06 Selección de Modelos.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 6. Selección de Modelos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Dataset\n",
    "\n",
    "Definimos datos de juguete:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[2, 2, 4],\n",
       "       [3, 4, 1],\n",
       "       [1, 2, 4],\n",
       "       [3, 4, 2],\n",
       "       [3, 2, 2],\n",
       "       [3, 1, 2],\n",
       "       [1, 4, 2],\n",
       "       [1, 1, 3]])"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = np.array([[2, 2,4], [3, 4, 1], [1, 2, 4], [3, 4, 2], [3, 2, 2], [3, 1, 2], [1, 4, 2], [1, 1, 3]])\n",
    "X"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Con la siguientes etiquetas:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['a', 'a', 'a', 'a', 'a', 'b', 'b', 'b'], dtype='<U1')"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y = np.array((\"a\",\"a\",\"a\",\"a\",\"a\",\"b\",\"b\",\"b\"))\n",
    "y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Validación Cruzada\n",
    "\n",
    "- [Cross-validation: evaluating estimator performance](https://scikit-learn.org/stable/modules/cross_validation.html)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### División K-Fold\n",
    "\n",
    "Podemos usar [K-Fold](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.KFold.html) para hacer cross-validation:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TRAIN: [0 1 3 4 5 7] VAL: [2 6] ['a' 'b']\n",
      "TRAIN: [0 2 3 4 5 6] VAL: [1 7] ['a' 'b']\n",
      "TRAIN: [1 2 4 5 6 7] VAL: [0 3] ['a' 'a']\n",
      "TRAIN: [0 1 2 3 6 7] VAL: [4 5] ['a' 'b']\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import KFold\n",
    "KFold??\n",
    "kf = KFold(n_splits=4, shuffle=True, random_state=0)\n",
    "\n",
    "for train_index, val_index in kf.split(X):\n",
    "    X_train, X_val = X[train_index], X[val_index]\n",
    "    y_train, y_val = y[train_index], y[val_index]\n",
    "    print(f\"TRAIN: {train_index} VAL: {val_index} {y_val}\")\n",
    "\n",
    "    #model.fit(X_train, y_train)\n",
    "    #model.predict(X_val)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### División K-Fold Estratificada\n",
    "\n",
    "Para que la división sea estratificada, usamos [StratifiedKFold](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.StratifiedKFold.html):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TRAIN: [0 1 3 6 7] VAL: [2 4 5] ['a' 'a' 'b']\n",
      "TRAIN: [1 2 4 5 7] VAL: [0 3 6] ['a' 'a' 'b']\n",
      "TRAIN: [0 2 3 4 5 6] VAL: [1 7] ['a' 'b']\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import StratifiedKFold\n",
    "StratifiedKFold??\n",
    "kf = StratifiedKFold(n_splits=3, shuffle=True, random_state=2)\n",
    "\n",
    "for train_index, val_index in kf.split(X, y):\n",
    "    X_train, X_val = X[train_index], X[val_index]\n",
    "    y_train, y_val = y[train_index], y[val_index]\n",
    "    print(f\"TRAIN: {train_index} VAL: {val_index} {y_val}\")\n",
    "\n",
    "    #model.fit(X_train, y_train)\n",
    "    #model.predict(X_val)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Búsqueda en Grilla (Grid Search)\n",
    "\n",
    "- [Tuning the hyper-parameters of an estimator](https://scikit-learn.org/stable/modules/grid_search.html)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Grilla de Parámetros"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from sklearn.tree import DecisionTreeClassifier\n",
    "DecisionTreeClassifier??"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "param_grid = {\n",
    "    'criterion': ['gini', 'entropy'],\n",
    "    'max_depth': [1, 2],\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Podemos listar todas las combinaciones para usarlas a mano con [ParameterGrid](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.ParameterGrid.html):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'criterion': 'gini', 'max_depth': 1}\n",
      "{'criterion': 'gini', 'max_depth': 2}\n",
      "{'criterion': 'entropy', 'max_depth': 1}\n",
      "{'criterion': 'entropy', 'max_depth': 2}\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import ParameterGrid\n",
    "\n",
    "for params in ParameterGrid(param_grid):\n",
    "    print(params)\n",
    "    model = DecisionTreeClassifier(**params, random_state=0)\n",
    "    #model.fit(...)\n",
    "    #model.predict(...)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Grilla de Parámetros + Validación Cruzada\n",
    "\n",
    "[GridSearchCV](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.GridSearchCV.html) nos sirve para hacer validación cruzada sobre una grilla de parámetros. Sklearn se encarga de todo el proceso y nos devuelve una tabla de resultados y el mejor clasificador obtenido.\n",
    "\n",
    "La búsqueda se puede configurar de varias maneras. Por defecto la validación cruzada es estratificada.\n",
    "\n",
    "- [scoring parameter](https://scikit-learn.org/stable/modules/model_evaluation.html#scoring-parameter9)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import GridSearchCV\n",
    "GridSearchCV??\n",
    "model = DecisionTreeClassifier(random_state=0)\n",
    "\n",
    "cv = GridSearchCV(model, param_grid, scoring='accuracy', cv=3)\n",
    "cv.fit(X, y);\n",
    "cv.fit??"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "crit.\tdepth\t| mean\tstd\trank\n",
      "gini\t1\t| 0.67\t0.27\t3\n",
      "gini\t2\t| 0.78\t0.16\t1\n",
      "entropy\t1\t| 0.67\t0.27\t3\n",
      "entropy\t2\t| 0.78\t0.16\t1\n"
     ]
    }
   ],
   "source": [
    "results = cv.cv_results_\n",
    "params = results['params']\n",
    "mean = results['mean_test_score']\n",
    "std = results['std_test_score']\n",
    "rank = results['rank_test_score']\n",
    "\n",
    "print(\"crit.\\tdepth\\t| mean\\tstd\\trank\")\n",
    "for p, m, s, r in zip(params, mean, std, rank):\n",
    "    print(f\"{p['criterion']}\\t{p['max_depth']}\\t| {m:0.2f}\\t{s:0.2f}\\t{r}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "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>param_criterion</th>\n",
       "      <th>param_max_depth</th>\n",
       "      <th>mean_test_score</th>\n",
       "      <th>std_test_score</th>\n",
       "      <th>rank_test_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>gini</td>\n",
       "      <td>1</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.272166</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>gini</td>\n",
       "      <td>2</td>\n",
       "      <td>0.777778</td>\n",
       "      <td>0.157135</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>entropy</td>\n",
       "      <td>1</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.272166</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>entropy</td>\n",
       "      <td>2</td>\n",
       "      <td>0.777778</td>\n",
       "      <td>0.157135</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  param_criterion param_max_depth  mean_test_score  std_test_score  \\\n",
       "0            gini               1         0.666667        0.272166   \n",
       "1            gini               2         0.777778        0.157135   \n",
       "2         entropy               1         0.666667        0.272166   \n",
       "3         entropy               2         0.777778        0.157135   \n",
       "\n",
       "   rank_test_score  \n",
       "0                3  \n",
       "1                1  \n",
       "2                3  \n",
       "3                1  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "df = pd.DataFrame(results)\n",
    "df[['param_criterion', 'param_max_depth', 'mean_test_score', 'std_test_score', 'rank_test_score']]\n",
    "#df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "best_model = cv.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'criterion': 'gini', 'max_depth': 2}"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cv.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgMAAAGFCAYAAABg2vAPAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/P9b71AAAACXBIWXMAAA9hAAAPYQGoP6dpAACCkklEQVR4nO3dd1xT1/8/8FeAACJTEBVkQ2WpDEEtVCjWURWLAyuIaNXWUTd2iHuPuqqtW9HiqqMoWvupE604SmU5cKEsFVFARVHm+f3BL/drTAIEExLI+/l45NFyz73nvm/MuXnn3nPP4THGGAghhBCistQUHQAhhBBCFIuSAUIIIUTFUTJACCGEqDhKBgghhBAVR8kAIYQQouIoGSCEEEJUHCUDhBBCiIqjZIAQQghRcRqKDoDIT1ZWFp49e6boMAj5YCYmJrC0tFR0GIQ0WpQMNFJZWVlwcnJCcXGxokMh5IPp6OggLS2NEgJC5ISSgUbq2bNnKC4uxq5du+Dk5KTocAips7S0NISFheHZs2eUDBAiJ5QMNHJOTk7w8PBQdBiEEEKUGHUgJIQQQlQcJQOEEEKIiqNkgBBCCFFxlAwQQgghKo6SAUIIIUTFUTJACCGEqDhKBgghhBAVR8kAqTdxcXHg8XhCr8mTJ8us/ufPn4vU7+/vL7P6CSGksaJkgNQ7LS0t+Pj4wMfHB7a2tiLlGRkZ2LZtG8aOHYsOHTpAS0sLPB4Pffr0qbZeDQ0Nrl4HBwd5ha+04uLisGzZMgQHB8PGxoZLiA4ePPhB9VpbW4skWe+/bt26JaOjIIQoAo1ASOpdy5YtceHCBYnla9aswc8//yx1vbq6uly9O3bswFdffVXnGBuioKAgvHjxQm71u7q6wsDAQGyZjo6O3PZLCJE/SgaI0jExMUGvXr3QoUMHdOjQASdPnsS6desUHZbSc3Z2hoODA/e+DRo0CDk5OTKrf926dXTbhZBGipIBonRmzpwp9PfVq1cVFEntVFZW4ty5c/D09IS+vr7C4rh48aLQ3+rq6gqKhBDS0FCfASK1adOmgcfjwcHBAa9evRIpT09Ph76+Png8Hn777TcFRFg/bt68ienTp8Pa2hoBAQEoKChQdEiEEFIndGWASG3JkiU4d+4c/vvvP4wbN07oC7+srAwhISEoKipCaGgowsPDFRip7D158gR79+5FdHQ0EhMTueUeHh7Q09NTYGTyt2nTJqxcuRLFxcUwNTXFxx9/jLCwMBgZGSk6NELIB6JkgEiNz+dj7969cHd3R3R0NLp3746wsDAAQGRkJBISEmBra4sNGzYoOFLZKC4uxpEjRxAdHY0TJ06goqICAGBra4vQ0FAMGTIEjo6OYrcNDg7G48ePpd5nr169EBkZ+UFxy9q+fftE/p45cya2bt2K4OBgBUVFCJEFSgZIndjb2+PXX3/FsGHDMG7cOHTu3Bn37t3DypUruWRBkffPP1RlZSXi4uIQHR2NQ4cOoaioCEBV58ZBgwZhyJAh+Pjjj2usJyEhAZmZmVLv397eXupt5KVz586YPn06unTpAktLS1RUVCAhIQELFizAuXPnEBISAgMDA3Tv3l3RoRJC6oiSAVJn4eHhOHHiBHbv3s31XGeMYcGCBfD29lZ0eHXy8OFDrFu3Drt37+Z64uvo6CAkJARDhgxBjx49oKFR+2aTkZEhp0jrz969e0WWde3aFf7+/ujbty+OHz+OiRMn0lgDhDRg1IGQfJANGzbA1tYWiYmJyMvLw2effYbvv/9e0WHV2cmTJ7Fs2TLk5OTA3NwcO3fuxJMnT7Bnzx707t1bqkSgsVNXV8eKFSsAALdv38b169cVHBEhpK4oGSAfRE9PDx07duT+HjVqFHg8ngIj+jCtWrWCrq4ugKqrBDNnzsSCBQuQmpqq4MiUk5OTE9eB8M6dOwqOhhBSV/Qzh3yQffv2Ye/evVBTU0NlZSUmT56MgIAANG/eXNGh1UmPHj3w5MkTHD58GNHR0Th58iSWL1+O5cuXw8XFBUOGDEFoaCisrKxqVV9j6kAoiaamJoCqJ0kIIQ0TJQOkzjIyMjBmzBgAwNatW7Fr1y6cOXMGX331FY4dO6bg6OpOR0cHoaGhCA0NRW5uLvcoYVJSEiIjIzFjxgz4+PhgyJAhCA4OhrGxscS6GkMHwurk5eUhLy8PAGBhYaHgaAghdUW3CUidlJeXIyQkBC9evEBISAi++uorREdHw8TEBH/++Wed5hZQRi1btsSUKVOQmJiI69ev44cffoC5uTkuXLiAsWPHolWrVggMDMS+fftQXl4usn1GRgYYY1K/duzYUf8HWwfLli0DYwxGRkbw8vJSdDiEkDqiZIDUyZw5c3D58mXY2Nhg48aNAAAzMzNERUUBAH744QekpKQoMkSZc3FxwdKlS5GZmYnTp09j+PDh0NbWxrFjxxASEiLTeQDq0+XLl2FtbQ1ra2uRY1ixYgXWrl2Lp0+fCi0vKipCZGQkVq1aBQCYNWsW+Hx+vcVMCJEtSgaI1OLi4rB06VJoaGhg9+7dQuMJ9OnTB+PHj0dJSQkGDx6M4uJiqeuPj4+HiYkJ91q+fDkA4MSJE2KX1zc1NTUEBAQgKiqKe9Lg888/V/hcABMmTBB6f7KzswEAI0aMELtc4O3bt8jMzERmZqbI1Y2cnBxMmjQJLVq0gI2NDTp27Ag3NzeYmJhgyZIlAIApU6ZgypQp9XOQhBC5oD4DRCoFBQUICwtDZWUl5s2bh86dO4uss2LFCpw/fx6pqamYNGkStmzZItU+ysrKkJ+fX+PyuiQastakSROEhIQgJCRE0aGgqKhI7PsmGDBJQDCCYm0MHjwYAPDvv/8iKysLqamp4PF4sLCwgI+PD0aPHl2rwZcIIcqNkgEilWbNmtV4OVxLS+uDbhH4+/uDMVbn7VXVjh076tTXoLr3u1OnTujUqdMHRkYIUXaUDJB6l5ubC19fXwDAoEGDMHHiRJnU++rVK/Ts2RMAuB7uhBBCakbJAKl3JSUliI+PBwB06NBBZvWWl5dz9RJCCKk9SgZIvZH35X9DQ0O6vUAIIXVATxMQQgghKo6SAUIIIUTFUTJACCGEqDhKBgghhBAVR8kAIYQQouIoGSByl5GRAR6PBx6PJ9N6BXVmZGTItF5CCFE1lAwQImfFxcWYN28eXFxcoKOjA2NjY3Tr1g1Hjx6V6X6mTZvGJUjDhw+vcf3Hjx/j+++/h4uLC/T09KCnpwcHBweEhYXh/PnzYrepqKjAtm3b4OfnByMjI2hra8POzg4TJkzA48ePZXo8hJD6Q+MMELnj8/lo06aNzOsV1KnMs+UVFBSgS5cuuHHjBvh8PlxcXPD8+XOcOnUKp06dwsyZM7FgwYIP3k9CQgLWrFlT6/VjY2MRFhaGoqIiNGnSBB999BGAqomJdu/ejZYtW6JLly5C27x9+xZffPEFTpw4AaBqemcLCwvcvn0bv/zyC3bv3o1Tp07Bw8Pjg4+HEFLPGGmUrl69ygCwq1evKjoUldavXz8GgLVv355lZWVxyw8cOMD4fD4DwI4fP/5B+ygtLWWurq5MT0+Pde3alQFgw4YNk7j+P//8w/h8PuPz+Wz16tXszZs3QuU3b95kV65cEdnu22+/ZQBY8+bN2cmTJ7nlRUVFbOjQoQwAa926NXv9+vUHHc/76LNMiPzRbQJC5CQlJQUxMTFQU1PDvn37YGFhwZUNHDgQ06ZNAwDMnj37g/azaNEiXL9+HYsXL0br1q2rXbe8vBwjRoxAWVkZNm7ciMmTJ0NbW1toHScnJ3h7ewstKywsxObNmwEAq1evxmeffcaV6erqYvv27XB0dEROTg7Wr1//QcdDCKl/lAwQqeTm5mLs2LGwsLCAtrY2rK2tMXnyZBQWFmLu3Lli71dX14HQ398fPB4PO3bswLNnzzB+/HhYWlpCS0sLlpaWmDRpEp4/fy42FmXvQHjgwAEAwKeffgpHR0eR8jFjxgAA/vvvP9y/f79O+7h27RoWL16Mjh07Yty4cTWuHxsbi7t376JNmzb46quvar2fy5cvo6ysDGpqahg4cKBIuYaGBvr16wcA2LNnT+0PgBCiFKjPAKm1e/fuwc/PD48ePYK6ujpcXV1RUVGBdevW4c8//0Tv3r3rXHdOTg48PDyQm5sLFxcXaGhoICMjA2vXrsXFixdx8eJFpe4bIM6lS5cAQOTeu4ClpSWsra2RkZGBS5cuwdbWVqr6KyoqMHLkSADAli1boKZWc25/+PBhAEDfvn3x8uVLbN26FefPn0dxcTGsrKzQt29f9O3bV2S7goICAICJiQm0tLTE1i248pGcnIzi4mLo6OhIdTyEEMWhZIDUWlhYGB49eoQOHTrg0KFDsLS0BADcvXsXQUFBH3R5eP78+ejRoweuXr2K5s2bAwDOnj2LPn364L///kN0dDRGjBghk+MIDg6uU8/3Xr16ITIystbr37lzBwBgb28vcR07OztkZGTg9u3bUsezatUqJCQkIDIyEm3btq3VNgkJCQAATU1NuLq6IicnR6h827Zt6N69Ow4ePAg9PT1uuaGhIQDg2bNnKCkpEZsQZGdnAwAYY7h9+zbc3d2lPiZCiGJQMkBqJS4uDleuXIGWlpZQIgAADg4OOHjwIFxdXetcv5GREXbv3g19fX1u2aeffoqRI0di3bp1OHr0qMySgYSEBGRmZkq9XXVf6uIIfk03a9ZM4jqCssLCQqnqvnv3LubMmQMHBwfMmjWr1tsJkqDly5eDz+djy5YtCA4OhpqaGmJjYzF+/HicOHECo0ePFrrc7+3tDTU1NVRWVuLgwYMYMmSIUL3l5eWIiYnh/pb2eAghikV9Bkit/O9//wMAdOvWTSgREHBycoKvr2+d6w8JCRFKBAQ6d+4MAEhPT69z3e/LyMgAY0zq144dO6Taz9u3bwFU/QqXRNB5782bN7WulzGGUaNG4c2bN9i0aZNIB8DqvHr1CgBQVlaGn3/+GaNGjYKBgQH09PQwZMgQbNmyBQCwd+9e3Lx5k9uuefPmGDx4MABgypQpOHXqlFCdI0aMwK1bt7hlxcXFtY6JEKJ4lAyQWhFcxm7fvr3Eddzc3Opcv4ODg9jlLVq0APB/X2INieBLurS0VOI6goShSZMmta53w4YNOH/+PL766it8+umndYqpWbNmYgcmGjBgAGxsbAAAf/31l1DZr7/+Cnd3dzx9+hTdunWDmZkZ2rVrBxMTE0RHR2P48OFcJ1FxiR0hRHlRMkBqpaioCACE7iO/70O+AJo2bSp2uaBTXGVlZZ3rVhQjIyMAQH5+vsR1BLcSBOvW5NGjR/jxxx9hamqKFStWSB2T4LaEg4MDNDRE7xLyeDzuyYcHDx4IlRkaGiI+Ph7Lly+Hp6cnXr58ifT0dLRv3x47d+7EjBkzwBgDALRq1Urq2AghikN9Bkit6OrqAqj+F/rLly/rK5wPUl8dCNu0aYOHDx/i3r17EtcR3P6o7QiNd+7cQVFREcrKyuDs7CxS/uLFCwDA77//zt3auXPnDpeoOTo6Ijs7u9pbC4KrFBUVFWLLvvvuO3z33XciZfv27QNQlTRI27+CEKJYlAyQWhF8WSUnJ0tcp7oyZVJfHQg7d+6MM2fOSBznPysrixsjoVOnTlLV/fbtW+4WQ03l715V8fHxwcmTJ6vtgyEoMzc3lyqm33//HQDwxRdfyHxSKkKIfNFtAlIrPXv2BACcPHkSWVlZIuW3bt3ChQsX6jusOqmvDoSCwXni4uKEOtcJbNy4EQDg6ekJOzu7WtXp7+9fbYzDhg0DAAwbNoxbJngsEAC+/PJL8Hg85OTk4OTJkyL1JyQkICUlBQCERhmsyT///IMjR45AXV0dkydPrvV2hBDlQMkAqRV/f3907NgRJSUlGDhwoFBCcO/ePQwYMADq6uoKjFD5uLm5ISgoCJWVlRg8eDD3HD4AHDx4kLvnP3fuXJFtL1++DGtra1hbW4uMBfAhHB0dMXToUADA2LFjkZaWxpU9ePCAe3wzICBA5GrFtWvXcOjQIZSUlHDLKioqsH//fvTt2xeMMXz//fcf1JGUEKIYdJuA1AqPx8OuXbvQpUsXJCQkwNbWFq6urqisrMSNGzdgY2ODsWPHYu3atZQUvGPr1q24ffs2UlJSYGdnx81aKLg9MH36dPTp00dku7dv33K3MsrLy2Ua06+//orbt2/jypUrcHFxgaurK9TU1HD9+nVUVFTA2dkZu3btEtkuPT0dAwcOhJaWFqysrKCvr4/79+9znSAnT56MRYsWyTRWQkj9oCsDpNbs7e2RmJiI0aNHo0WLFkhLS8OLFy8wbtw4/Pvvv9xwwfRY2f8xNjZGQkICZs+eDTs7O9y6dQvPnz9HQEAAjhw5gsWLF9d7TLq6uvjnn3+wYsUKuLm54cGDB7hz5w6cnZ2xYMECXLlyRezTAO3bt8e4cePQpk0bPH36FKmpqWjatCmGDBmC+Ph4rF69mvoKENJA8ZjgWSDSqCQmJsLT0xNXr16tt/nle/fujePHj2Pt2rWYMGFCveyTNH6K+CwTomroygCRiczMTG5UOkkT8xBCCFFOlAyQWktPT8fq1atFBtFJSUlBnz59UFpaii5dulQ7SiEhhBDlQx0ISa0VFRVh6tSpiIiIgJmZGczMzPD06VOuM5yVlRWioqIUGyQhhBCp0ZUBUmvW1taYNWsWOnbsiIqKCqSkpODZs2dwc3PDrFmzkJiYCFtbW0WHSQghREp0ZYDUmqGhIebPn4/58+crOhRCCCEyRFcGCCGEEBVHyQAhhBCi4igZIKQWeDweeDwe11mSEEIaE0oGCCFSKywsxJw5c+Dh4QE9PT3w+Xy0bNkSffr0weHDhxUdHiFEStSBkBAilXv37sHf3x8PHz4Ej8eDlZUVmjVrhoyMDPz555/4888/MWLECGzbtk3RoRJCaomuDBBCpDJmzBg8fPgQdnZ2SE5OxoMHD3D16lXk5eXhl19+AQBs374dhw4dUnCkhJDaomSAEFJrr169wpkzZwAAK1euRLt27bgydXV1fPvtt/jiiy8AAEePHlVIjIQQ6VEyQGrl5cuXmDt3Ltq3bw9dXV1oaWnBzMwMHTt2xPTp0/Ho0SOh9XNycvDzzz+jZ8+esLOzQ5MmTaCvrw9PT08sWrQIr169ErufuXPngsfjYfjw4SgtLcWCBQvg5OSEJk2aoHXr1hg/fjwKCwu59Q8cOABfX18YGBhAX18fvXr1Qmpqqti6/f39wePxsGPHDuTk5GDkyJEwNzeHtrY27O3tMXPmTLx+/bpO709CQgLCwsJgaWkJLS0tNGvWDJ999hkOHjwocZuzZ88iKCgIrVq1Ap/Ph6GhIezt7REcHIz9+/fXKQ55KykpgWBuMzs7O7HrCJbLeuplQogcMdIoXb16lQFgV69e/eC6ioqKmLOzMwPAeDwec3BwYF5eXszS0pLx+XwGgB09elRom4iICAaANWnShNnY2DAvLy9ma2vL1NXVGQDm6urKCgsLRfY1Z84cBoCFhoYyPz8/xuPxmLOzM3N0dGRqamoMAPP09GSlpaUsMjKSAWAWFhbMzc2NaWtrMwBMX1+f3b17V6RuPz8/BoDNnj2bmZiYMA0NDebm5sYcHR0ZAAaAubu7i41LUP7gwQORskWLFjEej8ft283NjbVs2ZLbZsyYMSLbbN26lSs3MjJi7u7uzNXVlRkZGXHHqKwsLCwYALZ+/XqRssrKSubt7c0AsHXr1slkf7L8LBNCxKNkoJGS5Ql0zZo1DABr27Yty8jIECp7/fo127VrF7t27ZrQ8lOnTrG4uDhWXl4utDwrK4v17duXAWDjxo0T2ZcgGeDz+czR0ZGlpaVxZQkJCczAwIABYIMGDWK6urosNjaWK8/Ly2Pu7u4MAAsLCxOpW5AM8Pl89vHHH7OcnByuLDExkbVu3ZoBYMOGDRPZVlIysHv3bu4Lfffu3ayyspIrO3HiBDM1NWUAWFRUFLe8vLycGRsbMwBs7dq1rKysTKjOa9eusc2bN4vEUJ1t27YxHx+fOr2ktXfvXsbj8Zienh7bsGEDe/jwISsuLmbXrl1joaGhDADr2LEje/v2rdR1i0PJACHyR8lAIyXLE+jo0aMZALZmzRoZRFaVQPD5fKanpyeSLAiSAQDsypUrIttOmzaNK1++fLlI+bFjx7gv5/e9mwxkZWVJ3FZNTY1lZmYKlYlLBsrKypilpSUDIJSUvOvQoUMMAHN0dOSWPX78mAFghoaGYrepi3ffN2lfdXH8+HHWqVMnkbp0dXXZ4sWL2Zs3b2R2bJQMECJ/1GeA1MjS0hIAEBsbi6Kiolpv9/LlS2zbtg0jRoxAjx498Mknn8DX1xfdu3eHmpoaioqKcPfuXbHburm5wdvbW2R5hw4duP//5ptvJJYXFhaioKBAbN39+/eHhYWFyPJevXrBzs4OlZWV+Pvvv2s8vsuXLyMrKwvm5uYIDAwUu05gYCD4fD5u3brF9ato3rw5tLW18eLFCxw7dqzG/dTG3LlzwaqSe6lfdfHgwQMUFBSAx+OhdevWcHd3h4GBAV69eoXt27fj5MmTMjkuQkj9oHEGSI1GjBiBVatW4cyZMzAzM0O3bt3g4+MDX19feHl5QU1NNKeMj4/HgAED8OTJk2rrzs/PF7vc3t5e7HJTU1MAgImJCQwMDCSWA1U935s1ayayjrOzs9i6eTwenJyckJ6ejrS0tGrjBoCUlBQAwOvXr+Hr6ytxPR6PB6CqU6WZmRnU1dUxdepULF68GIGBgXB1dUW3bt3QsWNHBAQEoHnz5jXuW5EmTZqEtWvXom3btkhMTISbmxsAoLKyEtu2bcO3336LoKAgHDlyBH369FFssISQWqFkgNSoZcuWuHz5MubNm4fDhw8jJiYGMTExAABzc3P8+OOP+Pbbb7kvvaKiIvTv3x95eXno2rUrfvzxR7Rr1w5GRkbg8/kAqq42ZGdno6ysTOw+mzZtKna5YB81lQNVX07itGjRQuKxCspqcwXk+fPn3H/j4+NrXL+4uJj7/4ULF8LS0hK//vorrl27huvXrwOoejyvd+/eWLlypcSESJGuXbuGdevWQV1dHYcOHYKDgwNXpqamhq+//ho5OTmYP38+vv/+e0oGCGkgKBkgtWJvb4/o6GiUl5cjKSkJ58+fR2xsLM6fP48JEyagtLQUU6dOBQAcP34ceXl5sLCwwLFjx6CtrS1UF2NM4iX8+lDd1QpBmZ6eXo316OrqAgC6du2KU6dOSRUDj8fD6NGjMXr0aDx58gQXLlzA6dOnsX//fsTGxiIpKQmpqakwNDSsVX3bt2/H9u3bpYpB4MKFC1KtyxjDRx99JJQIvKtXr16YP38+0tLSUFRUVKv3khCiWJQMEKloaGjAy8sLXl5eiIiIwNy5czFv3jysX7+eSwbu378PAPDy8hJJBAAgNTW1zs/zy8LNmzcllt26dQsA4OTkVGM9bdu2BQDcuHEDjDGhqxLSaNGiBQYMGIABAwZg3rx5cHFxQXZ2No4dO4awsLBa1ZGVlVWrqxMf6uXLl1KtX1JSQskAIQ0AdSAkH0Rwr/zdQYd0dHRElr3rp59+kn9g1YiJicHDhw9Flh8/fhz37t2DmpoaunfvXmM9vr6+aNWqFXJzc7Fr1y6ZxNa8eXN89NFHACS/f+LUVwfCNm3aAADu3LmD9PR0sev89ddfAKr6dRgbG0tVPyFEMSgZIDWaPn06Nm7ciKdPnwotz8/Px8qVKwFUXQUQ6NKlC4Cq3vYbNmzglpeUlCAyMhJ79uyBpqZmPUQuHmMMgwcPxuPHj7llycnJGDNmDABgyJAhsLKyqrEeTU1NLF++HAAwevRorF+/HiUlJULrFBYWIjo6Gt999x237ObNmxg1ahTi4+NF+jUcPnwYCQkJAITfU2XRvXt3tGzZEhUVFRgwYIDQaI+VlZXYvn07lixZAgAYNmxYna+WEELqmfyfXiSKIMtns7/44gvuOXJra2vm7e3NnJ2dmaamJgPAmjVrxpKTk4W2GTp0KLeNmZkZ69ChAzdg0MKFC5mVlRUDwM6ePSu0neB5eXED/zDG2NmzZxkAZmVlJTFewX7fHyBIMM7ArFmzmLGxMePz+czd3Z05OTlx27Rr144VFBTUuk7GGFuxYgU3sqKOjg5zc3Nj3t7ezMbGhhuZ0M/Pj1s/KSmJq69p06asffv2zMvLS2jUwpEjR0o8PkU7e/Ys09PT40aktLS0ZO7u7ty/LwD2ySefsNevX8tkfzTOACHyR1cGSI1mzZqFGTNmwMfHB2VlZUhOTkZGRgbs7e0xdepUXL9+He3btxfaJioqCsuWLUObNm3w9OlTpKenw8PDAzExMZgxY4aCjqSKra0trl69itDQUOTm5iI9PR22traYPn064uPjYWRkJFV9ERERSElJwZgxY9C6dWvcuXMHN27cgIaGBnr06IG1a9cK3Ub46KOPsG3bNoSGhsLCwgKZmZlISkpCZWUlPv/8cxw4cABbt26V9WHLjL+/P27cuIGIiAi4uroiPz8f165dA5/PR0BAALZs2YKzZ89yt4sIIcqPx1gdRx0hSi0xMRGenp64evUqPDw8FB2OUvD398e5c+cQFRWF4cOHKzocUkv0WSZE/ujKACGEEKLiKBkghBBCVBwlA4QQQoiKo2SAEEIIUXE0AiFRGXFxcYoOgRBClBJdGSCEEEJUHCUDhBBCiIqjZIAQQghRcZQMEIWwtrYGj8ej+/jv2LFjB3g8ntBrzZo1ig7rgwUFBYkcV0ZGhqLDIoS8g5IBQpSMvr4+fHx84OPjA3Nzc5Fyf39/kS/X91//+9//ZBrTrVu3MHPmTPTo0QO2trbQ19eHlpYWLCwsMGDAgGr35+zszB0PIUQ50dMEhCgZd3f3Wl0xcXBwgKmpqdgyaedXqMmpU6ewaNEi8Hg8NG/eHHZ2digrK0NGRgb++OMP/PHHHxg9ejQ2btwosu3ixYu5/6dZDAlRTpQMENJARUZG1tscCx4eHtizZw+6du0qlIC8efMGq1atwsyZM7Fp0yYEBARg0KBB9RITIUR26DYBIaRGH3/8MUJCQkSuRDRp0gQzZszA559/DgA4dOiQIsIjhHwgSgZU3P3798Hj8aChoYHHjx9LXG/37t3g8Xiwt7cXWv7PP//gu+++g7e3N1q1agVNTU20aNECvXv3xrFjx6SOZ/jw4eDxeJg7d67Y8oyMDO6+uCQnT55E//79uXiaN2+OwMBAnDlzRup4SO04OTkBqLpSQAhpeCgZUHG2trbw8fFBRUUF9u7dK3G96OhoAMDQoUOFlvfr1w8rVqxAeno6jI2N0bZtWzDGcPz4cQQGBmLWrFlyjf9djDGMHTsW3bt3R0xMDMrKyuDq6grGGI4dO4auXbti6dKl9RaPvB08eBD9+vVDQEAAgoODsWzZMjx8+LDe42CM4eLFiwAALy+vet8/IeTDUZ8BgvDwcMTHxyM6OhpTp04VKc/NzcWpU6cAiCYDS5cuRUBAAGxtbYWWnzp1CkOGDMHChQsRGBgIb29v+R3A/7dkyRJs3LgRlpaW2LRpE3r27MmV7dmzB9988w0iIyPRsWNHfPrpp7Wud8KECUhKSpI6Hnd3d6xbt07q7Wrrzz//FPr74MGDmD17Nn766SdMnDhRbvsVKCoqwp07d/DTTz/h8uXLcHR0xKRJk+S+X0KI7FEyQDBo0CBMmjQJycnJuH79OlxdXYXK9+zZg4qKCvj6+op86Y8aNUpsnZ999hkWLVqEr7/+Gjt27JB7MlBYWIjFixdDQ0MDMTEx8PDwECoPDQ1Fbm4uIiIisHz5cqmSgWvXriE+Pl7qmDQ05NO82rZti8DAQHTv3h1WVlbQ0NBASkoKfvrpJ8TExGDSpEnQ0dGR+G/zIcrLy8Hn84WWNW3aFJGRkfj++++hr68v830SQuSPkgECQ0NDBAYG4sCBA4iOjsayZcuEygW3CMLDw8Vun5aWhgMHDiA1NRX5+fkoKysDALx48QIAkJiYKMfoq/z55594/fo1fHx8RBIBgQEDBiAiIgLnzp1DRUUF1NXVa1W3sg2MJO5qQ+fOnfHHH39gzJgx2LRpE3744QeEhISgadOmMt03j8fjxgvIz89HZmYmXr9+jYMHD8LLywtBQUEy3R8hpH5QMkAAVH3RHzhwAHv27MGSJUugplbVneT69etITk6Gtra22EfGZs6ciSVLlqCyslJi3fn5+XKLWyAlJQUAcO/ePfj6+opdhzEGoKqTW35+vsRn9BuypUuXIioqCgUFBTh79iz69Okj0/rV1dVx4cIF7u+3b9/i119/xfTp09G/f38cPHgQ/fv3l+k+CSHyR8kAAQD07NkTpqamyMnJQVxcHAICAgAAv/32GwCgb9++MDAwENpm//79WLRoEdTU1DBnzhz0798fNjY2aNq0KdTU1HDmzBl07dqVu1IgT8+fPwcAPHnyBE+ePKlx/eLiYjlHpBiGhoZwcXFBUlIS7ty5I/f9aWtrIyIiAqWlpYiMjMSPP/5IyQAhDRAlAwRA1f3twYMHY+3atYiOjkZAQAAqKyuxZ88eAKIdBwEgKioKABARESH2UcC6XBEQPDIo+BX/vtevX4tdrqurCwAYOXIktm7dKvV+q6OsHQgl0dTUBIB6ScIE+vbti8jISNy9excvXrwQSRwJIcqNkgHCCQ8Px9q1a3Ho0CGsX78e8fHxePjwIUxNTYV65gvcv38fAPDJJ5+IrU/wuJk0BPe4Jf26l/Rrt23btgCqOvvJmrJ1IKxOeXk5bt26BQCwsLCot/2+m3hUd8uIEKKcKBkgHE9PT7i4uODGjRs4fPgwN/lMSEiI2C82HR0dAMCjR49EyvLy8rBz506pY3BwcAAAXLlyRWz5pk2bxC7v06cPmjRpgn///Rfnz59Hly5dpN63JMrWgbA6GzZswIsXL6ChocHd6qkPBw8eBFA1G6Ws50UghMgfDTpEhAhuB2zevBl//PEHAMlPEQi+cBcvXsz9GgWqrhj07t27TqPR9e7dGzweD8nJyVi9ejW3vLS0FLNnz5Y4iqCpqSkiIyMBVA2EtG/fPlRUVAitk5ubiw0bNjTogYd+++03LFy4ENnZ2ULLS0pKsHr1akRERAAAxo0bh5YtW4psL5jxUNo5Db7++mvExcWJvKdFRUVYunQp956KG6eCENIAMNIoXb16lQFgV69elWq7nJwcpqamxgAwAMzFxUXiug8fPmQtWrRgAJiGhgZzdnZmbdu2ZWpqaszQ0JCtW7eOAWBWVlYi21pZWTEA7OzZsyJlEydO5PbfokUL1qFDB2ZgYMA0NTXZxo0bubL3VVZWsilTpnDl+vr6zNPTk3l5ebHWrVtzy4cNGybVe1JfoqKiGADm5+cncZ3Vq1dzx2Fubs68vLyYh4cH09HR4ZZ/+eWXrLS0VOz2fn5+dXoPDAwMGACmra3NXF1dWadOnZizszPj8/kMAOPxeGzq1KmssrKy2noEMT548KDW+67rZ5kQUnt0ZYAIMTc3F7q8LOmqAACYmZnhypUrGDJkCIyMjHD37l08f/4cw4YNQ1JSksjgRbW1Zs0arF69Gs7OzigsLMT9+/fh5+eHS5cuoUePHhK34/F4WLVqFS5duoTw8HAYGxvjxo0buH37NvT09BAUFIStW7dixYoVdYpLGXTv3h3ff/89/Pz8oKamhuvXr+PGjRswMTFBcHAw/vzzT+zbt09kYCCB3NxcAJA4FoMkW7duxTfffIM2bdogLy8P//33H7KysvDRRx/hm2++wZUrV7By5UqaopiQBorHmIRu26RBS0xMhKenJ65evSr1iZ8oxo4dO/DVV1/Bz89PLv0Unj59ClNTU5iYmCAzM5Pr81GfBMnCgwcPYG1tXatt6LNMiPxRB0JClExSUhI3cNKkSZMQHBwsk3r/+ecfrs76TAQiIyNx/vz5etsfIUR6lAwQomRevnzJPco4cOBAmdV74cIF6Orq4ttvv5VZnbVx8+bNOj2aSQipP3SboJGiS6uksaDPMiHyRx0ICSGEEBVHyQAhhBCi4igZIIQQQlQcJQOEEEKIiqNkgBBCCFFxlAwQQgghKo6SAUIIIUTFUTJACCGEqDhKBgghhBAVR8MRN3JpaWmKDoGQD0KfYULkj5KBRsrExAQ6OjoICwtTdCiEfDAdHR2YmJgoOgxCGi2am6ARy8rKwrNnzxS2/zt37iAkJARbtmzhxpQvKSnByJEjUVBQgF27dqFZs2YKi68+PX78GM+fP1d0GA2WoaEhWrVqpegwCKp+aFhaWio6DCJjdGWgEbO0tFRoo01ISIC6ujpCQ0Oho6MDxhi++uorPHjwABcuXICnp6fCYqtPWVlZ+OSTT1BcXKzoUAj5YDo6OkhLS6OEoJGhZIDIzZUrV9C+fXvo6OgAANatW4edO3ciOjpaZRIBAHj27BmKi4uxa9cuODk5KTocQuosLS0NYWFhePbsGSUDjQwlA0RuLl++DH9/fwDA2bNnMXXqVEydOlVl+zE4OTnRFLyEEKVEjxYSuXj+/DnS0tLQqVMnZGRkIDg4GJ9++imWLVvGrUPdVQghRDlQMkDk4t9//wUAtG/fHv369YO+vj727dsHDQ0NPH36FHPnzoWpqSlWrFih4EgJIYTQbQIiF1euXIGRkRGWLl2KO3fu4NKlSygsLMSsWbMQFRUFNTU1jBw5EsOGDVN0qIQQovIoGSBycfnyZZiammLfvn1YtGgR5s+fjz/++APNmzfHjBkzMHbsWBgbGys6TEIIIaBkgMgBYwwXLlzAy5cvYWFhgRkzZsDBwQEbN27E0KFD0aRJE0WHSAgh5B2UDBCZS01NxcuXLwEA5ubm+Pnnn9G3b1+oq6srODJCCCHiUAdCInMWFhb45JNP8Ndff+HSpUvo168fJQJKJi4uDjweT+g1efJkmdX//PlzkfoFj5kSQpQPJQNE5po1a4bz58+jZ8+eig6F1EBLSws+Pj7w8fGBra2txPViY2PRrVs3GBsbQ0dHBy4uLpg3bx7evHkjdn0NDQ2uXgcHB3mFr7Ti4uKwbNkyBAcHw8bGhkuIDh48+EH1WltbiyRZ779u3bolo6MgqoRuExCiwlq2bIkLFy5Uu8706dOxdOlSAFVfRpaWlrhx4wbmzp2LQ4cO4fz58zA0NBTaRldXl6t3x44d+Oqrr+QSv7IKCgrCixcv5Fa/q6srDAwMxJYJRvwkRBoyTwYUPTkOIbJCE7JUXRFYunQpNDU1sWfPHgwYMAAAkJ2djT59+iA1NRWjR4/G77//ruBIlYuzszMcHBzQoUMHdOjQAYMGDUJOTo7M6l+3bh3ddiEyJdNkICsrC05OTjQhC2kUaEIWYO7cuQCAadOmcYkAUNUv5Pfff4eLiwv279+PWbNmwdXVVUFR/p/KykqcO3cOnp6e0NfXV1gcFy9eFPqb+swQZSfTZIAmZCGNRUOZkGXatGlYuXIl7O3tkZSUBF1dXaHy9PR0uLu7o6ioCDt37kR4eHit67537x6SkpIAAGPGjBEpd3R0hJ+fH86ePYv9+/crNBm4efMmoqOjsXv3bmRnZ+PBgwcKTQYIaWjk0meAJmQhpH4sWbIE586dw3///Ydx48bht99+48rKysoQEhKCoqIihIaGSpUIAMClS5cAADY2NrCwsBC7jiAZEKxbn548eYK9e/ciOjoaiYmJ3HIPDw/o6enVezz1adOmTVi5ciWKi4thamqKjz/+GGFhYTAyMlJ0aKSBog6EhDRgfD4fe/fuhbu7O6Kjo9G9e3duVsjIyEgkJCTA1tYWGzZskLruO3fuAADs7e0lrmNnZwcAuH37dh2il15xcTGOHDmC6OhonDhxAhUVFQAAW1tbhIaGYsiQIXB0dBS7bXBwMB4/fiz1Pnv16oXIyMgPilvW9u3bJ/L3zJkzsXXrVgQHBysoKtKQUTJASANnb2+PX3/9FcOGDcO4cePQuXNn3Lt3DytXruSShbpcMi8oKABQ9aioJIKywsLCugVfC5WVlYiLi0N0dDQOHTqEoqIiAFUdPAcNGoQhQ4bg448/rrGehIQEZGZmSr3/6pKh+ta5c2dMnz4dXbp0gaWlJSoqKpCQkIAFCxbg3LlzCAkJgYGBAbp3767oUEkDQ8kAIY1AeHg4Tpw4gd27d3M91xljWLBgAby9vetU59u3bwEAmpqaEtfR1tYGAInjDXyIhw8fYt26ddi9ezfXE19HRwchISEYMmQIevToAQ2N2p/CMjIyZB5jfdu7d6/Isq5du8Lf3x99+/bF8ePHMXHiRBprgEiNBh0ipJHYsGEDbG1tkZiYiLy8PHz22Wf4/vvv61yf4Iu+tLRU4jqChEEe802cPHkSy5YtQ05ODszNzbFz5048efIEe/bsQe/evaVKBBo7dXV1bjrw27dv4/r16wqOiDQ0lAwQ0kjo6emhY8eO3N+jRo0Cj8erc32Czmj5+fkS1xHcSpBHx7VWrVpxT0c8fPgQM2fOxIIFC5CamirzfTUGTk5O3L+DoL8HIbVFqTUhjcS+ffuwd+9eqKmpobKyEpMnT0ZAQACaN29ep/ratGkDoOoRQ0nS09OF1pWlHj164MmTJzh8+DCio6Nx8uRJLF++HMuXL4eLiwuGDBmC0NBQWFlZ1aq+xtSBUBLBLZ2ysjIFR0IaGkoGCGkEMjIyuLEAtm7dil27duHMmTP46quvcOzYsTrV2blzZ67u7OxssY8Xnjt3TmhdWdPR0UFoaChCQ0ORm5vLPUqYlJSEyMhIzJgxAz4+PhgyZAiCg4NhbGwssa7G0IGwOnl5ecjLywMAiY+CEiIJ3SYQIyMjg5v0Q5YEdTaGjkxEeZSXlyMkJAQvXrxASEgIvvrqK0RHR8PExAR//vknfv755zrVa29vj/bt2wMANm7cKFJ+69YtLhmoj8fZWrZsiSlTpiAxMRHXr1/HDz/8AHNzc1y4cAFjx45Fq1atEBgYiH379qG8vFxk+4yMDDDGpH7t2LFD7scmC8uWLQNjDEZGRvDy8lJ0OKSBoWSAiCguLsa8efPg4uICHR0dGBsbo1u3bjh69OgH1VteXo41a9bAw8MDurq6MDQ0hI+PD3bu3CmjyFXTnDlzcPnyZdjY2HBf2mZmZoiKigIA/PDDD0hJSalT3fPmzQMArFixAocOHeKWZ2dn48svv0RlZSUGDhyItm3bfuBRSMfFxQVLly5FZmYmTp8+jeHDh0NbWxvHjh1DSEiITOcBqE+XL1+GtbU1rK2tRY5hxYoVWLt2LZ4+fSq0vKioCJGRkVi1ahUAYNasWeDz+fUWM2kkmAxdvXqVAWBXr16VZbX1Licnh7Vp04a1adNGpvUK6szJyZFpvbKUn5/PXFxcGADG5/OZm5sbs7a2ZgAYADZz5sw61fv27Vvm7+/PADA1NTXWtm1b1qZNG67eoUOHssrKShkfTd3J8rMsz3Zx9uxZpqamxjQ0NNjFixdFysePH88AMEdHR/b69Wuh7QAwKyurGvfx3Xffcf9ONjY2zM3NjfH5fAaAubi4sPz8/Gq3j4qKYgCYn5+ftIcnleLiYrZnzx72+eefs6ysLLnuqybjx49nxsbG3EtNTY0BYHp6ekLL349T8O8CgD148ECobNKkSQwA4/F4zNramnl7e7P27dszTU1NbpspU6bI9bgayzmeiKJkgAjp168fA8Dat28vdKI6cOAA9wVw/PhxqeudMmUKA8AsLS3ZtWvXuOXnzp1jBgYGDABbv369TI5BFhpCMpCfn8/Mzc0ZALZgwQKx67x9+5a1a9eOAWCjRo3ilkuTDDDGWExMDAsICGCGhoZMW1ubOTo6stmzZwslGJLUVzKgTIYNG8Z9QVf3ev8Lv7pk4NKlS2zSpEmsc+fOzNzcnGlra7MmTZowOzs7Fh4ezuLj4+V+XHSOb7woGSCc5ORk7pd7WlqaSPn06dMZANahQwep6s3Ly+N+vZw4cUKkfNOmTQwAa9myJSsvL69z/LLUEJKBDyFtMvAhVDEZaKyU8bNMZKPR9xnIzc3F2LFjYWFhAW1tbVhbW2Py5MkoLCzE3LlzwePxMHz4cKFtqutA6O/vDx6Phx07duDZs2cYP348LC0toaWlBUtLS0yaNAnPnz8XG4uydyA8cOAAAODTTz8VO767oLf6f//9h/v379e63iNHjqC0tBR2dnbo1q2bSHl4eDh0dHSQm5vLdUgj9SM3Nxe+vr7w9fXF2rVrZVbvq1evuHoXL14ss3oJIfLRqJOBe/fuwdPTExs3bsTjx4/h6OgIPT09rFu3Dt7e3hK/tGsjJycHHh4e2Lx5M4yNjWFubo6cnBysXbsW3bp1a5DP+QpmnuvSpYvYcktLS1hbWwutK4t6tbW1uSFzFTH7nSorKSlBfHw84uPjpUrwalJeXs7Ve/fuXZnVSwiRj0Y9zkBYWBgePXqEDh064NChQ9y89Hfv3kVQUBDWr19f57rnz5+PHj164OrVq9ygLmfPnkWfPn3w33//ITo6GiNGjJDJcdTXYCm1naUuIyNDqlnqaltvXFxcvc1+p+r8/f3BGJNb/YaGhnKtnxAiW402GYiLi8OVK1egpaUllAgAgIODAw4ePAhXV9c6129kZITdu3cLzQb36aefYuTIkVi3bh2OHj0qs2SgvgZLkdcsdcoy+x0hhBDxGu1tgv/9738AgG7dugklAgJOTk7w9fWtc/0hISFip4UVjMQmGKZVFuprsBR5zVKn6NnvCCGEVK/RJgOCy82CEdTEcXNzq3P9Dg4OYpe3aNECQFUHqoZGXrPUKXr2O0IIIdVrtMlAUVERgKqZ3CQR98u+tpo2bSp2uZpa1VtaWVlZ57oVRV6z1Cl69jtCCCHVa7R9BgRTn1b3C/3ly5f1Fc4Hqa8OhG3atMHDhw9lPktdmzZtEB8fr7DZ7wghhFSv0SYDgi+V5ORkietUV6ZM6qsDYefOnXHmzBmcP39ebHlWVhY3RkKnTp2kqnf79u34559/xJa/ffsW//77L7cukZ+MjAzY2NgAgEx7+wvG5Hjw4AH3+CkhpOFotLcJevbsCQA4efIksrKyRMpv3bqFCxcu1HdYdVJfHQgHDhwIoOpJjFu3bomUCybB8fT0hJ2dXa3r7du3LzQ1NZGeno6TJ0+KlP/2228oLi5GixYtJI5FQIgsyHoSrpKSEsTGxmLcuHFwd3eHvr4+NDU1YW5ujv79++PEiRMStxUMYFabl7gfAxUVFdiyZQv8/f1hbGwMPp8PIyMj+Pj44Oeff662jw4hImQ5nKEyDVVZWVnJOnbsyAAwLy8vlpmZyZXdvXuXOTs7c2PtDxs2TGjbBw8ecOODv8/Pz48BYFFRUWL3W90wr5Aw5rgyCQoKqnFugqNHj4psd+nSJWZlZcWsrKxYdna2SPnkyZNrnJvgl19+kc9B1UFjHY6YJuGS7SRcM2fO5LbX0NBgjo6OzN3dnenp6XHLx44dK3YSrvHjxzMfHx+JL2NjY67NVFRUCG1bXFzMTfwFgLVo0YJ5enqy1q1bc8s8PDxYYWFhXd8usZTps0xkq9EmA4xVfem3atWKAWDq6uqsffv2rG3btkxNTY3Z2dmxiRMnMgBsxIgRQtupcjLw7Nkz5uTkJPGEOX36dLHbVTfBCmOMvXnzhnXp0kXirIWhoaE0ayGRK3lMwjVjxgz28ccfs927d7OioiJueWlpKVu6dCn3+d64caNU9VZUVHCTUM2aNUukfMGCBQwA09LSYocOHRIqi4uLY0ZGRgwAGz9+vFT7rQl9lhuvRnubAKi6Z56YmIjRo0ejRYsWSEtLw4sXLzBu3Dj8+++/3JzfH/JUQWNjbGyMhIQEzJ49G3Z2drh16xaeP3+OgIAAHDlypM7jzGtra+P06dNYuXIl2rVrh/T0dDx69IjrT7B7926xc0EQIgspKSmIiYmBmpoa9u3bBwsLC65s4MCBmDZtGgBg9uzZUtU7efJkxMfHIzQ0lOu0DAB8Ph8//PADRo0aBeD/brHV1t9//42HDx+KnTsFqJrvAwDGjh2L/v37C5X5+flxx1HX2x9EBckys2hoWWOvXr0YALZ27VpFh0KUTEO4MvD48WM2ZswY1rp1a6alpcWsrKzYpEmTWEFBAZszZ84H3QJ7+vQp+/bbb5mFhQXT1NRkFhYWbOLEiRIvO0PJr3rNmDGDAWBdu3YVW56ZmckdQ3p6usz2+8cffzAATFtbW6rtgoODGQDm7+8vttzV1ZUBYOvWrRNbHhsbywAwc3NzqWOuTkM7x5Paa9RXBqqTmZmJU6dOAZA8gQ4hyoom4ZKOvCbhqolgMC0dHZ1ab1NQUIDY2FgAkDikubu7OwBIfEJH8ESQNE/9ENXWaB8tBKqeXY+NjUV4eDiMjY255SkpKQgLC0NpaSm6dOlS7SiFhCgjmoRLOSbhqsnevXsBVF26r609e/agpKQE+vr6GDBggNh1Zs6cidjYWOzfvx+tW7fG2LFj0bp1azx58gS7du3C6tWrYWpqStNHk9qT5WUGZbuElJSUxAAwHo/HzM3NmZeXl1BnOCsrK5leEiSNhzLfJhB01tTS0hJ6Skbg5s2bTE1Nrc63CUxNTdmLFy9EyidMmMAAsKCgIJEy1PE2gZWVFbetNK/3j6smOjo6DAD766+/JK4juDQvq053MTEx3Pnn8uXLtd7Ow8ODAWBff/11tevdunWL9e/fn2loaAi9Nzwej33zzTdCnSRlRdnO8UR2GvVtAmtra8yaNQsdO3ZERUUFUlJS8OzZM7i5uWHWrFlITEyEra2tosMkRCo0CZfyTMIlyY0bNzBs2DAAQEREBDp27Fir7VJTU5GYmAgA+Oqrr6pdNycnB8+ePUNFRQWMjY3h4eGBFi1agDGGgwcPIjo6mqaRJrXWqG8TGBoaYv78+Zg/f76iQyFEZmo7CZekkSRr0lgn4SouLq6XybLS09PRvXt3vHz5EkFBQViyZEmtt42KigIAODo6Vjsa58GDBzF48GDo6enhwIEDQrcT4uLiEB4ejhkzZuD58+dYvnx53Q+GqIxGfWWAkMaIJuGSXn1NlpWRkYGAgAA8evQIn3/+OX7//XdoaNTuN1dZWRl2794NoPqrAmVlZZg4cSIqKiqwatUqkX4F/v7+2LlzJwBg9erVePjwYR2PhqiSRn1lgJDGiCbhUp5JuN6VnZ2NgIAAZGVlISAgAH/88Ue1tyXed/ToUTx9+hQaGhoIDw+XuN69e/e496x3795i1/Hz80PTpk3x+vVrJCUlwdzcXLqDISqHkgElRRO/EEloEi7lmYRL4OHDh/j000/x4MED+Pr6IjY2luuDUFuCWwSff/45WrZsKXG92iZ6gv4CgtsfhFSHbhMQpcYYw6VLlxAZGYlPPvkEJiYm4PP5MDY2hr+/PzZv3ozy8nJFh1mvaBIu5ZmECwByc3MREBCA9PR0dOzYEcePH5d4q6W6OgQdQ2vqOGhvb8/dsjl+/LjYdc6fP4/i4mIAVf0PCKkJJQNEqZ05cwYff/wxlixZggsXLkBPTw9ubm7g8/k4d+4cRo8ejU8++eSDBtlpaPz9/dGxY0eUlJRg4MCBQgnBvXv3MGDAAKirqyswQuXj5uaGoKAgVFZWYvDgwcjOzubKDh48iBUrVgAA5s6dK7Lt5cuXYW1tDWtra+Tk5AiVPX36FF27dsWdO3fQoUMH/P3339X25ZAkOjoa5eXlaN68Ofr06VPtusbGxtw6U6dO5YYmFjh//jw3hLGnpydcXV2ljoeoHrpNQJQaYwxWVlaYOHEiQkJC0KpVK67s6NGjGDZsGC5fvoyxY8dyA7w0djweD7t27UKXLl2QkJAAW1tbuLq6orKyEjdu3ICNjQ3Gjh2LtWvXUlLwjq1bt+L27dtISUmBnZ0dXFxc8Pz5c+72wPTp08V+Eb99+5a7lfH+VaiZM2fi5s2b3HqS7uEDVUmHpMv/glsEYWFh3Jwp1dm4cSNu376N27dvIygoCCYmJrC0tMTjx4+5/gRmZmZch0RCakLJAFFq3t7euHv3rtgTZGBgINatW4ewsDAcOHAAv/zyi9BIk42ZYBKuuXPn4ujRo0hLS0PLli0xbtw4zJs3jxt5jibh+j+CSbiWL1+O/fv349atW9DW1kZAQAAmTZqEvn37Sl1nSUkJ9//Xr1+vdl1J9+6vXLmCtLQ0AJKHH35fq1atkJiYiA0bNuCPP/7AzZs3kZKSgqZNm6JDhw7o06cPJkyYgGbNmtXySIjKk+UIRrIcnerFixdszpw5rF27dqxp06ZMU1OTtWrVinl7e7Mff/yRPXz4UGj97OxstmbNGtajRw9ma2vLtLW1mZ6eHvPw8GALFy4Uml70Xe9O6FJSUsLmz5/PHB0dmba2NjM3N2fffvstKygo4Nbfv38/8/HxYfr6+kxPT499/vnnLCUlRWzd7078kp2dzUaMGMHMzMyYlpYWs7OzYzNmzGCvXr0Suy1qGNHt33//ZUOGDOEmkjEyMmJdu3ZlBw4ckPienjlzhn3xxResZcuWTENDgxkYGDA7Ozs2cOBA9vvvv0vcTpkVFBRw75U0o7zVRJlHIKwNmoSLyAONQNh4KWUyUFRUxJydnbmhNR0cHJiXlxeztLTk5h0/evSo0DYREREMAGvSpAmzsbFhXl5ezNbWlqmrqzMAzNXVVeyMa4JkIDQ0lPn5+TEej8ecnZ2Zo6MjN6Srp6cnKy0tZZGRkQwAs7CwYG5ubkxbW5sBYPr6+uzu3bsidQuSgdmzZzMTExOmoaHB3NzcmKOjI/cF5u7uLjau6pKBRYsWMR6Px+3bzc2NtWzZkttmzJgxItts3bqVKzcyMmLu7u7M1dWVm/fc09Oz1v8+yuTRo0fccaWmpsqs3oacDGRkZDBNTU0GgCUnJ9fLPolqoGSg8VLKZGDNmjUMAGvbti3LyMgQKnv9+jXbtWsXu3btmtDyU6dOsbi4OFZeXi60PCsri/Xt25cBYOPGjRPZlyAZ4PP5zNHRkaWlpXFlCQkJzMDAgAFggwYNYrq6uiw2NpYrz8vLY+7u7gwACwsLE6lbkAzw+Xz28ccfs5ycHK4sMTGRtW7dWuI465KSgd27d3Nf6Lt372aVlZVc2YkTJ5ipqSl3NUKgvLycGRsbc78Uy8rKhOq8du0a27x5s0gM1dm2bRvz8fGp00uWVq5cyQAwExMTkX/7D6HsycC9e/fYqlWr2LNnz4SWJycnc9PbdunSRWb7I4QxSgYaM6VMBkaPHs0AsDVr1sgkrtevXzM+n8/09PREvjAEyQAAduXKFZFtp02bxpUvX75cpPzYsWPcl/P73k0GxE0aIthWTU1NZMIZcclAWVkZs7S0ZACEkpJ3HTp0iAFgjo6O3LLHjx8zAMzQ0FDsNnXx7vsm7UtWMjIyuGRN3L/Nh1D2ZIAm4SKKQMlA46WUjxYKJl+JjY3lhl6tjZcvX2Lbtm0YMWIEevTogU8++QS+vr7o3r071NTUUFRUhLt374rd1s3NDd7e3iLLO3TowP3/N998I7G8sLCQG870ff3794eFhYXI8l69esHOzg6VlZX4+++/azy+y5cvIysrC+bm5ggMDBS7TmBgIPh8Pm7duoVHjx4BAJo3bw5tbW28ePECx44dq3E/tTF37tw6PR/OZDRxyqtXr/DFF1/gxYsX6NSpE6ZMmSKTehsKmoSLECJLSvk0wYgRI7Bq1SqcOXMGZmZm6NatG3x8fODr6wsvLy9uwI13xcfHY8CAAXjy5Em1dUsam1zSiGampqYAABMTExgYGEgsB6q+oMT13nV2dhZbN4/Hg5OTE9LT07nexNVJSUkBALx+/braWekEoxfm5OTAzMwM6urqmDp1KhYvXozAwEC4urqiW7du6NixIwICArg56xuKN2/eIDAwECkpKXBwcEBMTEytx39vLGgSLkKILCnlGbRly5a4fPky5s2bh8OHDyMmJgYxMTEAAHNzc/z444/49ttvuS+9oqIi9O/fH3l5eejatSt+/PFHtGvXDkZGRtwjaZaWlsjOzkZZWZnYfUoaMUywj5rKAckTuAhme6uurDZXQAQD6zx//hzx8fE1ri8YgQwAFi5cCEtLS/z666+4du0a9xiUuro6evfujZUrV0o9xKsilJSUICgoCHFxcbC2tsbp06erHbqVEEJIzZQyGQCqfqkLRuVKSkrC+fPnERsbi/Pnz2PChAkoLS3F1KlTAVQNyZmXlwcLCwscO3ZMZExwxpjES/j1obqrFYKy2oxaJpigpmvXrjh16pRUMfB4PIwePRqjR4/GkydPcOHCBZw+fRr79+9HbGwskpKSkJqaCkNDw1rVt337dmzfvl2qGATqOlRuaWkp+vfvjxMnTsDc3BynT58We/uFEEKIdJQ2GRDQ0NCAl5cXvLy8EBERgblz52LevHlYv349lwzcv38fAODl5SV2cpDU1FS8fv26XuN+l2CEMnEE46Q7OTnVWE/btm0BADdu3ABjTOiqhDRatGiBAQMGYMCAAZg3bx5cXFyQnZ2NY8eOISwsrFZ1ZGVl1erqhKyUlZUhODgYx48fR8uWLXHmzBm6J04IITKilB0IqyO4Vy7oHAcAOjo6Isve9dNPP8k/sGrExMSInVP8+PHjuHfvHtTU1NC9e/ca6/H19UWrVq2Qm5uLXbt2ySS25s2b46OPPgIg+f0Tpz47EJaXl+PLL79EbGwsmjdvjtOnT3MxE9XD4/HA4/G4YYQJIR9OKZOB6dOnY+PGjXj69KnQ8vz8fKxcuRJA1VUAgS5dugCo6m2/YcMGbnlJSQkiIyOxZ88eqeYVlzXGGAYPHiw0b3tycjLGjBkDABgyZAisrKxqrEdTUxPLly8HAIwePRrr168XGg4VqHqqITo6Gt999x237ObNmxg1ahTi4+NF+jUcPnwYCQkJAITfU2VRWVmJoUOHIiYmBs2aNcOpU6ckdsgkRJUdPHgQffr0QatWraCpqYmWLVvC19cXs2bNUrmZPUkdyPI5RVk9g/rFF19wz0tbW1szb29v5uzszI2q1qxZM5GR1YYOHcptY2Zmxjp06MA9g75w4UJmZWXFALCzZ88KbffucMTinD17lntuWxLBft8fIEgwzsCsWbOYsbEx4/P5zN3dnTk5OXHbtGvXTmi445rqZIyxFStWcCMr6ujoMDc3N+bt7c1sbGy4kQn9/Py49QXPpANgTZs2Ze3bt2deXl5CoxaOHDlS4vEp0p49e7gYLSwsqh3M6Pjx4zLbr7KPM6DKqmsbquj169fs888/594XS0tL5uXlxWxsbLgRWyUNxy4t+iw3XkrZZ2DWrFlwdXVFXFwcMjIykJycDA0NDdjb26Nnz56YNm2a0Ox1QNWsX66urti+fTvu37+PN2/ewMPDAxMnTkRQUBC2bNmioKMBbG1tcfXqVcyZMwcnTpxAfn4+bG1t8eWXXyIyMpLrGFhbERER6NmzJ3755RecOXMGd+7cAY/Hg5mZGXr06IFevXqhX79+3PofffQRtm3bhtOnTyMxMRGZmZncY5Cff/45RowYwc33rmzevfKRnZ0tNPXs+2p6rJSQxmjgwIH466+/8PHHH2Pjxo1c3yKg6omikydPQktLS4ERkgZBlpkFZY3C3p2oiDQsdGVAeYGuDHB27NjBALA2bdqw169fy31/9FluvJSyzwAhRLKXL19i7ty5aN++PXR1daGlpQUzMzN07NgR06dPF+kImpOTg59//hk9e/aEnZ0dmjRpAn19fXh6emLRokV49eqV2P3MnTsXPB4Pw4cPR2lpKRYsWAAnJyc0adIErVu3xvjx41FYWMitf+DAAfj6+sLAwAD6+vro1asXUlNTxdbt7+8PHo+HHTt2ICcnByNHjoS5uTm0tbVhb2+PmTNn1vkJoISEBISFhcHS0hJaWlpo1qwZPvvsMxw8eFDiNmfPnkVQUBBatWoFPp8PQ0ND2NvbIzg4GPv3769THPVh1apVAIA5c+ZwHakJqRNZZhaUNQqjKwMNl7JeGaAZPWlGT4H09HQGgKmrq7MXL16wc+fOsa+//pp17dqVffHFF2zBggXs0aNHMt0nneMbL0oG5IiSgYZLWZMBmtGTZvQU2LdvHwPAbG1t2XfffSd2UjAdHR126NAhqeqtDp3jGy9KBuSIkoGGS1mTAZrRk2b0FFi9ejX3HgJgPXr0YKmpqaykpITdunWL9e/fnwFgWlpaLDU1VSbHRuf4xov6DMhRXFwcGGMYPny4okMhjQTN6CmeKs7oKejrUVZWBhsbG8TGxqJt27bQ1NREmzZtcODAAbRt2xYlJSVYuHChTI6NNF5K+WghIUQ8mtFTPFWc0fPdodcnTJggMrCampoaIiIiMHz4cPz999+orKwU+/kgBKBkgJAGhWb0FE8VZ/R8N7mSNLeJYPmLFy9QUFAAExOTeomNNDyUDBDSwNCMnqJUcUZPR0dH7v/FTdAGAE2aNOH+v6Kiok4xEdWgUteMrK2twePxEBcXp+hQlMaOHTu4iV8ErzVr1ig6rA8WFBQkclyNbWIbwYyeEREROHfuHObMmQMAWL9+PbeOqs7oWVeCGT3Xr1+PtLQ0NG/enJvRs7YEM3rW5SUNd3d3bmyB9PR0sesIlmtra8PY2Fiq+olqUalkgEimr68PHx8f+Pj4wNzcXOJ6O3bsgI+PDwwMDKCrqwsPDw+sWbNGLr86Hj58iF9++QX9+/eHjY0NtLS00LRpU7i4uGDq1KnVDk3s7OzMHY+qoBk9VWtGzyZNmqBv374AIPFKhGAYdj8/P2ho0IVgUg1ZPpqg7I+dSJqsSJVFRUWJTGwkTmVlJQsNDeUegWrTpg1r27YtN8BL165dWUlJiUxjEzxrjv//6Je7uztzcHDgBsvR19dnJ06cqLEeQR3SDF+rrI8W/vjjj2zDhg0sLy9PaPmzZ89Yz549GQDWpUsXbnliYiJ3/OvXr+eWv337lk2fPp3xeDxuAjBFTOLF5/OZr6+v0OA4SUlJzMLCggFgQ4cOrXWd0dHR3OBKv/76K3v79q1QeUFBAfvtt9/YtGnTuGU3btxgI0eOZBcuXGAVFRVC68fExHDvzZkzZyQeoyLdunWLe7Rw1qxZ3DgJFRUVbMWKFdzgVHFxcTLZn7Kf40ndUTKg4mqbDPz888/cl/K5c+e45deuXeNO3N9//71MY7O1tWVff/01u3LlitCJOjMzkwUEBDAAzMDAgD158qTaehpTMkAzetKMnu/bvXs309DQYACYiYkJ8/b2Zi1atOASgVWrVslsX8p+jid1R7cJSI3Ky8u555SXL1+OLl26cGWurq7YunUrAGDt2rUSH0+riytXrmDz5s3w9vYWeiTK0tIShw4dgomJCV68eIG9e/fKbJ/KbtasWZgxYwZ8fHxQVlaG5ORkZGRkwN7eHlOnTsX169fRvn17oW2ioqKwbNkytGnTBk+fPkV6ejo8PDwQExODGTNmKOhIqghm9AwNDUVubi7S09Nha2uL6dOnIz4+HkZGRlLVFxERgZSUFIwZMwatW7fGnTt3cOPGDWhoaKBHjx5Yu3at0G0EwYyeoaGhsLCwQGZmJpKSklBZWYnPP/8cBw4c4D7fyio0NBQJCQkICQkBn89HUlISGGPo168f/vnnH0yZMkXRIZKGQJaZRW2zxnfH1K5u7Oxdu3YxAMzOzk5o+fnz59m0adO4DJ7P5zNTU1PWq1cvkXHZ3yXpF9CwYcMYADZnzhyx2z148KDGEcJOnDjB+vXrx8VjYmLC+vTpw06fPi1xG2VQmysDJ0+e5H45vXnzRuw6tra2DADbtm2bnCIVJRhKV9x48+9CNb8kJVHWKwONBY3O2TDRZ7nxUsiVAVtbW/j4+KCioqLaX3XR0dEAgKFDhwot79evH1asWIH09HQYGxujbdu2YIzh+PHjCAwMxKxZs+Qa/7sYYxg7diy6d++OmJgYlJWVwdXVFYwxHDt2DF27dsXSpUvrLR55uHTpEgDA29tb4iNMgqsFgnXrw9u3bwGAZmsjhJAPpLDupeHh4YiPj0d0dDT3TPS7cnNzueeF308Gli5dioCAANja2gotP3XqFIYMGYKFCxciMDBQ7BCqsrZkyRJs3LgRlpaW2LRpE3r27MmV7dmzB9988w0iIyPRsWNHfPrpp7Wud8KECUhKSpI6Hnd3d6xbt07q7apz584dAJJHogMAOzs7AMDt27dlum9JHj9+zD0i6ufnVy/7JISQxkphycCgQYMwadIkJCcn4/r163B1dRUq37NnDyoqKuDr6yvypT9q1CixdX722WdYtGgRvv76a+zYsUPuyUBhYSEWL14MDQ0NxMTEwMPDQ6hccB80IiICy5cvlyoZuHbtmtTPHQOQy+NDgkFpxA0nKyAoe3d+e3kaN24cSktL4erqij59+tTLPgkhpLFSWDJgaGiIwMBAHDhwANHR0Vi2bJlQueAWQXh4uNjt09LScODAAaSmpiI/P58bSvXFixcAgMTERDlGX+XPP//E69ev4ePjI5IICAwYMIAbFKaiogLq6uq1qluZBkYSXI5/f+zzdwluH7x580bu8SxYsACHDx+GpqYmfvvtNxpvnRBCPpBCR6EIDw/HgQMHsGfPHixZsoQ7qV+/fh3JycnQ1tbGoEGDRLabOXMmlixZInG8c0DypCuyJJgc5d69exInR2H/fyCRN2/eID8/X2jyloZC8EVfWloqcR1BwvDu8KfysHHjRsyePRvq6uqIjo6Gu7u7XPdH5EOZkl1CiIKTgZ49e8LU1BQ5OTmIi4tDQEAAAOC3334DAPTt21dkNrT9+/dj0aJFUFNTw5w5c7jR6Zo2bQo1NTWcOXMGXbt2lTjpiiwJJkd58uRJjTPCAcKTozQkgse7qkuwBLcSpH0UTBrbt2/HuHHjwOPxsH37drGJIiGEEOkpNBnQ0NDA4MGDsXbtWkRHRyMgIACVlZXYs2cPANGOg0DVM9NA1fPEc+fOFSmvyxUBwexqgl/x75M0drtgcpSRI0fK/FlkZepA2KZNGwBVV0AkEYyBLlhX1n777Td8/fXXYIxh06ZNEm8fEUIIkZ7CB6sODw/H2rVrcejQIaxfvx7x8fF4+PAhTE1NhXrmCwgmXvnkk0/E1nfx4kWpYxBMwSrp172gN/37BJOjXLt2Tep91kSZOhB27twZAJCQkIC3b9+Kfbzw/PnzQuvK0p49e/DVV1+hsrISP//8M7755huZ70PVWVtbIzMzE2fPnoW/v7+iw1EKO3bswFdffSW0bPXq1Zg8ebJiApKRoKAgHDlyRGjZgwcPYG1trZiAiFJQeM8rT09PuLi4oKioCIcPH+Y6DoaEhIj9Yqtu4pW8vDzs3LlT6hgcHBwAVI14J86mTZvELu/Tpw+aNGmCf//9l/sylJW4uLg6TXQij3uxfn5+aN68OV6/fs39+7zrxIkTuH//PrS0tLiJU2Rl//79CA8PR2VlJX766SdMnDhRpvUTUpOaJvHy9/cXmSHz/df//vc/mcb05s0bHDlyBDNnzkTPnj1hYmLC7evZs2fVbquKk3iRmik8GQD+73bA5s2b8ccffwCQ/BSBYHCbxYsXc1OcAlVXDHr37l2n3uy9e/cGj8dDcnIyVq9ezS0vLS3F7NmzcebMGbHbmZqaIjIyEkDVQEj79u0Tmb0vNzcXGzZsaNADD/H5fG7Y2u+//14o8bl+/Tr3qOeECRNgYmIisr3gZDl8+HCp9nvkyBEMGTIEFRUVWLhwIaZNm1b3gyCkjtzd3XHhwgVcuHABwcHBEtdzcHDgvmTff8m6L83t27cRFBSERYsW4e+//5bq9ujixYu54yFEQOG3CQAgLCwMkZGR3K9aFxcXiY/q/fDDD/j999+RlZWFtm3b4qOPPoK6ujpu3LgBfX19/PTTT5gwYYJU+7e3t8eECROwdu1aTJ06FcuWLYOFhQXu3r2LN2/eYN26dRgzZozYbWfMmIGCggKsXr0aISEhGD16NBwcHKCmpobHjx8jJycHADBs2DCpYlI2EydOxKVLl/D777/Dz88Pbdq0gaamJm7cuIHKykr4+/tz8xfIyqBBg1BeXg4tLS389ddf+Ouvv8SuJ49+EoRIKzIyUuqEt674fD68vb3h5eWFDh06oEWLFujVq1e97Js0TkqRDJibmyMgIIAbcbC6zmFmZma4cuUKZsyYgRMnTuDu3bto2bIlhg0bhtmzZyMjI6NOMaxZswY2NjbYsmUL7t27h7KyMvj5+WHOnDnVDrbD4/GwatUqDBo0CBs2bMA///yDGzduQFNTE+bm5ggKCkKfPn3wxRdf1CkuZcHj8bB37150794dW7ZswY0bN1BRUYF27dph6NChmDhxosT+Crm5uQAgMcGTRPAoY0lJSbX9J2iedqJqXFxchG5r1vW8R4iA0pxFT548Wet1rayshGYee5e1tbXEpwKqazA8Hg+TJ0+W2DlIUp0CnTp1QqdOnapdp6Hj8XgYMWIERowYUettnj59itu3b8PExETiyJGS1PSeN0b379+HnZ0d1NXVkZ2djVatWoldb/fu3QgLC4OdnZ3QUx7//PMPYmNjce7cOWRnZyM/Px9GRkbo0KEDxo4dK/VojcOHD8fOnTsxZ84csU/vZGRkwMbGBoDkf6+TJ09iw4YNuHTpEvLz82FgYIBOnTphypQp3OPEhBDFUoo+A0TxkpKS4OvrC19fXxw4cEBm9f7zzz8AgEmTJtXrhEKRkZHc8TQkNIlXw3Xw4EH069cPAQEBCA4OxrJly/Dw4UNFh0VIrSjNlQGiWC9fvuQuxQ8cOFBm9V64cAG6urr49ttvZVZnbdy8ebNOj2YqA5rEq3rKNAbHu/7880+hvw8ePIjZs2fTUzCkYZDlfMg01zVpLGT5WZa2rsLCQqatrc0AsGvXromUr1y5kgFgvr6+UsWxZcsWBoCNHTtWpMzKyooBYGfPnhVaPmzYMAaAzZkzR2ydDx48YADY+6eSgoIC1rRpU6ahoSHxuAXH0bNnT6mOw8/Pj9unNC8/Pz+p9hMVFVWr7caPH89WrFjBUlNT2YsXL9jr16/ZxYsXWb9+/bh9b9myRap9S+vdf4enT5/WejvBNg8ePKjV+nSOb7zoygAhSoYm8aqess1rIO5qQ+fOnfHHH39gzJgx2LRpE3744QeEhIRwA5wRomwoGSBECdEkXo3D0qVLERUVhYKCApw9e5am2yZKi5IBQpQQTeLVOBgaGsLFxQVJSUkShzUnRBlQMkCIEqJJvCRT1g6EkmhqagJAvSRhhNQVJQOEKCmaxEs8ZZrEqybl5eXcsOkWFhb1vn9CaouSAUKUlGASrxs3buDw4cPcZDcNcRIvwZwisqBsHQirs2HDBrx48QIaGho0wBJRajToECFKjCbxUm6//fYbFi5ciOzsbKHlJSUlWL16NSIiIgAA48aNQ8uWLUW2r+skXoTIGiUDhCixsLAwqKmpIS4uDq9evapxEq8WLVpwk3i5uLigXbt2cHBwwL179/DTTz9JvX/BJF4AMHXqVLRs2RJeXl4wNTXFsmXLqr0HP2PGDEyZMgUFBQUICQlBs2bN0KFDB3h7e8PCwgKtWrXCuHHjhBKXhqagoACzZs2CpaUlWrduDW9vb3h6eqJZs2aYOnUqysrK8OWXX2LFihUy37eHhwdMTExgYmIi9Jlo06aN2OWEVIeSAUKUmGASL4HaTOI1ZMgQGBkZ4e7du3j+/DmGDRuGpKQkuLq61imGNWvWYPXq1XB2dkZhYSHu378PPz8/XLp0CT169JC4nWASr0uXLiE8PBzGxsa4ceMGbt++DT09PQQFBWHr1q1y+aKsL927d8f3338PPz8/qKmp4fr167hx4wZMTEwQHByMP//8E/v27QOfzxe7fV0n8QKqEpH8/Hzk5+ejsLBQ7PKCgoK6HRhROXLpM5CWliaPagmpN8r0GaZJvJSXs7OzyKBQtfUhk3gBNFMhkS2ZJgMmJibQ0dFBWFiYLKslRCF0dHRgYmKi6DCIggkm8QKqJtwKDg6WSb2KnMTr/Pnz9bY/0jDINBmwtLREWloanj17JstqCVEIExMTWFpaKjoMomA0iRdRBTxW0zU+QsgHSUxMhKenJ65evUodukiDRp/lxos6EBJCCCEqjpIBQgghRMVRMkAIIYSoOEoGCCGEEBVHyQAhhBCi4igZIIQQQlQcJQOEEEKIiqNkgBBCCFFxlAwQQgghKk4uExURQkQp0+RHhNQFfYYbL0oGCJEzmsCLNCY0gVfjRHMTEFIPsrKyFDqBV2VlJT799FOEh4dj5MiR3PKdO3di7dq1WLZsGT777DOFxUeq3L17F8OHD8cnn3yCJUuWgMfjAQAuXryICRMm4PDhw7CwsFBojDSBV+NEVwYIqQeWlpYKPYGmpaXh1atX6N+/PzfBzP/+9z+sW7cOkZGR+P777xUWG/k/Hh4eUFdXR3BwMD777DPu38XKygoTJkzAq1evaIIgIhfUgZAQFXDlyhXweDx4eXkBAO7du4eQkBB8/vnnmD9/voKjI+8aOHAgIiMj8eOPP+J///sfAMDY2BgODg64cuWKgqMjjRUlA4SogMuXL8PZ2Rn6+vooKipCUFAQmjdvjt27d0NdXV3R4ZH3zJ8/H59//jlCQkJw7949AECnTp1w+fJlBUdGGitKBghRAZcvX0anTp1QWVmJYcOGISsrC0eOHIGhoaGiQyNiqKurY/fu3TA1NUVQUBCKiorQqVMnJCcn4+3bt4oOjzRClAwQ0si9fv0a165dQ8eOHbF48WLExMRg165dcHJyAgD8+++/+PLLLzFjxgwFR6raoqOj0bNnTxw9ehSVlZUwNDTE4cOHkZWVhWHDhsHLywtlZWVISkpSdKikEaJkgJBG7r///kNlZSXevn2L2bNnY968eejTpw+OHTsGPz8/dOzYEYmJifDx8VF0qCrNzc0NRUVF6Nu3L1xdXbF9+3bY2tpi165diImJwfHjx6GtrU23CohcUDJASCN3+fJl6OjoYMaMGejTpw/Mzc3Rtm1bBAYGorS0FH/88Qdu3bqFXr16KTpUlda2bVvEx8fjwoUL+OijjzBq1CjY2Njg5s2b+PHHHzFv3jzY2tpSMkDkgsYZIKSR69OnD06dOgUDAwOoqakhNzcXffv2xXfffQcfHx/uWXaiXG7fvo2VK1fit99+A5/PR4sWLZCZmYnmzZvj0aNHig6PNDKUDBDSiDHGoKOjg7dv34LP5yM8PBwRERFcfwGi/J48eYJ169bh119/xfPnzwEA9+/fh42NjWIDI40KJQOENGKMMdja2qJTp05YtWoVWrVqpeiQSB29evUKS5cuxebNm/Hvv//C2tpa0SGRRoSSAUIIIUTFUQdCQgghRMXR3ASNmKInxyFEVmQ5OQ61C9JYyLJdUDLQSGVlZcHJyQnFxcWKDoWQD6ajo4O0tLQPPvFRuyCNiazaBUDJQKP17NkzFBcXC400R0hDlJaWhrCwMDx79uyDT3rULkhjIct2AVAy0Og5OTnRlKeEvIfaBSHCqAMhIYQQouIoGSCEEEJUHCUDhBBCiIqjZIAQQghRcZQMEEIIISqOkgFCCCFExVEyQAghhKg4SgaIUsvIyACPxwOPx5NpvYI6MzIyZFovIfWB2gWRNUoGCFFixcXFmDdvHlxcXKCjowNjY2N069YNR48erXOd1tbW3Elf0uvWrVsyPApCZEse7QIAysvLsWbNGnh4eEBXVxeGhobw8fHBzp07ZRS58qIRCIlS4/P5aNOmjczrFdTJ5/NlXresFBQUoEuXLrhx4wb4fD5cXFzw/PlznDp1CqdOncLMmTOxYMGCOtfv6uoKAwMDsWU6Ojp1rpfIH7UL2beLkpIS9OzZE3FxcVBTU4OLiwtKS0tx8eJFXLx4EadPn8bOnTtlfjVGaTDSKF29epUBYFevXlV0KKSO+vXrxwCw9u3bs6ysLG75gQMHGJ/PZwDY8ePHpa7XysqKAWBnz56VYbTyI8vPMrWLhk9e7WLKlCkMALO0tGTXrl3jlp87d44ZGBgwAGz9+vUyOQZZkPVnmW4TEKKEUlJSEBMTAzU1Nezbtw8WFhZc2cCBAzFt2jQAwOzZsxUVIiH1Tl7t4unTp/j1118BAFu3boWrqytX1qVLFyxfvhwAMH/+fFRUVHzoYSglSgZIvcnNzcXYsWNhYWEBbW1tWFtbY/LkySgsLMTcuXPB4/EwfPhwoW2q6yjl7+8PHo+HHTt24NmzZxg/fjwsLS2hpaUFS0tLTJo0Cc+fPxcbi7J3lDpw4AAA4NNPP4Wjo6NI+ZgxYwAA//33H+7fv1+vsRHZonZRe/JqF0eOHEFpaSns7OzQrVs3kfLw8HDo6OggNzcX586dq2P0yo2SAVIv7t27B09PT2zcuBGPHz+Go6Mj9PT0sG7dOnh7e0s8OdVGTk4OPDw8sHnzZhgbG8Pc3Bw5OTlYu3YtunXrhrKyMtkdSD25dOkSgKpfJeJYWlrC2tpaaF1pbdq0CYGBgejatStCQkKwbt06FBYW1qkuUjfULqQjr3ZRU73a2trw9vaWut6GhDoQknoRFhaGR48eoUOHDjh06BA3//bdu3cRFBSE9evX17nu+fPno0ePHrh69SqaN28OADh79iz69OmD//77D9HR0RgxYoRMjiM4OBiPHz+WertevXohMjKy1uvfuXMHAGBvby9xHTs7O2RkZOD27dtSxwMA+/btE/l75syZ2Lp1K4KDg+tUJ5EOtQvlaBe1rTcuLq7O7U3ZUTJA5C4uLg5XrlyBlpaW0AkPABwcHHDw4EGhe3TSMjIywu7du6Gvr88t+/TTTzFy5EisW7cOR48eldlJLyEhAZmZmVJvV91JRpyCggIAQLNmzSSuIyiT9td8586dMX36dHTp0gWWlpaoqKhAQkICFixYgHPnziEkJAQGBgbo3r27VPUS6VC7UJ52Ic/21lDQbQIid//73/8AAN26dRM64Qk4OTnB19e3zvWHhIQInfAEOnfuDABIT0+vc93vy8jIAGNM6teOHTuk2s/bt28BAJqamhLX0dbWBgC8efNGqrr37t2L0aNHw8nJCU2bNoW+vj66du2K06dPo1evXqioqMDEiROlqpNIj9qF8rQLeba3hoKSASJ3gstq7du3l7iOm5tbnet3cHAQu7xFixYAgFevXtW5bkURnHhKS0slriM4gTVp0kQm+1RXV8eKFSsAVP2bXb9+XSb1EvGoXUhPXu1CEe1N2VAyQOSuqKgIAKCnpydxHXG/YGqradOmYperqVV9vCsrK+tct6IYGRkBAPLz8yWuI7i0KVhXFpycnLj6BPdRiXxQu5CevNqFotqbMqE+A0TudHV1AVT/S+Tly5f1Fc4Hqa+OUm3atMHDhw9x7949iesILvPKeiQ6waXShtjbvCGhdqE87aJNmzaIj49XSHtTFpQMELkTNJ7k5GSJ61RXpkzqq6NU586dcebMGZw/f15seVZWFvcseKdOnaSOR5K8vDzk5eUBgNCALkT2qF0oT7vo3Lkztm/fjn/++Uds+du3b/Hvv/9y6zZGdJuAyF3Pnj0BACdPnkRWVpZI+a1bt3DhwoX6DqtO6quj1MCBAwFU9TgXN2nQxo0bAQCenp6ws7P74OMSWLZsGRhjMDIygpeXl8zqJaKoXShPu+jbty80NTWRnp6OkydPipT/9ttvKC4uRosWLSSORdDQUTJA5M7f3x8dO3ZESUkJBg4cKHTiu3fvHgYMGAB1dXUFRqh83NzcEBQUhMrKSgwePBjZ2dlc2cGDB7mOfnPnzhXZ9vLly7C2toa1tTVycnKEylasWIG1a9fi6dOnQsuLiooQGRmJVatWAQBmzZql1JPVNAbULqQnr3ZhamqKcePGAQBGjRol1Hn2/Pnz+P777wFUtQsNjcZ5Qb1xHhVRKjweD7t27UKXLl2QkJAAW1tbuLq6orKyEjdu3ICNjQ3Gjh2LtWvX0snvHVu3bsXt27eRkpICOzs7bnY2wWXQ6dOno0+fPiLbvX37lrtkW15eLlSWk5ODn3/+GZMnT4aVlRVMTU1RUlKCtLQ0rif1lClTMGXKFPkeHKF2UUfyaBcAsGTJEiQmJuL8+fNo3749N2uh4KmP0NBQLmFojOjKAKkX9vb2SExMxOjRo9GiRQukpaXhxYsXGDduHP7991/uV+iH9J5ubIyNjZGQkIDZs2fDzs4Ot27dwvPnzxEQEIAjR45g8eLFUtc5ePBgTJo0CZ06dUJZWRlSU1Nx584dWFhYIDw8HPHx8dzVASJ/1C6kJ492AVQ9Xnj69GmsXLkS7dq1Q3p6Oh49esT1J9i9e3fjnb4YoCmMG6uGNlVrr169GAC2du1aRYdClIwqT2FM7YJIQlMYk0YnMzMTp06dAiB5ohBCVA21C1KfKBkg9SI9PR2rV68WGdQjJSUFffr0QWlpKbp06VLtaGyENDbULoiyoA6EpF4UFRVh6tSpiIiIgJmZGczMzPD06VOu04+VlRWioqIUGyQh9YzaBVEWdGWA1Atra2vMmjULHTt2REVFBVJSUvDs2TO4ublh1qxZSExMhK2traLDJKReUbsgyoKuDJB6YWhoiPnz52P+/PmKDoUQpUHtgigLujJACCGEqDhKBgghhBAVR8kAIYQQouIoGSCkFng8Hng8HtfLmxBC7aIxoWSAEFJrjDFcunQJkZGR+OSTT2BiYgI+nw9jY2P4+/tj8+bNYsd9J0QV/fnnn1zCZG1trehwqkVPExBCau3MmTP47LPPuL+tra1hY2OD7OxsnDt3DufOnUNUVBT++usvGBoaKi5QQhTs5cuXGDNmjKLDqDW6MkAIqTXGGKysrLBy5Uo8evQIDx48QEJCAnJzcxEbGwsjIyNcvnwZY8eOVXSohCjUd999h5ycHPTr10/RodQKJQOEkFrz9vbG3bt3MXXqVLRq1UqoLDAwEOvWrQMAHDhwQGSIXUJUxdmzZ7FlyxYMGDAAffv2VXQ4tULJAKmVly9fYu7cuWjfvj10dXWhpaUFMzMzdOzYEdOnT8ejR4+E1s/JycHPP/+Mnj17ws7ODk2aNIG+vj48PT2xaNEivHr1Sux+5s6dCx6Ph+HDh6O0tBQLFiyAk5MTmjRpgtatW2P8+PEoLCzk1j9w4AB8fX1hYGAAfX199OrVC6mpqWLr9vf3B4/Hw44dO5CTk4ORI0fC3Nwc2trasLe3x8yZM/H69es6vT8JCQkICwuDpaUltLS00KxZM3z22Wc4ePCgxG3Onj2LoKAgtGrVCnw+H4aGhrC3t0dwcDD2799fpzjkTV9fn5tWV5xevXoBACoqKnDv3r36CkthqF1UT1XaxbuKi4vx9ddfQ19fn0uOGwSZzH1IlI4sp7csKipizs7ODADj8XjMwcGBeXl5MUtLS8bn8xkAdvToUaFtIiIiGADWpEkTZmNjw7y8vJitrS1TV1dnAJirqysrLCwU2decOXMYABYaGsr8/PwYj8djzs7OzNHRkampqTEAzNPTk5WWlrLIyEgGgFlYWDA3Nzemra3NADB9fX129+5dkbr9/PwYADZ79mxmYmLCNDQ0mJubG3N0dGQAGADm7u4uNi5B+YMHD0TKFi1axHg8HrdvNzc31rJlS26bMWPGiGyzdetWrtzIyIi5u7szV1dXZmRkxB1jQ/To0SPuuFJTU2VSp7JOYUztgtqFOFOmTGEA2MaNGxljjEVFRTEAzMrKSqb7kfUUxpQMNFKy/KCsWbOGAWBt27ZlGRkZQmWvX79mu3btYteuXRNafurUKRYXF8fKy8uFlmdlZbG+ffsyAGzcuHEi+xKc9Ph8PnN0dGRpaWlcWUJCAjMwMGAA2KBBg5iuri6LjY3lyvPy8pi7uzsDwMLCwkTqFpz0+Hw++/jjj1lOTg5XlpiYyFq3bs0AsGHDholsK+mkt3v3bu7EtXv3blZZWcmVnThxgpmamjIALCoqilteXl7OjI2NuXnqy8rKhOq8du0a27x5s0gM1dm2bRvz8fGp00uWVq5cyQAwExMTkX/7ulLWZIDaBbWL912+fJmpqakxX19f7pgpGSAKJcsPyujRoxkAtmbNGhlEVnWi5PP5TE9PT+SkKDjpAWBXrlwR2XbatGlc+fLly0XKjx07xp2E3vfuSS8rK0vitmpqaiwzM1OoTNxJr6ysjFlaWjIAQiffdx06dIgBYI6Ojtyyx48fMwDM0NBQ7DZ18e77Ju1LVjIyMrgvJXH/NnWlrMkAtQtqF+8qKSlhzs7OTFNTk928eZNb3lCSAeozQGpkaWkJAIiNjUVRUVGtt3v58iW2bduGESNGoEePHvjkk0/g6+uL7t27Q01NDUVFRbh7967Ybd3c3ODt7S2yvEOHDtz/f/PNNxLLCwsLUVBQILbu/v37w8LCQmR5r169YGdnh8rKSvz99981Ht/ly5eRlZUFc3NzBAYGil0nMDAQfD4ft27d4u4fN2/eHNra2njx4gWOHTtW435qY+7cuWBVyb3UL1l49eoVvvjiC7x48QKdOnXClClTZFKvMqN2IZ6qtosFCxbg5s2bmD59OpycnGQSf32icQZIjUaMGIFVq1bhzJkzMDMzQ7du3eDj4wNfX194eXlBTU00p4yPj8eAAQPw5MmTauuW1OPc3t5e7HJTU1MAgImJCQwMDCSWA1VfUM2aNRNZx9nZWWzdPB4PTk5OSE9PR1paWrVxA0BKSgoA4PXr1/D19ZW4Ho/HA1DVeczMzAzq6uqYOnUqFi9ejMDAQLi6uqJbt27o2LEjAgIC0Lx58xr3rUzevHmDwMBApKSkwMHBATExMdDQaPynFmoX4qliu0hJScGyZcvg6OiIyMhIRYdTJ42/xZIP1rJlS1y+fBnz5s3D4cOHERMTg5iYGACAubk5fvzxR3z77bdc4y4qKkL//v2Rl5eHrl274scff0S7du1gZGTE9US3tLREdnY2ysrKxO6zadOmYpcL9lFTOQBUVlaKXadFixYSj1VQVptfes+fP+f+Gx8fX+P6xcXF3P8vXLgQlpaW+PXXX3Ht2jVcv34dAKCuro7evXtj5cqVEk/8yqSkpARBQUGIi4uDtbU1Tp8+jZYtWyo6rHpB7UI8VWwXI0aMQHl5OTZv3gxNTU1Fh1MnlAyQWrG3t0d0dDTKy8uRlJSE8+fPIzY2FufPn8eECRNQWlqKqVOnAgCOHz+OvLw8WFhY4NixY9DW1haqizEm8VJlfajuV5mgTE9Pr8Z6dHV1AQBdu3bFqVOnpIqBx+Nh9OjRGD16NJ48eYILFy7g9OnT2L9/P2JjY5GUlITU1NRaj+K3fft2bN++XaoYBC5cuFCn7UpLS9G/f3+cOHEC5ubmOH36tNjLzI0ZtQtRqtguEhMToaamhuDgYJGyN2/eAACys7O5RHnHjh3o2bNnneKSF0oGiFQ0NDTg5eUFLy8vREREYO7cuZg3bx7Wr1/PnfTu378PAPDy8hI54QFAampqnZ9bloWbN29KLLt16xYA1OqeX9u2bQEAN27cAGNM6NeXNFq0aIEBAwZgwIABmDdvHlxcXJCdnY1jx44hLCysVnVkZWXV6leYrJSVlSE4OBjHjx9Hy5YtcebMGdja2tbb/pUNtYv/o6rtorKystqE6t3yt2/f1ktM0qAOhOSDCO4Jvju4io6Ojsiyd/3000/yD6waMTExePjwocjy48eP4969e1BTU0P37t1rrMfX1xetWrVCbm4udu3aJZPYmjdvjo8++giA5PdPnPrsKFVeXo4vv/wSsbGxaN68OU6fPs3FTKpQu1CtdlFdPVFRUQAAKysrbllQUJBU9dcHSgZIjaZPn46NGzfi6dOnQsvz8/OxcuVKAFW/dgS6dOkCoKpX8YYNG7jlJSUliIyMxJ49exR6X40xhsGDB+Px48fcsuTkZG5SkSFDhsDKyqrGejQ1NbF8+XIAwOjRo7F+/XqUlJQIrVNYWIjo6Gh899133LKbN29i1KhRiI+PF7l/e/jwYSQkJAAQfk+VRWVlJYYOHYqYmBg0a9YMp06dktjxrLGjdiGeKraLRuEDHkskSkyWz6B+8cUX3LO31tbWzNvbm3ueFgBr1qwZS05OFtpm6NCh3DZmZmasQ4cO3DPoCxcuZFZWVgwAO3v2rNB2gueCxQ1wwhhjZ8+erfGZXcF+3x8IRfA89axZs5ixsTHj8/nM3d2dOTk5cdu0a9eOFRQU1LpOxhhbsWIFN4Kcjo4Oc3NzY97e3szGxoYbgc3Pz49bPykpiauvadOmrH379szLy0todLaRI0dKPD5F2rNnDxejhYVFtYO2HD9+XCb7VNZxBqhdULuojYYyzgD1GSA1mjVrFlxdXREXF4eMjAwkJydDQ0MD9vb26NmzJ6ZNmyYyaU1UVBRcXV2xfft23L9/H2/evIGHhwcmTpyIoKAgbNmyRUFHA9ja2uLq1auYM2cOTpw4gfz8fNja2uLLL79EZGQk1wGqtiIiItCzZ0/88ssvOHPmDO7cuQMejwczMzP06NEDvXr1Epq57KOPPsK2bdtw+vRpJCYmIjMzk3vc6/PPP8eIESMwcOBAWR+2TLz7Cy87OxvZ2dkS163p8bmGjtpF9VSpXTQKMkkpiNKRddbYGAh+Ab07BCpRfsp6ZaCxoHbRMNEIhIQQQgiRKUoGCCGEEBVHyQAhhBCi4igZIIQQQlQcPU1AVEZcXJyiQyBE6VC7IABdGSCEEEJUHiUDhBBCiIqjZIAohLW1NXg8Hl2ifMeOHTvA4/GEXmvWrFF0WB8sKChI5LgyMjIUHZbSorYhitqG/FEyQIiS0dfXh4+PD3x8fGBubi5SfvPmTaxfvx4jRoxAu3btoKGhAR6Ph/Hjx8stpjdv3uDIkSOYOXMmevbsCRMTE+7k9ezZs2q3dXZ25o6HkA9RU9sQ55dffuE+q/7+/nKLLTMzE9988w0sLS2hpaUFMzMzDBkyBNevX5e4jTK1DepASIiScXd3r/ZXYWRkJI4cOVJ/AQG4fft2nWdaW7x4Mff/dZ3OlhCg5rbxvszMTEyfPl1+Af1/CQkJ+Oyzz/Dy5UsYGBigbdu2yMzMxJ49e3Do0CH88ccf6NWrl8h2ytQ26MoAIQ2MmZkZ+vXrh0WLFuHvv//GgAED5L5PPp8Pb29vfPvtt4iKisLx48flvk9CPtTo0aPx5s0bBAYGym0fb968Qf/+/fHy5UuEhYXh0aNH+O+///Do0SN8//33KCkpweDBg5V+rg66MkBIA7N+/Xqhv/fs2SP3fbq4uODKlSvc33TPnyi7HTt24O+//0ZERAR0dXVx9OhRuexny5YtyMnJgY2NDbZu3QotLS0AVQn00qVLceHCBVy8eBErVqzATz/9JJcYZIGuDKi4+/fvg8fjQUNDQ2ge8/ft3r0bPB4P9vb2Qsv/+ecffPfdd/D29karVq2gqamJFi1aoHfv3jh27JjU8QwfPhw8Hg9z584VW56RkcHd/5Pk5MmT6N+/PxdP8+bNERgYiDNnzkgdD1Fd1DYartzcXEydOhXW1taYP3++XPe1f/9+AMCIESO4RECAx+Nh9OjRQuspK0oGVJytrS18fHxQUVGBvXv3SlwvOjoaADB06FCh5f369cOKFSuQnp4OY2NjtG3bFowxHD9+HIGBgZg1a5Zc438XYwxjx45F9+7dERMTg7KyMri6uoIxhmPHjqFr165YunRpvcVDGjZqGw3XuHHjUFhYiA0bNkBHR0du+6moqMB///0HAOjSpYvYdfz8/AAAWVlZePjwodxi+VB0m4AgPDwc8fHxiI6OxtSpU0XKc3NzcerUKQCiJ7ylS5ciICAAtra2QstPnTqFIUOGYOHChQgMDIS3t7f8DuD/W7JkCTZu3AhLS0ts2rQJPXv25Mr27NmDb775BpGRkejYsSM+/fTTWtc7YcIEJCUlSR2Pu7s71q1bJ/V2RHlQ26ieMraNAwcOICYmBqGhoULHKQ8ZGRkoKSkBAJErQwIWFhbQ1NREaWkpbt++XeunIOobJQMEgwYNwqRJk5CcnIzr16/D1dVVqHzPnj2oqKiAr6+vyIlt1KhRYuv87LPPsGjRInz99dfYsWOH3E94hYWFWLx4MTQ0NBATEwMPDw+h8tDQUOTm5iIiIgLLly+X6oR37do1xMfHSx2ThgY1r4aO2kb1lK1t5OfnY/z48WjWrFm9jENQUFDA/X+zZs3ErqOmpgYDAwM8ffoUhYWFco+pruhsRWBoaIjAwEAcOHAA0dHRWLZsmVC54DJoeHi42O3T0tJw4MABpKamIj8/H2VlZQCAFy9eAAASExPlGH2VP//8E69fv4aPj4/IyU5gwIABiIiIwLlz51BRUQF1dfVa1U2Dv6guahvVU7a2MXnyZOTl5SEqKgrNmzeX+/7evn3L/b+mpqbE9bS1tQFUPXmgrCgZIACqTmYHDhzAnj17sGTJEqipVXUnuX79OpKTk6GtrY1BgwaJbDdz5kwsWbIElZWVEuvOz8+XW9wCKSkpAIB79+7B19dX7DqMMQBVDTI/Px+mpqZyj4s0fNQ2Gobjx49j165dCAgIwPDhw+tln4IveQAoLS0V+vtdgqShSZMm9RJXXVAyQAAAPXv2hKmpKXJychAXF4eAgAAAwG+//QYA6Nu3LwwMDIS22b9/PxYtWgQ1NTXMmTMH/fv3h42NDZo2bQo1NTWcOXMGXbt25X4NydPz588BAE+ePKnV87zFxcVyjog0FtQ2lF9xcTFGjx4NbW1tbNq0qd72a2RkxP1/fn6+2P4AlZWV3JWgd9dXNpQMEABV9/AGDx6MtWvXIjo6GgEBAaisrOSeYX+/cxQAREVFAQAiIiLEPu5Ul189gseiBL9U3vf69Wuxy3V1dQEAI0eOxNatW6Xeb3WUsZMUqT/UNiRTlraRl5eHnJwc8Pl8sVc/Xr16BQC4ePEiWrZsCQA4ffo0XFxcPmi/NjY2XOfAe/fuiU0GsrOzUVpaCgBo06bNB+1PnigZIJzw8HCsXbsWhw4dwvr16xEfH4+HDx/C1NRUbK/c+/fvAwA++eQTsfVdvHhR6hiaNm0KABJ/wdy5c0fs8rZt2wKo6tAka8rWSYrUP2ob4ilb2ygrK6v26se75bK4KqOurg4vLy/Ex8fj/Pnz3GOE7zp37hyAqqcKlPVJAoDGGSDv8PT0hIuLC4qKinD48GGuc1RISIjYxit4fvfRo0ciZXl5edi5c6fUMTg4OACA0Gh375J0CbBPnz5o0qQJ/v33X5w/f17q/VYnLi4OjDGpX8rWuYrUHbUN8ZSlbVhbW1e7vzlz5gCoeuZfsMzNzU0m+x44cCCAqqtBgscMBRhj3L+LuH4lyoSSASJEcMlz8+bN+OOPPwBI7iktGGRj8eLFuHXrFrf8/v376N27d516zvbu3Rs8Hg/JyclYvXo1t7y0tBSzZ8+WOFKaqakpIiMjAVQN9rJv3z5UVFQIrZObm4sNGzY0qsFVpOXv7w8ej1dvHawaE2objVtd28Y333wDMzMzPHjwAKNGjeL6XJSVleHHH3/ExYsXoauri2nTpskhahlipFG6evUqA8CuXr0q1XY5OTlMTU2NAWAAmIuLi8R1Hz58yFq0aMEAMA0NDebs7Mzatm3L1NTUmKGhIVu3bh0DwKysrES2tbKyYgDY2bNnRcomTpzI7b9FixasQ4cOzMDAgGlqarKNGzdyZe+rrKxkU6ZM4cr19fWZp6cn8/LyYq1bt+aWDxs2TKr3pL5ERUUxAMzPz6/a9fbu3cuMjY25l5aWFgPAtLW1hZbv3btXZFs/P786vwfu7u5c3UZGRtz72axZM265u7t7tXUItnnw4EGt91vXz7Ks66K2oTi1bRuSzJkzp8btP6RtXL58menp6TEAzMDAgHl6ejITExMGgGlqarLY2Nga65C2bciyXTDGGF0ZIELMzc253tKA5F8+QNXseVeuXMGQIUNgZGSEu3fv4vnz5xg2bBiSkpJEBmiprTVr1mD16tVwdnZGYWEh7t+/Dz8/P1y6dAk9evSQuB2Px8OqVatw6dIlhIeHw9jYGDdu3MDt27ehp6eHoKAgbN26FStWrKhTXMri7du3yM/P516CS5PvL3/3GWiB3NxcAJD4vHl1CgoKuLrfHTzl3eXvDsLS2FDbaNw+pG107NgRqampGDlyJPT09HDt2jWu4+l///0n11kTZUYmKQVROrLOGon8feivn5rk5eUxAMzExIS9fv1aLvuoCRrwlQGiONQ2RMn6s0zdnQlRMklJSdzjUZMmTUJwcLBM6v3nn3+4OuU5ecv7IiMjZd5xjagmahvyQ8kAIUrm5cuX3ONagp7KsnDhwgXo6uri22+/lVmdtXHz5s06PX5GyPuobcgP7/9fniCNTGJiIjw9PXH16tU63QMjRFnI8rNM7YI0FrL+LFMHQkIIIUTFUTJACCGEqDhKBgghhBAVR8kAIYQQouIoGSCEEEJUHCUDhBBCiIqjZIAQQghRcZQMEEIIISqOkgFCCCFExdFwxI1cWlqaokMg5IPI4zNM7YI0dLL+DFMy0EiZmJhAR0cHYWFhig6FkA+mo6MDExOTD66H2gVpTGTVLgCam6BRy8rKwrNnzxQdBiEfzMTEBJaWljKpi9oFaSxk2S4oGSCEEEJUHHUgJIQQQlQcJQOEEEKIiqNkgBBCCFFxlAwQQgghKo6SAUIIIUTFUTJACCGEqDhKBgghhBAVR8kAIYQQouIoGSCEEEJUHCUDhBBCiIqjZIAQQghRcZQMEEIIISqOkgFCCCFExVEyQAghhKg4SgYIIYQQFUfJACGEEKLiKBkghBBCVBwlA4QQQoiKo2SAEEIIUXGUDBBCCCEqjpIBQgghRMX9P5sdoSacyHuCAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.tree import plot_tree\n",
    "\n",
    "plot_tree(best_model);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Podemos inspeccionar las clases:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['a', 'b'], dtype='<U1')"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_model.classes_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Podemos realizar una predicción con el mejor modelo encontrado:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['b'], dtype='<U1')"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "best_model.predict([[1,1,2]])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Búsqueda Aleatoria\n",
    "\n",
    "- [Randomized Parameter Optimization](https://scikit-learn.org/stable/modules/grid_search.html#randomized-parameter-optimization)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "jupyter": {
     "outputs_hidden": true
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "from sklearn.linear_model import SGDClassifier\n",
    "SGDClassifier??"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = SGDClassifier(random_state=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.stats import loguniform\n",
    "\n",
    "param_dist = {\n",
    "    'loss': [\n",
    "        'hinge',        # SVM\n",
    "        'log_loss',          # logistic regression\n",
    "        #'preceptron',  # perceptron (not supported)\n",
    "    ],\n",
    "    'alpha': loguniform(1e-4, 1e2),  # de 0.0001 a 100.0\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Al parámetro alpha lo va a obtener desde muestras de una función de probabilidad tipo loguniform:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f22168ad110>]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "a= 1e-4\n",
    "b=100\n",
    "fig, ax = plt.subplots(1, 1)\n",
    "x = np.linspace(a, b, 100000)\n",
    "plt.xlim(-1, 5)\n",
    "plt.ylim(-0.50,4)\n",
    "ax.plot(x, loguniform.pdf(x, a, b), 'r-', lw=2, alpha=0.6, label='loguniform pdf')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Podemos muestrear mano con [ParameterSampler](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.ParameterSampler.html):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'alpha': 0.19628224813442813, 'loss': 'log_loss'}\n",
      "{'alpha': 11.630395714480311, 'loss': 'log_loss'}\n",
      "{'alpha': 0.1859084363016964, 'loss': 'log_loss'}\n",
      "{'alpha': 0.5512926225087429, 'loss': 'hinge'}\n",
      "{'alpha': 0.042220489831498315, 'loss': 'hinge'}\n",
      "{'alpha': 0.00021891618132748297, 'loss': 'hinge'}\n",
      "{'alpha': 0.019982467392329464, 'loss': 'hinge'}\n",
      "{'alpha': 7.46470024356059, 'loss': 'hinge'}\n",
      "{'alpha': 0.2560161525002872, 'loss': 'log_loss'}\n",
      "{'alpha': 10.386580256500284, 'loss': 'hinge'}\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import ParameterSampler\n",
    "\n",
    "for params in ParameterSampler(param_dist, 10, random_state=0):\n",
    "    print(params)\n",
    "    model = SGDClassifier(**params, random_state=0)\n",
    "    #model.fit(...)\n",
    "    #model.predict(...)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Búsqueda Aleatoria + Validación Cruzada"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "También podemos dejar que Sklearn se encargue de todo con [RandomizedSearchCV](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.RandomizedSearchCV.html):\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "\n",
    "model = SGDClassifier(random_state=0)\n",
    "\n",
    "cv = RandomizedSearchCV(model, param_dist, n_iter=10, cv=3, random_state=0)\n",
    "cv.fit(X, y);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "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>param_loss</th>\n",
       "      <th>param_alpha</th>\n",
       "      <th>mean_test_score</th>\n",
       "      <th>std_test_score</th>\n",
       "      <th>rank_test_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>log_loss</td>\n",
       "      <td>0.196282</td>\n",
       "      <td>0.611111</td>\n",
       "      <td>0.078567</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>log_loss</td>\n",
       "      <td>11.630396</td>\n",
       "      <td>0.611111</td>\n",
       "      <td>0.078567</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>log_loss</td>\n",
       "      <td>0.185908</td>\n",
       "      <td>0.611111</td>\n",
       "      <td>0.078567</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>hinge</td>\n",
       "      <td>0.551293</td>\n",
       "      <td>0.611111</td>\n",
       "      <td>0.078567</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>hinge</td>\n",
       "      <td>0.04222</td>\n",
       "      <td>0.611111</td>\n",
       "      <td>0.078567</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>hinge</td>\n",
       "      <td>0.000219</td>\n",
       "      <td>0.611111</td>\n",
       "      <td>0.078567</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>hinge</td>\n",
       "      <td>0.019982</td>\n",
       "      <td>0.611111</td>\n",
       "      <td>0.078567</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>hinge</td>\n",
       "      <td>7.4647</td>\n",
       "      <td>0.611111</td>\n",
       "      <td>0.078567</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>log_loss</td>\n",
       "      <td>0.256016</td>\n",
       "      <td>0.611111</td>\n",
       "      <td>0.078567</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>hinge</td>\n",
       "      <td>10.38658</td>\n",
       "      <td>0.611111</td>\n",
       "      <td>0.078567</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  param_loss param_alpha  mean_test_score  std_test_score  rank_test_score\n",
       "0   log_loss    0.196282         0.611111        0.078567                1\n",
       "1   log_loss   11.630396         0.611111        0.078567                1\n",
       "2   log_loss    0.185908         0.611111        0.078567                1\n",
       "3      hinge    0.551293         0.611111        0.078567                1\n",
       "4      hinge     0.04222         0.611111        0.078567                1\n",
       "5      hinge    0.000219         0.611111        0.078567                1\n",
       "6      hinge    0.019982         0.611111        0.078567                1\n",
       "7      hinge      7.4647         0.611111        0.078567                1\n",
       "8   log_loss    0.256016         0.611111        0.078567                1\n",
       "9      hinge    10.38658         0.611111        0.078567                1"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "results = cv.cv_results_\n",
    "df = pd.DataFrame(results)\n",
    "df[['param_loss', 'param_alpha', 'mean_test_score', 'std_test_score', 'rank_test_score']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {\n",
       "  /* Definition of color scheme common for light and dark mode */\n",
       "  --sklearn-color-text: black;\n",
       "  --sklearn-color-line: gray;\n",
       "  /* Definition of color scheme for unfitted estimators */\n",
       "  --sklearn-color-unfitted-level-0: #fff5e6;\n",
       "  --sklearn-color-unfitted-level-1: #f6e4d2;\n",
       "  --sklearn-color-unfitted-level-2: #ffe0b3;\n",
       "  --sklearn-color-unfitted-level-3: chocolate;\n",
       "  /* Definition of color scheme for fitted estimators */\n",
       "  --sklearn-color-fitted-level-0: #f0f8ff;\n",
       "  --sklearn-color-fitted-level-1: #d4ebff;\n",
       "  --sklearn-color-fitted-level-2: #b3dbfd;\n",
       "  --sklearn-color-fitted-level-3: cornflowerblue;\n",
       "\n",
       "  /* Specific color for light theme */\n",
       "  --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, white)));\n",
       "  --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, black)));\n",
       "  --sklearn-color-icon: #696969;\n",
       "\n",
       "  @media (prefers-color-scheme: dark) {\n",
       "    /* Redefinition of color scheme for dark theme */\n",
       "    --sklearn-color-text-on-default-background: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-background: var(--sg-background-color, var(--theme-background, var(--jp-layout-color0, #111)));\n",
       "    --sklearn-color-border-box: var(--sg-text-color, var(--theme-code-foreground, var(--jp-content-font-color1, white)));\n",
       "    --sklearn-color-icon: #878787;\n",
       "  }\n",
       "}\n",
       "\n",
       "#sk-container-id-1 {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 pre {\n",
       "  padding: 0;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 input.sk-hidden--visually {\n",
       "  border: 0;\n",
       "  clip: rect(1px 1px 1px 1px);\n",
       "  clip: rect(1px, 1px, 1px, 1px);\n",
       "  height: 1px;\n",
       "  margin: -1px;\n",
       "  overflow: hidden;\n",
       "  padding: 0;\n",
       "  position: absolute;\n",
       "  width: 1px;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-dashed-wrapped {\n",
       "  border: 1px dashed var(--sklearn-color-line);\n",
       "  margin: 0 0.4em 0.5em 0.4em;\n",
       "  box-sizing: border-box;\n",
       "  padding-bottom: 0.4em;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-container {\n",
       "  /* jupyter's `normalize.less` sets `[hidden] { display: none; }`\n",
       "     but bootstrap.min.css set `[hidden] { display: none !important; }`\n",
       "     so we also need the `!important` here to be able to override the\n",
       "     default hidden behavior on the sphinx rendered scikit-learn.org.\n",
       "     See: https://github.com/scikit-learn/scikit-learn/issues/21755 */\n",
       "  display: inline-block !important;\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-text-repr-fallback {\n",
       "  display: none;\n",
       "}\n",
       "\n",
       "div.sk-parallel-item,\n",
       "div.sk-serial,\n",
       "div.sk-item {\n",
       "  /* draw centered vertical line to link estimators */\n",
       "  background-image: linear-gradient(var(--sklearn-color-text-on-default-background), var(--sklearn-color-text-on-default-background));\n",
       "  background-size: 2px 100%;\n",
       "  background-repeat: no-repeat;\n",
       "  background-position: center center;\n",
       "}\n",
       "\n",
       "/* Parallel-specific style estimator block */\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item::after {\n",
       "  content: \"\";\n",
       "  width: 100%;\n",
       "  border-bottom: 2px solid var(--sklearn-color-text-on-default-background);\n",
       "  flex-grow: 1;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel {\n",
       "  display: flex;\n",
       "  align-items: stretch;\n",
       "  justify-content: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  position: relative;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item:first-child::after {\n",
       "  align-self: flex-end;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item:last-child::after {\n",
       "  align-self: flex-start;\n",
       "  width: 50%;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-parallel-item:only-child::after {\n",
       "  width: 0;\n",
       "}\n",
       "\n",
       "/* Serial-specific style estimator block */\n",
       "\n",
       "#sk-container-id-1 div.sk-serial {\n",
       "  display: flex;\n",
       "  flex-direction: column;\n",
       "  align-items: center;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  padding-right: 1em;\n",
       "  padding-left: 1em;\n",
       "}\n",
       "\n",
       "\n",
       "/* Toggleable style: style used for estimator/Pipeline/ColumnTransformer box that is\n",
       "clickable and can be expanded/collapsed.\n",
       "- Pipeline and ColumnTransformer use this feature and define the default style\n",
       "- Estimators will overwrite some part of the style using the `sk-estimator` class\n",
       "*/\n",
       "\n",
       "/* Pipeline and ColumnTransformer style (default) */\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable {\n",
       "  /* Default theme specific background. It is overwritten whether we have a\n",
       "  specific estimator or a Pipeline/ColumnTransformer */\n",
       "  background-color: var(--sklearn-color-background);\n",
       "}\n",
       "\n",
       "/* Toggleable label */\n",
       "#sk-container-id-1 label.sk-toggleable__label {\n",
       "  cursor: pointer;\n",
       "  display: block;\n",
       "  width: 100%;\n",
       "  margin-bottom: 0;\n",
       "  padding: 0.5em;\n",
       "  box-sizing: border-box;\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 label.sk-toggleable__label-arrow:before {\n",
       "  /* Arrow on the left of the label */\n",
       "  content: \"▸\";\n",
       "  float: left;\n",
       "  margin-right: 0.25em;\n",
       "  color: var(--sklearn-color-icon);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {\n",
       "  color: var(--sklearn-color-text);\n",
       "}\n",
       "\n",
       "/* Toggleable content - dropdown */\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content {\n",
       "  max-height: 0;\n",
       "  max-width: 0;\n",
       "  overflow: hidden;\n",
       "  text-align: left;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content pre {\n",
       "  margin: 0.2em;\n",
       "  border-radius: 0.25em;\n",
       "  color: var(--sklearn-color-text);\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-toggleable__content.fitted pre {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {\n",
       "  /* Expand drop-down */\n",
       "  max-height: 200px;\n",
       "  max-width: 100%;\n",
       "  overflow: auto;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {\n",
       "  content: \"▾\";\n",
       "}\n",
       "\n",
       "/* Pipeline/ColumnTransformer-specific style */\n",
       "\n",
       "#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-label.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator-specific style */\n",
       "\n",
       "/* Colorize estimator box */\n",
       "#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-estimator.fitted input.sk-toggleable__control:checked~label.sk-toggleable__label {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-label label.sk-toggleable__label,\n",
       "#sk-container-id-1 div.sk-label label {\n",
       "  /* The background is the default theme color */\n",
       "  color: var(--sklearn-color-text-on-default-background);\n",
       "}\n",
       "\n",
       "/* On hover, darken the color of the background */\n",
       "#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "/* Label box, darken color on hover, fitted */\n",
       "#sk-container-id-1 div.sk-label.fitted:hover label.sk-toggleable__label.fitted {\n",
       "  color: var(--sklearn-color-text);\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Estimator label */\n",
       "\n",
       "#sk-container-id-1 div.sk-label label {\n",
       "  font-family: monospace;\n",
       "  font-weight: bold;\n",
       "  display: inline-block;\n",
       "  line-height: 1.2em;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-label-container {\n",
       "  text-align: center;\n",
       "}\n",
       "\n",
       "/* Estimator-specific */\n",
       "#sk-container-id-1 div.sk-estimator {\n",
       "  font-family: monospace;\n",
       "  border: 1px dotted var(--sklearn-color-border-box);\n",
       "  border-radius: 0.25em;\n",
       "  box-sizing: border-box;\n",
       "  margin-bottom: 0.5em;\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-0);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-estimator.fitted {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-0);\n",
       "}\n",
       "\n",
       "/* on hover */\n",
       "#sk-container-id-1 div.sk-estimator:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-2);\n",
       "}\n",
       "\n",
       "#sk-container-id-1 div.sk-estimator.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-2);\n",
       "}\n",
       "\n",
       "/* Specification for estimator info (e.g. \"i\" and \"?\") */\n",
       "\n",
       "/* Common style for \"i\" and \"?\" */\n",
       "\n",
       ".sk-estimator-doc-link,\n",
       "a:link.sk-estimator-doc-link,\n",
       "a:visited.sk-estimator-doc-link {\n",
       "  float: right;\n",
       "  font-size: smaller;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1em;\n",
       "  height: 1em;\n",
       "  width: 1em;\n",
       "  text-decoration: none !important;\n",
       "  margin-left: 1ex;\n",
       "  /* unfitted */\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted,\n",
       "a:link.sk-estimator-doc-link.fitted,\n",
       "a:visited.sk-estimator-doc-link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "div.sk-estimator:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link:hover,\n",
       ".sk-estimator-doc-link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "div.sk-estimator.fitted:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover,\n",
       "div.sk-label-container:hover .sk-estimator-doc-link.fitted:hover,\n",
       ".sk-estimator-doc-link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "/* Span, style for the box shown on hovering the info icon */\n",
       ".sk-estimator-doc-link span {\n",
       "  display: none;\n",
       "  z-index: 9999;\n",
       "  position: relative;\n",
       "  font-weight: normal;\n",
       "  right: .2ex;\n",
       "  padding: .5ex;\n",
       "  margin: .5ex;\n",
       "  width: min-content;\n",
       "  min-width: 20ex;\n",
       "  max-width: 50ex;\n",
       "  color: var(--sklearn-color-text);\n",
       "  box-shadow: 2pt 2pt 4pt #999;\n",
       "  /* unfitted */\n",
       "  background: var(--sklearn-color-unfitted-level-0);\n",
       "  border: .5pt solid var(--sklearn-color-unfitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link.fitted span {\n",
       "  /* fitted */\n",
       "  background: var(--sklearn-color-fitted-level-0);\n",
       "  border: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "\n",
       ".sk-estimator-doc-link:hover span {\n",
       "  display: block;\n",
       "}\n",
       "\n",
       "/* \"?\"-specific style due to the `<a>` HTML tag */\n",
       "\n",
       "#sk-container-id-1 a.estimator_doc_link {\n",
       "  float: right;\n",
       "  font-size: 1rem;\n",
       "  line-height: 1em;\n",
       "  font-family: monospace;\n",
       "  background-color: var(--sklearn-color-background);\n",
       "  border-radius: 1rem;\n",
       "  height: 1rem;\n",
       "  width: 1rem;\n",
       "  text-decoration: none;\n",
       "  /* unfitted */\n",
       "  color: var(--sklearn-color-unfitted-level-1);\n",
       "  border: var(--sklearn-color-unfitted-level-1) 1pt solid;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 a.estimator_doc_link.fitted {\n",
       "  /* fitted */\n",
       "  border: var(--sklearn-color-fitted-level-1) 1pt solid;\n",
       "  color: var(--sklearn-color-fitted-level-1);\n",
       "}\n",
       "\n",
       "/* On hover */\n",
       "#sk-container-id-1 a.estimator_doc_link:hover {\n",
       "  /* unfitted */\n",
       "  background-color: var(--sklearn-color-unfitted-level-3);\n",
       "  color: var(--sklearn-color-background);\n",
       "  text-decoration: none;\n",
       "}\n",
       "\n",
       "#sk-container-id-1 a.estimator_doc_link.fitted:hover {\n",
       "  /* fitted */\n",
       "  background-color: var(--sklearn-color-fitted-level-3);\n",
       "}\n",
       "</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>SGDClassifier(alpha=0.19628224813442813, loss=&#x27;log_loss&#x27;, random_state=0)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator fitted sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label fitted sk-toggleable__label-arrow fitted\">&nbsp;&nbsp;SGDClassifier<a class=\"sk-estimator-doc-link fitted\" rel=\"noreferrer\" target=\"_blank\" href=\"https://scikit-learn.org/1.4/modules/generated/sklearn.linear_model.SGDClassifier.html\">?<span>Documentation for SGDClassifier</span></a><span class=\"sk-estimator-doc-link fitted\">i<span>Fitted</span></span></label><div class=\"sk-toggleable__content fitted\"><pre>SGDClassifier(alpha=0.19628224813442813, loss=&#x27;log_loss&#x27;, random_state=0)</pre></div> </div></div></div></div>"
      ],
      "text/plain": [
       "SGDClassifier(alpha=0.19628224813442813, loss='log_loss', random_state=0)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cv.best_estimator_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'alpha': 0.19628224813442813, 'loss': 'log_loss'}"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cv.best_params_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Referencias\n",
    "\n",
    "\n",
    "Scikit-learn:\n",
    "\n",
    "- [Tuning the hyper-parameters of an estimator](https://scikit-learn.org/stable/modules/grid_search.html#grid-search)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}