZRnJSG5D6qQ8bX+gF9XAgQMVFBSkN998U76+vsqVK5fu3r2rsLAw/etf/1KbNm00cOBAo2PapORWBZjNZsuG6BEREVq8eLHKli2byelsm5eXl6ZPn55gz73ksCFwYl5eXtq5cye3rraC2WzWsmXLtGLFCh0/ftxS+JUuXVrvvvuu2rRpwy82rHT48GGKgyT8/XLHp32N8fNfYrt27VK1atUSXEaKZ7dv3z4FBwdrw4YNunfvnooUKaJ3331XzZs3V548eYyOZ3Nq166tBg0aqFGjRqxyBDIY5VQmGjFihH755Re98847cnFx0erVq9WhQwd9/PHHGjNmjJYuXarixYtr4MCBql69utFx8Rzbtm2bVqxYoYMHD+rmzZvKlSuXKlSooHfffVdvvfWW0fFsVrt27ZI87uzsrBw5cqh06dJq0aKFChQokMnJbF9q96IxmUy8OUuCl5dXqosT5vd0jx490q1bt5QrVy5W61np8uXLCgkJUVBQkM6cOcPXXxL4xZp19u7dm+pz+eVG0v766y8FBwcrJCRE58+f18svv6w6deroxx9/VEhICJdHArAJlFOZqGbNmurbt6+aNm0q6fFG1F999ZV8fX0VGhqqXr16qX379ok2rMb/pHY/C5PJpE8//TSD0wBILS8vL+3YsSPJPZPwdF5eXvr2229TtfKMN7fIaA8ePNCmTZsUHBys3bt3y2w2q2bNmmrTpo1q1apldDw8Z/jlhnU++OADHThwQK+++qreeustBQQEWFY4ent7U049xYABA1J97ujRozMwCfD8Y8+pTHTz5s0EbxqqVaum69evKzw8XKtWrVKRIkUMTGcfdu/eneLjf/31ly5duiRnZ2fKqSRcvHgx1edy15vELl68KA8PjxRXsMTExGjjxo1q3LhxJiazfVwuZZ0nt56n3Hs2qV15ZjKZ9Mcff2RCIvu0d+9eBQUFaePGjbp//75KlSols9msRYsWydfX1+h4Nqt9+/aaPn26cubMaTm2Zs0a+fv7K1u2bJKk69ev66233tLRo0eNimmzuKGQdY4eParChQvrjTfeUKVKlbjpDQCbRTmViWJjY5UlS5YEx5ydnTV48GCKqVRavHhxksfv37+vb7/9VmFhYfL19VVgYGDmBrMT/v7+Cd6g/X3h5D/fuPHbx8QCAgK0ffv2BAVB165dNWLECMudWu7cuaOvvvqKcuofWKRrHeZnnUWLFiX72JUrVzR58mRduHBBDRo0yMRU9mPatGlatWqVLl26pAoVKqh79+6qV6+eihYtKm9vb+XOndvoiDZtz549evToUYJjgwcPVsWKFS3llNlsVmxsrBHxbF54eHiqVk/98MMP6tixY8YHsjO7du1SaGio1qxZo8WLF8vJyUk1atRQQEAAd8FOhdSuhoqPj8/gJMDzj3LKBrBCxTqbNm3S6NGjFRMToxEjRlgum0RiW7duTfax48ePa/To0bp8+bI++uijTExlP5IqCPbu3auHDx8akMa+LFq0KFWXpCFpPXr0sLyJRdoldamj2WzW4sWLNW3aNL300kv6/vvv2e8xGTNnzpSnp6dGjhypgIAA5ciRw+hIdi+p7yeUBElr37695s6dq0qVKiX5+NmzZ9W/f38dPHiQcioJWbNmVcOGDdWwYUPduXNHGzdu1Lp16zRo0CDFxcVp5MiRatOmjWrXri0nJ94a/tP06dPVo0ePFM/5888/1b9/f61YsSKTUgHPJ/4FymSRkZGJ3shevnw50T5TFFZPd/78eQ0fPlzbt29Xq1at1Ldv3wRL5pFYoUKFEh27f/++pk6dqqVLl8rX11dz58613CIcSC8mk0lhYWGpOpcNbRN72g/GSJtDhw4pMDBQp0+fVrdu3dS5c2c5OzsbHctmLVy4UGvXrtXo0aM1aNAg+fj4qG7dugoICDA6Gl4AtWrV0ocffqiZM2cmKpB/+OEHTZ06Ve7u7po3b55BCW3b37ckyJkzp1q1aqVWrVrp2rVrWr9+vdauXauePXvK3d1dO3bsMDquzZk9e7aioqLUv3//RI/Fx8drzpw5mjFjRpI/YwNIG8qpTNayZcsEH5vNZn3wwQeW35aZzWY2dHyKR48eae7cuZYS5aeffuLW1c9o/fr1GjNmjOLi4jRq1Cg1adLE6Eh4TiV3p8Mn/r5igH//EmPPpPRx584dTZgwQcuXL9dbb72lb7/9VoULFzY6ls2rVq2aqlWrpsGDB2v79u1at26dpk6darncZfXq1Wrfvj17oiFDjBs3TmPHjlW3bt00adIk1alTR2fPntWAAQN06NAhtW/fXr1795arq6vRUW1SUlsSSFK+fPnUrl07tWvXTn/99ZfWrVtnUELbNnPmTPXu3VtRUVEaPny45Xvxf//7X/Xv31///e9/1bFjR/Xs2dPgpID9o5zKRCldUoXU2bVrl4YOHarr16/ryy+/1Pvvv88y+Gdw7tw5DR06VLt27dJ7772nzz77jMs0kKFS2tD23//+t4YPH66oqCj17ds3E1PZD/ZMst7KlSs1YcIEubq66ttvv1WdOnWMjmR3nJycVKtWLdWqVUsPHz5UaGio1q1bpx9++EHz589XrVq1Un1X3RdJUvv68LNL2vTr10/58uVTnz591LRpU61Zs0bFixfXL7/8Im9vb6Pj2bTU7FlYuHBhde3aNRPS2J8333xTP/zwg7p166aoqCiNGTNG8+fP16xZs1S6dGn9+uuvqb6jJICUUU5loqCgIH300UfKmjWr0VHs1ocffihJypUrl77//nt9//33yZ5LGZhYTEyM5syZo3nz5unVV1/V8uXL+aEujcLCwhLsnWQ2m3X48GFFRkZKkm7fvm1UNLsTGRmpESNGKDQ0VE2bNtUXX3yhvHnzGh3LJrFnknXef/99HThwQIUKFVKnTp0UFRWl4ODgJM9l38LUyZIli+rXr6/69esrKipKmzdv1tq1a42OZZPMZrNatGghBwcHy7Ho6Gi1a9fOsq1DXFycUfHsxkcffSR3d3cNGjRIlStX1g8//JBgpkgeZah1KlasqGXLlqlz587y8/NTXFyc+vbtqw4dOvA1CKQjk5lbAGWaMmXKJLmsFqm3cuXKVH+DbdasWQansT9169bVX3/9pUKFCumdd95JcZbscZNYan8zxqW5KYuLi9P333+vGTNmqHDhwhoyZAi3oU8j9kxKG39//1SdZzKZ+MUG0l1QUFCqz+Vnl6f7z3/+o969e+vDDz/kUqpU8PLyko+PT6q+R6S0SheP9wnu3LmzXF1dtWDBAq46ANIZ5VQm8vLy0o4dOyinYJin7fvzhMlk4gcUZIh9+/YpMDBQFy5cUI8ePdSxY8dEN4RA8v65Z9KgQYPYMykdxcTEyMXFxegYNie1e55J7BmH9Ofv75/o6+/mzZu6f/++PDw8EqxcoVxOzMvLSx9++KGyZ8/+1HP5xWRie/fuTfDx7du3FRgYqHz58ql///4Jfobhhi6AdSinMpGXl5dWrFihPHnyPPVc7taXtOT2snB2dlaOHDnk7e2tihUrZnIqvCgGDBiggQMHys3NzegodufGjRsaN26cQkJCVLduXQ0cOFAFChQwOpZd+fueSV9//TV7JqWjgwcPKigoSBs2bNDu3buNjmNz9uzZk+Bjs9msrl27asSIEYn+O07qEtQX3T/f3KaEN7eJsfLMOvxy3DqsmgcyD+VUJkrNbx65W1/Kklv5YzabdefOHZ0+fVrly5fXnDlzWGqbgpiYGP3nP//R4cOHdevWLeXOnVsVKlRQzZo1WTWQAi7NfXavvfaa7t69Kw8PD1WrVi3Fc5/cAQz/8889k7Jly5bsueyZlDqXLl1SSEiIgoODdfbsWeXMmVPNmzdXv379jI5mF3x8fLRq1SoVKVLE6Cg2jze3GWP//v0qX748P7c8BeUUAHvBhuiZ7JdffmHDXyssXrw4xcfv3Lmjnj17avz48Ro2bFgmpbIvv//+uwYMGKDLly/rlVdeUc6cOXXlyhXNmzdPL730ksaMGaPXX3/d6Jg2iS7/2SV1WQZS79KlS/Lw8FB8fLzmzZuX7Hkmk4lyKgXR0dHauHGjgoODLauBzGaz+vXrp7Zt2/ImFxkipbuV4tl16dJFISEhFKRPUbVqVfYkzACBgYHq1asX7+uAdEQ5lYlMJpMKFizIby4yUM6cOdWjRw99/vnnRkexSeHh4erevbtatGihTz75JME31KtXr2rWrFnq3r27fv75Z7366qsGJrVd/7xbX3K4NCOhMWPGGB3BroWGhhodwa79/vvvCgkJ0YYNGxQbG6vq1atr6NChCggIUM2aNeXn50cxBdgZfmGUOk/7xS6ezapVq/TRRx9RTgHpiHIqE/FNNHMUKlRId+7cMTqGTZo5c6YaN26sQYMGJXrspZde0uDBg+Xg4KAZM2Zo6tSpBiS0fanZLJRLMxILDg5WgwYNUiwAoqKiNHbsWA0fPjwTk9mHixcvysPDI8XVZzExMdq4caMaN26cicnsQ8eOHeXp6akhQ4aoXr16KV4WCaS3AQMGpOo8k8mkUaNGZXAavGjat2+fqvNMJpMWLlyYwWmeH7yvA9If5VQmGj16NPsgZYI///xTL7/8stExbNK+ffu0YMGCFM9p3bq1OnbsmDmB7BD7NjybAQMGqGbNmglmV7ly5QSXZDx48EC//vor5VQSAgICEu139mRD6vz580t6fFnzV199RTmVhE8//VRr167VwIEDtWjRIvn7+6tOnTqp3gvoRZfUzUgePXqkRYsWJVpJyt2+0m7v3r3666+/UrUqF//z8ccfM7NUSOkmBfHx8QoJCdGFCxdUtGjRTEwFAIlRTmWiatWq6dq1a4mOOzk5KWfOnHJ1dTUg1fPj7t27CgsL07Bhw9S8eXOj49ik+/fvP/VWwlmzZlVMTEwmJbIv7Jn07JL6DSO/dUy9pGa1d+9ePXz40IA09qdnz57q2bOnjh07pjVr1mjFihWaMWOGChYsKLPZrDNnzqhkyZJGx7RZSd3B0MfHJ9FeSvwbmbTkbvJw+fJljRw5Un/99ZeaNGnCZvypEBMTozNnzujevXvy9/fnZ+dUSK4wPnbsmAIDA3XlyhV1795dH3/8cSYns2/r1q3jrsNAOqOcykRP2xDY09NTnTp1UuvWrTMxlX152h0Ps2TJotatW6t79+6ZmMp+vPrqq9q2bZs++OCDZM/Ztm0b+00lI7VlyuHDh1WhQoUMTgMgrby9veXt7a1+/fpp3759Wrt2rTZu3KiePXuqZMmSatWqVaovgXmRsGdN+oqPj9fChQs1ffp0eXh4aNGiRSmuboF08uRJTZw4Udu3b1dsbKzl7tbOzs7y9/dXnz599Morrxgd0y5ERUVpypQp+vHHH/X6669r9erVzC4Vrl69qm3btikiIkL37t1T9uzZVbp0ab311lvKkyeP0fGA5wLlVCbaunVrksfj4+N19+5dHTx4UJMmTZKDg4NatmyZyensw6JFi5I8/mT1maenJ3ckSUGHDh30zTffqGTJkknekW/r1q2aPHmyxo4da0A625fUJSxPXL58WSEhIQoKCtKZM2fYcwqwIQMGDNDAgQPl5uZmOebr6ytfX19988032rlzp9auXavp06dTTqXB1atXdeDAAbm7u8vX19foOHYhLCxMgYGBOnfunLp3765OnTrJyYkfx1Ny5MgRtW/fXt7e3hoxYoRKliypHDlyKCoqSuHh4QoKClKLFi30008/qVSpUkbHtWlr1qzRmDFjZDKZNH78eDVo0MDoSHZh+vTpmjNnjpycnFS4cGHL19/SpUvl4OCgTz75RF27djU6JmD3+G6YiQoVKpTi42XLllW2bNk0f/58yqlk8JtF6zRs2FCnTp1Sp06dVK5cOZUvX145c+bUtWvXdOzYMf33v/9Vr169VKdOHaOj2qR/fv09ePBAmzZtUnBwsHbv3i2z2ayaNWtyaQZgY4KDg/XFF18kKKeecHBwkJ+fn/z8/LikORmPHj3S2LFjtWLFCgUFBemVV17Rtm3b1Lt3b0mSo6OjSpQooXnz5ilnzpwGp7VNt27d0rhx4xQUFKTatWtr1qxZKliwoNGx7MLkyZP1f//3f0ne9bVs2bJq3ry5+vbtqxkzZmjKlCmZH9AOnD59WkOHDtXevXvVtm1b9e7dO8l/D5HYokWLtGDBAg0ZMkRNmjRJcGOXmJgYrVq1SqNHj1b+/PnVtGlT44ICzwHKKRtTuXJlBQYGGh3D5m3evFmbN2/WyZMnde/ePbm5uenVV1/V22+/rbfeesvoeDatZ8+e8vf314oVK3TkyBHdvn1buXLlUuXKlTVq1Cg2CE6FvXv3KigoSBs3btT9+/dVqlQpmc1mLVq0iNUDKZg/f36Cu6T9c0Pl+/fvGxXNLoSFhSVYuWc2m3X48GFFRkZKkm7fvm1UNJuX2ktyU7qb5Its7ty52rx5s4YOHSoPDw/FxMRo4MCBKly4sJYuXaqsWbPqiy++0JQpUzR48GCj49qc5cuXa+LEiXJzc9PMmTNVu3ZtoyPZlcOHD2vJkiUpnvPRRx+xZ1IyJk+erO+//1758+fX1KlTVbZsWd25cyfJO1tTmCb2888/a8CAAUkuHHBxcVHLli314MEDLVu2jHIKsJLJzI60NiU8PFydOnXSzp07jY5ik+7du6dPPvlE+/fvl6+vr0qWLCk3NzdFRUXpxIkT2r9/v9544w1Nnz6dTTKR7qZNm6ZVq1bp0qVLqlChggICAlSvXj0VLVpU3t7eCgkJYVPlZLRr1y7V57K/TWKpLY1NJhOXlCbBy8tL06dPT9WdvapWrZoJiezL//3f/6lfv37y9/eX9Hhvwm7dumnEiBGWN2z79u1Tnz59tH37diOj2qS///f7tE3j+e83sTJlymjbtm2WO5Mm5cqVK6pdu7aOHTuWicnsQ2q+/p7s4cXXX2IVKlTQ2rVrLXcWTsq5c+fUokUL7d27NxOTAc8fVk7ZkLi4OH333XesvEjB5MmTFRkZqdWrV6tYsWKJHj9z5oy6du2qefPmcTvrZNy8eVMLFy7UBx98oHz58lmOT548WfHx8ercuTO3Zk7GzJkz5enpqZEjRyogIEA5cuQwOpLdoHCyzj/vioa0S833BN6cJe3ixYsJ3uDu2rVLJpMpwUplDw8PVu8lI7n9MpE6ZrNZjo6OKZ7j4OCg+Pj4TEpkX5Lb8xapExMTk2DVd1KyZcumqKioTEoEPL8opzLRgAEDkjxuNpt19+5dHTlyRCaT6alLl19kmzdv1vDhw5MspiTplVde0VdffaVJkyZRTiUhMjJSH3zwgeLi4vT2228nKKfc3d31ww8/aNOmTVq8eHGKv6F8US1cuFBr167V6NGjNWjQIPn4+Khu3boKCAgwOppdOX/+fKJLcrmUABltx44dcnd3NzqGXcqbN6+uXr1q+e9027ZtKlOmjF566SXLOf/9738TfIz/Sc1+mYcOHVJQUBB7aybjn5c1/xPFaPKetuftE5cvX87gJPbJZDI9dcUjgPRBOWUDnJ2d5enpqXr16qlu3bo07ym4fv36U+/E4u3trYsXL2ZSIvsyZcoUeXp6asaMGYkue2zfvr1atGihrl27aurUqRo5cqRBKW1XtWrVVK1aNQ0ePFjbt2/XunXrNHXqVI0ePVqStHr1arVv3543wMnYtWuXRo8erT///DPBHkAmk0ne3t7q378/K0eTkdwvN5Ly5OsR/8MbC+vUq1dPEyZM0KBBg7Rz506dPn1a33zzjeXx69eva9KkSZbL/pA6kZGRCgkJUXBwsE6fPq2XXnqJfUeTwS8cn11AQIB+/fVX5cmTx3Js7ty5eu+99yw3MLh27Zpq1arFytEkmM1mjRgxQlmyZEn2nIcPH2ZiIuD5RTmViVJ6wxATE6PNmzerV69e2rVrF9fMJyM2NvapG9Y6OzvzTSIZO3bsSLKYeiJ79uzq3bu3vvrqq0xOZl+cnJxUq1Yt1apVSw8fPlRoaKjWrVunH374QfPnz1etWrU0ffp0o2PalO3bt6tbt25q2LChBg8enOhW4CtWrNCHH36oRYsWycfHx+i4eM6kdnvNw4cPq0KFChmcxv706dNH/fr1U9OmTWUymdSiRQu9//77kqTZs2dr5syZKlmypHr16mVwUtv34MEDbdy4McFdXn19fdW7d2/ulJsMLmu2zoULFxJd8jh79mzVr18/wd012YY4aU/+3UtJlixZ2AwdSAeUUwbbv3+/goODtWHDBkVFRalEiRL6+uuvjY5ls1haa52oqKgEvzlLCvuGpE2WLFlUv3591a9fX1FRUdq8ebPWrl1rdCybM2PGDHXs2FFffvllguO5cuWyrEjLlSuXZs2apblz5xqU0naxGso6f78r5D9dvnxZISEhCgoK0pkzZ1g5kITs2bNr+vTplpXdf78FfeXKlTVx4kTVrl1bTk78WJmcPXv2KCgoSJs2bVJ0dLQqVaqkL7/8UhMmTNCQIUO4mUY6+PXXX5O8oxoSS6qI4ufrpI0ZM8boCMALg58iDHDhwgUFBwcrJCRE58+fV86cORUVFaWJEyeqQYMGRsezaWazWS1atJCDg0Oy58TFxWViIvtSqlQp7dmzJ8U7jjztcSTPzc1NzZo1U7NmzYyOYnPCw8M1fPjwFM9p1aqVOnXqlEmJ7FNUVJQcHR2VNWvWRI9dvXpV48eP17hx4wxIZtv+uY/PgwcPtGnTpgSrV2rWrKl+/foZlNA+/L2UeoI9kp7O399fd+7cUbVq1dS/f3/Vrl3bsufjhAkTDE5n+2JjYzV37lxt2bJFjo6Oevvtt9WpUydLmXL48GENGzZMx44do5xCuktuqxAnJyflzJmTu4MD6YhyKhOtWLFCwcHB2rdvn/Lnzy9/f3/Vq1dPVatWVcWKFfXqq68aHdHmsXrAOm3bttXYsWNVvHjxJC+dCgsL0/jx4/XJJ58YkM72eXl5JfmbRUdHR+XMmVNlypTRhx9+KD8/PwPS2bYHDx489S6QefLk0Y0bNzIpkX2JjIxU//79tXv3bknSm2++qXHjxilXrlyKi4vTDz/8oBkzZsjZ2dngpLZt7969CgoK0saNG3X//n2VKlVKZrNZixYtYr+zFLRr1y5VqypMJpMWLlyYCYnsT5YsWeTk5KSHDx8qJibG6Dh2ZcyYMfrll1/0zjvvyMXFRXPmzNGDBw/08ccfa8yYMVq6dKmKFy+uBQsWGB0VzyF/f/8U//3z9PRUp06d1Lp160xMBTyfKKcy0cCBA+Xp6amxY8eqSZMmRsexS6xIsU7Tpk11/PhxtW3bVhUrVlS5cuWUI0cO3blzR8eOHdPhw4fVsmVLtW/f3uioNim524HHx8fr7t27OnjwoHr16qXx48dzB79/MJvNKa54lB6/sWXPi6QNGzZMFy5c0Lhx4+Ts7Ky5c+dq9OjR+uyzz9S9e3eFh4erZcuW+uyzz4yOapOmTZumVatW6dKlS6pQoYK6d++uevXqqWjRovL29lbu3LmNjmjTqlWrluxj8fHxCgkJ0YULF1S0aNFMTGU/QkNDdfDgQa1Zs0azZ8/WyJEjVbp0acseU1xOlbKNGzdq2LBhlj196tWrp6+++kqnTp1SaGiovvzyS7Vv316Ojo7GBrVRSW2Jwddc6m3dujXJ43//2W/SpElycHBg5R5gJZOZdwKZZuXKlVq7dq1+//135cyZU7Vq1VKdOnXk5+enypUrKyQkhD0HniI4ODjV57IxYfL27dunkJAQhYeH686dO8qTJ4+8vb3VuHFjVapUyeh4dm3RokVavXq1li9fbnQUm+Ll5aVBgwYleVnQE3fv3tWoUaPY8ycJ1apV05QpU1S9enVJ0rlz59SsWTMVKVLEcieh8uXLG5zSdnl5ecnT01Pdu3dXQECAcuTIYXnM29ub77/P6NixYwoMDFR4eLg6d+6sjz/+OMU7WuFxUf/7779rzZo12rJli27fvq0SJUqoVatWatKkifLmzWt0RJtTrlw5bdq0SQULFkxwzNPTU7Nnz2YrgqdIatW32WxOcOzJx3z/fTbBwcGaP3++Vq9ebXQUwK5RThngxo0bWr9+vdatW6cDBw7I1dVVDx480KBBg9S6dWsuy0jB025Tfe/ePd25c0eS+AYLQ5w6dUotWrRQWFiY0VFsSlpuMR8aGpqBSexTmTJltG3bNuXPn99yrEKFCqpZs6amTJnC942n2L17t9auXauNGzfq3r178vHxUd26dRUQEKB69epRTqVRVFSUpkyZoh9//FGvv/66vvnmG73yyitGx7I7jx490m+//aZ169YpNDRUsbGxOnLkiNGxbI6Xl5d27Nghd3d3yzEfHx/Nnj07xVV9eGzPnj2pPpc95J7NuXPn1KRJEx08eNDoKIBdo5wyWGRkpNasWaN169bpjz/+UO7cufXOO+9owIABRkezK/Hx8Vq2bJmmTp2qvHnzavDgwapRo4bRsezGk5V7/PbRehEREWrbtq1lbyAgPST35uzHH3+Ul5eXgcnsS2xsrLZv365169Zp69atun//viSpa9euat++fYL5Imlr1qzRmDFjZDKZNGDAAG7kkk4ePHig0NBQ5pmE5P79W7VqFT+3wCaEh4erU6dO2rlzp9FRALvGnlMGe/nll9W5c2d17txZZ86csRRVlFOpd+TIEQUGBioiIkJdunRRly5d5OLiYnQsu0JHnX5+/vlnVahQwegYeEFkz57d6Ah2xcnJSbVq1VKtWrX08OFDhYaGat26dfrhhx80f/581apVS9OnTzc6pk06ffq0hg4dqr1796pt27bq3bt3ipfp4n/4mrJeZGSkHj58mODY5cuXE+0z9fdL//AYW2JkrLi4OH333XfcVANIB5RTNuSVV15Rjx491KNHD6Oj2IWoqChNnDhRP//8s9544w2tWrWKzViRoZJ7g2E2my2bYkZERGjx4sWZnMz2pfZuX1LyG8+/6NavX5+gDIiPj9emTZsSrfbhzUXqZMmSRfXr11f9+vUVFRWlzZs3s19IMiZPnqzvv/9e+fPn19SpU1W2bFnduXPHchn931EOJDZ9+nQ5ODioTJkyyp49e7K/EGKT6uT9c6Nps9msDz74wDIz9kxK3rRp01J1nslk4vtHEpJbMPDkZ78jR47IwcGBn/2AdEA5BbsUEhKicePGycnJSZMmTdLbb79tdCS75uvryya2qZDcpXrOzs7KkSOHatWqpW+//VYFChTI5GS2j31BrFOwYMFEt0l3d3fX0qVLExzjzUXSpk+fnuIvftzc3OTt7a0lS5ZkYir7MWfOHEnSX3/9pZ49eyZ5DuVA8oYMGaItW7bo4MGDqlq1qgICAhQQEMDm56mU3N3SkDrs45gxnJ2d5enpqXr16qlu3brKli2b0ZEAu8eeU7ArERERCgwMVFhYmNq1a6eePXvyzQAAkKJy5crpgw8+UP/+/RM9Fh8frzlz5mjmzJkqWLCgNm7caEBC23bhwoVUn1uoUKEMTGLfoqKitG3bNm3evFk7d+7Uq6++qjp16qhu3brMLQ0ePXqk27dvK1euXNwMIh39+uuviVao4X+uXbum3Llzy8np8dqOY8eOaffu3cqbN6/q1avH+xEgHVBOwa6UK1dOsbGxKlCggDw9PVM8l0uDknb48GGtXLlSvXr1Ut68eXXjxg1988032rlzp/LmzatOnTrp/fffNzqmzbt9+7aOHTummzdvKkeOHCpbtqzy5csnSTp48KBcXV3ZqDoJu3btUqVKlZQ1a1bLsXXr1ilHjhyqWbOmgcnsx5OvvVu3bilPnjwqU6aMcufObXQsm/bbb7+pd+/eatiwoYYPH265FOi///2v+vfvr//+97/q2LGjevbsySpSZIqYmBjt2rVLW7du1b/+9S/ly5dPderU0aeffmp0NJu1bNkyLV++XOHh4ZZjpUuXVuvWrdW2bVsDk9m22NhYzZ07V1u2bJGjo6PefvttderUyfLv4OHDhzVs2DAdO3aMlY9JuH//vj7//HNt27ZNa9asUYkSJRQUFKSBAweqQIECcnV1VUxMjJYuXaqXX37Z6LiAXeOyPtiVbt26sSeDFXbu3KmuXbvqtddeU2xsrCTp888/16FDh9SvXz/lyJFDEyZMkKurq1q0aGFwWtt05coVjRo1Slu2bLHMUHp8OVWtWrXUv39/ff311+rVqxfl1N88fPhQvXv31m+//abFixerSpUqlsd27NihoKAg1alTRxMmTOCGBslI7mvPyclJderU0cCBA/XSSy8ZmNB2vfnmm/rhhx/UrVs3RUVFacyYMZo/f75mzZql0qVL69dff+W/1xSk5SYto0ePzsAkzw8XFxfVrFlT2bJlU7Zs2bR8+XJ99913lFNJiIuLU/fu3bVv3z41b95cXbp0Ua5cuXTlyhUdOXJEY8eO1bZt2zRr1iw5ODgYHdfmjBkzRr/88oveeecdubi4aM6cOXrw4IE+/vhjjRkzRkuXLlXx4sUTXTqOx6ZNm6YLFy5oyZIlKl68uO7fv68RI0aoQoUKWrx4sZydnTVkyBBNmDBBEyZMMDouYNdYOYXnTkxMjLZs2cLtmJPQrl07VatWzbL3yp9//qnGjRura9eu+vzzzyU9XsUyZ84chYSEGBnVJt24cUOtW7dW7ty51a1bN1WpUkU5c+bUnTt3dODAAX333Xc6duyYqlWrpvnz5xsd16Z8++23Wr16tb799luVLl060ePh4eHq3r27Wrdure7duxuQ0Lb9/Wuva9eulq+9J2/OZs+eraioKC1fvlx58uQxOq7NOnXqlDp37qw7d+4oLi5OvXr1UocOHXhD+xSUU+nn3r17+s9//qPQ0FD99ttvkqRatWrJ399ffn5+XBqUhAULFmjJkiVaunSpPDw8Ej1+6dIldejQQW3btlXHjh0zP6CNq1mzpvr27WvZj3D37t366quv5Ovrq9DQUPXq1Uvt27dPdOdDPObv769Ro0bp9ddflyRt2rRJvXr10sSJE9WwYUNJ0qFDh9StWzf9/vvvRkYF7B4rp/DcCAsLU1BQkDZs2KC7d+9STiXh6NGjGj58uOXj3377TSaTSf/3f/9nOVauXDmdOXPGgHS2b+bMmfLw8NCCBQsS7HORN29e1alTRy4uLuratSt3q0rC6tWrNXDgwCSLKUny8vLSV199palTp1JOJSG5r71ChQqpUKFCqlOnjrp27apZs2bp66+/NjCpbStevLh+/PFHde7cWa6urmrZsiXFVCqktnCKj4/P4CT2KTIyUlu3blVoaKj27t2rAgUKyN/fX9OmTVOVKlUoBZ4iKChIX375ZZLFlCR5eHjoyy+/1LRp0yinknDz5k299tprlo+rVaum69evKzw8XKtWrVKRIkUMTGf7rl69muBu4Dt37pSjo6P8/Pwsx/Lly6fo6Ggj4gHPFcop2LVLly4pODhYwcHBOnfunLJnz64mTZqoTZs2RkezSSaTKcEtrJ/sM+Xt7W05dvfuXbm6uhoRz+aFhoZqzJgxyW7AOn78eLVp00b/+te/MjmZ7bt8+bJKliyZ4jnlypVTZGRkJiWyL0/72nNyctInn3yir776inIqCXv37k3wce/evRUYGKh27dqpf//+CcqBqlWrZnY8m/e0ux1Kj1fi9u/fXytWrMikVPajdu3acnJyUtWqVdWvXz+9+uqrlscOHDiQ4Fy+/hI7d+6cKlSokOI55cqV0/nz5zMpkX2JjY1NtJees7OzBg8eTDGVCgUKFND58+dVsGBBmc1mbdu2TRUrVlSuXLks54SFhSVbngJIPcop2J3o6Ght3LhRQUFB2rt3r5ydnfXGG2/o/PnzWrJkCfuGpMDHx0cbNmxQ9+7dde7cOe3evTvRnVl+/PFHlS9f3qCEtu3GjRsp/iA3YcIEZcuWTUFBQZmYyj54eHgoIiIixTtSnT592rKpPBK6du3aU99EFCpUSDdu3MikRPalXbt2SR6/du1agpUWJpOJDYGT8OSy0ZTudjhjxgzuOJcMs9msR48eaefOndq5c2ey5/H1l7QcOXLo8uXLKX59Xbx4UXnz5s3EVPaPVd6p884772jkyJHq3bu3fv/9d126dEl9+/a1PB4eHq5JkyapSZMmBqYEng+UU7Ar/fr10+bNm+Xs7KyaNWtq4sSJeuutt5QtWzZ5e3tbbu+KpH322Wfq2LGjNm3apAsXLih37tyWS6h27dqlJUuW6LffftPChQsNTmqbChcurD/++CPZ346VLl1aW7Zs4Q1aEho1aqRJkybJx8dHOXLkSPR4VFSUJk+erICAAAPS2T4PDw8dO3Ysxd/MHjt2TIULF87EVPbj73f3QtrNnDlTvXv3VlRU1FPvdojE+PqzTu3atTVjxgzNmzcvyZvimM1mzZw5U/7+/gaksw+RkZF6+PBhgmOXL19OdEkphVVi3bt3V1RUlL7++muZTCb16tVLjRo1kiSNHTtW33//vWrVqsWWBEA6YEN02BUvLy95enr+f3v3HhVlnYBx/BnESyduZdgiKbZqYkwYWnhJWhi0NYotk9Q4AuYWLFZmW0abdtPUyrazaziaa5SXdEmEARECkbSicmmlTFLaLl7CQMmEDBUZZ//wNCcSTUx4uXw/53jifd9p5pEz58z4vL+LJk2apKFDhzYYGh8QEKDMzMxfnTrU0VVWVio/P18uLi66+eabnXcaly5dqpKSEsXHxysoKMjglK3TsmXLlJaWptWrVzd6h7ayslKxsbEaP368/vznPxuQsPWqq6tTTEyMysvLNW7cOJnNZrm5uammpkalpaVKT0+Xt7e3li9fLg8PD6PjtjpWq1WZmZlas2ZNo++9AwcOKCYmRhMmTNCUKVMMSIj27qcFf4cNG3babodz585l1DKazcGDB3XnnXeqV69eio+Pl9lslqenpw4ePKjS0lJZrVZVV1crNTWV0VON8Pf3P21Zh8Ywcq/pysrKZLfbdfXVVxsdBWgXKKfQpnzzzTfKyclRdna2/ve//6lnz54aNWqUwsPDNWXKFNlsNsqp34jdDs/MbrcrPj5eO3fuVFRUlAIDA3XxxRersrJSO3bs0Lp16xQcHMx21mdw4sQJLVu2TDabTXv27HGe79+/vyIjIxUXF3fauhg4pa6uTgkJCSorK3OWex4eHqqqqlJpaanefPNNBQcHy2q18t5Ds2G3QxiloqJCs2fPPm1NRxcXF40aNUozZ85Ujx49DErXupWXlzc4rqurU01NjTw8PNSlS5cG1xj5DcBIlFNos7744gtlZ2crNzfX+Q/dsWPHKiYmhjsY5+GXux1y96xxDodDq1evVlpaWoPfkb+/v8aPH6+77rqr0WkHaOjYsWOqqamRl5fXaV+O0Ti73a6VK1cqLS1NX375pfMu+FVXXaUJEyYoOjqa9x6aXWVlpXO3w5SUlEan6QLN5adCvqamRp6enjKbzYyWOkerV6/W2rVrG0wzHTBggMaPH6/o6GgDkwHAKZRTaBe2b9+uDRs26K233lJlZaUGDhzIotTn4Gy7Hfbv39/oeK1OcXGxgoKCnGub1dXV6fDhww0KlqNHj2r58uX6y1/+YmTUVmvPnj3q2bNng13nPvjgA/Xo0UN9+/Y1MFnbcvToUWe5x2gzNLdf7nZYXV2tp59+Wpdddhm7HQKtnN1uV2Jioj766CPdcccdGjx4sDw9PXXgwAF9+umnWrdunYYNG8aobwCGo5xCu+JwOFRcXKwNGzbomWeeMTpOq3Sm3Q63bNmi9PR01g05i4EDB+q9995T9+7dneciIyO1dOlS50LVVVVVCgkJYeTZLzgcDs2dO1erV6/W66+/ruDgYOe1qVOn6u2331ZcXJySkpIY/dMEv3z/Ac3hXD8XWLMGzcFisZzT54LJZFJBQUELJGpbUlJStGrVKr3xxhuNflZ8++23iouLU3R0dIPdSwGgpbG1GdqFQ4cOKTc3Vw6HQ2FhYRRTZ8Buh79NY13+N998o/r6egPStC0rVqxQTk6OFi1a1KCYkk4t9l1YWKi//e1v6t27N9MLmoD3H1oCu83BSGfbBbK2tlYpKSkqLy9nM5czyMjI0IwZM854E8PHx0czZszQwoULKacAGIp/iaJNOXr0qF544QXl5ORIkm677TbFxMRo4sSJOnr0qBwOhxYsWKBly5YxtaARmZmZZ9ztEGhOb775pp544gmFhYU1et1iseiRRx7RihUrKKcAAE5jx45t9PymTZv08ssvq7a2Vs8++6yioqJaOFnbsHfvXgUGBp71MWazWfv27WuhRADQOCYWo02ZN2+etm3bpqeeekrz58/XF198ofHjx2vEiBHaunWriouLddttt2nhwoVGR22VCgoKNG7cOK1du1a33XabwsPDNX/+fP3nP/9hKhWaVXl5+a9+OR42bBhfjpvI19eXEY8AOpTy8nIlJibqgQce0IgRI/TWW29RTJ2Fu7u7Kisrz/qY/fv3s7A8AMNRTqFN2bRpk+bMmaOIiAhZLBa9+OKL+v777zVp0iR17txZrq6umjJlinbs2GF01FbpiiuuUHx8vLKysrR+/XpFRkZq8+bNio2NVX19vV599VV99tlnRsdEO9S9e/fTtrP+pYqKCnl5ebVMoHYiOzub9aYAdAj19fV65ZVXdMstt2j//v1644039Oyzz/K58SvCwsK0aNGiRpcmkE4tWWC1WmWxWFo4GQA0xO1WtCmHDh3S7373O+fxpZdeqosuukiXXHKJ85ybm5uOHTtmRLw2pV+/fpo+fbqmT5/eYLfDjIwMdjs8i9zcXLm5uTmPT548qY0bNzrvOP7www9GRWvVRo8erZdfflkpKSkNdur7SX19vZKTkzVy5EgD0rV+NpvtnB97++23N1sOADDC1q1bNXv2bFVWVmr69OmKjY1lZ7lzNG3aNN15552KjY1VfHy8zGazPD09dfDgQZWWlspqtaq6uloLFiwwOiqADo7d+tCm+Pv7q6ioqMFuaUFBQcrKylKvXr0ksVvab8Fuh2fXlLuKhYWFzZik7ampqVFUVJS6du2qmJgYmc1mubu7q7q6WqWlpVq1apV+/PFHrVmzRpdffrnRcVudc33vmUwmbdq0qZnTAEDLeeSRR7Rhwwb5+vpq+vTpZ/2MYL3RxlVUVGj27Nl6++23G5x3cXHRqFGjNHPmTPXo0cOgdABwCuUU2hR/f38lJyfL09PTee7ee+/V3LlznR+q1dXVeuCBByinmuCXux36+voaHQnt0OHDh/Xiiy8qJydHtbW1MplMcjgccnd3V0REhB544AFddtllRscEALQi/v7+5/Q4k8nEd79f8d1336m0tFTV1dXy9PSU2WxmrSkArQblFNoUvqD8Nuey2+HJkyfZ7RDNqq6uTvv27VNNTY28vLzUu3dvderUyehYbVpdXZ0KCgpks9m0dOlSo+MAAAAATUI5BXQgTzzxhLZv366EhAR169ZNq1at0s6dOzVy5EjNmzdPJpNJs2fP1tdff62VK1caHRft1E9T+Q4fPiwvLy9dffXVLGh7nrZt2yabzabc3Fz98MMPMpvNSktLMzoWAAAA0CSUU0AHMmLECC1ZskSBgYGSTk3nGzFihFJTUzVo0CBJ0u7duzV27FiVlJQYGRXt0IEDBzRv3jwVFBSovr7eed7V1dW55oW3t7eBCduG/fv3y2azKTMzU3v27JHJZFJERIQmT56sa665xuh4AAAAQJOxWx/QgbDbIYxy6NAhRUdHy8vLSy+99JKGDBkiDw8PHThwQJ9++qmWLFmiu+66S2vXrm3wfsQptbW1ysvLU3p6uj766CO5ubkpNDRUDz/8sB566CElJiaqX79+RscEAAAAzgvlFNDBNLa2j8lkMiAJOhKr1SofHx+lpKSoc+fOzvO+vr7y9fXVqFGjFB8fr8WLF+vxxx83MGnrdMMNN6h79+6yWCxKTExUcHCwXF35CAcAAED7wDdboIMpKSlpsNuhw+HQ9u3bVVFRIenUekDAhVZYWKjnnnuuQTH1c66urpo6daoeffRRyqlGmM1mlZSUaNu2berUqZM6d+7MpgUAAABoNyingA7m/vvvP+3cww8/3OCYkVS40KqqqtSrV6+zPsbX11eHDh1qoURty8qVK1VZWanc3FxlZ2frtddek5eXl8LCwiSdKpkBAACAtopyCuhAdu3aZXQEdFA+Pj4qLS2Vj4/PGR9TWlqqK664ogVTtS2XX365Jk+erMmTJ2vv3r3Kzs5WTk6O7Ha7Jk2apMjISEVFRcnf39/oqAAAAECTsFsfAKDZWa1WZWZmas2aNbr00ktPu37gwAHFxMRowoQJmjJligEJ265du3YpJydHOTk5Ki8v186dO42OBAAAADQJ5RQAoNnV1dUpISFBZWVlGjdunMxmszw8PFRVVaXS0lK9+eabCg4OltVqlYuLi9Fx26xPPvlEgwYNMjoGAAAA0CSUUwCAFmG327Vy5UqlpaXpyy+/dK6TdNVVV2nChAmKjo5mvbOz2L9/vzZs2KAJEybIw8NDx48f19///nd98MEHuuSSSzRlyhSFhoYaHRMAAABoMsopAECLO3r0qGpqauTl5aWuXbsaHafVKy0tVWxsrLy9vfXaa6/Jx8dH06dPV0FBge6++265u7vr1Vdf1fz582WxWIyOCwAAADQJ5RQAwDCRkZFaunTpWRdKh3TvvfeqZ8+eeuaZZyRJ+/bt0+jRoxUdHa0nn3xSkpSWlqa1a9cqNTXVyKgAAABAk7GwBwDAMN98843q6+uNjtHqlZSUKCYmxnm8ZcsWmUwmRUREOM8NGTJEZWVlRsQDAAAAfhPKKQAAWrn6+voG0x/ff/99ubu7a/Dgwc5zJ06cUOfOnY2IBwAAAPwmlFMAAMP4+vrK1dXV6Bit3sCBA1VUVCRJOnTokIqKihQaGtpgZ8P169fL39/fqIgAAADAeWPNKQAAWrmioiLdd999GjlypMrKylRVVaW0tDT17dtXZWVlSk9P16pVq5ScnKywsDCj4wIAAABNQjkFAGh2NpvtnB97++23N1uOtmzHjh3Kzs6WyWRSVFSU+vbtK0l6/vnnVVRUpKlTp2rMmDEGpwQAAACajnIKANDsLBbLOT3OZDJp06ZNzZym/dq+fbsCAwONjgEAAAA0CeUUAABtWGVlpTIzM5WRkaHdu3dr586dRkcCAAAAmoRVaAEAhqurq1NBQYFsNpuWLl1qdJxW79ixY8rPz5fNZtPWrVvlcDgUEhKipKQko6MBAAAATUY5BQAwzLZt22Sz2ZSbm6sffvhBZrPZ6EitWnFxsTIyMpSXl6fa2lr1799fDodDK1as0HXXXWd0PAAAAOC8UE4BAFrU/v37ZbPZlJmZqT179shkMikiIkKTJ0/WNddcY3S8VmnhwoXKysrSt99+q8DAQCUmJuqmm25S7969FRAQIC8vL6MjAgAAAOeNcgoA0Oxqa2uVl5en9PR0ffTRR3Jzc1NoaKgefvhhPfTQQ0pMTFS/fv2MjtlqWa1W+fn5ae7cuQoPD5e7u7vRkQAAAIALhnIKANDsbrjhBnXv3l0Wi0WJiYkKDg6WqysfQedq+fLl2rBhg+bPn69Zs2YpKChIo0ePVnh4uNHRAAAAgN/MxegAAID2z2w2q6KiQtu2bdO7776rkpISoyO1KUOHDtXs2bNVVFSk5ORk+fj46J///KdGjRqlkydPav369fruu++MjgkAAACcF5PD4XAYHQIA0P5VVlYqNzdX2dnZ2rFjh7y8vBQWFqasrCzZbDb179/f6IhtyvHjx1VYWKicnBy98847stvtCg0NVXJystHRAAAAgCahnAIAtLi9e/cqOztbOTk5+uKLL+Tp6anIyEhFRUXJ39/f6HhtzpEjR7Rx40atX79eKSkpRscBAAAAmoRpfQCAFte7d29NnTpV2dnZstlsmjBhgjZv3qyxY8caHa1V+rXRUG5ubgoICFB1dXULJQIAAAAuHEZOAQBajU8++USDBg0yOkarYzabNWnSJD322GOnXTt58qReeeUVWa1W9ezZU3l5eQYkBAAAAM4fI6cAAC1i//79+te//qWamhpJp9ZMmjdvniIjIxUbG6vNmzdTTJ2B1WpVamqqZs2apZ/fU/r8888VFRWlRYsWKS4uTllZWQamBAAAAM4P5RQAoNmVlpYqMjJS69at048//ihJSkpK0urVqxUaGqqRI0cqKSlJhYWFBidtnW688Ua9/vrrKigo0EMPPaRjx45p0aJFuuOOO2QymZSWlqZHHnlEXbt2NToqAAAA0GRM6wMANLt7771XPXv21DPPPCNJ2rdvn0aPHq3o6Gg9+eSTkqS0tDStXbtWqampRkZt1b766ivdc889qqmpkd1u17Rp0xQXFycXF+41AQAAoO3i2ywAoNmVlJQoJibGebxlyxaZTCZFREQ4zw0ZMkRlZWVGxGszfv/732vNmjXy8fFRv379FBUVRTEFAACANs/V6AAAgPavvr6+wZSz999/X+7u7ho8eLDz3IkTJ9S5c2cj4rV6xcXFDY4ffPBBPf3004qJidFjjz2mTp06Oa9df/31LR0PAAAA+E0opwAAzW7gwIEqKirSxIkTdejQIRUVFemPf/xjg1E/69evl7+/v4EpW6+fjzr7uaqqKk2ePNl5bDKZtHPnzhZKBQAAAFwYlFMAgGZ3//3367777tN7772nsrIyubi4KCEhQZJUVlam9PR0rVq1SsnJyQYnbZ127dpldAQAAACg2bAgOgCgRezYsUPZ2dkymUyKiopS3759JUnPP/+8ioqKNHXqVI0ZM8bglAAAAABaGuUUAKDV2L59uwIDA42OAQAAAKAFMa0PAGCoyspKZWZmKiMjQ7t372bNJAAAAKCDoZwCALS4Y8eOKT8/XzabTVu3bpXD4VBISIiSkpKMjgYAAACghVFOAQBaTHFxsTIyMpSXl6fa2lr1799fDodDK1as0HXXXWd0PAAAAAAGoJwCADS7hQsXKisrS99++60CAwOVmJiom266Sb1791ZAQIC8vLyMjggAAADAIJRTAIBmZ7Va5efnp7lz5yo8PFzu7u5GRwIAAADQSrgYHQAA0P4tX75cQ4cO1fz58zV8+HDFxMRoxYoVKi8vNzoaAAAAAIOZHA6Hw+gQAICOob6+Xu+9955ycnK0adMm1dbWSpLi4+MVGxur7t27G5wQAAAAQEujnAIAGOL48eMqLCxUTk6O3nnnHdntdoWGhio5OdnoaAAAAABaEOUUAMBwR44c0caNG7V+/XqlpKQYHQcAAABAC2LNKQBAs/u10VBubm4KCAhQdXV1CyUCAAAA0FpQTgEAmt2SJUv03HPPNXrt5MmTWrx4scaNG6cjR460cDIAAAAARqOcAgA0O6vVqtTUVM2aNUs/n03++eefKyoqSosWLVJcXJyysrIMTAkAAADACKw5BQBoEZ988okSEhI0bNgwPffcc3r11Ve1ePFiDRgwQHPnzpW/v7/REQEAAAAYgHIKANBivvrqK91zzz2qqamR3W7XtGnTFBcXJxcXBvICAAAAHRXlFACgRVVWVuqee+5Rt27dlJKSInd3d6MjAQAAADAQ5RQAoNkVFxc3OK6urtbTTz+tyy67TI899pg6derkvHb99de3dDwAAAAABqKcAgA0u3NdT8pkMmnnzp3NnAYAAABAa0I5BQAAAAAAAMOwAi0AAAAAAAAMQzkFAAAAAAAAw1BOAQAAAAAAwDCUUwAAAAAAADAM5RQAAB2QxWLRgAEDnH8CAgI0ZswYvf766xf8tdLT02WxWC748wIAAKB9cDU6AAAAMMbjjz+uiIgISVJ9fb0+/PBDzZw5U15eXrr99tsv2OtEREQoNDT0gj0fAAAA2hdGTgEA0EG5u7vL29tb3t7e8vHx0dixYzV8+HDl5+df0Nfp1q2bLr300gv6nAAAAGg/KKcAAICTq6urOnfurJiYGM2ZM0fh4eEKDQ3VkSNHVFFRoQcffFDBwcEaOnSonn32WdXV1enkyZMKCQnRunXrnM/jcDh04403KjMzs8G0vq1bt8pisWj16tUKCQnRtddeqxkzZqiurs75/2ZmZmrMmDEaNGiQJk6cqM8++8x57d///rcsFouCgoIUExOjsrKylvvlAAAAoFlQTgEAAJ04cUL5+fkqKipSeHi4pFNrRS1YsEDJycnq0qWL4uLidPToUa1cuVL/+Mc/tHnzZr3wwgtycXHRmDFjtHHjRufzffzxxzp8+LDzuX7uwIEDysvL07Jly/Tyyy8rPz9fNptNkvTuu+9q5syZiouLU1ZWlsxmsxISElRXV6fCwkIlJyfriSeeUEZGhoYMGaLY2FhVV1e3yO8IAAAAzYNyCgCADuqpp55SUFCQgoKCFBgYqKSkJMXFxelPf/qTJCk0NFSDBw+W2WzWu+++q8rKSi1YsEADBgzQ8OHD9eSTT2rNmjX68ccfdcstt6ioqEhHjhyRJOXl5ekPf/iD3NzcTnvdEydOaNasWRowYIBCQkIUEhKiTz/9VJKUmpqqW2+9VXfddZf8/Pz06KOP6tZbb1V1dbWWLVumhIQEhYWFqU+fPpo+fbp8fX2VlZXVcr80AAAAXHAsiA4AQAc1bdo03XTTTZKkrl27ytvbW506dXJe9/X1df785Zdfqk+fPvL09HSeGzx4sOrr67V3715de+218vb21pYtW3TLLbcoPz9fM2bMOONr+/n5OX92c3NTfX29JOnrr7/WxIkTnde6dOmipKQkZ4YFCxbopZdecl4/fvy4du/efZ6/AQAAALQGlFMAAHRQ3bt3b1AS/VLXrl0b/fkndru9wX8jIiKUl5cnPz8/ff/992fdoa9Lly4Njh0Oh6RTa16did1u1+OPP67hw4c3ON/Y6CwAAAC0HUzrAwAAv+rKK6/U7t27dfjwYee5jz/+WK6ururdu7ckOaf25eXlyWKx6KKLLmry6/j5+WnXrl3OY7vdLovFov/+97+68sorVVFRIT8/P+efJUuW6OOPP/6tfz0AAAAYiHIKAAD8qhtuuEG9evXSo48+qrKyMn344YeaM2eObr31Vnl4eEiSBg4cqB49emjVqlW6+eabz+t1YmJilJWVpYyMDO3Zs0fz58+Xw+FQQECA7r77bi1fvlw2m0179+7VggULlJubq759+17IvyoAAABaGNP6AADAr+rUqZOsVqvmzJmj8ePH6+KLL1ZkZKT++te/NnhcRESEli9frhtvvPG8Xuf666/XU089pUWLFungwYMym81asmSJunXrpoiICFVVVWnhwoWqqqpSv379tHjxYvXp0+cC/A0BAABgFJPjp0UeAAAAAAAAgBbGtD4AAAAAAAAYhnIKAAAAAAAAhqGcAgAAAAAAgGEopwAAAAAAAGAYyikAAAAAAAAYhnIKAAAAAAAAhqGcAgAAAAAAgGEopwAAAAAAAGAYyikAAAAAAAAYhnIKAAAAAAAAhqGcAgAAAAAAgGEopwAAAAAAAGCY/wMoudCdK1+YtgAAAABJRU5ErkJggg=="
     },
     "metadata": {},
     "output_type": "display_data",
     "jetTransient": {
      "display_id": null
     }
    }
   ],
   "execution_count": 33
  },
  {
   "cell_type": "code",
   "id": "24a4ad40319f441b",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T18:07:02.981141Z",
     "start_time": "2025-09-25T18:06:14.115049Z"
    }
   },
   "source": [
    "# Explode identified_name into words\n",
    "df_names = (\n",
    "    df.assign(\n",
    "        name=(\n",
    "            df['identified_name']\n",
    "            .fillna('')\n",
    "            .str.replace(r\"[^\\w'\\-]+\", \" \", regex=True)  # keep letters, digits, _, -, '\n",
    "            .str.strip()\n",
    "            .str.split()\n",
    "        )\n",
    "    )\n",
    "    .explode('name', ignore_index=True)\n",
    "    .dropna(subset=['name'])\n",
    ")\n",
    "\n",
    "# Clean up tokens\n",
    "df_names['name'] = df_names['name'].str.strip()\n",
    "df_names = df_names[df_names['name'].ne('')]\n",
    "\n",
    "df_names = df_names[['name', 'province', 'sex']].reset_index(drop=True)"
   ],
   "outputs": [],
   "execution_count": 34
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T18:01:37.658949Z",
     "start_time": "2025-09-25T18:01:33.224026Z"
    }
   },
   "cell_type": "code",
   "source": "df_names.describe().T",
   "id": "5ce61cb4109c1cee",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "             count  unique       top     freq\n",
       "name      10015595  644336    ilunga    82342\n",
       "province  10015595      12  KINSHASA  2106077\n",
       "sex       10015595       2         m  6033856"
      ],
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>unique</th>\n",
       "      <th>top</th>\n",
       "      <th>freq</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>name</th>\n",
       "      <td>10015595</td>\n",
       "      <td>644336</td>\n",
       "      <td>ilunga</td>\n",
       "      <td>82342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>province</th>\n",
       "      <td>10015595</td>\n",
       "      <td>12</td>\n",
       "      <td>KINSHASA</td>\n",
       "      <td>2106077</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sex</th>\n",
       "      <td>10015595</td>\n",
       "      <td>2</td>\n",
       "      <td>m</td>\n",
       "      <td>6033856</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 29
  },
  {
   "cell_type": "code",
   "id": "77abf83a2f6360f5",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T17:48:50.735911Z",
     "start_time": "2025-09-25T17:48:49.779413Z"
    }
   },
   "source": [
    "# Province counts before exploding\n",
    "prov_counts_orig = (\n",
    "    df['province'].value_counts()\n",
    "    .rename_axis(\"province\")\n",
    "    .reset_index(name=\"count_original\")\n",
    ")\n",
    "\n",
    "# Province counts after exploding into df_names\n",
    "prov_counts_names = (\n",
    "    df_names['province'].value_counts()\n",
    "    .rename_axis(\"province\")\n",
    "    .reset_index(name=\"count_names\")\n",
    ")\n",
    "\n",
    "# Merge both into one table\n",
    "comparison = (\n",
    "    prov_counts_orig\n",
    "    .merge(prov_counts_names, on=\"province\", how=\"outer\")\n",
    "    .fillna(0)\n",
    ")\n",
    "\n",
    "# Add augmentation percentage\n",
    "comparison[\"augmentation\"] = (\n",
    "        (comparison[\"count_names\"] - comparison[\"count_original\"])\n",
    "        / comparison[\"count_original\"]\n",
    "        * 100\n",
    ").round(2)\n",
    "\n",
    "# Sort and display top 12 provinces\n",
    "comparison = (\n",
    "    comparison\n",
    "    .sort_values(\"count_original\", ascending=False)\n",
    "    .reset_index(drop=True)\n",
    ")\n",
    "\n",
    "display(comparison.head(12))"
   ],
   "outputs": [
    {
     "data": {
      "text/plain": [
       "            province  count_original  count_names  augmentation\n",
       "0           KINSHASA         1140620      2106077         84.64\n",
       "1             AUTRES         1035751      1644326         58.76\n",
       "2            KATANGA          836220      1370359         63.88\n",
       "3           BANDUNDU          809949       604374        -25.38\n",
       "4     KASAI-ORIENTAL          434497       802757         84.76\n",
       "5          NORD-KIVU          394999       695494         76.07\n",
       "6   KASAI-OCCIDENTAL          367626       543156         47.75\n",
       "7           EQUATEUR          356404       624252         75.15\n",
       "8           SUD-KIVU          346152       408708         18.07\n",
       "9          ORIENTALE          322756       541646         67.82\n",
       "10         BAS-CONGO          295155       536702         81.84\n",
       "11           MANIEMA          127813       137744          7.77"
      ],
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>province</th>\n",
       "      <th>count_original</th>\n",
       "      <th>count_names</th>\n",
       "      <th>augmentation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>KINSHASA</td>\n",
       "      <td>1140620</td>\n",
       "      <td>2106077</td>\n",
       "      <td>84.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AUTRES</td>\n",
       "      <td>1035751</td>\n",
       "      <td>1644326</td>\n",
       "      <td>58.76</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>KATANGA</td>\n",
       "      <td>836220</td>\n",
       "      <td>1370359</td>\n",
       "      <td>63.88</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BANDUNDU</td>\n",
       "      <td>809949</td>\n",
       "      <td>604374</td>\n",
       "      <td>-25.38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>KASAI-ORIENTAL</td>\n",
       "      <td>434497</td>\n",
       "      <td>802757</td>\n",
       "      <td>84.76</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>NORD-KIVU</td>\n",
       "      <td>394999</td>\n",
       "      <td>695494</td>\n",
       "      <td>76.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>KASAI-OCCIDENTAL</td>\n",
       "      <td>367626</td>\n",
       "      <td>543156</td>\n",
       "      <td>47.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>EQUATEUR</td>\n",
       "      <td>356404</td>\n",
       "      <td>624252</td>\n",
       "      <td>75.15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>SUD-KIVU</td>\n",
       "      <td>346152</td>\n",
       "      <td>408708</td>\n",
       "      <td>18.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>ORIENTALE</td>\n",
       "      <td>322756</td>\n",
       "      <td>541646</td>\n",
       "      <td>67.82</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>BAS-CONGO</td>\n",
       "      <td>295155</td>\n",
       "      <td>536702</td>\n",
       "      <td>81.84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>MANIEMA</td>\n",
       "      <td>127813</td>\n",
       "      <td>137744</td>\n",
       "      <td>7.77</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data",
     "jetTransient": {
      "display_id": null
     }
    }
   ],
   "execution_count": 13
  },
  {
   "cell_type": "markdown",
   "id": "98ba6e48-a7c4-4aae-8b46-872489c95faf",
   "metadata": {},
   "source": "## Transition probabilities"
  },
  {
   "cell_type": "code",
   "id": "e324e1c9e8da4bd8",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T19:59:00.238808Z",
     "start_time": "2025-09-25T19:59:00.225716Z"
    }
   },
   "source": [
    "def build_transition_probs(\n",
    "        names: pd.Series,\n",
    "        *,\n",
    "        letters: str = \"abcdefghijklmnopqrstuvwxyz\",\n",
    "        start_token: str = \"^\",\n",
    "        end_token: str = \"$\",\n",
    "        alpha: float = 0.0,  # Laplace smoothing; e.g., 1.0 to avoid zeros\n",
    ") -> dict:\n",
    "    # 1) Normalize\n",
    "    names = (\n",
    "        names.astype(str)\n",
    "        .str.lower()\n",
    "        .str.replace(fr\"[^{letters}]\", \"\", regex=True)\n",
    "    )\n",
    "    names = names[names.str.len() > 0]\n",
    "\n",
    "    # 2) Prepare sequences\n",
    "    sequences = (start_token + names + end_token).tolist()\n",
    "\n",
    "    # 3) Tokens and indices\n",
    "    tokens = [start_token] + list(letters) + [end_token]\n",
    "    index = {t: i for i, t in enumerate(tokens)}\n",
    "    V = len(tokens)\n",
    "\n",
    "    # 4) ASCII lookup table (O(1) char -> idx); others -> -1\n",
    "    lut = np.full(128, -1, dtype=np.int32)\n",
    "    for ch, i in index.items():\n",
    "        lut[ord(ch)] = i\n",
    "\n",
    "    # 5) Concatenate with a separator that’s not in vocab to kill cross-boundary pairs\n",
    "    concat = (\" \".join(sequences)).encode(\"ascii\", errors=\"ignore\")\n",
    "\n",
    "    # 6) Map bytes to indices\n",
    "    arr = np.frombuffer(concat, dtype=np.uint8)\n",
    "    idx = lut[arr]\n",
    "\n",
    "    # 7) Build bigram pairs; drop invalid ones (separator & OOV)\n",
    "    a = idx[:-1]\n",
    "    b = idx[1:]\n",
    "    mask = (a >= 0) & (b >= 0)\n",
    "    a, b = a[mask], b[mask]\n",
    "\n",
    "    # 8) Count with a single bincount\n",
    "    lin = a * V + b\n",
    "    counts = np.bincount(lin, minlength=V * V).reshape(V, V)\n",
    "\n",
    "    # 9) Optional Laplace smoothing\n",
    "    if alpha and alpha > 0:\n",
    "        counts = counts + alpha\n",
    "\n",
    "    # 10) Row-normalize to probabilities\n",
    "    row_sums = counts.sum(axis=1, keepdims=True)\n",
    "    # avoid division by zero\n",
    "    probs = np.divide(counts, np.where(row_sums == 0, 1.0, row_sums), where=True)\n",
    "\n",
    "    # 11) DataFrames\n",
    "    df_counts = pd.DataFrame(counts, index=tokens, columns=tokens)\n",
    "    df_probs  = pd.DataFrame(probs, index=tokens, columns=tokens)\n",
    "\n",
    "    return {\n",
    "        \"tokens\": tokens,\n",
    "        \"index\": index,\n",
    "        \"counts\": counts,\n",
    "        \"df_counts\": df_counts,\n",
    "        \"probs\": probs,\n",
    "        \"df_probs\": df_probs,\n",
    "    }\n"
   ],
   "outputs": [],
   "execution_count": 133
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T19:59:41.506588Z",
     "start_time": "2025-09-25T19:59:03.640342Z"
    }
   },
   "cell_type": "code",
   "source": [
    "transitions_both   = build_transition_probs(df_names['name'])\n",
    "transitions_male   = build_transition_probs(df_names.loc[df_names['sex'].str.lower() == 'm', 'name'])\n",
    "transitions_female = build_transition_probs(df_names.loc[df_names['sex'].str.lower() == 'f', 'name'])\n",
    "\n",
    "# Access the probability matrices\n",
    "P_both   = transitions_both['probs']\n",
    "P_male   = transitions_male['probs']\n",
    "P_female = transitions_female['probs']"
   ],
   "id": "12a7094d94ad519f",
   "outputs": [],
   "execution_count": 134
  },
  {
   "cell_type": "code",
   "id": "67b3e0f724af646b",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T20:02:32.577754Z",
     "start_time": "2025-09-25T20:02:32.569329Z"
    }
   },
   "source": [
    "def plot_transition_matrix(ax, df_probs, title=\"\", letters=\"abcdefghijklmnopqrstuvwxyz\"):\n",
    "    hm = sns.heatmap(\n",
    "        df_probs.loc[list(letters), list(letters)],\n",
    "        cmap=\"Reds\",\n",
    "        annot=False,\n",
    "        cbar=False,\n",
    "        ax=ax\n",
    "    )\n",
    "    ax.set_title(title, fontsize=12)\n",
    "    return hm"
   ],
   "outputs": [],
   "execution_count": 135
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T20:02:36.087443Z",
     "start_time": "2025-09-25T20:02:34.953458Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# Create 1x3 grid\n",
    "fig, axes = plt.subplots(1, 3, figsize=(20, 6), sharex=True, sharey=True)\n",
    "\n",
    "hm1 = plot_transition_matrix(axes[0], transitions_male['df_probs'],   \"Male\")\n",
    "hm2 = plot_transition_matrix(axes[1], transitions_female['df_probs'], \"Female\")\n",
    "hm3 = plot_transition_matrix(axes[2], transitions_both['df_probs'],   \"All\")\n",
    "\n",
    "# Add one shared colorbar\n",
    "cbar = fig.colorbar(hm3.collections[0], ax=axes, orientation=\"vertical\", fraction=0.03, pad=0.02)\n",
    "cbar.set_label(\"Transition probability\")\n",
    "\n",
    "plt.show()"
   ],
   "id": "f15ad6eb4df27527",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 2000x600 with 4 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data",
     "jetTransient": {
      "display_id": null
     }
    }
   ],
   "execution_count": 136
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T20:03:42.518370Z",
     "start_time": "2025-09-25T20:03:42.145944Z"
    }
   },
   "cell_type": "code",
   "source": [
    "diff = transitions_male['df_probs'] - transitions_female['df_probs']\n",
    "\n",
    "plt.figure(figsize=(6, 5))\n",
    "sns.heatmap(diff.loc[list(LETTERS), list(LETTERS)], cmap=\"RdBu_r\", center=0, annot=False)\n",
    "plt.title(\"Male – Female Transition Probability Differences\")\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ],
   "id": "51ec0792317364fc",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 600x500 with 2 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data",
     "jetTransient": {
      "display_id": null
     }
    }
   ],
   "execution_count": 142
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T19:54:41.378255Z",
     "start_time": "2025-09-25T19:54:41.361591Z"
    }
   },
   "cell_type": "code",
   "source": [
    "import numpy as np\n",
    "from scipy.spatial.distance import euclidean\n",
    "from scipy.stats import entropy\n",
    "\n",
    "P_m = transitions_male['probs'].flatten()\n",
    "P_f = transitions_female['probs'].flatten()\n",
    "\n",
    "# L2 distance\n",
    "l2 = euclidean(P_m, P_f)\n",
    "\n",
    "# KL divergence (use smoothing to avoid log(0))\n",
    "kl_mf = entropy(P_m + 1e-12, P_f + 1e-12)  # D_KL(male || female)\n",
    "kl_fm = entropy(P_f + 1e-12, P_m + 1e-12)\n",
    "\n",
    "# Symmetric Jensen-Shannon divergence\n",
    "jsd = 0.5 * (kl_mf + kl_fm)\n",
    "\n",
    "print(f\"L2 distance: {l2:.4f}\")\n",
    "print(f\"KL(male||female): {kl_mf:.4f}\")\n",
    "print(f\"KL(female||male): {kl_fm:.4f}\")\n",
    "print(f\"JSD: {jsd:.4f}\")\n"
   ],
   "id": "dae836cd8a6c26e6",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "L2 distance: 0.3189\n",
      "KL(male||female): 0.0432\n",
      "KL(female||male): 0.0215\n",
      "JSD: 0.0324\n"
     ]
    }
   ],
   "execution_count": 117
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "The transition probabilities of characters in male and female names from your dataset are very similar. There are measurable but small differences: male names diverge slightly more from the female pattern than the reverse. However, the overall structure of how characters follow each other is largely shared between the two groups.",
   "id": "f0bbe49cf3e65f10"
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T20:20:11.270081Z",
     "start_time": "2025-09-25T20:20:11.244279Z"
    }
   },
   "cell_type": "code",
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from scipy.stats import chi2_contingency\n",
    "\n",
    "def chi2_row_test(trans_male, trans_female, row_token, alpha=0.5, min_expected=1e-12):\n",
    "    \"\"\"\n",
    "    Chi-square test comparing next-letter distributions for a given starting letter.\n",
    "    Handles zeros by dropping all-zero columns and adding pseudocount alpha.\n",
    "\n",
    "    Args:\n",
    "        trans_male, trans_female: dicts returned by build_transition_probs (need 'counts','index','tokens')\n",
    "        row_token: e.g., 'a'\n",
    "        alpha: pseudocount added to every cell (0.5 is common; set 0 to disable)\n",
    "        min_expected: tiny floor to avoid expected==0 after drop (defensive)\n",
    "    \"\"\"\n",
    "    i_m = trans_male['index'][row_token]\n",
    "    i_f = trans_female['index'][row_token]\n",
    "    cm = trans_male['counts'][i_m].astype(float)\n",
    "    cf = trans_female['counts'][i_f].astype(float)\n",
    "\n",
    "    # Stack into 2xK\n",
    "    table = np.vstack([cm, cf])\n",
    "\n",
    "    # Drop columns with zero total across both groups\n",
    "    keep = table.sum(axis=0) > 0\n",
    "    table = table[:, keep]\n",
    "\n",
    "    # If everything got dropped or only 1 column left, test is undefined\n",
    "    if table.size == 0 or table.shape[1] < 2:\n",
    "        return np.nan, np.nan, 0, None\n",
    "\n",
    "    # Add pseudocounts (helps with sparsity and zero expected)\n",
    "    if alpha and alpha > 0:\n",
    "        table = table + alpha\n",
    "\n",
    "    chi2, p, dof, expected = chi2_contingency(table, correction=False)\n",
    "    # Defensive floor (optional)\n",
    "    if np.any(expected < min_expected):\n",
    "        # fall back: jitter + rerun (very rare)\n",
    "        table = table + 1e-9\n",
    "        chi2, p, dof, expected = chi2_contingency(table, correction=False)\n",
    "    return chi2, p, dof, expected\n",
    "\n",
    "# Example: a single row\n",
    "chi2, p, dof, expected = chi2_row_test(transitions_male, transitions_female, row_token='a', alpha=0.5)\n",
    "print(f\"Row 'a' → χ²={chi2:.3f}, dof={dof}, p={p:.3e}\")\n",
    "\n",
    "# Run for all letters and assemble a summary (with FDR correction)\n",
    "from statsmodels.stats.multitest import multipletests\n",
    "\n",
    "tokens = [t for t in transitions_both['tokens'] if t not in ('^','$')]\n",
    "results = []\n",
    "for t in tokens:\n",
    "    chi2, p, dof, _ = chi2_row_test(transitions_male, transitions_female, row_token=t, alpha=0.5)\n",
    "    results.append((t, chi2, dof, p))\n",
    "\n",
    "df_tests = pd.DataFrame(results, columns=['row_token','chi2','dof','p'])\n",
    "# FDR (Benjamini–Hochberg)\n",
    "mask = df_tests['p'].notna()\n",
    "rej, p_adj, _, _ = multipletests(df_tests.loc[mask, 'p'], method='fdr_bh')\n",
    "df_tests.loc[mask, 'p_adj'] = p_adj\n",
    "df_tests.loc[mask, 'significant'] = rej\n",
    "\n",
    "df_tests = df_tests.sort_values('chi2', ascending=False).reset_index(drop=True)"
   ],
   "id": "d029bbd85794df95",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Row 'a' → χ²=66636.402, dof=26, p=0.000e+00\n"
     ]
    }
   ],
   "execution_count": 153
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T20:20:17.405785Z",
     "start_time": "2025-09-25T20:20:17.395818Z"
    }
   },
   "cell_type": "code",
   "source": "df_tests.head(7)",
   "id": "80e7f52285a073ea",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "  row_token          chi2  dof    p  p_adj significant\n",
       "0         a  66636.401639   26  0.0    0.0        True\n",
       "1         i  36497.202246   26  0.0    0.0        True\n",
       "2         n  32505.512092   25  0.0    0.0        True\n",
       "3         u  31904.897504   26  0.0    0.0        True\n",
       "4         g  24595.637377   26  0.0    0.0        True\n",
       "5         m  23254.272134   26  0.0    0.0        True\n",
       "6         e  19945.183545   26  0.0    0.0        True"
      ],
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>row_token</th>\n",
       "      <th>chi2</th>\n",
       "      <th>dof</th>\n",
       "      <th>p</th>\n",
       "      <th>p_adj</th>\n",
       "      <th>significant</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a</td>\n",
       "      <td>66636.401639</td>\n",
       "      <td>26</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>i</td>\n",
       "      <td>36497.202246</td>\n",
       "      <td>26</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>n</td>\n",
       "      <td>32505.512092</td>\n",
       "      <td>25</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>u</td>\n",
       "      <td>31904.897504</td>\n",
       "      <td>26</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>g</td>\n",
       "      <td>24595.637377</td>\n",
       "      <td>26</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>m</td>\n",
       "      <td>23254.272134</td>\n",
       "      <td>26</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>e</td>\n",
       "      <td>19945.183545</td>\n",
       "      <td>26</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 155,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 155
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": [
    "Male and female names in your dataset have very similar character transition patterns overall (global JSD ≈ 0.03),\n",
    "But for certain letters — especially a, i, n, u, g, m, e — the differences are statistically huge.\n",
    "These letters likely correspond to prefixes or suffixes that carry strong gender information in local naming traditions."
   ],
   "id": "d5cd70c832e5ac2"
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "## Letters frequency",
   "id": "c49254df7eee9ae2"
  },
  {
   "cell_type": "code",
   "id": "ffb0b937-fb9e-47d5-94c7-20afdf99560c",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T20:27:29.403600Z",
     "start_time": "2025-09-25T20:27:29.397414Z"
    }
   },
   "source": [
    "def letter_freq(df: pd.DataFrame, col: str = \"name\") -> pd.DataFrame:\n",
    "    s = (\n",
    "        df[col].astype(str).str.lower()\n",
    "        .str.replace(r'[^a-z]', '', regex=True)  # keep only a–z\n",
    "        .str.cat(sep='')                         # one big string\n",
    "    )\n",
    "    ser = pd.Series(list(s))\n",
    "    out = (\n",
    "        ser.value_counts(normalize=False)\n",
    "        .reindex(list(LETTERS), fill_value=0)\n",
    "        .rename_axis(\"letter\").reset_index(name=\"count\")\n",
    "    )\n",
    "    total = out[\"count\"].sum()\n",
    "    out[\"freq\"] = out[\"count\"] / (total if total > 0 else 1)\n",
    "    return out\n"
   ],
   "outputs": [],
   "execution_count": 159
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T21:39:15.252765Z",
     "start_time": "2025-09-25T21:39:15.191830Z"
    }
   },
   "cell_type": "code",
   "source": [
    "def plot_letter_grid(df_all: pd.DataFrame, use=\"freq\", sort_values=False):\n",
    "    \"\"\"\n",
    "    Plot a 1×3 grid of letter distributions for Male, Female, and Both.\n",
    "    `use` ∈ {\"freq\",\"count\"} controls y-axis.\n",
    "    \"\"\"\n",
    "    # Slice datasets (adapt values if your labels are 'M'/'F', etc.)\n",
    "    df_male   = df_all[df_all['sex'].str.lower() == str('m').lower()]\n",
    "    df_female = df_all[df_all['sex'].str.lower() == str('f').lower()]\n",
    "\n",
    "    L_m = letter_freq(df_male,   col='name')\n",
    "    L_f = letter_freq(df_female, col='name')\n",
    "\n",
    "    y = \"freq\" if use == \"freq\" else \"count\"\n",
    "    if sort_values:\n",
    "        L_m = L_m.sort_values(y, ascending=False)\n",
    "        L_f = L_f.sort_values(y, ascending=False)\n",
    "\n",
    "    ymax = max(L_m[y].max(), L_f[y].max()) * 1.10\n",
    "\n",
    "    fig, axes = plt.subplots(1, 2, figsize=(18, 5), sharey=True, constrained_layout=True)\n",
    "    for ax, data, title in zip(axes, [L_m, L_f], [\"Male\", \"Female\"]):\n",
    "        ax.bar(data[\"letter\"], data[y], color=\"steelblue\", alpha=0.8)\n",
    "        ax.set_title(title)\n",
    "        ax.set_xlabel(\"Letter\")\n",
    "        ax.set_ylim(0, ymax)\n",
    "        ax.grid(axis=\"y\", alpha=0.3)\n",
    "    axes[0].set_ylabel(\"Frequency\" if y == \"freq\" else \"Count\")\n",
    "    plt.show()"
   ],
   "id": "4a0d8e2a74a6406c",
   "outputs": [],
   "execution_count": 207
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T21:39:45.123899Z",
     "start_time": "2025-09-25T21:39:17.350875Z"
    }
   },
   "cell_type": "code",
   "source": "plot_letter_grid(df_names, use=\"freq\")",
   "id": "9265e47639c4319d",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 1800x500 with 2 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data",
     "jetTransient": {
      "display_id": null
     }
    }
   ],
   "execution_count": 208
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T21:40:08.366997Z",
     "start_time": "2025-09-25T21:39:45.174312Z"
    }
   },
   "cell_type": "code",
   "source": "plot_letter_grid(df_names, use=\"freq\", sort_values=True)",
   "id": "395dace523f1bed4",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 1800x500 with 2 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data",
     "jetTransient": {
      "display_id": null
     }
    }
   ],
   "execution_count": 209
  },
  {
   "cell_type": "code",
   "id": "98c7ed61-5c2d-4d9a-9d85-e4efe4bfc402",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T21:35:00.849876Z",
     "start_time": "2025-09-25T21:35:00.831997Z"
    }
   },
   "source": [
    "def count_ngrams(df: pd.DataFrame, n: int, col: str = \"name\") -> pd.DataFrame:\n",
    "    # Normalize and clean\n",
    "    _names = (\n",
    "        df[col]\n",
    "        .astype(str)\n",
    "        .str.lower()\n",
    "        .str.replace(r\"[^a-z]\", \"\", regex=True)\n",
    "    )\n",
    "\n",
    "    # Collect n-grams from all names\n",
    "    ngrams = []\n",
    "    for name in _names:\n",
    "        if len(name) >= n:\n",
    "            ngrams.extend(name[i:i+n] for i in range(len(name) - n + 1))\n",
    "\n",
    "    # Count\n",
    "    counter = Counter(ngrams)\n",
    "\n",
    "    # Build DataFrame\n",
    "    df_ngrams = (\n",
    "        pd.DataFrame(counter.items(), columns=[f\"{n}-gram\", \"count\"])\n",
    "        .sort_values(\"count\", ascending=False)\n",
    "        .reset_index(drop=True)\n",
    "    )\n",
    "    df_ngrams[\"freq\"] = df_ngrams[\"count\"] / df_ngrams[\"count\"].sum()\n",
    "    return df_ngrams\n",
    "\n",
    "def plot_ngrams_grid(df: pd.DataFrame, n: int, top_k: int = 20,\n",
    "                     gender_col=\"sex\", male_value=\"m\", female_value=\"f\", name_col=\"name\"):\n",
    "    \"\"\"\n",
    "    Plot top n-grams for Male, Female, and All in a 1×3 grid.\n",
    "    \"\"\"\n",
    "    # Split datasets\n",
    "    df_male   = df[df[gender_col].str.lower() == str(male_value).lower()]\n",
    "    df_female = df[df[gender_col].str.lower() == str(female_value).lower()]\n",
    "\n",
    "    # Compute n-grams\n",
    "    ng_male   = count_ngrams(df_male, n, col=name_col)\n",
    "    ng_female = count_ngrams(df_female, n, col=name_col)\n",
    "\n",
    "    # Plot in a grid\n",
    "    fig, axes = plt.subplots(1, 2, figsize=(18, 6), sharey=True, constrained_layout=True)\n",
    "    for ax, data, title in zip(\n",
    "            axes, [ng_male, ng_female], [\"Male\", \"Female\"]\n",
    "    ):\n",
    "        sns.barplot(\n",
    "            data=data.head(top_k),\n",
    "            x=f\"{n}-gram\", y=\"count\",\n",
    "            ax=ax\n",
    "        )\n",
    "        ax.set_title(f\"{title} – Top {top_k} {n}-grams\")\n",
    "        ax.set_xlabel(f\"{n}-gram\")\n",
    "        ax.set_ylabel(\"Count\")\n",
    "        ax.tick_params(axis=\"x\", rotation=45)\n",
    "\n",
    "    plt.show()"
   ],
   "outputs": [],
   "execution_count": 202
  },
  {
   "cell_type": "code",
   "id": "412883b2-c1fd-483f-b61a-ba2bfc8b04e7",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T21:35:45.360163Z",
     "start_time": "2025-09-25T21:35:03.778209Z"
    }
   },
   "source": "plot_ngrams_grid(df_names, n=3, top_k=20, gender_col=\"sex\")",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 1800x600 with 2 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data",
     "jetTransient": {
      "display_id": null
     }
    }
   ],
   "execution_count": 203
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T21:36:19.354244Z",
     "start_time": "2025-09-25T21:35:45.428602Z"
    }
   },
   "cell_type": "code",
   "source": "plot_ngrams_grid(df_names, n=4, top_k=20, gender_col=\"sex\")",
   "id": "b4eda839e6cc5d8f",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 1800x600 with 2 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data",
     "jetTransient": {
      "display_id": null
     }
    }
   ],
   "execution_count": 204
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T21:36:46.475564Z",
     "start_time": "2025-09-25T21:36:21.247703Z"
    }
   },
   "cell_type": "code",
   "source": "plot_ngrams_grid(df_names, n=5, top_k=20, gender_col=\"sex\")",
   "id": "55605ab05b9a10bb",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 1800x600 with 2 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data",
     "jetTransient": {
      "display_id": null
     }
    }
   ],
   "execution_count": 205
  },
  {
   "cell_type": "markdown",
   "id": "28e65639-9082-464d-9330-71e2a028c878",
   "metadata": {},
   "source": [
    "## Name generation"
   ]
  },
  {
   "cell_type": "code",
   "id": "28fe77f9-f9b0-4267-9dbb-3969e04b83fb",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T21:43:42.826584Z",
     "start_time": "2025-09-25T21:43:42.798968Z"
    }
   },
   "source": [
    "# tokens must include '^' (start) and '$' (end)\n",
    "tokens = [\"^\"] + list(LETTERS) + [\"$\"]\n",
    "token_to_idx = {t: i for i, t in enumerate(tokens)}\n",
    "idx_to_token = np.array(tokens)\n",
    "V = len(tokens)\n",
    "\n",
    "# Prepare a well-formed probability matrix (rows sum to 1, no NaNs/negatives)\n",
    "def prepare_prob_matrix(df_probs, tokens):\n",
    "    import pandas as pd, numpy as np\n",
    "    if isinstance(df_probs, pd.DataFrame):\n",
    "        P = df_probs.loc[tokens, tokens].to_numpy(dtype=float, copy=True)\n",
    "    else:\n",
    "        P = np.array(df_probs, dtype=float, copy=True)\n",
    "        if P.shape[0] != len(tokens) or P.shape[1] != len(tokens):\n",
    "            raise ValueError(\n",
    "                f\"Matrix shape {P.shape} does not match tokens length {len(tokens)}. \"\n",
    "                \"Use the same tokens used to build the matrix (e.g., transitions['tokens']).\"\n",
    "            )\n",
    "    # clean & renormalize\n",
    "    P[~np.isfinite(P) | (P < 0)] = 0.0\n",
    "    rs = P.sum(axis=1, keepdims=True)\n",
    "    P = np.divide(P, np.where(rs == 0, 1.0, rs), out=np.zeros_like(P), where=True)\n",
    "    return P\n",
    "\n",
    "\n",
    "def generate_name(P, token_to_idx, idx_to_token, *,\n",
    "                  target_len=None,   # exact character length (letters only), if provided\n",
    "                  min_len=1,         # minimum character length\n",
    "                  max_len=12,        # hard cap on steps\n",
    "                  temperature=1.0):\n",
    "    start = token_to_idx['^']\n",
    "    end   = token_to_idx['$']\n",
    "    cur = start\n",
    "    out = []\n",
    "\n",
    "    for _ in range(max_len):\n",
    "        row = P[cur]\n",
    "\n",
    "        # Temperature scaling (τ<1 = sharper, τ>1 = flatter)\n",
    "        if temperature != 1.0:\n",
    "            row = np.power(row, 1.0 / temperature)\n",
    "            s = row.sum()\n",
    "            row = row / s if s > 0 else row\n",
    "\n",
    "        row_mod = row.copy()\n",
    "\n",
    "        # 1) Prevent early stop before min_len\n",
    "        if len(out) < min_len:\n",
    "            row_mod[end] = 0.0\n",
    "\n",
    "        # 2) If target_len reached or exceeded, force end\n",
    "        if target_len is not None and len(out) >= target_len:\n",
    "            row_mod[:] = 0.0\n",
    "            row_mod[end] = 1.0\n",
    "\n",
    "        s = row_mod.sum()\n",
    "        if s == 0.0:\n",
    "            # Fallback: uniform over valid next tokens\n",
    "            candidates = np.arange(V)\n",
    "            # exclude '^'\n",
    "            candidates = candidates[candidates != start]\n",
    "            # exclude '$' if below min_len\n",
    "            if len(out) < min_len:\n",
    "                candidates = candidates[candidates != end]\n",
    "            probs = np.ones(len(candidates)) / len(candidates)\n",
    "            nxt_idx = np.random.choice(candidates, p=probs)\n",
    "        else:\n",
    "            row_mod = row_mod / s\n",
    "            nxt_idx = np.random.choice(V, p=row_mod)\n",
    "\n",
    "        if nxt_idx == end:\n",
    "            break\n",
    "        out.append(idx_to_token[nxt_idx])\n",
    "        cur = nxt_idx\n",
    "\n",
    "    return \"\".join(out).capitalize()"
   ],
   "outputs": [],
   "execution_count": 210
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T21:43:58.987200Z",
     "start_time": "2025-09-25T21:43:57.635944Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# Example preparation and usage\n",
    "P = prepare_prob_matrix(transitions_male['df_probs'], tokens)\n",
    "generated_var = [generate_name(P, token_to_idx, idx_to_token, min_len=5, max_len=12, temperature=0.5) for _ in range(10)]\n",
    "\n",
    "names = pd.DataFrame(generated_var, columns=[\"name\"])"
   ],
   "id": "a01eb547985d5a62",
   "outputs": [],
   "execution_count": 212
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T21:44:03.354829Z",
     "start_time": "2025-09-25T21:44:03.336042Z"
    }
   },
   "cell_type": "code",
   "source": "names",
   "id": "c5bcbff0204ceaa4",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "        name\n",
       "0    Tolemba\n",
       "1    Kalanga\n",
       "2      Mbayo\n",
       "3     Kukasa\n",
       "4     Kadamu\n",
       "5  Bongalula\n",
       "6      Kanga\n",
       "7   Mangonga\n",
       "8   Kayalela\n",
       "9      Mbalo"
      ],
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Tolemba</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Kalanga</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Mbayo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Kukasa</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Kadamu</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Bongalula</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Kanga</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Mangonga</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Kayalela</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Mbalo</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 213,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 213
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T21:44:29.411892Z",
     "start_time": "2025-09-25T21:44:29.403639Z"
    }
   },
   "cell_type": "code",
   "source": [
    "P = prepare_prob_matrix(transitions_female['df_probs'], tokens)\n",
    "generated_var = [generate_name(P, token_to_idx, idx_to_token, min_len=5, max_len=12, temperature=0.5) for _ in range(10)]\n",
    "\n",
    "names = pd.DataFrame(generated_var, columns=[\"name\"])"
   ],
   "id": "5d5ec37da189b7a8",
   "outputs": [],
   "execution_count": 216
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-09-25T21:44:33.290488Z",
     "start_time": "2025-09-25T21:44:33.284314Z"
    }
   },
   "cell_type": "code",
   "source": "names",
   "id": "f9da3dad4317ac10",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "          name\n",
       "0        Ngone\n",
       "1        Kandi\n",
       "2     Kangonga\n",
       "3  Dangangandi\n",
       "4       Kukale\n",
       "5        Mando\n",
       "6       Kasaba\n",
       "7        Mambo\n",
       "8     Mpolanga\n",
       "9      Kangena"
      ],
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Ngone</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Kandi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Kangonga</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Dangangandi</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Kukale</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Mando</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Kasaba</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Mambo</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Mpolanga</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Kangena</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ]
     },
     "execution_count": 217,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 217
  },
  {
   "metadata": {},
   "cell_type": "code",
   "outputs": [],
   "execution_count": null,
   "source": "",
   "id": "c0524e989ea1c7b9"
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}