notebooks/mod_06_topic_12_kmeans.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Практика виконання кластеризації. EDA датасету NCI-60"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Знайомство з даними"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "import pickle\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from scipy.cluster.hierarchy import dendrogram\n",
    "from matplotlib import pyplot as plt\n",
    "import seaborn as sns\n",
    "from yellowbrick.cluster import KElbowVisualizer\n",
    "from kneed import KneeLocator\n",
    "from sklearn.cluster import KMeans, AgglomerativeClustering\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.decomposition import PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(64, 6830)"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "with open('../datasets/mod_06_topic_12_nci_data.pkl', 'rb') as fl:\n",
    "    data_dict = pickle.load(fl)\n",
    "\n",
    "data = data_dict['data']\n",
    "target = data_dict['labels']\n",
    "\n",
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "label\n",
       "BREAST         7\n",
       "CNS            5\n",
       "COLON          7\n",
       "K562A-repro    1\n",
       "K562B-repro    1\n",
       "LEUKEMIA       6\n",
       "MCF7A-repro    1\n",
       "MCF7D-repro    1\n",
       "MELANOMA       8\n",
       "NSCLC          9\n",
       "OVARIAN        6\n",
       "PROSTATE       2\n",
       "RENAL          9\n",
       "UNKNOWN        1\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "target['label'].value_counts().sort_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Практика виконання кластеризації. Підготовка й обробка даних"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = StandardScaler().fit_transform(data)\n",
    "\n",
    "pca = PCA(random_state=42).fit(X)\n",
    "pve = pca.explained_variance_ratio_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set_theme()\n",
    "\n",
    "kneedle = KneeLocator(\n",
    "    x=range(1, len(pve) + 1),\n",
    "    y=pve,\n",
    "    curve='convex',\n",
    "    direction='decreasing')\n",
    "\n",
    "kneedle.plot_knee()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[Text(0.5, 0, 'number of components'),\n",
       " Text(0, 0.5, 'cumulative explained variance')]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "n_components = kneedle.elbow\n",
    "\n",
    "ax = sns.lineplot(np.cumsum(pve))\n",
    "\n",
    "ax.axvline(x=n_components,\n",
    "           c='black',\n",
    "           linestyle='--',\n",
    "           linewidth=0.75)\n",
    "\n",
    "ax.axhline(y=np.cumsum(pve)[n_components],\n",
    "           c='black',\n",
    "           linestyle='--',\n",
    "           linewidth=0.75)\n",
    "\n",
    "ax.set(xlabel='number of components',\n",
    "       ylabel='cumulative explained variance')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Практика виконання кластеризації. Кластеризація набору даних"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Визначення кількості кластерів за допомогою KMeans"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = pca.transform(X)[:, :n_components]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgwAAAFsCAYAAACgmuHIAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8fJSN1AAAACXBIWXMAAAsTAAALEwEAmpwYAABiZUlEQVR4nO3dd1yVdf/H8dc5cFiCDJmCguBAcYs5AfcEB2k50qzMsqGN242a5UhvzewuK73Vcpal4gLMkXtrORJcOEAQkBFDZJ3z+8NbfpIiosAFh8/z8ehh5zrXeH/PAc7nXN/v9b1UOp1OhxBCCCHEE6iVDiCEEEKI8k8KBiGEEEIUSQoGIYQQQhRJCgYhhBBCFEkKBiGEEEIUSQoGIYQQQhRJCgZR4URHR1O/fn369u1L3759CQgIYNCgQYSEhOSvs2jRIoKDg5+4n6+//ppdu3YV+/gPb/c0xymOvXv38vLLL9OnTx969+7N2LFjuX37dont/2lt3LiRFi1a5L/GD/4bP348ABMnTmTZsmUA1KtXj6SkpFLNEx4eTpcuXQgMDCQ6OvqZ9nHs2DH8/f0LLFuxYgW+vr5ERERw7Ngx6tWrx4QJEx7ZdtiwYTRr1uyZjluSfv/9d4YNG0bfvn3p3bs3H3zwAbGxscD99+ytt9565n0/6+/Dm2++yZUrV575uKLiMFQ6gBDPwsTEhM2bN+c/vnXrFiNGjMDAwIDu3bszduzYIvdx7NgxateuXexjP7zd0xznacXFxTFhwgQ2btyIs7MzAN9++y0ffPABP/30U4kd52l5e3vz/fffl/lxH2f37t20atWKWbNmldg+Fy5cyG+//ca6detwdnbm2LFj2NnZ8fvvv5OZmYmpqSlw/2fr2rVrJXbcZ7V161a+/fZbvv32W1xdXdHpdCxZsoThw4ezffv2597/s/4+LF269LmPLSoGKRiEXnB2dmbMmDEsW7aM7t27M3HiROrUqcMbb7zBV199xc6dO9FoNFhbWzNnzhx27tzJ+fPnmTdvHgYGBrRu3ZoZM2YQERGBSqXCx8eHjz76CENDQxo2bEjnzp2JiIggICCgwHa7d+/OP87JkyeZN28emZmZaDQaPvjgA3x9fdm4cSM7d+5ErVZz48YNTExMmDt3Lh4eHgXakJycTE5ODnfv3s1f9uqrr+Lp6Zn/+Pvvv2fTpk0YGhri6urK559/joWFBd988w3bt2/HwMCAWrVqMXXqVOzs7Bg2bBiWlpZERkYyePBg+vXrx6xZs7h06RI5OTm0adOG8ePHY2j4fH8KvvzyS86dO4dWq+WDDz6gY8eOAI/NdebMGZYvX87atWsB6N69O71792bMmDHcvn2bAQMGsH//ftTq+ydAt2zZwrp168jLy+PevXssWLDgqds7bNiwR7JqtVo+/fRTIiIiWLt2LdbW1vnPWVlZUaNGDXbt2kVAQAAAwcHBBAQEFCjafvnlF9atW4dWq8XKyoqpU6fi4eHBtWvX+PTTT8nIyCAhIQFPT0++/PJLjI2NadSoEaNGjeLQoUPEx8czcuRIhgwZQkJCAhMmTCA5ORkAPz8/Pvjgg0dyL1y4kM8++wxXV1cAVCoVo0aNwsnJiezs7ALrDhs2jKFDh9KjR49HHj/N74Ofnx/z58/nxIkT5OXl0aBBA4KCgjA3N6dTp040btyYixcv8tFHHzFnzhwWLVrE3bt3WbhwITVq1ODy5cvk5uYyY8YMWrRoQVJSEpMmTeLmzZtYWVlhZ2dHnTp1eP/995/p500oQ7okhN7w9PTk0qVLBZbFxsby448/smHDBjZu3Ei7du04e/YsQ4cOpWHDhowfP56uXbsyc+ZMrKys2Lp1Kxs2bODixYssX74cgJycHDp27MiOHTt47733Cmz3QHJyMmPGjGHKlCls3bqVuXPnMm7cOKKiogA4ceIEU6dOZdu2bTRp0oQlS5Y8Nv9LL71E//796dWrF0FBQfz+++/4+PgA979lb9y4kZ9//plt27bh4uLC6tWr2bBhAwcOHODXX39l69at1KlTh4kTJ+bvt2rVqoSEhDBs2DBmz56Nl5cXGzduJDg4mOTkZFasWPHY1/PkyZOPdEls2LDhseu6uLiwadMm/v3vfzNx4kSSkpIKzdW+fXsuXrxIamoq0dHRZGRkcPjw4fw2dunSJb9YAOjTpw+DBg2iV69eLFiwoFjt/afc3FzGjRvHunXrGD16dIFi4YF+/foVOHsVGhpaoCvj+PHjBAcHs2bNGoKDgxk5ciTvvfceAOvXr6dfv36sX7+e3377jejoaPbu3QtAdnY21tbW/PTTT3z11VfMmTOHrKws1q9fn//6rVmzhhs3bpCWllYgU3JyMrdu3aJ58+YFlqtUKvr06YO5uflj35d/etrfhyVLlmBgYMDGjRvZsmUL9vb2zJ8/P38/derUITQ0tMDvAMDZs2d5/fXXCQ4OJjAwkIULFwIwc+ZMateuTWhoKIsWLeL06dNPlVeUL3KGQegNlUqFiYlJgWUODg54enrSv39/fH198fX1pU2bNo9su3//ftatW4dKpcLIyIhBgwbx448/MmrUKOD+6fknOXv2LDVr1qRJkybA/T+ozZs35/jx46hUKry8vHB0dASgQYMG7Ny587H7mThxIm+99RbHjx/nxIkTzJs3j1WrVrFmzRqOHDlCjx49sLS0BGDSpEnA/W6RwMBAzMzMABg+fDjfffdd/rfOh7Pv3buXc+fO8euvvwJw7969QttUnC6JwYMHA1C3bl08PDz4448/2L9//2NzqdVq2rZty6FDh0hOTubll1/m559/Ji0tjT179jBy5MgnHquw/T6uvf907do1mjVrxty5c5k4cSIbN27EycmpwDodO3bkk08+4c6dO9y4cQN3d/f81xzuv4Y3btxg0KBB+ctSU1NJSUlh3LhxHDp0iKVLl3L9+nXi4+MLnDHq3LkzAF5eXmRnZ3P37l18fHwYNWoUsbGxtG3blo8//hgLC4sCmR4UUFqt9omvTVGe9vdh7969pKWl5RdyOTk5VKtWLf/5wl7j6tWrU79+feD+z/mmTZsA2LdvX/7/29vb55/5EBWLFAxCb5w7d466desWWKZWq1m9ejXnzp3jyJEjzJ49Gx8fn/zBew9otVpUKlWBx7m5ufmPH3w4FSYvL6/A9gA6nY7c3Fw0Gk2BQkalUvG4W7js3r2blJQUXnzxRbp370737t358MMP8fPz48KFCxgYGBQ4RmpqKqmpqcXKrtVqWbRoUX53SGpq6iO5n8XDZwS0Wi2GhoZPzNWlSxf2799PamoqI0eOJDIykl27dnHp0iVeeOGFJx7red4rNzc35syZA8Dp06d5//33Wbt2LUZGRvnrGBkZ0a1bN7Zv386VK1fo37//I8fv27cv48aNy38cHx+PpaUlH374IXl5efTs2ZMOHToQGxtb4L02NjYGyM+v0+lo3Lgxu3fv5siRIxw9epSBAweydOlSGjZsmL+dpaUlbm5unDlzhrZt2xbIM3bsWEaPHv1IWx8+bk5ODlC834fJkyfj5+cHQEZGBllZWUW+xoX9nBsaGhbI8/DPi6g45F0TeuHatWssXryY119/vcDyiIgI/P398fDw4K233mLEiBGcO3cOAAMDg/wPmvbt27N69Wp0Oh3Z2dmsX7/+kT/MDzy83QNNmzYlMjKSs2fPAnD58mVOnDhR5Iffw6pUqcIXX3xRYMR5VFQUBgYG1KxZk7Zt27Jz507S09MB+M9//sMPP/yAj48PGzZsyP8mu2rVKlq2bFngQ/CB9u3b88MPP+S3c/To0axevfqpMxbmwbfHv/76i5s3b9KkSZMn5urUqRNHjhwhPDycxo0b065dOxYtWoSvry8GBgZPPFZx2vtPGo0m//+nTJlCXl4eM2bMeGS9fv36sWnTJk6cOJHfJfRA+/bt2b59O/Hx8QCsW7eOV199FYCDBw/y7rvv0qtXLwDOnDlDXl7eEzPNnz+fxYsX06VLF6ZMmULt2rW5fPnyI+u99957zJo1ixs3bgD3i9TFixcTERGBu7t7gXVtbGw4f/48AFeuXOHixYtA8X4f1qxZQ3Z2NlqtlqlTp/LFF188sR1P4ufnl39WKzk5mV27dpVIoSrKlpxhEBXSvXv36Nu3L3D/24qxsTEfffQRHTp0KLCep6cnPXv25MUXX8TMzAwTExOCgoIA6NSpE1988QU5OTkEBQUxc+ZMAgICyMnJwcfHh7fffvuxx354uwdsbGxYtGgRn332Gffu3UOlUjFnzhxq1arFH3/88VRtat26NVOnTmXChAmkpaVhYGCAnZ0dS5cuxdLSEj8/P65cuZJ/+r927dp89tlnmJmZERsby8CBA9Fqtbi6uhbob37YlClTmDVrVn4727ZtW2gXwIMxDA970K/9T1FRUfTr1w+VSsUXX3yBlZUVAwYMKDSXhYUFHh4emJqaYmBggI+PD1OmTKFbt25Fvk5P2m9xGBsbs2jRIvr370/jxo1xc3PLf65Zs2ZkZmbSqVOnRwaEtm/fnjfffJPXX38dlUqFubk5X3/9NSqVig8//JB3330XMzMzzM3NadmyJTdv3nxijldffZWJEyfi7++PkZER9erVo3fv3o+sFxAQgE6n46OPPiI3N5esrCy8vLz48ccfHymWRo8ezcSJE9m3bx/u7u75XQhP+/vwzjvvMHfuXPr3709eXh7169cvME6kuCZNmkRQUBABAQFYWVlRvXr1R7oPRfmnkttbCyGEKE1r1qyhQYMGNGvWjOzsbIYMGcL777+f3+UhKgY5wyCEEKJUPTgbptVqycnJoUePHlIsVEByhkEIIYQQRZJBj0IIIYQokhQMQgghhCiSFAxCCCGEKJIUDEIIIYQoklwlUYTk5Ay02pIZF1qtmjmJieklsq/yTNqpPz766D0MDQ2YN2+R0lFKXWV4P0HaqW9Ksp1qtQpr6yqFPi8FQxG0Wl2JFQwP9lcZSDv1Q1xcPBqNgd638wFpp36RdpYs6ZIQQgghRJHkDIMQolAODo6YmGiKXlEIofekYBBCFGr+/EXY2VmQkJCmdBQhhMKkYBBCPJO8vFySkxPIzc1WOkqJiI9Xo9VqlY5R6qSd+uVZ2mloaIS1tR0GBsUrAaRgEEIUatasGZiaavjoo8mPPJecnICJiRlVqjjqxa2KDQ3V5Obq/weMtFO/FLedOp2OjIxUkpMTsLV1Kt6xihtOCFF5RERcQKMxeOxzubnZelMsCFFZqFQqqlSpSnp6SrG3lYKhjAQfiKRKFWO6NndWOooQJUaKBSEqnmf9vZWCoQwEH4hky6HrAGRkZNHPx13ZQEIIIUQxyTwMpezhYgFgy6HrBB+IVC6QEEII8QykYChF/ywWHpCiQVQUbm618PDwUDpGpRUY2JtLlyKUjiEEIF0SQogn+Oyzz2UeBoWkpqaSmHgHV9daJbbP994bxYUL5zEwuD+Q1dbWjnXrNpbY/sur2NgYFiz4nPPnz2FkZESHDp0YM+ZjtFotCxZ8zsmTx0lNTcXFxYVRo96lTZt2+dtu2PAzISHbiIy8Qpcu3Zky5ZPHHiMq6iavvjqIDh06M23aZ8Xa9oHU1L+ZM+czTpw4iqWlFW+99R7duvUgOzu7yJxdu/oU2FdWVhb9+w/gww/HP9uL9hhyhqEU9fNxp087t0eW+7d1lXEMQijg9OmTDBv20iP/r5QPP3yXlJSUxz4XGXkFZ2cXjI2NS/SYH388gZ07D7Bz54FyVSzk5eWV2r4XLPgca2sbNm8OY8WKNfz552k2bfqVvLw87O0d+PrrJezYsZeRI0czbdokYmNj8re1tbXj1VffoHfvPk88xhdfzMXTs0GBZU+77f/nnItGo2HLlt+YNm0mCxbMITLy6lPlfPCe7tx5gC1bfsPY2JiOHbsU41UqmhQMpexxRYNaRpaLCmLq1Il88MEHSsfQWydOHCv0uatXL+Pufr876N69e3zyyRQmTx7H3bt3yypeAbduRTN+/Af07t2Z7t39+OCDd/Kf27kzjNdff4Xu3f146aW+nD59Ep1Ox+rVP/Dii/706NGBqVMnkp7+/3dV3Lo1mA8/fJc5cz6lR4+O/PzzGgC2bNnEK68MpHt3Pz7+eAzJyUnPnT02NoZOnbpgbGxMtWq2tGrVlmvXrmJqasobb7yFk1N11Go17dr5UL16dS5eDM/f1s+vE76+Haha1bLQ/e/atQNzcwtatGhZYPnTbPtAZmYm+/btYeTItzEzM6NJk6a0b+/Ljh0hT5XzYXv37sbKyoYmTZo95Sv0dKRLogw8OJtQpYox12+lsP3IDVrUs6eGvbnCyYR4suvXrxU6D0N5d/Dgfn78cRm5uTmYmJjw7rsfPLJOZmYmQUHjiY6OxsLCgnHjJlOzpisAmzdv5Ndff0KtNsDGxoYPPxzPtGmTeO+9D/D2foGdO8OYM+dTQkP3YGxswueff0a9evXp339A/v61Wi1fffUFf/11jszMu+h0OiZMCKJx46bMnj0DgDFj3uLf/16Eg4NjgWxXrlzB3b02MTG3CAoaT/v2frz22psFLokbP/4Dzp7987Htb9y4KfPmffnI8m+//Q+LF39FzZquvPnmOzRv7v1Ur+fMmdPp0qU7c+YsIDc3N//Dat261WzfvoWpUz+lTp26XLsWiZmZGUuXfsvZs3/y/fc/YGFhwfTpk1ixYinvv/8hcL8gOn/+HNOnv8SECUHk5uaycuVy9u7dw+eff4GDgyMLFnzOkiXfMmHClOdq88CBg9i16zeaNfMmLS2Vo0cPMXLk6EfWS0pKJCrqJrVqPf24nYyMdP773+9ZtGgx27Ztfurt/ikq6gZqtUH+zx+Ah0dd/vzzdLFzhoZuo0ePXiV+2bMUDGWkn487dnYWXLuZxF/XklgeEk7Q8BYYqOUkj9Afw4a9/MiyHj16M3TocDIzMxk1asQjz/fvP4DAwIEkJSUxduyjf8QHD36FXr0CiI2Nwcmp+lPliIq6yZIl3/Cf/3yPpaUVkZFX+fDDd5gwYWqB9eLj45g+fSaNGjVh27ZNfPbZNJYu/ZFTp06wdu1KvvtuBdbW1oSEbGXy5H/RqVNXjh49jLf3Cxw7dgQLCwvOnPmTli1bceTIId58s2D+CxfOc+dOAt9/vwK1Ws2qVT+wevWPzJvXlMmTpxMSspWvvvoeKyurR9oQGXkFtVrF2LGjGTPmI3x8OjyyzuM+HJ9k9Ogx1K7tgUplwO7dvzFhwkf88MNanJ1ditw2JiYarVaLVqvF2NiYxo2bkpyczIoVS/nmm6XUqVMXAA+P2iQlJbJhw8+sWfMrtra2AHTo0LnAB+qVK5cZPPgV2rf3A+5/8P744zJWrFiLi0sNAPz9+7JgwefP1WaApk1bsGVLMN27+5GXl0fPnv74+nYosE5ubi4zZkylR4/euLq6PfW+ly79Dn//Po8UfMWVmZmJuXnBL5Hm5ubcvZtRrJy3b9/mzz9PM3Hi1Eeee17yaVXGzE01DO1Wjxu30/jteJTScYTQSydOHCMx8Q5jx77DiBFD+PTTqahUaqKjC/7OeXjUplGjJgD07t2HixfDSU9P59ixw3Tq1BVra2sAevUKICEhHj+/Thw9ehidTseZM3/y8stDOXHiGH/9dQ5nZxeqVbMtsP+GDRszatRoNm/eyNdff8nevbvJzCy6S0Gn0xEZeZX9+/fSt2/gY4uFZ+Hl1ZAqVapgZGREz57+NGrUhCNHDj7VttOmzeTgwX3069eDOXM+JTX1b06ePIaHh0d+sfDAmTN/4O5eG1tbu/xlf/+dQrVq1fIfX7lyuUAf+8mTx8nNzWXUqFfp0aMDPXp04OOP36dKlec7E6vVavnoo/fw8+vIzp0H2L59F2lpqXz77VcF1vnss6loNIZ89NGEp9735csXOXnyOC+/PPS5MgKYmpqSkZFeYFlGRgZmZlWKlTMsbBuNGzelevWSnyRQzjAowLueHS3q2rHpwDWa1rHFqVqVojcSogJYternQp8zNTV94vM2NjZPfP5pzy4AaLV5tGjxAp9+Oid/WVzcbaKibhZYT60u2N2iUqkwNDQkL0+LoWHB07k6HWg0GnJzczh4cB81atSgXTtfpk2bhIGBAR06dH4kx+HDB1m0aD6DBr2Cj48frq5u7NgRUmT+mJhbAHz55WLGjh2Nt/cLjwyoA/j44zGcPfvHY/fRuHEzFiz46rHPPdxena7IOAC0aNGSFi1akpycxL/+NZaQkK1oNEaYm1s8sm5KSsoj35YPHNhHx473X6Pbt2PJy8st8A05NTUVH58OzJw594k5itvm1NRU4uPjePHFlzEyMsLIyIhevfqwdOli3nlnLDqdjs8//4ykpCTmz1+EoeHTfyz+8ccpbt+O4cUX/QHIzLxLXp6W69cjWb58zVPvB6BGDVfy8vKIirpJjRo1Abhy5RK1at3v0n7anGFhIbzyyqvFOvbTkoJBASqVile61SXiv8dYERrBxKHNZSCkKJc8PRtgaqpROkaxtWjxAv/97/fcuHEdV1c3jhw5yIwZU5k69dMC6125conLly9Sp049Nm3aQKNGTTAxMaF16zbMn/85AwcOxtramu3bt2BpaYmLSw18fTvw3Xdf4+/fD1dXNzIy0vntt1C+/XbZIzlOnDhGu3Y+9O8/gKyse6xZ82OBOwsaGBiQm5v7yHZXr16hdu3aeHjUZvz4+4Mdlyz5Mf/0/gNFFQQPS0tL48KF83h7e6PTqdizZydnzpxm7NiPitx23749uLvXxsWlBnfv3iUtLZU6dephZGTEkiXfcPnyJWrXrkN0dBR5eXnUr9+AJUsWc+tWNNbWNqxdu5Lk5CR69+6b/7q7u9dG/VCXbL16nixf/j0XL0ZQr54nGRnpnD59kvbt/Qr0xRenzQBWVlY4OTmzadOvDB78CpmZmYSGbqN27ftnRebPn8P169f48svFGBubPLJ9bm4ueXl5/+uOySMrKwsDAwMMDQ3p0yeQzp275a+7bt1qbt+O4eOPJxW57T+Zmpri59eR//73OyZOnMrlyxc5eHAf3367/KlyApw7d4Y7d+Lp1Klkr454QAoGhViaGzOocx2WbQ9nz6lounjXUDqSEI+YMmV6hZyHoVYtd8aPn8L06ZPR6XQYGBgwd+4Xj1y65+ZWi+XLlxITcwsbGxuCgu4PRGzZsjUvvTSEsWPfRqvVYWVlxdy5C1Gr1fj6dmTt2lW0bNnqf+u24sqVy4/tw+7X70U++WQyw4e/TF5eHi1btmbfvj1otVrUajUdOnTmvfdGMXv2PNzda+dvFxl5BQ+POgD4+nYgMvIKkyZ9zNdfL3nmyyxzc3NZuvRbpk69jlqtpmZNN+bMmU/Nmm756/zrX2Po1+/F/HEFD5w9+ydffDGPu3czsLW145VXRuRfEfDqq28wYcKHpKWl4ujoRFDQp3h6NmD48Nd55503yMrKwtu7FYsWfYeJyf0PuitXLj/SjdGwYWNGjHiToKDxpKQkY25uQbt2PiXSHTN79jwWLVrAmjU/YmCgplkzb8aM+Yjbt2PZvHkjRkZG9O3bPX/9ceMm061bT4D/jatYmv/cjh2hvPbam7zxxluYmJjktwnuf+gbGRnnd2U9aVu4f7akSZOmDB/++v8eT2TOnE8JCOhK1aqWfPzxJNzdPZ6Ys1ev3vmPQ0O34efXsUA3RklS6XRPe0KqckpMTEerLZmX6J9/eHU6HQt/OcPlqL/57I0XsLUyLZHjKK0ifsA8i8reztu3b+Do6PqYLSomuR3y/Usa7e0daN26bRmnKnnyfj7Z435/1WoV1aoVPmZEBj0qSKVS8Wp3T1DBD2ERSO0mypt//Wssb731ltIxRBkxMDDA2/sFpWOIckoKBoVVszThpY61uXA9mQNnY5WOI0QBcXG3iYmJKXpFoRd69+5TrEF/onKRgqEc8GtanXo1rPh5zxWS07KUjiOEEEI8QgqGckCtUjGilyd5eVpW7bgoXRNCCCHKHSkYygkHazP6+7rz55U7HAuPUzqOEEIIUYAUDOVIV+8auFevytqdl0m9m610HCFo2rQ5LVu2LPR5ORsmRMXzrL+3UjCUI2q1itd6enIvO5e1Oy8pHUcIPv54AtOmTXvsc2q1AXl5j046JIQo3/Lych+Z5fRpSMFQzjjbmRPQ1o3j4fGcvpSgdBwhCmVqak5aWgo6nf5f6y6EvtDptKSlJWNqWvx7dMj1M+VQz9aunLyYwKodF6lX04oqJhVval6hH95//y2MjAxZsOCbR54zN7ckOTmBuLhooOJ3TajV6gLTNusraad+KX47VRgZmWBublnsY0nBUA4ZGqh5vVd9PvvxJD/vvsLrvesrHUlUUikpKWg0jz91qVKpsLGxL+NEpaeyz9ypb6SdJU+6JMopV0cLerauycFzsZy/lqh0HCGEEJWcFAzlWJ92bjhVM+PH0Agys2RwmRBCCOVIwVCOaQwNeK1nfZJSs9iw76rScYQQQlRiMoahnKvtYklnbxd2nYympac99WpaKx1JVCKtW7elSpVnu52yEEK/yBmGCuBFXw9sLU34ITSC7Jw8peOISuTdd8cybtw4pWMIIcoBKRgqAGMjA17r6UlccibBB68pHUcIIUQlJAVDBVHfzQbfJtXZcfwm12JTlY4jKomRI4czcOBApWMIIcoBKRgqkJc61sbK3JjlIeHk5un/hCRCeVlZWdy7d0/pGEKIckAKhgrEzMSQ4d3rcSshg22HrysdRwghRCUiBUMF06S2LW28HNh+5AZR8elKxxFCCFFJlGrBkJ6ejr+/P9HR0QWWr169mmHDhuU/Dg8PJzAwkO7duzNlyhRyc+9PUhQTE8PQoUPp0aMHo0ePJiMjA4DU1FRGjRpFz549GTp0KAkJ92/SlJ2dzbhx4+jZsyf9+/fn6lX9nLtgcJe6VDExZHlIOHmVYK50IYQQyiu1guHMmTMMHjyY69evF1h+5coVlixZUmDZuHHjmDZtGjt27ECn07F+/XoAZsyYwZAhQwgLC6Nhw4YsXrwYgC+//BJvb29CQ0MZOHAgs2bNAmDVqlWYmpoSGhrK5MmTmTRpUmk1T1HmphqGdqvHjdtp7DgepXQcocc6dOhEt27dlI4hhCgHSq1gWL9+PdOnT8fe/v9vTpOdnc20adMYM2ZM/rJbt25x7949mjZtCkBgYCBhYWHk5ORw4sQJunfvXmA5wN69ewkICADA39+f/fv3k5OTw969e+nTpw8ALVu2JCkpiZiYmNJqoqK869nRoq4dwQeuEZuYoXQcoafeeOMt3n//faVjCCHKgVKb6fHBt/6HLViwgBdffBEXF5f8ZfHx8djZ2eU/trOzIy4ujuTkZMzNzTE0NCyw/J/bGBoaYm5uTlJS0mP3dfv2bapXr/7M7ahWrfj3DH8SOzuLEtvX2MHNeWfeHlbvvMzn77ZHrVaV2L6fV0m2szyTduoXaad+kXaWrDKbGvrQoUPExsYyadIkjh07lr9cq9WiUv3/B51Op0OlUuX/+7B/Pn54G7Va/cg2D5Y/j8TEdLRa3XPt44HSuA3py51qs2x7OD/vCKeLd40S3fezktvK6o9hw15GozFg+fK1SkcpdZXh/QRpp74pyXaq1aonfkkus6sktm3bxuXLl+nbty9BQUGcP3+eDz74AEdHx/xBiwB37tzB3t4eGxsb0tLSyMu7PxVyQkJCfveGvb09d+7cASA3N5eMjAysrKxwcHAgPj7+kX3ps7YNHWnobsOGfZEkpGQqHUcIIYSeKrOCYc6cOYSGhrJ582ZmzpxJw4YN+fLLL3F2dsbY2JhTp04BsHnzZnx9fdFoNHh7exMSEgJAcHAwvr6+APj5+REcHAxASEgI3t7eaDQa/Pz82Lx5MwAnT57E2Nj4ubojKgKVSsWr3T1BBT+GRaDTlczZECGEEOJh5WIehvnz5zNnzhx69OjB3bt3GT58OADTp09n/fr19OrVi5MnT/LBBx8AMHbsWP7880969+7N2rVrmTZtGgDDhg0jOzub3r17M2vWLObNm6dUk8pUNUsTXupYmwvXkzlwNlbpOEIIIfSQSidfSZ+ovI9heECr0/HvtX9wMz6dmSNbYW2h3C2Jpe9Qf8gYBv0j7dQvejmGQZQutUrFiF6e5OVpWbXjonRNiBLRo0dv+vXrp3QMIUQ5IAWDHnGwNqO/rzt/XrnDsfA4peMIPTB06HDeeOMNpWMIIcoBKRj0TFfvGrhXr8ranZdJzchWOo6o4DIzM7l7967SMYQQ5YAUDHpGrVbxWk9PMrNyWbvrktJxRAU3atQIXn75ZaVjCCHKASkY9JCznTl92rlxPDye05cSit5ACCGEKIIUDHqqZ2tXatibs2rHRTLu5SgdRwghRAUnBYOeMjRQ83qv+qTdzeHn3VeUjiOEEKKCk4JBj7k6WtCzdU0OnovlfGSi0nGEEEJUYFIw6Lk+7dxwqmbGj2ERZGblKh1HVDD9+w9g8ODBSscQQpQDUjDoOY2hAa/1rE9SahYb9l1VOo6oYAIDBzJkyBClYwghygEpGCqB2i6WdPZ2Yc/pW1y8max0HFGBJCUlkZgo3VlCCCkYKo0XfT2wtTThh9AIsnPylI4jKoixY0czYsQIpWMIIcoBKRgqCWMjA0b09CQuOZPgg9eUjiOEEKKCkYKhEmngZoNvk+rsOH6Ta7GpSscRQghRgUjBUMm81LE2VubGLN8eTk6uVuk4QgghKggpGCoZMxNDhnevx607GWw/cl3pOEIIISoIQ6UDiLLXpLYtrb0c2H7kBi3q2VPD3lzpSKKcGjz4FSwsTJSOIYQoB+QMQyU1pEtdqpgYsjwknDytdE2Ix+vVK4DAwEClYwghygEpGCopc1MNQ7vV48btNHYcj1I6jiinYmNjiI6OVjqGEKIckIKhEvOuZ0fzunYEH7hGbGKG0nFEOTR+/IeMHj1a6RhCiHJACoZKTKVS8Uq3uhhr1KwIjUCr0ykdSQghRDklBUMlZ2VuzKDOdbgS/Td7TsmpZyGEEI8nBYOgbUNHGrrbsGFfJAkpmUrHEUIIUQ5JwSBQqVS82t0TVPBjWAQ66ZoQQgjxDzIPgwCgmqUJL3XwYNVvlzhwNhbfJtWVjiTKgddeexNLS1OlYwghygE5wyDy+TVzpl4NK37ec5nktCyl44hyoFOnLvTo0UPpGEKIckAKBpFPrVIxopcneXk6Vu24KF0TgsjIq1y+fFnpGEKIckAKBlGAg7UZ/Xzc+fPKHY6FxykdRyhs+vTJfPTRR0rHEEKUA1IwiEd0a1mDWk5VWbvzMqkZ2UrHEUIIUQ5IwSAeoVareL2XJ5lZuazddUnpOEIIIcoBKRjEYznbmdOnnRvHw+M5fSlB6ThCCCEUJgWDKFTP1q7UsDdn1Y6LZNzLUTqOEEIIBUnBIAplaKDm9V71Sbubw8+7rygdRyhg9Oj3+fjjj5WOIYQoB6RgEE/k6mhBz9Y1OXgulvORiUrHEWWsbdv2dOjQQekYQohyQAoGUaQ+7dxwqmbGj2ERZGblKh1HlKHw8L84d+6c0jGEEOWAFAyiSBpDA17rWZ+k1Cw27LuqdBxRhmbP/pTJkycrHUMIUQ5IwSCeSm0XSzp7u7Dn9C0u3kxWOo4QQogyJgWDeGov+npga2nCitAIsnLylI4jhBCiDEnBIJ6asZEBI3p6Ep+cyeYD15SOI4QQogxJwSCKpYGbDb5NqrPjxE2uxaYqHUcIIUQZkYJBFNtLHWtjZW7M8u3h5ORqlY4jStGHH44jKChI6RhCiHKgVAuG9PR0/P39iY6OBuDnn3/G39+fgIAAJk2aRHb2/RsbhYeHExgYSPfu3ZkyZQq5ufcv3YuJiWHo0KH06NGD0aNHk5GRAUBqaiqjRo2iZ8+eDB06lISE+1MXZ2dnM27cOHr27En//v25elVG9JcGMxNDhnWvx607GWw/cl3pOKIUNW/uTatWrZSOIYQoB0qtYDhz5gyDBw/m+vXrAFy7do1ly5bx008/sWXLFrRaLWvXrgVg3LhxTJs2jR07dqDT6Vi/fj0AM2bMYMiQIYSFhdGwYUMWL14MwJdffom3tzehoaEMHDiQWbNmAbBq1SpMTU0JDQ1l8uTJTJo0qbSaV+k1rW1Lay8Hth+5QVR8utJxRCk5ffokx44dUzqGEKIcKLWCYf369UyfPh17e3sAjIyMmD59Oubm5qhUKurWrUtMTAy3bt3i3r17NG3aFIDAwEDCwsLIycnhxIkTdO/evcBygL179xIQEACAv78/+/fvJycnh71799KnTx8AWrZsSVJSEjExMaXVxEpvSJe6VDExZHlIOHla6ZrQRwsX/puZM2cqHUMIUQ6UWsEwa9YsvL298x87OzvTrl07AJKSklizZg2dO3cmPj4eOzu7/PXs7OyIi4sjOTkZc3NzDA0NCywHCmxjaGiIubk5SUlJj93X7du3S6uJlZ65qYah3epx43YaO45HKR1HCCFEKTIs6wPGxcUxcuRIXnzxRVq1asWpU6dQqVT5z+t0OlQqVf6/D/vn44e3UavVj2zzYPnzqFbN/Lm2/yc7O4sS3Z/Setqa8+fVRDYfvEbnVq642N9vn761szD63k6NxgDQ/3Y+IO3UL9LOklWmBcPVq1cZOXIkw4YN4/XXXwfA0dExf9AiwJ07d7C3t8fGxoa0tDTy8vIwMDAgISEhv3vD3t6eO3fu4OjoSG5uLhkZGVhZWeHg4EB8fDw1a9YssK/nkZiYjlare659PGBnZ0FCQlqJ7Ks8GejnztnLCSxYc4r6rtaYVzGma3NnpWOVOn19Px+Wk5OHRmOg9+2EyvF+grRT35RkO9Vq1RO/JJfZZZXp6em88cYbjB07Nr9YgPtdFcbGxpw6dQqAzZs34+vri0ajwdvbm5CQEACCg4Px9fUFwM/Pj+DgYABCQkLw9vZGo9Hg5+fH5s2bATh58iTGxsZUr169rJpYaVmZGzOocx2uRP/N1kPXWffbRYIPRCodSwghRAlS6XS6kvn6XIhOnTqxcuVKdu3axfz58/Hw8Cjw3NixY4mIiCAoKIj09HS8vLyYM2cORkZG3Lp1i4kTJ5KYmIiTkxNffPEFlpaWpKSkMHHiRKKiorCwsGD+/Pm4uLiQlZXFtGnTOH/+PEZGRsycORMvL6/nyi9nGJ5O8IFIthy6XmBZn3Zu9PNxVyZQGdDn9/OB8PC/sLaugqOjm9JRSl1leD9B2qlvyvIMQ6kXDBWdFAxFe1yx8IA+Fw36+n7+k7RTv0g79UtZFgxlPuhRCFFxHD58EEtLU7y8WigdRQihMCkYxHN7cAbhcWcZcvK05OZpMTSQWcgrom+//Q8ajQHLl69VOooQQmFSMIgS8c+ioVfrmqRn5hJ69CZ/XUtiVIAX1W2rKJhQCCHE85CCQZSYB0VDlYcuq2ziUY0VoRHM+OEEL3WsTafmzoXOpyGEEKL8koJBlKh+Pu4FBuE0q2uHe/WqLAsJZ83OS5y9msjrvTyxNDdWOKkQQojikI5lUeoszY35cGAThnatS8TNZKYuO84flxKK3lAIIUS5IWcYRJlQqVR0buFCfVdrlmz9i/9sPIdvk+oM6lwbEyP5MSyvZsyYjY2NjD0RQsgZBlHGqttWIWi4Nz1b1+TAmRg+WXGCqzF/Kx1LFMLd3YM6deooHUMIUQ4UWTBotVr++9//MmHCBNLT0/n+++/Jy8sri2xCTxkaqBnYoTbjhzQjL0/LnFWn2XLwmtwiuxzas2dX/m3lhRCVW5EFw7x587h06RJnz54F4MCBA8yZM6fUgwn9V6+mNTNef4EXGtgTfPAan685TXzyXaVjiYesWLGUb775RukYQohyoMiC4ciRI3z++ecYGxtjbm7O8uXLOXToUFlkE5WAmYmGUQFejOrTgJg7d5m+4gQHzsYgM5YLIUT5UmTBYGhoiFr9/6sZGRlhaCiD1ETJat3AkU9ffwE3BwtWhESweNN50jNzlI4lhBDif4osGOrWrcuaNWvIy8sjMjKSadOm4enpWRbZRCVTzdKEcYObMbCjB39eucPUZcc4fy1R6VhCCCF4ioJhypQp/PXXXyQmJjJkyBDu3r3L5MmTyyKbqITUahU9W7ky9VVvzIwN+eLnM6zddYnsHBloK4QQSiqyb2Hz5s3Mnj27LLIIka+mgwXTR7Tkl71X2XUymvDrybwZ0ICaDhZKR6tU5s1bKPMwCCGApzjDsG7durLIIcQjjDQGDO1alw9fakJ6Zg4zV54k7NhNtDIgssw4OVXHxcVF6RhCiHKgyDMMtWrVIigoCG9vb8zMzPKXd+vWrVSDCfFAI/dqzHjjBX4MjWD971c4e/UOI/0bYFPVROloei8kZCsWFib4+HRVOooQQmFFFgwpKSmkpKRw48aN/GUqlUoKBlGmqpoZ8V5gIw6cjWXdrstMW3ac4T3q8UJ9B6Wj6bV161aj0RhIwSCEKLpgWLVqFQC5ubnodDo0Gk2phxLicVQqFb5NqlOvphVLt17gu81/ceZKIkO71sXMRC71FUKI0lTkGIbExERGjhxJ06ZNady4McOHDycuLq4ssgnxWA7WZkx6pTl92rlx7EIc05cf51JUitKxhBBCrxVZMHz66ac0bdqUw4cPc/jwYby9vfnkk0/KIJoQhTNQq+nn487EV5pjoFYxd81pNuy7Sm6e3I9CCCFKQ5EFw/Xr13nvvfeoWrUq1tbWjBkzhps3b5ZFNiGKVNvZkumvtaR9Yye2H7nBrFWniE3MUDqWEELonSILhtzcXLKysvIfZ2ZmolKpSjWUEMVhamzIa73q827/RiT+fY8ZK07w++louR9FCVi06Ft++OEHpWMIIcqBIkeK9erVixEjRhAYGIhKpWLDhg107969LLIJUSwt6tnh4VyV5dvDWfXbJc5cTeS1XvWxrGKkdLQKy8bGhmrVLEhISFM6ihBCYUUWDO+++y5OTk7s378frVZLYGAgAwYMKItsQhSblbkxH7zUhD2noln/+1WmLTvGaz3r07SOrdLRKqSNG3/BwsKErl0DlI4ihFBYkV0S6enpJCQk8OWXXzJhwgQuXbpEZmZmWWQT4pmoVSq6eNdg+ghvrMyN+WrDWVaGRZCVLfejKK5Nm36V2V6FEMBTFAyTJk0iJSUFgKpVq6JSqZg6dWpp5xLiuTnbmRM03JuerWqy788YPllxnGuxqUrHEkKICumprpKYMGECABYWFkyePJnLly+XejAhSoLGUM3AjrX51+BmZOdqmb3qFFsPXSNPK5dfCiFEcTzVVRLp6en5jzMyMmT0uahw6rta8+kbL9Cinh2bDlxj7po/iE+RrjUhhHhaRQ567NevHwMHDqRHjx6oVCp27txJYGBgWWQTokRVMdHwdt+GNK19m1W/XeST5ccZ2rUubRs6yqXCQghRBJXuKU4X7N69myNHjmBoaEibNm3w8/Mri2zlQmJiOlptyZxRsbOrHJenVYR23vk7k/9uC+dSVAre9ewY3sMTc9Pi3SelIrTzeWVmZmJra05Ghv4PGK0M7ydIO/VNSbZTrVZRrZp5oc8/1R17WrVqRefOnfnrr7+4fPkyOTk5chMqUaHZWpoyfnAzwo7fZNP+SK7cOsYbvRvgVctG6WjliqmpKWZmZmRk6P8fXiHEkxU5hmHRokVMnz6dmJgY3nzzTTZu3Cj3khB6Qa1W0au1K0HDvTE1NmTBz3+ybtdlcnL1/9v001qzZiXLli1TOoYQohwosmDYt28fM2fO5LfffqN3796sXLmSiIiIssgmRJlwdbRg2oiWdGruzM6TUXz640mi4tOL3rASCAvbTnBwsNIxhBDlQJEFA9w/LXn48GFat24NQHZ2dqmGEqKsGWsMeKVbPT4Y2Ji0uzl89uMJdhy/iVauCBJCCOApCgZra2s++eQTzp8/T9u2bZk/fz729vZlkU2IMtfYw5ZP33iBRu7V+HnPFRb89CdJqfeUjiWEEIorsmCYO3cu9vb2fP/995iamqJSqZg7d25ZZBNCEVXNjHgvsBGv9qjH1Zi/mb78OCci4pWOJYQQiiryKglbW1veeeed/Mcff/xxqQYSojxQqVT4NXXGs6Y1S7Ze4Nvg85xp6MjQrnUxNTYk+EAkVaoY07W5s9JRhRCiTDzVZZVCVFYONmZMeqU5Ww9dZ9uR61yKSqFeTSsOnbsNQEZGFv183BVOWXpWrfq50lzPLoR4sqca9ChEZWZooKa/rzuThrYg415OfrEAsOXQdYIPRCqYTgghyoacYRDiKZ2/lkhm1qNzNGw5dB1AL880LFv2PVWqGDNo0AilowghFFZkwRAZGcnSpUtJSUkpcNOp7777rlSDCSGUt3fvHjQaAykYhBBFd0lMnDiRKlWq0LVrV7p3757/39NIT0/H39+f6OhoAA4fPkxAQADdunVj4cKF+euFh4cTGBhI9+7dmTJlCrm5uQDExMQwdOhQevTowejRo8nIyAAgNTWVUaNG0bNnT4YOHUpCQgJwf36IcePG0bNnT/r378/Vq1eL92oI8QT9fNzp087tkeUGahUezpZlH0gIIcpQkQVDZmYmQUFBBAYG0r9///z/inLmzBkGDx7M9evXAbh37x6TJ09m8eLFhISEcP78efbt2wfAuHHjmDZtGjt27ECn07F+/XoAZsyYwZAhQwgLC6Nhw4YsXrwYgC+//BJvb29CQ0MZOHAgs2bNAmDVqlWYmpoSGhrK5MmTmTRp0jO9KEIU5p9FQ1dvF5xtq/DlL2fYczpauWBCCFHKiiwYXF1diY8v/jXo69evZ/r06fmTPJ09exZXV1dq1KiBoaEhAQEBhIWFcevWLe7du0fTpk0BCAwMJCwsjJycHE6cOJF/NuPBcoC9e/cSEBAAgL+/P/v37ycnJ4e9e/fSp08fAFq2bElSUhIxMTHFzi7EkzwoGgZ3q8fgLnWZ+EpzGrtXY/Vvl1i363KJ3d1UCCHKkyLHMGi1Wvz9/fHy8sLY2Dh/eVFjGB58638gPj4eOzu7/Mf29vbExcU9stzOzo64uDiSk5MxNzfH0NCwwPJ/7svQ0BBzc3OSkpIeu6/bt29TvXr1oppZqCfd6vNZ2NlZlOj+yit9b+ebgU0KPJ7xdjuWbTnP1gORpGbm8PHQFpgaV/wxxRYWVQD9fz8fkHbqF2lnySryL1rXrl3p2rXrcx9Iq9WiUqnyH+t0OlQqVaHLH/z7sH8+fngbtVr9yDYPlj+PxMT0EvvGWFmuZ6+s7ezfzo2qJoas3XWJfy3ax9gBTbC2MH7CHsq/xYuXV9r3U19JO/VLSbZTrVY98UtykZ+m/fv354UXXgAgNzeX5s2bP9UYhn9ydHTMH5wIkJCQgL29/SPL79y5g729PTY2NqSlpZGXl1dgfbh/duLOnTv5mTIyMrCyssLBwaFA98mDfQlRVjq3cGHMi42JS85k5sqT3IzT/z9YQojKociC4cCBA7z44ovs2rWL3bt3M2DAAHbt2lXsAzVp0oRr165x48YN8vLy2LZtG76+vjg7O2NsbMypU6cA2Lx5M76+vmg0Gry9vQkJCQEgODgYX19fAPz8/PJvuRsSEoK3tzcajQY/Pz82b94MwMmTJzE2Nn6u7gghnkWT2rZMGtocgDlrTnPmyh2FEz27b75ZxL///W+lYwghygGVTvfk+/cOGDCAzz//nNq1awNw+fJlxo0bl/+BXZROnTqxcuVKXFxcOHLkCHPmzCErKws/Pz8mTZqESqUiIiKCoKAg0tPT8fLyYs6cORgZGXHr1i0mTpxIYmIiTk5OfPHFF1haWpKSksLEiROJiorCwsKC+fPn4+LiQlZWFtOmTeP8+fMYGRkxc+ZMvLy8nusFki6J4pN23peclsVXv57lZnwagzvXoYt3jTJMVzKGDXsZjcaA5cvXKh2l1MnPrX6RdhZfUV0SRRYMffv2zf/W/kBAQABbt24tkYDlnRQMxSft/H9Z2Xl8v+Uv/rxyh84tXBjcuQ5q9ePH4pRHUjDoH2mnfilXYxhMTEw4d+5c/uNz585hampaIuGE0HfGRga8F9iIbi1rsPtUNP/ZcJZ72blKxxJCiGIr8iqJcePG8fbbb+Pq6grAtWvXWLRoUakHE0JfqNUqBnWug4O1Kat3XuLz1acZM6AxNlVNlI4mhBBPrciCwdvbm+3bt3PmzBm0Wi1NmzbF2tq6LLIJoVc6NnehmqUp324+z8yVJxk7oAmujuX7OnErKyuMjCr+fBJCiOdX6BiGzZs307dvX1asWPHYDV977bVSDVZeyBiG4pN2PllUfDqLfj1DemYOb/dpSNM6tqWQruTI+6lfpJ36pSzHMBT61eHGjRsAXLp0qUSCCCHuq2FvTtBwb7769Sz/2XCWQZ3r0MXbpdCJyYQQojwotGAYM2YMAJ07d6ZLly4FnnvaSyqFEI9nZW7MhCHNWbrtAut2XyYu+S6Du9TB4DlnJi1pCxbMxczMiNGjP1Q6ihBCYYUWDHv27CE3N5d58+ah0+l40HORm5vLf/7zH/r161dWGYXQS8ZGBrzTvyG//n6VsOM3SUi5x9t9vcrVPSj+/PM0Go0Bo0crnUQIobRC/zKFh4dz9OhREhMTWbly5f9vYGjIiBEjyiKbEHpPrVLxUqfa2Fubsvq3S8xZfZoPBsoVFEKI8qfQguHdd9/l3Xff5euvv+a9994ry0xCVDodmjlja2XCt8Hn+WzlScYOaIybY1WlYwkhRL4iO0y3b99eFjmEqPQa1qrGpFdaYKhW8/ma0/xxKaHojYQQoowUWTDUq1ePrVu3EhMTQ0pKSv5/QoiS52JnTtDwFjjbVuHrjefYcfwmRczeXqocHBzlBm5CCOApJm7avXs3YWFhBZapVCrCw8NLLZQQlZmluTHjhzTnv9su8POeK8QlZzK0qzJXUMyfv6jSXM8uhHiyIguGh+8jIYQoG8YaA0b3a8iGvVcJPXaTOymZjO7XsFxdQSGEqFyK/Mqi1WpZunQpw4YNY/DgwXz99dfk5srNc4QobWqVioEda/Nqj3qE30hm9upT3Pk7s0wzzJo1g0mTJpXpMYUQ5VORBcOCBQs4evQor776Kq+99hp//PEHc+fOLYtsQgjAr6kzH7zUhKTULGauPMW12NQyO3ZExAXOnz9fZscTQpRfRZ7fPHDgABs2bECj0QDQoUMH+vTpU+rBhBD/z8vNhsnDWrDolzPMXXOaNwMa0KKevdKxhBCVSJFnGHQ6XX6xAGBkZFTgsRCibDjbVmHKcG9c7M1ZvOk8YceUvYJCCFG5FFkweHp6Mnv2bG7evElUVBRz5syhbt26ZZFNCPEPllWMGD+4GS087Vn/+xVW7rhIbp5W6VhCiEqgyIJh+vTppKamMmjQIAYOHEhSUhLTpk0ri2xCiMcw0hjwdl8verdxZd+fMSz65Qx375XOQGQ3t1p4eHiUyr6FEBVLkWMYjh49yueff15gWXBwsNx8SggFqVUqXvTzwN7alJVhF5m9+hQfDGiMrZVpiR7ns88+l3kYhBCA3K1SiArNp3F1bKua8M2m88xceZL3BzTGo7ql0rGEEHpI7lYpRAVX382GKcNbsHD9Geat/YM3/Rvg7VkyV1BMnToRExMNU6Z8ViL7E0JUXEXerXLNmjUMHTq0LDMJIYrJqVoVgl715j8bzrI4+DwDOnjQs1VNVCrVc+33+vVraDQGJZRSCFGRFTnosVu3buzevRuA+fPn8+qrrxIREVHqwYQQxVPV7P4VFC/Ut+fXvVf5ITRCrqAQQpSYIguGSZMmERUVxZEjR9i/fz99+/Zl5syZZZFNCFFMGkMDRvXxwr+tGwfOxrJw/Rnu3stROpYQQg8UWTCkpKQwYsQI9u/fj7+/P4GBgWRmlu189kKIp6dWqQj0deeN3vW5FJXCrFWnSEiR31khxPMpsmDIyckhJyeHAwcO0LZtWzIzM7l7925ZZBNCPId2jZz4+OWmpGZkM3PlSa7c+rvY+/D0bEDDhg1LIZ0QoqIpsmDo3Lkzbdq0wdramoYNGzJw4ED8/f3LIpsQ4jl5ulozeVgLTI0Mmbf2D46HxxVr+ylTpjNnzpxSSieEqEhUuqeYjP727ds4ODigUqmIiIjA09OzLLKVC4mJ6Wi1JTNff2WZAEfaWf6k3c3mPxvPcSX6bwJ93endxvWpr6CoSO18HtJO/SLtLD61WkW1auaFPl/oZZWbN2+mb9++rFix4pHnjhw5wmuvvVYiAYUQpc/CzIhxg5qxIiScjfsjiUu+y6s9PDE0ePJJxn/9aywmJhpmzpxfRkmFEOVVoQXDjRs3ALh06VKZhRFClB6NoZo3Axpgb23KlkPXSfz7Hu8GNqKKSeF3n42Luy3zMAghgCcUDGPGjAGQ/ksh9IhKpaKfjzv21qasCIlg1spTfDCwMfbWZkpHE0KUc4UWDMOGDXtiH+fD00ULISqWtg2dqFbVhK83nmPmylO8/2Ij6rhYKR1LCFGOFdqB+corrzB06FDs7e0xMzNj2LBhjBgxAmtra1xdXcsyoxCiFNSraU3QcG+qmBjy73V/cPTCbaUjCSHKsULPMHTv3h2AZcuW8dNPP6FW368tOnTowMsvv1w26YQQpcrBxowpw735euM5lmy5QHxyJgFt3fLPLjZt2hwzMyOFUwohyoMi52FITk4mKysr/3FGRgZ//138CWCEEOWTuamGj19uShsvB4IPXGPZ9nBycu/fg8LjhYHUbvWSwgmFEOVBoWcYHvD39+ell16ia9eu6HQ6wsLCeOkl+QMihD7RGKoZ6d8AB2szgg9e487f93CvXpWwYzcByMjIop+Pu8IphRBKKrJgGDt2LF5eXhw9ehSAiRMn4ufnV+rBhBBlS6VS0ad9LeytTVm67QKXolKI2PP1/559D0CKBiEqsSILBoAuXbrQpUuX0s4ihCgHbifd5cH8rzlZ6fnLtxy6DkjRIERlVeQYBiGEEEIIKRiEEAX083GnTzu3R5YbqFVoDNXk5OaVfSghhOIUKRg2b95M79696d27N3PnzgXg8OHDBAQE0K1bNxYuXJi/bnh4OIGBgXTv3p0pU6aQm5sLQExMDEOHDqVHjx6MHj2ajIwMAFJTUxk1ahQ9e/Zk6NChJCQklH0Dhajg/lk0dG7uTGOPamzYF8mUpcc4ERHPU9y3TgihR8q8YMjMzGTWrFmsWrWKzZs3c/LkSfbs2cPkyZNZvHgxISEhnD9/nn379gEwbtw4pk2bxo4dO9DpdKxfvx6AGTNmMGTIEMLCwmjYsCGLFy8G4Msvv8Tb25vQ0FAGDhzIrFmzyrqJQuiFfj7utGrVFh8fX4Z2q8f7LzbmX4OaYmJkwLfB5/l8zWmu305VOqYQooyUecGQl5eHVqslMzOT3NxccnNzMTc3x9XVlRo1amBoaEhAQABhYWHcunWLe/fu0bRpUwACAwMJCwsjJyeHEydO5E8u9WA5wN69ewkICADuXxK6f/9+cnJyyrqZQuiFr+d9wqrv5+Y/buBmwyevvcDwHvW4nXSXz344ybLtF0hOy3rCXoQQ+uCprpIoSebm5owdO5aePXtiampKy5YtiY+Px87OLn8de3t74uLiHlluZ2dHXFwcycnJmJubY2hoWGA5UGAbQ0NDzM3NSUpKwsHBoQxbKYT+UqtVdGjqzAueDmw/cp2dJ6M4GZFArzaudG9ZAyO5u6UQeqnMC4aIiAg2bNjA77//joWFBf/617+4fv16gRtd6XQ6VCoVWq32scsf/Puwwm6UpdPp8qe1fhbVqpk/87aPY2dnUaL7K6+knfph4MCBAPzyyy+Pff6dGtb071SXFdv+YtP+SA6ei2VE7wb4NHV+4s3ryit9fz8fkHbql7JqZ5kXDAcPHqRNmzZUq1YNuN+dsGzZMgwM/v9bSUJCAvb29jg6OhYYtHjnzh3s7e2xsbEhLS2NvLw8DAwM8teH+2cn7ty5g6OjI7m5uWRkZGBlZfXMeRMT09FqS2Zwl52dBQkJaSWyr/JM2qk/0tIy0GgMnthOQ+DN3vXxaejIT7sv8+/Vp9j0+xUGda6De/WqZRf2OVWG9xOknfqmJNupVque+CW5zMcweHp6cvjwYe7evYtOp2PPnj00adKEa9eucePGDfLy8ti2bRu+vr44OztjbGzMqVOngPtXV/j6+qLRaPD29iYkJASA4OBgfH19AfDz8yM4OBiAkJAQvL290Wg0Zd1MISodT1drpo1oyYiensSnZDJz5UmWbpXxDULoizI/w9C+fXsuXLhAYGAgGo2GRo0a8f7779OuXTvef/99srKy8PPzo0ePHgDMnz+foKAg0tPT8fLyYvjw4QBMnz6diRMn8u233+Lk5MQXX3wB3J/KeuLEifTu3RsLCwvmz59f1k0UotJSq1X4NqlOS097th+5wW8nbnLqUjy9WrnSvVVNjGV8gxAVlkonF1M/kXRJFJ+0U38MG/YyGo0By5evfabtE1Iy+eX3K5y8mIC1hTEDO3jQqoFDuRzfUBneT5B26puy7JIo8zMMQoiKo0OHTlSpYvzM29tZmfJO/0ZcvJnMT7uvsGTrBXafimZQ5zp4OFuWYFIhRGmTMwxFkDMMxSft1C8l1U6tTsfhc7fZsO8qf2dk09rLgQF+HthUNSmBlM9P3k/9Iu0sPjnDIIQoF9QqFe0bO9Ginh2hx24QdiyK0xcT6NGqJj1buWJsJOMbhCjP5OZTQohCDRv2cv7MqSXF1NiQQF8PZr/ZiqZ1bNly6DqTlx7l8PlYtHLCU4hySwoGIYQibK1MebtvQya90hzLKkb8d1s4s1ae5Er030pHE0I8hhQMQghF1XGxIuhVb97oXZ/ktCxmrz7Fd5vPc+fvTKWjCSEeImMYhBCKU6tUtGv0v/ENR28Sdvwmf1y+Q/cXatKrdU1MjORPlRBKk99CIUS5YWJkSH9fd3ybVGfDvqtsO3ydA2djGODnQZuGjqjL4fwNQlQWUjAIIQrVo0dvLCzK/rLHapYmjOrjRacWLqzbdZll28PZdSqawZ3rULeGVZnnEUJIwSCEeIKhQ4crej17bWdLpgxvwbELcfy69yqfrzlNS097BnbwwNbKVJFMQlRWUjAIIQqVmZnJ3bvKzo+gVqlo4+VI8zp2hB2/SejRG/8b31CDXq1dMTWWP2NClAX5TRNCFGrUqBHPdS+JkmRsZEDf9rXwaezEhn1X2X7kBgfPxhLo6067Rk6o1TK+QYjSJJdVCiEqFJuqJrwZ4MWU4S2wtTRhRWgEn/54gos3k5WOJoRek4JBCFEheVS3ZPKwFrzVx4v0zBzmrv2DbzadIz5F5m8QojRIl4QQosJSqVS0auBA0zq2/Hb8JtuP3uDMlTt0bVkD/zZuMr5BiBIkv01CiArPWGNAQLtatG9cnY37rhJ69CaHzsYS6OdBexnfIESJkIJBCFGo/v0HKDIPw7OytjDmDf8G9+dv2H2ZH0Ij2H0qmkGd61Df1VrpeEJUaFIwCCEKFRg4UNF5GJ5VLaeqTBranBMR8fzy+1X+ve4Pmte146WOHthbmykdT4gKSQoGIUShkpKSUKuzASOloxSbSqXihfoONK1ty28noth+5AZTrtyhq3cN/Nu6YWYif/6EKA75jRFCFGrs2NHlZh6GZ2WkMcC/rRvtGzuxcX8kO47f5ND5WPr7uOPTxAkDtZrgA5FUqWJM1+bOSscVotySgkEIUSlYmRvzeq/6dG7uwrpdl1i54yJ7TkdTw96cI3/FAZCRkUU/H3eFkwpRPknBIISoVFwdLZgwtDmnLiawIjSc6ISM/Oe2HLoOIEWDEI8hEzcJISodlUpFdEI6mVl5jzy35dB1gg9EKpBKiPJNCgYhhPiH3/+4xYGzMeTkPlpQCFFZSZeEEKJQgwe/UqHmYSiOB90OD7ohHmhSuxp3/r7HipAIft17lY7NnOnYzBlLc2MFUgpRfkjBIIQoVK9eARVyHoan9c+ioU87N/r5uKPT6Yi4kczOk9FsPXSd7Udu0KqBA129a+DqaKFgYiGUIwWDEKJQsbExZGVVwdjYUukopeZB0fDwZZUqlYr6bjbUd7MhLukuu05Fc/BsLIfP36auiyVdW9agWR07mXJaVCoqnU6nUzpEeZaYmI5WWzIvkT5/U3uYtFN/DBv2coWfh+FpFfV+3r2Xw4Gzsew6GU1i6j1sLU3o3MIFn8bVK9QkUJXh5xaknc9CrVZRrZp5oc9XnJ9yIYRQkJmJhu4v1KSLtwt/Xr7DzhNR/LznCsEHr9G+kRNdWrjgYCPTTgv9JQWDEEIUg4FaTYt69rSoZ8+N22nsPBnF3j9usedUNE1q29LF24X6rtaoVNJdIfSLFAxCCPGMXB0tGOnfgIEdPPj9j1v8/sct/vzpDs52VejqXYPWDRww0hgoHVOIEiEFgxBCPCdLc2P6+bjTu40rRy/EsfNEND+E3r8ss8P/Lsu0tpDLMkXFJgWDEKJQr732JpaWpkrHqDA0hgb4NK5O+0ZORNxMYeeJKLYfvk7o0Ru8UN+eri1r4OZYVemYQjwTKRiEEIXq1KlLpRltXpJUKhX1Xa2p72pNfPL9yzIPnI3lyF9x1HGxpKt3DZrVtcVALZPtiopDCgYhRKEiI6+SklIFKytHpaNUWPbWZgzpUpd+7d05eC6WXSejWBx8nmpVjencoga+TZwwM9EoHVOIIknBIIQo1PTpkyvNPAylzczEkG4ta9ClhQt/Xrl/Web636+w+eA12jVypIt3DRzlskxRjknBIIQQZUitVtG8rh3N69pxM+7+ZZn7z8Sw5/QtGntUo6t3DRq4yWWZovyRgkEIIRRS08GCN3o3YECH2uz94xa/n45mwc9/4mxbhS7eLrTxcpTLMkW5IQWDEEIozLKKEX3b16JXa1eOh8ex80QUP4ZdZMO+SPyaVqdTcxe5LFMoTgoGIYQoJzSGato1cqJtQ0cuRaXw24koQo7cIOzYTVp62tPFuwbu1eWyTKEMKRiEEIUaPfp9mYdBASqVino1ralX05r4lEz2nIpm/5kYjl6Iw8O5Kl29a9Cinp1clinKlBQMQohCtW3bXuZhUJi9lSmDOtehb/taHDwXy+6T0Xy3+S9sqhrTubkLPk2qY24ql2WK0qdIebpnzx4CAwPp2bMnM2fOBODw4cMEBATQrVs3Fi5cmL9ueHg4gYGBdO/enSlTppCbmwtATEwMQ4cOpUePHowePZqMjAwAUlNTGTVqFD179mTo0KEkJCSUfQOF0BPh4X9x7tw5pWMIwNTYkK7eNZg9qjXvv9gIeytTftl7lX8tPsSqHReJTcxQOqLQc2VeMERFRTF9+nQWL17Mli1buHDhAvv27WPy5MksXryYkJAQzp8/z759+wAYN24c06ZNY8eOHeh0OtavXw/AjBkzGDJkCGFhYTRs2JDFixcD8OWXX+Lt7U1oaCgDBw5k1qxZZd1EIfTG7NmfMnnyZKVjiIeo1Sqa1bFj/JDmzHj9BV6o78CBs7FMWXqMhevPcP5aIjqdTumYQg+VecGwc+dOevXqhaOjIxqNhoULF2Jqaoqrqys1atTA0NCQgIAAwsLCuHXrFvfu3aNp06YABAYGEhYWRk5ODidOnKB79+4FlgPs3buXgIAAAPz9/dm/fz85OTll3UwhhCh1NezNeb1Xfea/05Z+PrW4EZfGFz+fIei/x9j7xy2ycvIKrB98IJK1OyIUSisqujIfw3Djxg00Gg1vv/02sbGxdOjQgTp16mBnZ5e/jr29PXFxccTHxxdYbmdnR1xcHMnJyZibm2NoaFhgOVBgG0NDQ8zNzUlKSsLBweGZ8larZv6sTX0sOzuLEt1feSXt1A+a/80BoO/tfKCittPODjzcqjHc34sDf8awef9VVu64yKYDkXRv7UbvdrX47dgNthy6nr/NkO6eygUuIxX1/SyusmpnmRcMeXl5nDx5klWrVmFmZsbo0aMxMTEpMKuZTqdDpVKh1Wofu/zBvw8rbFY0nU6H+jlGEicmpqPVlszpvcoyeEzaqT9ycvLQaAz0vp2gP+9nI1crGr7SnMvRf7PzRBQbfr/Mr3suF1hn3W8XycjIop+Pu0IpS5++vJ9FKcl2qtWqJ35JLvMuCVtbW9q0aYONjQ0mJiZ06dKFw4cPFxicmJCQgL29PY6OjgWW37lzB3t7e2xsbEhLSyMvL6/A+nD/7MSdO3cAyM3NJSMjAysrq7JroBBCKEylUlG3hhXvBjaic3OXx66z5dB1gg9ElnEyUZGVecHQsWNHDh48SGpqKnl5eRw4cIAePXpw7do1bty4QV5eHtu2bcPX1xdnZ2eMjY05deoUAJs3b8bX1xeNRoO3tzchISEABAcH4+vrC4Cfnx/BwcEAhISE4O3tjUYjlxwJ8Sw+/HAcQUFBSscQz8HMpPATyScvxnM+MrHEzqIK/abSKTCc9tdff+WHH34gJyeHdu3aERQUxLFjx5gzZw5ZWVn4+fkxadIkVCoVERERBAUFkZ6ejpeXF3PmzMHIyIhbt24xceJEEhMTcXJy4osvvsDS0pKUlBQmTpxIVFQUFhYWzJ8/HxeXx1fYT0O6JIpP2qlfpJ0VX/CByALjFwBqOVoQn5JJxr1crC2MadvQkXaNnPTmjpn6/H4+rCy7JBQpGCoSKRiKT9qpP06fPomVlRnu7g2UjlLq9P39fLho6NPOjX4+7uTkajlz5Q4Hz8VyLjIRnQ5qO1vSvrETLT3tMTWuuHP76fv7+UBZFgwV96dBCFHqFi78NxqNAcuXr1U6inhODwY4VqliTNfmzsD9e1d4e9rj7WlPSnoWR87f5uC5WH4IjWDtzku0qGdHu0ZOeLpao5bbbVd6UjAIIUQl0c/HvdBvpFbmxvRs7UqPVjWJjE3l0LnbHLsQx5G/4qhW1Zi2DZ1o19gJeyu5t0hlJQWDEEKIfCqVCo/qlnhUt2RQp9r8cfl+l8W2w9fZevg69WpY0a6RE96edpgYyUdIZSLvthBCiMcy0hjQqoEDrRo4kJR6j8Pnb3PoXCzLQ8JZs/MS3p52tG/kRN0aVoXOhSP0hxQMQgghimRT1QT/tm70buPKlVt/c/BsLMcj4jl07jZ2Via0a+RE24aO2Mrt0PWWFAxCiEJNnjwNa+sqSscQ5YhKpaKOixV1XKwY0qUupy7dLxqCD1xj84FreLpa076xE83r2mH8v6nFhX6QgkEIUaj69b0qzeVpoviMjQxo29CJtg2duJOSyeH/XWWxdOsFTI0NaOnpQPtGTng4V5UuCz0gBYMQolCHDx/E0tIUL68WSkcR5ZytlSl92tfCv50bl6NSOHg2lqMXbrP/TAwONma0b+RI24ZOWFsYKx1VPCMpGIQQhfr22//IPAyiWNQqFfVqWlOvpjVDutbl5MV4Dp2NZcO+SDbuj8TLzYb2jZ1oVscWjaF0WVQkUjAIIYQoFabGhvg0ro5P4+rEJd/l0LnbHD4fy3eb/8LM2JBWDRxo39gJN0cL6bKoAKRgEEIIUeocrM0I9HWnn08twm8kc+hsLAfPxfL7H7dwtq1Cu0ZOtPFywNJcuizKKykYhBBClBm1SoWXmw1ebjbcvZfL8Yg4Dp2LZf3vV/h171UaudvQrpETTevYYmhQ5jdUFk8gBYMQQghFmJkY0qGpMx2aOhObmJHfZXHmaiLmphpaN3CgXSMnXB0tlI4qkIJBCPEEM2bMxsZG5mEQpc+pWhUGdPAg0Nedv64ncfBsLHv/vMWuU9HUsDenXSMnWns5UNXMSOmolZYUDEKIQrm7e8g8DKJMqdUqGrlXo5F7NdIzczgeHsfBs7H8tPsyv/x+hcYe1Wjf2IlG7tWky6KMScEghCjUnj27sLQ0pUWLdkpHEZWQuamGTs1d6NTcheiEdA6di+XI+dv8cfkOVc00tPZypH1jJ1zszAtsF3wgssBtvEXJkIJBCFGoFSuW/m8eBikYhLJc7Mx5uVMdXvTz4HxkEofOxbL7VDS/nYjC1dGC9o2caNXAgV0no9hy6DoAGRlZ9PNxVza4HpGCQQghRIVhaKCmaR1bmtaxJe1uNkcvxHHobCxrdl5i7a5L6HT/v+6DwkGKhpIhHUBCCCEqJAszI7p61+CT11/Ar2n1AsXCA1sOXee7zedJvZtd9gH1jJxhEEIIUeFZVin86onj4fEcD4/H1tIE9+pVca9uiXv1qrg6mMv01MUgBYMQQogK70G3w4NuiAd6t3alobsNkbGpRMakcjn6b46HxwNgoFZRw978f0XE/ULCwdpUpqkuhBQMQohCzZu3UOZhEBXGP4uGPu3c8pfVq2mdv15yWhaRMalExv7NtZhUDp27zZ7TtwCoYmJILaeqBc5EmJtqyrYh5ZQUDEKIQjk5VZd5GESF8qBAeNJlldYWxrSoZ0eLenYAaLU6Yu5kEBmbytVbfxMZm8rWQ9d5MCTC3tr0fgHhdL+IqGFvjsaw8g0BlIJBCFGokJCtWFiY4OPTVekoQjy1fj7uxSp01WoVLvbmuNib49ukOgCZWblcv51GZMzfRMakEn4jmaN/xQFgaKCipoPF/wqI+//ZWel/V4YUDEKIQq1btxqNxkAKBlHpmBobUt/Vmvqu97sydDrd/3dlxKQSGfM3+8/EsOtUNHB/kqn8sxDO9/81M9GvrgwpGIQQQogiqFQqbKqaYFPVBG9PewDytFpuJWT8fxERm8q5q4n5XRmONmYPDaisioudeYWezloKBiGEEOIZGKjV1HSwoKaDBR2a3R8vcfdeLtdvp3I1JpVrMamcj0zk8PnbAGgM1bg6WPx/EeFUlWqWJhWmK0MKBiGEEKKEmJkY0sDNhgZuNsD9rozEv+/lX9YZGZPK73/c4rcTUQBUrWJUYCxELaeqmBo/3UdzWd8zQwoGIYQQopSoVCpsrUyxtTLlhfoOAOTmaYlOSH9oPEQqf165c399wMm2SoEiwtmuCgbqgl0ZwQciy/yeGVIwCCEKtWjRt9jamqPVKp1ECP1haKDGzbEqbo5V6dT8/rKMezlciy1YQBw8FwuAkUaNm4NF/rwQl6NT2HkyOn9/ZXXPDCkYhBCFsrGxoVo1mYdBiNJWxURDw1rVaFirGnC/KyPh73v3L+u8dX9A5a5TUeQef8wNMyibokEKBiFEoTZu/AULCxO6dg1QOooQlYpKpcLeyhR7K1NaN3AEICdXy+rfLnLgbKwimaRgEEIUatOmX9FoDKRgEKIc0Biqea1XfawtjB+5Z8bD02CXlop7QagQQghRCfXzcadPO7f8x2VRLICcYRBCCCEqnKe5Z0ZJk4JBCCGEqICKe8+M5yVdEkIIIYQokpxhEEIUasmSH7C1NScjI0/pKEIIhckZBiFEoUxNTTEzM1M6hhCiHJAzDEKIQq1ZsxILCxP69HlJ6ShCCIXJGQYhRKHCwrYTHBysdAwhRDkgBYMQQgghiiQFgxBCCCGKJAWDEEIIIYokgx6LoFaryvX+yitpp35wcLDH0NBA79v5gLRTv0g7S3Y/Kp1O9/h7ZQohhBBC/I90SQghhBCiSFIwCCGEEKJIUjAIIYQQokhSMAghhBCiSFIwCCGEEKJIUjAIIYQQokhSMAghhBCiSFIwCCGEEKJIUjAIIYQQokhSMAghhBCiSFIwCCGEEKJIUjAIIYQQokhyt8oy8PXXXxMaGgqAn58f48ePVzhR6Vm0aBE7duxApVIxYMAAXnvtNaUjlZq5c+eSnJzM559/rnSUUjNs2DCSkpIwNLz/p+LTTz+lSZMmCqcqWXv27OHrr78mMzOTdu3aERQUpHSkUvHLL7+wevXq/MfR0dH07duXadOmKZiqdGzevJklS5YA4Ovry4QJExROVPKWLFnChg0bMDIyolevXowePbr0D6oTperQoUO6l19+WZeVlaXLzs7WDR8+XPfbb78pHatUHDt2TDdo0CBdTk6OLjMzU9exY0fd1atXlY5VKg4fPqxr1aqVbsKECUpHKTVarVbXvn17XU5OjtJRSs3Nmzd17du318XGxuqys7N1gwcP1u3du1fpWKXu0qVLuq5du+oSExOVjlLi7t69q2vZsqUuMTFRl5OToxswYIDu0KFDSscqUYcOHdL5+/vr0tLSdLm5ubq33npLt2PHjlI/rnRJlDI7OzsmTpyIkZERGo0GDw8PYmJilI5VKl544QVWrlyJoaEhiYmJ5OXlYWZmpnSsEpeSksLChQt5++23lY5SqiIjIwF4/fXX6dOnT4Fvp/pi586d9OrVC0dHRzQaDQsXLtS7MyiP88knn/Dhhx9iY2OjdJQSl5eXh1arJTMzk9zcXHJzczE2NlY6Vom6cOEC7du3x9zcHAMDA3x8fNi1a1epH1cKhlJWp04dmjZtCsD169cJDQ3Fz89P2VClSKPR8NVXX9G7d2/atGmDg4OD0pFK3LRp0/jwww+pWrWq0lFKVWpqKm3atOGbb77hhx9+4KeffuLQoUNKxypRN27cIC8vj7fffpu+ffuydu1aLC0tlY5Vqg4fPsy9e/fo2bOn0lFKhbm5OWPHjqVnz574+fnh7OxM8+bNlY5Vory8vDh48CApKSlkZWWxZ88e7ty5U+rHlYKhjFy+fJnXX3+d8ePH4+bmpnScUjVmzBiOHDlCbGws69evVzpOifrll19wcnKiTZs2Skcpdc2aNWPevHlYWFhgY2PDgAED2Ldvn9KxSlReXh5Hjhxh9uzZ/Pzzz5w9e5ZNmzYpHatU/fTTT3o9tigiIoINGzbw+++/c+DAAdRqNcuWLVM6Volq06YNgYGBDBs2jJEjR9KiRQs0Gk2pH1cKhjJw6tQpRowYwccff0z//v2VjlNqrl69Snh4OACmpqZ069aNixcvKpyqZIWEhHDo0CH69u3LV199xZ49e5g9e7bSsUrFyZMnOXLkSP5jnU6XP/hRX9ja2tKmTRtsbGwwMTGhS5cunD17VulYpSY7O5sTJ07QqVMnpaOUmoMHD9KmTRuqVauGkZERgYGBHD9+XOlYJSo9PZ1u3bqxdetWVq1ahZGRETVq1Cj140rBUMpiY2N59913mT9/Pr1791Y6TqmKjo4mKCiI7OxssrOz2b17Ny1atFA6VolasWIF27ZtY/PmzYwZM4ZOnToxefJkpWOVirS0NObNm0dWVhbp6els2rSJrl27Kh2rRHXs2JGDBw+SmppKXl4eBw4cwMvLS+lYpebixYu4ubnp5diiBzw9PTl8+DB3795Fp9OxZ88eGjVqpHSsEhUdHc0777xDbm4uaWlp/Prrr2XSxaRfXxfKoWXLlpGVlVXg0rtBgwYxePBgBVOVDj8/P86ePUu/fv0wMDCgW7duel8k6bOOHTty5swZ+vXrh1arZciQITRr1kzpWCWqSZMmjBw5kiFDhpCTk0O7du148cUXlY5VaqKionB0dFQ6Rqlq3749Fy5cIDAwEI1GQ6NGjRg1apTSsUqUp6cn3bp1o0+fPuTl5TFixIgy+XKm0ul0ulI/ihBCCCEqNOmSEEIIIUSRpGAQQgghRJGkYBBCCCFEkaRgEEIIIUSRpGAQQgghRJGkYBBClBvHjh3D399f6RhCiMeQgkEIIYQQRZKCQQhRLp08eZIOHTpw+vRppaMIIZCZHoUQ5dDRo0eZOnUq3333HZ6enkrHEUIgZxiEEOXM7du3efvtt+nSpYsUC0KUI1IwCCHKFQMDA5YvX86mTZs4c+aM0nGEEP8jBYMQolyxs7OjefPmTJgwgfHjx5OZmal0JCEEUjAIIcqp/v37U6tWrQJ3ehVCKEfuVimEEEKIIskZBiGEEEIUSQoGIYQQQhRJCgYhhBBCFEkKBiGEEEIUSQoGIYQQQhRJCgYhhBBCFEkKBiGEEEIUSQoGIYQQQhTp/wAMW2u5iu8c+gAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Axes: title={'center': 'Distortion Score Elbow for KMeans Clustering'}, xlabel='k', ylabel='distortion score'>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_kmn = KMeans(random_state=42)\n",
    "\n",
    "visualizer = KElbowVisualizer(\n",
    "    model_kmn,\n",
    "    k=(2, 10),\n",
    "    timings=False)\n",
    "\n",
    "\n",
    "with warnings.catch_warnings():\n",
    "    warnings.simplefilter('ignore')\n",
    "\n",
    "    visualizer.fit(X)\n",
    "\n",
    "visualizer.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ієрархія кластерів на основі AgglomerativeClustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_dendrogram(model, **kwargs):\n",
    "\n",
    "    counts = np.zeros(model.children_.shape[0])\n",
    "    n_samples = len(model.labels_)\n",
    "    for i, merge in enumerate(model.children_):\n",
    "        current_count = 0\n",
    "        for child_idx in merge:\n",
    "            if child_idx < n_samples:\n",
    "                current_count += 1  # leaf node\n",
    "            else:\n",
    "                current_count += counts[child_idx - n_samples]\n",
    "        counts[i] = current_count\n",
    "\n",
    "    linkage_matrix = np.column_stack(\n",
    "        [model.children_, model.distances_, counts]\n",
    "    ).astype(float)\n",
    "\n",
    "    dendrogram(linkage_matrix, **kwargs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_agg = AgglomerativeClustering(\n",
    "    distance_threshold=0,\n",
    "    n_clusters=None)\n",
    "\n",
    "model_agg = model_agg.fit(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Number of points in node (or index of point if no parenthesis)')"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAewAAAFsCAYAAAAOmoq4AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8fJSN1AAAACXBIWXMAAAsTAAALEwEAmpwYAAA6UUlEQVR4nO3deVhUZf8/8PcwLGZgKoK4oD93zEpMexQ3XBEVQtESF9yXSq3cRVHKxNAo1/TJpU0tNRUxd820EMVSMXEpN8QVEBUBA2a5f3/45TwgArMzB96v6+q6nJlz7s99nzOn99xnDnMUQggBIiIismo2pd0BIiIiKhkDm4iISAYY2ERERDLAwCYiIpIBBjYREZEMMLCJiIhkgIFNVq1JkyZ48OBBgee2b9+OcePGAQCWLl2KHTt2WLxfM2fOxLp16577WkBAAB4/fmxQu8HBwdi3b99zX7t69SomTpwIf39/vPnmmxgyZAj+/PNPAMCtW7fQokULg2oCwJEjR7B06VK91zPl9o+Li8Nrr72GgIAABAQEwN/fH0OHDkVsbKxJ2s+zbt06zJw506RtElmCbWl3gMgYH3zwQWl3oZDo6GiTt3nt2jUMGzYMn376KTp06AAAOH78ON555x38+OOPeOGFF4xq/9y5c0hPT9d7PVNv/zp16hTYfpcuXcKoUaOwcuVKNG/e3KS1iOSGgU2yNnPmTDRq1AijRo3C1atXER4ejkePHkGj0SA4OBj9+/dHXFwcwsPDUbFiRWRlZWHbtm1YtGgRzp49i6ysLAghMH/+fLRs2RIzZ87Eo0ePcPPmTXTq1Anvvfce5s+fj9OnT0OpVKJbt26YNGkSAODMmTMICgrC/fv30ahRI3z++eeoWLEimjRpguPHj6Nq1ar46quvEBUVBVtbW9StWxcRERFQKpX46KOPcOPGDTx69AgvvvgiIiMjUb9+/SLHuWbNGvTr108KawDw8vLC559/jgoVKhRYdvny5Xj48CHmzp1b6PGBAwewatUqKBQKKJVKTJ8+Hfb29ti0aRM0Gg2cnJwwadIk/PTTT/jxxx+h1WpRuXJlzJkzBw0aNCi0fdLS0qTt/+qrr2Ls2LE4duwYUlJSMHr0aAwaNAgajQaLFi3C4cOH4eTkhNdeew1Xr17F+vXrS9y/Hh4eCA4OxrfffovFixcjIyMD4eHh+Oeff6BSqeDl5YXp06fD1ta2yPoqlQrz589HbGwsnJ2d4ezsDCcnJwBPz2i89NJLuHbtGgYOHIju3bvjo48+wu3btyGEQJ8+fTB69GgAT8/srF69GhUqVECbNm3w/fff48KFC1i+fDni4+ORkpKCJk2aYObMmZg7dy7S0tKQmpqKWrVqYcmSJXB2dkaXLl3g5+eHEydOID09HaNHj8bp06dx/vx52NraYtWqVahevbp+BwGVGwxssnrDhg2Djc3/vr1JT09HkyZNCiyjVqvx/vvvY9GiRWjWrBkyMjIwYMAANGzYEABw+fJlHDp0CLVq1cKZM2eQkpKCzZs3w8bGBqtXr8aaNWvQsmVLAEB2djZ2794NAPj000+Rk5ODPXv2QKPRYOTIkTh58iQAIDk5Gd9//z3s7e3x1ltv4cCBA+jTp4/Up19++QXbt2/Hli1b8NJLL+HTTz/Fhg0bUK9ePVSqVAmbN28GAMydOxcbN27EnDlzitwGCQkJmDp1aqHnvb29ATw9Ja6LRYsWITIyEp6enoiJiUFcXBwmTJiAoKAgPHz4EJMmTcLJkyexY8cObNy4ES+88AJiYmIwYcIE7N27t9D2yX9qOTc3F1WqVMGmTZuQkJCAgQMHol+/foiKisL58+exa9cuKBQKvPvuuzr1NY+Hhwd+/vlnAMCCBQvQrFkzREREQKPRYObMmfjmm28wZsyYIutv2rQJiYmJ2L17N9RqNYYMGSIFNgBUqlQJe/bsAQAMGTIEXbt2xYgRI5CRkYHBgwejRo0aaNKkCSIjI7F9+3a4ublhxYoV0Gg0Uhu3b9/Grl27YGtri++++w6enp4YO3YshBAYO3YsoqOjMXLkSABATk4OtmzZgj179mDKlCmIioqCh4cHxo8fj6ioKLzzzjt6bR8qPxjYZPW+++47VK1aVXq8fft27N+/v8AyiYmJSEpKwqxZs6TnsrOzceHCBTRo0AA1atRArVq1AAAtWrTASy+9hE2bNuHmzZuIi4vDiy++KK2XF9wAEBsbi5CQECiVSiiVSmzYsAEAEBUVhW7dukmnohs1alTou/bjx4/D19cXL730EgAgJCREes3d3R3r16/HjRs3cPLkyRK/f1YoFNBqtSVvrBL07t0bEyZMgLe3N9q1a4cxY8YUWubIkSO4ceMGgoKCpOceP36MR48eASi4fZ7VtWtXAECzZs2Qm5uLJ0+e4OjRowgICICDgwMAYMCAATrNrvMoFArpLMKRI0dw7tw5bN26FcDTfVxS/ePHj8PPzw/29vawt7eHv78//v77b2mdVq1aAQCePHmC06dP4+uvvwYAODk5ITAwEL/99htSU1PRrl07uLm5AXga7MuXL5fa8PT0hK3t0/+dDhs2DH/++Se++eYbJCYm4vLlywVO5/v4+AB4+h6oVq0aPDw8ADz9OsCQryWo/GBgU5mQdzo3//ef9+/fh5OTE+Lj41GxYkXp+SNHjiA8PBwjRoxA165dUb9+fezcuVN6Pf+ytra2UCgU0uO7d+9K4ZH3P2jgaag8+7P8SqWywLqPHz/G48eP8dtvv2HLli0YPHgw/P39Ubly5RJnyJ6enoiPj0fnzp0LPL9ixQrUqVMHr7/+epF9UalU0r8nTZqEfv364dixY9i+fTu+/vprKfzyaLVaBAQEYNq0adLjlJQU6YNH/u3zrLxQzhu3EKLAdgJQ4GyJLs6dO4fGjRtLfVm6dCkaNGgA4Ok2zb+Nn1f/WUqlssDjvPFotdpCy2u1WqjVaiiVygKvFdUGAHz22Wf466+/0K9fP7Ru3RpqtbrAuvb29tK/7ezsSho+kYRXiVOZUK9ePVSoUEEK7Lt378LPzw8JCQmFlj127Bg6d+6MQYMG4ZVXXsGhQ4cKnN7Mz8vLC1FRUdBqtcjNzcX777+PP/74Q6c+tW3bFgcPHkRmZiaAp98lf/vtt4iJiUHfvn3x1ltvoV69ejh8+HCR9fOMGjUKP/30E2JiYqTnfvvtN6xfv16aoeWpUqUKzp8/DyEEMjMz8euvvwJ4+rVBly5d8O+//2LgwIEICwvD33//jdzcXCiVSqjVagBA+/btsXv3bqSkpAAAfvzxRwwbNkynMT+Pt7c3du7cidzcXKjVakRFRem87l9//VWgfvv27fHtt99CCIHc3Fy8++670lmPonTo0AE7duxATk6O9PXG8zg6OqJ58+bYuHEjACAjIwM7duxA27Zt0b59exw/fhzJyckAgJ9++qnIejExMRg2bBj69OkDZ2dnxMbGlrh/iXTBGTaVCfb29li5ciXCw8Oxdu1aqNVqfPDBB2jZsiXi4uIKLBsUFIQpU6bA398farUa7dq1w4EDB557ynnChAkIDw9HQEAANBoNevXqBR8fHxw+fLjEPnl7e+PKlSsYOHAgAKBhw4b45JNPcOnSJcydO1ea2Xp6euKff/4ptq26deviv//9L5YsWYKFCxdCq9WiatWqWLVqFRo3blxghv7mm2/i999/h4+PD6pXr47//Oc/0kx31qxZmDp1qnTmYMGCBbC3t0ebNm0wdepUfPLJJ5gzZw7GjBmDkSNHQqFQwNHREStWrCgwk9VHYGAgrl+/jj59+qBixYqoXbt2kVe1JyUlISAgAMDTmbijoyMiIyOlDyWzZ89GeHg4/P39oVKp0LZtW+misKIEBQUhKSkJfn5+qFy5MurWrVvkspGRkZg3bx62b9+O3Nxc+Pv7IzAwEAqFAiEhIRg1ahTs7e3RtGnTIscwfvx4LFq0CEuXLoWdnR1ef/11JCUl6bKpiIql4O01icicYmJikJaWJgXx/Pnz4eDgIJ1yl4ObN28iOjoa7733HmxsbHDgwAGsWbOm2Jk2kalxhk1EZtWoUSOsW7cOa9euhVarhYeHBz766KPS7pZe3NzckJKSAn9/fyiVSjg5OWHBggWl3S0qZzjDJiIikgFedEZERCQDDGwiIiIZYGATERHJAAObiIhIBqz+KvGHD7Og1ep+XZyzsyPS0jLN2CPWkUutslbHkrVYx/prsY7119K3jo2NAlWqvFjk61Yf2Fqt0Cuw89axBNax/lplrY4la7GO9ddiHeuvZco6PCVOREQkAwxsIiIiGWBgExERyQADm4iISAYY2ERERDLAwCYiIpIBBjYREZEMMLCJiIhkgIFNREQkAwxsIiIiGbD6nyal8uVI/G3EnU82SVt29kqocjUmacsa6liyFutYfy1D6rRuVh2dPGuZqUdkbpxhk1WJO5+MpBTL3ACAqDxJSsk02YdhKh2cYZPVqePqiBmDXze6HRcXJ6SmZpigR9ZRx5K1WMf6a+lbZ+HG02bsDVkCZ9hEREQywMAmIiKSAQY2ERGRDDCwiYiIZICBTUREJAMMbCIiIhlgYBMREckAA5uIiEgGGNhEREQywMAmIiKSAQY2ERGRDDCwiYiIZICBTUREJAMMbCIiIhlgYBMREckAA5uIiEgGGNhEREQywMAmIiKSAVtdFlqxYgX27t0LAPD29sb06dMREhKCU6dO4YUXXgAATJgwAd27d8fFixcxe/ZsZGVloVWrVvj4449ha6tTGSIiIipCiUkaGxuLmJgYREVFQaFQYPTo0Th48CASEhKwYcMGuLq6Flh+2rRpmD9/Pjw9PTFr1ixs2bIFgwYNMtsAiIiIyoMST4m7uLhg5syZsLe3h52dHRo0aIA7d+7gzp07mDVrFvz9/bFs2TJotVrcvn0b2dnZ8PT0BAAEBgZi37595h4DERFRmVfiDLtRo0bSvxMTE7F3715s3LgRJ0+eRFhYGJycnDBu3Dhs3boVjRo1gouLi7S8i4sLkpOTjeqgs7Oj3uu4uDgZVZN1Sq+Wnb3SpH0pT9uOdayjjiVr6VPHmGPLGscjl1qmrKPzl8uXL1/GuHHjMH36dNSvXx9ffvml9FpwcDB27NiBBg0aQKFQSM8LIQo8NkRaWia0WqHz8i4uTkhNzTCqJuuUXi1VrgYATNKX8rbtWKf061iylr51DD22rHU8cqilbx0bG0Wxk1SdrhI/deoUhg8fjilTpqBv3774+++/sX//ful1IQRsbW3h5uaG1NRU6fn79+8X+o6biIiI9FdiYN+9exfjx49HZGQkevfuDeBpQC9YsADp6elQqVTYvHkzunfvjlq1asHBwQGnTp0CAERHR6Njx47mHQEREVE5UOIp8XXr1iEnJwcRERHSc0FBQRg7diwGDhwItVoNHx8f+Pn5AQAiIyMRGhqKzMxMNGvWDEOHDjVf74mIiMqJEgM7NDQUoaGhz31t8ODBhZ7z8PDA1q1bje8ZERERSfhLZ0RERDLAwCYiIpIBBjYREZEMMLCJiIhkgIFNREQkAwxsIiIiGWBgExERyQADm4iISAYY2ERERDLAwCYiIpIBBjYREZEMMLCJiIhkgIFNREQkAwxsIiIiGWBgExERyQADm4iISAYY2ERERDJgW9odICKyNkfibyPufLJB69rZK6HK1Zi4R8bXSUrJAAAs3HjarHUMZUid1s2qo5NnLTP1yPpwhk1E9Iy488lISsks7W6YVB1XJ9RxdSrtbphMUkqmwR+q5IozbCKi56jj6ogZg1/Xez0XFyekpmaYoUesk5++ZwrKAs6wiYiIZICBTUREJAMMbCIiIhlgYBMREckAA5uIiEgGGNhEREQywMAmIiKSAQY2ERGRDDCwiYiIZICBTUREJAMMbCIiIhlgYBMREckAA5uIiEgGGNhEREQywMAmIiKSAQY2ERGRDDCwiYiIZICBTUREJAMMbCIiIhmwLe0OkO6OxN9G3Plks7RtZ6+EKldjlrb1qZWUkgEAWLjxtFnrmJIhdVo3q45OnrXM1CMiKos4w5aRuPPJSErJLO1umFUdVyfUcXUq7W6YVVJKptk+eBFR2cUZtszUcXXEjMGvm7xdFxcnpKZmmLzd0qxlrXVMcfaAiMofzrCJiIhkQKfAXrFiBXr37o3evXtj0aJFAIDY2Fj4+/vDx8cHixcvlpa9ePEiAgMD0aNHD8yePRtqtdo8PSciIipHSgzs2NhYxMTEICoqCjt27MD58+exa9cuzJo1CytXrsSePXuQkJCAo0ePAgCmTZuGuXPnYv/+/RBCYMuWLWYfBBERUVlXYmC7uLhg5syZsLe3h52dHRo0aIDExETUrVsX7u7usLW1hb+/P/bt24fbt28jOzsbnp6eAIDAwEDs27fP3GMgIiIq80q86KxRo0bSvxMTE7F3714MGTIELi4u0vOurq5ITk5GSkpKgeddXFyQnGzc1bDOzo56r+PiYpmrjC1dx85eada6lhqPJWtZYx1j96M1jqms1eE+sv465XEf6XyV+OXLlzFu3DhMnz4dSqUSiYmJ0mtCCCgUCmi1WigUikLPGyMtLRNardB5eWu9MtgUdfL+1tccdXmVuOXqGLMfrXVMZa0O95H11ymL+8jGRlHsJFWni85OnTqF4cOHY8qUKejbty/c3NyQmpoqvZ6amgpXV9dCz9+/fx+urq46d5aIiIier8TAvnv3LsaPH4/IyEj07t0bANC8eXNcv34dN27cgEajwa5du9CxY0fUqlULDg4OOHXqFAAgOjoaHTt2NO8IiIiIyoEST4mvW7cOOTk5iIiIkJ4LCgpCREQEJk6ciJycHHh7e8PX1xcAEBkZidDQUGRmZqJZs2YYOnSo+XpPRERUTpQY2KGhoQgNDX3uazt37iz0nIeHB7Zu3Wp8z4iIiEjCXzojIiKSAQY2ERGRDDCwiYiIZICBTUREJAMMbCIiIhlgYBMREckAA5uIiEgGGNhEREQywMAmIiKSAQY2ERGRDDCwiYiIZICBTUREJAMMbCIiIhlgYBMREckAA5uIiEgGGNhEREQywMAmIiKSAQY2ERGRDDCwiYiIZICBTUREJAMMbCIiIhlgYBMREckAA5uIiEgGGNhEREQywMAmIiKSAQY2ERGRDDCwiYiIZICBTUREJAO2pd0BImsRc/sE/kyO13s9OzslVCqNzsvfyqwNAFhy+qTZawFAq+qeaF+rjd61iMi6MLCJ/s+fyfG4lXkHtR1rmrVOgzdumbX9/G5l3gEABjZRGcDAJsqntmNNfPj6O3qt4+LihNTUDDP1yLhaS07/14y9ISJL4nfYREREMsDAJiIikgEGNhERkQwwsImIiGSAgU1ERCQDDGwiIiIZYGATERHJAAObiIhIBhjYREREMsBfOiMiIpO5EH8Hl8+n6L2enb0Sqlzdfyf/fkomACB6Y7zZawFAo2aueNnTvD9bXBIGNhERmczl8ym4n5KJaq6OZq3T3rWSWdvPL+/DAQObiIjKlGqujggY7KnXOtb8m/yGzOLNQafvsDMzM+Hn54dbt57eZSgkJAQ+Pj4ICAhAQEAADh48CAC4ePEiAgMD0aNHD8yePRtqtdp8PSciIipHSgzss2fPYuDAgUhMTJSeS0hIwIYNGxAdHY3o6Gh0794dADBt2jTMnTsX+/fvhxACW7ZsMVvHiYiIypMSA3vLli0ICwuDq6srAODff//FnTt3MGvWLPj7+2PZsmXQarW4ffs2srOz4enpCQAIDAzEvn37zNp5IiKi8qLE77DDw8MLPL5//z7atGmDsLAwODk5Ydy4cdi6dSsaNWoEFxcXaTkXFxckJyebvsdERETlkN4Xnbm7u+PLL7+UHgcHB2PHjh1o0KABFAqF9LwQosBjQzk763+loYuLk9F1rbGOnb3SrHUtNR5L1tKnjp2d4dvXWredHMZkjXWMPdascUyWqmPMtrPa48hKxqR3YP/9999ITExEjx49ADwNZltbW7i5uSE1NVVa7v79+9JpdGOkpWVCqxU6L2+pKw1Lo07e3w2ao641X6FpqToqlWHb15q3nbWPyVrrGHOsWeuYLFXH0G1n1ceRhcZkY6ModpKq9y+dCSGwYMECpKenQ6VSYfPmzejevTtq1aoFBwcHnDp1CgAQHR2Njh076ts8ERERPYfeM2wPDw+MHTsWAwcOhFqtho+PD/z8/AAAkZGRCA0NRWZmJpo1a4ahQ4eavMOWknvxCNRXjhf5+h07W6hU5v+ztfx1NGlNAQBPft5v1jr6sG3oBfumnUzeHyIiKkjnwD58+LD078GDB2Pw4MGFlvHw8MDWrVtN07NSpr5yHJq0JCid65R2VyQf1rtY2l0oQJOWBAAMbCIiC+AvnRVD6VwHFf1DnvuatX5/ZMk6T37+1Ey9ISKiZ/FuXURERDLAwCYiIpIBBjYREZEMMLCJiIhkgIFNREQkAwxsIiIiGWBgExERyQADm4iISAYY2ERERDLAwCYiIpIBBjYREZEM8LfEiYio3LgQfweJl9Oke1zr4n5KJgAgemO8XrVatK6DOg2r6rVOcTjDJiKicuPy+RTcu/1Yr3WquTqimqujXuvcT8lEwpnbeq1TEs6wiYioXHGrVQm93nrVrDX0nY3rgjNsIiIiGWBgExERyQADm4iISAYY2ERERDLAwCYiIpIBBjYREZEMMLCJiIhkgIFNREQkAwxsIiIiGWBgExERyQADm4iISAYY2ERERDLAwCYiIpIBBjYREZEMMLCJiIhkgIFNREQkAwxsIiIiGWBgExERyQADm4iISAYY2ERERDLAwCYiIpIBBjYREZEMMLCJiIhkgIFNREQkAwxsIiIiGWBgExERyQADm4iISAYY2ERERDLAwCYiIpIBnQI7MzMTfn5+uHXrFgAgNjYW/v7+8PHxweLFi6XlLl68iMDAQPTo0QOzZ8+GWq02T6+JiIjKmRID++zZsxg4cCASExMBANnZ2Zg1axZWrlyJPXv2ICEhAUePHgUATJs2DXPnzsX+/fshhMCWLVvM2nkiIqLyosTA3rJlC8LCwuDq6goA+Ouvv1C3bl24u7vD1tYW/v7+2LdvH27fvo3s7Gx4enoCAAIDA7Fv3z6zdp6IiKi8sC1pgfDw8AKPU1JS4OLiIj12dXVFcnJyoeddXFyQnJxsdAednR31XsfFxcnounfsbEtsyxR1dGGtdXTZRqaqZSh96tjZKfVex5A6xiprY7LGOnb2hm83Y9YrC3WM2XaWGI+x+7Y065QY2M/SarVQKBTSYyEEFApFkc8bKy0tE1qt0Hl5FxcnpKZmGF1XpXr6/XtRbZmqTkksUSf34hEobvwhjVlXmrQkAMCNr2fptV4Vz07IcffSax1D6LvtVCoNgKL3uanqGKOsjcla66hyDdtuhtQylLXWMXTbWWo8qlwN7OyVZq9lSB0bG0Wxk1S9rxJ3c3NDamqq9Dg1NRWurq6Fnr9//750Gp2sm/rKceQmX9d7PaVzHSid6+i1jiYtCZnnY/SuRURU3uk9w27evDmuX7+OGzduoHbt2ti1axf69euHWrVqwcHBAadOnULLli0RHR2Njh07mqPPZAb21evBznea2es8+flTs9cgItO4EH8HiZfTpFmzLu6nZAIAojfG61WrRes6qNOwql7rlDd6B7aDgwMiIiIwceJE5OTkwNvbG76+vgCAyMhIhIaGIjMzE82aNcPQoUNN3mEiIrKMy+dTkJaaBWeXF3Vep5qr/tcd3U/JRMKZ2wzsEugc2IcPH5b+7eXlhZ07dxZaxsPDA1u3bjVNz4iIqNS51aqEXm+9atYa+s7Gyyv+0hkREZEMMLCJiIhkgIFNREQkAwxsIiIiGWBgExERyQADm4iISAYY2ERERDLAwCYiIpIBBjYREZEMMLCJiIhkgIFNREQkAwxsIiIiGWBgExERyQADm4iISAYY2ERERDLAwCYiIpIBBjYREZEMMLCJiIhkgIFNREQkAwxsIiIiGWBgExERyQADm4iISAYY2ERERDLAwCYiIpIBBjYREZEMMLCJiIhkgIFNREQkAwxsIiIiGWBgExERyQADm4iISAZsS7sDRES6enT0CDLijuu93j17W6hy1Tovn6NtCgC4uWi/2WsBgFNrL1T27qR3LSpfGNhEJBsZcceRczMJDu51zFpnhM1Fs7afX87NJABgYFOJGNhEJCsO7nXgPj1Er3VcXJyQmpphph4ZV+vmok/N2BsqS/gdNhERkQxwhk0kEzG3T+DsuXNQqTQ6r3Mr4w4AYMnp/+pVq3PDNmheyVOvdYjIvDjDJpKJP5Pjkfjoll7r1HaqidpONfVa51bmHcTc+EOvdYjI/DjDJpKR/1e5Nsa/OsasNfSdjRORZXCGTUREJAMMbCIiIhlgYBMREckAA5uIiEgGGNhEREQywMAmIiKSAaP+rCs4OBgPHjyAre3TZubNm4esrCx8+umnyMnJQc+ePTFp0iSTdJSIiKg8MziwhRBITEzEr7/+KgV2dnY2fH19sX79etSoUQPjxo3D0aNH4e3tbbIOExERlUcGB/a1a9cAACNHjsSjR4/w9ttvo3Hjxqhbty7c3d0BAP7+/ti3bx8Dm4iIyEgGB/bjx4/h5eWFOXPmQKVSYejQoRg9ejRcXFykZVxdXZGcnGxUB52dHfVex8XFyaiaAHDHzrbEtkxRRxfmrqPLWOVYS986dnZKvdcxpI6hjOmfNdbJo0+de/aGv38sNR59a1nzmOzsLfSes1AdS9YyRx2DA7tFixZo0aKF9Lh///5YtmwZWrZsKT0nhIBCoTCqg2lpmdBqhc7Lm+o2eirV0xvQF9WWpW7XZ4k6KpUadna2FhmPJWvpu+3ybqqhb98s9V5QqTSws1Na4P1gmTqAAfsot/jj0lR1jFGWxqTK1cDO3gLvOQvVsWQtQ+rY2CiKnaQaHNh//vknVCoVvLy8ADwN51q1aiE1NVVaJjU1Fa6uroaWICKSlUdHj+DemT+kENZFzs0kAPrfF1vTrROUr3vptQ7Jm8F/1pWRkYFFixYhJycHmZmZiIqKwuTJk3H9+nXcuHEDGo0Gu3btQseOHU3ZXyIiq5URdxxZ16/rtY6Dex04uNfRa52cm0lI/S1Gr3VI/gyeYXfu3Blnz55Fnz59oNVqMWjQILRo0QIRERGYOHEicnJy4O3tDV9fX1P2l0gnvHc0lZYX69WD24fTzFpD39k4lQ1G/R32hx9+iA8//LDAc15eXti5c6cxzRIZ7c/keNzOuotaL9bQeR197xsN/O/e0c1f9dR7XSIiffB+2FRm8d7RRFSW8KdJiYiIZICBTUREJAMMbCIiIhlgYBMREckAA5uIiEgGGNhEREQywMAmIiKSAQY2ERGRDDCwiYiIZICBTUREJAMMbCIiIhlgYBMREckAA5uIiEgGGNhEREQywMAmIiKSAd4Pm4hKxaOjR3DvzB9Q5ap1XifnZhIA4OaiT/WqpenWCcrXvfRah8jacIZNRKUiI+44sq5f12sdB/c6cHCvo9c6OTeTkPpbjF7rEFkjzrCJqNS8WK8e3D6cZtYa+s7GiawVZ9hEREQywMAmIiKSAQY2ERGRDDCwiYiIZICBTUREJAMMbCIiIhlgYBMREckAA5uIiEgGGNhEREQywMAmIiKSAQY2ERGRDDCwiYiIZICBTUREJAMMbCIiIhkoU7fXzL14BHf2/QGVSm10W5q0JADAk5+ff2u+O3a2etexbegF+6adjO0aERGVQ2UqsNVXjkM8uAlFVXej21I61zFBj/4n7wMAA5uIiAxRpgIbAOyr14Od7zSz13FxcUJqaobOyxc1UyciItIFv8MmIiKSAQY2ERGRDJS5U+KWYMjFbSVdxFaUx56dAHcvvdYhIqKyhzNsA6ivHEdu8nW91lE619H7QjZNWhIyz8fotQ4REZVNnGEbyBIXt/FCNSIiysMZNhERkQwwsImIiGTALIH9888/o1evXvDx8cHGjRvNUYKIiKhcMfl32MnJyVi8eDG2b98Oe3t7BAUFoXXr1mjYsKGpSxEREZUbJp9hx8bGok2bNqhcuTIqVqyIHj16YN++faYuQ0REVK6YfIadkpICFxcX6bGrqyv++usvg9uzsVHovKzSsTKUtrZ6rWMI1rH+WpUrVIKdnbLM1LFkLUvVsa9SGbZ25n8vWKqOJWtZqo5jJQfYWuC9YKk6lqxlSJ2SllUIIYSxHctv1apVyMnJwYcffggA2LJlCxISEjBv3jxTliEiIipXTH5K3M3NDampqdLj1NRUuLq6mroMERFRuWLywG7bti2OHz+OBw8e4N9//8WBAwfQsWNHU5chIiIqV0z+HXb16tUxadIkDB06FCqVCv3798drr71m6jJERETlism/wyYiIiLT4y+dERERyQADm4iISAYY2ERERDLAwCYiIpIBBjYREZEMMLCJiIhkgIFNREQkAwxsIiIiGZB1YG/YsAG9e/eGn58f3n33XaSlpRV4fcKECSa56UhRdVq3bo2AgADpv507d5qlzsaNG9G3b1/07NkTU6dORW5urtFj+uyzzxATE4Po6Gi8+eabCAgIQFBQEM6dOwcAOHToEL788kuT1SlqbKaqExERgU6dOkn7Iu/mMwBw9+5ddOjQAQ8ePDC6DmC5MVm6TlHb0Nz7yNTHEWD597el6uQ5dOgQWrRoUeCxoXXyty2EwIwZM7Bu3TrpdY1Gg/DwcPj6+qJ79+748ccfpdemTp2Kq1evWlWdouqa4zgqaUzZ2dkICQmBn58fevfujZCQEGRnZxs+JiFT586dE507dxaPHz8WQggREREh5syZI72+evVq0bp1a/Hxxx+bpc7Vq1eFj4+PUW3rUmf//v3C19dXPHz4UGg0GjFhwgTx1VdfGVXrzJkzYty4ceLq1auiXbt2Ijk5WQghxJEjR4S3t7e03PDhw8WFCxeMrlPSvjK2jhBCvP322+LUqVOFno+KihKdO3cWjRs3FmlpaUbVEMJyY7J0HSGK3oamqFNU+6Y+joSw/PvbUnXyXL9+XXTr1k14enoWWM6QOvnbvnLliggODhbNmzcXa9eulZbZsGGDGD16tFCpVOLRo0eiR48e4uzZs0IIIZKSksRbb70ltFqtVdQpqq45jiNdxvTFF1+IadOmCY1GI9RqtZg0aZJYsmSJwWOS7Qz7lVdewf79++Hk5IScnBwkJyejcuXKAIC4uDj8/vvvCAoKMludM2fOwMbGBoMGDYK/vz9WrFgBjUZj8jo7duzAyJEjUblyZdjY2ODjjz9GQECAUWNavnw5BgwYAHt7e8yfP1+6m9orr7yC+/fvSzP4/v37Y8WKFUbXKW5fmaJObm4uLly4gLVr18Lf3x8TJ07EnTt3kJycjEOHDhX4xGssS43J0nWK2oamqlNU+6Y+jvKPyVLvb0vVAYB///0X06ZNw8yZMwstZ0id/G1v3LgRb731Fnx9fQssc+jQIQQGBsLW1hYvvfQSevfuLZ0FcXd3h5OTE3755RerqFNUXXMcR7qM6Y033sC7774LGxsbKJVKNG3aVDquDBqTXh8prNDBgwfFf/7zH9G+fXtx/fp1ce/ePeHv7y+Sk5PFsmXLjJ5hF1Vn8+bNYt68eSIrK0ukp6eLAQMGiG+++cbkdXr27ClWrVolRo4cKfz8/MRHH30ksrKyDG4/PT1dNG/eXOTk5BR4XqvViilTpoiJEydKzz18+FC88sor4t9//zVJnWfHZoo6Qjz9pDp69Gjx999/C61WK9asWSMCAgIKfHI1xQzbUmMqjTolbUNz7SNTH0el+f62RJ2pU6eKn376Sdy8ebPQDFvfOkWNYcaMGQVmiT169BBnzpyRHm/ZskWMHz9eevz111+L6dOnl3odXeqa6jjSdUz53bp1S7Rr104cPnzY4DHJdoadp1u3boiLi8PEiRMxcuRITJo0CSEhISa/B3f+OqNGjUL//v0xZ84cVKxYEZUqVcKIESNw6NAhk9dRq9U4duwYli5dim3btiE9PR2LFy82uP0bN27AxcUF9vb20nNPnjzBBx98gKSkJMyfP196vnLlynBwcMDt27dNUufZsWm1WqPrAE8/qa5ZswaNGzeGQqHAqFGjkJSUhFu3bhnUXlEsNabSqFPSNjTXPvLy8jLpcVSa729z19m4cSNsbW3Rv3//5y6rb53njeF5hBBQKBQFHtvY/C86ateujevXr5d6HV3qmuo40nVMeRISEjB48GAMGTIEnTt3NnhMsg3sGzdu4M8//5Qe9+vXD3fv3sXFixcRERGBgIAAbNq0CXv27MHs2bNNWufOnTuIjo7GpUuXpOeFELC1NfxupUXVcXBwgI+PDxwdHWFvb48333wT8fHxBtdRKBTSmxQA7ty5g6CgICiVSnz//feoVKlSgeWVSiWUSqVRdYoaW3p6utF1AODSpUvYsWNHgeeEELCzszOovaJYakylUUeXbWiOfXTq1CmTHkel8f62VJ2oqCicO3cOAQEBGDt2LLKzsxEQEIDk5GSD6jw7hqLUqFEDKSkp0uOUlBS4ublJj21tbQsEa2nVKa6uqY8jXccEALt378bIkSMxZcoUvPPOOwVe03dMsg3s1NRUTJ48Wbry9+eff0ajRo1w5swZREdHIzo6GkFBQejVqxfCw8NNXufKlStYtmwZNBoNsrOzsXHjRvTq1cvkdd5++23s3bsX2dnZEELg0KFDePXVVw2uU6dOHaSlpSEnJweZmZkIDg6Gj48PFi9ejAoVKhRYNiMjA7m5uahZs6ZRdYoaW5UqVYyuAwA2NjYIDw/HzZs3AQA//PADmjRpUuBgNwVLjak06pS0Dc21jy5fvmzS46g03t+WqrN161bs2rUL0dHRWL16NSpUqIDo6GhUr17doDr52y5O165dsW3bNqjVajx+/Bi7d+9Gt27dpNdv3bqF+vXrl3qd4uqa+jjSdUyHDx/G/PnzsW7dOvj7+xd6Xd8xGf5RtpS1atUK77zzDoYOHQqlUglXV1eT/PmErnWqVauGefPmwd/fH2q1Gr6+vnjrrbdMXqdmzZpIT09HYGAgNBoNmjVr9twLTnRVqVIltGzZEidOnMClS5dw584dHDx4EAcPHpSW+fbbb1GlShXExMSgU6dOOp/2KaqOt7d3sfvKmDoA0LhxY4SGhuLdd9+FRqOBm5sbvvjiC4PaKo6lxlRadYrbhubaR1WqVDHpcVQa729LHkfF0beOrm0PHDgQSUlJCAgIgEqlwoABA/Cf//xHev3333/HkCFDSr1OSXVNeRzpOqaFCxdCCIHQ0FDpuddffx1hYWEGjUn2F52R/k6dOiXGjBlT4nLBwcHi4sWLVl/HksratuM+sv5tZ846urZdlBs3buj0p0mWqmNoXWvedvnJ9pQ4Ge71119HvXr18NtvvxW5zMGDB9GqVSt4eHhYfR1LKmvbjvvI+redOevo0nZxlixZgvnz5xe4WKw06xhS19q3XX4KIYQwqBoRERFZDGfYREREMsDAJiIikgEGNhERkQwwsK3QrVu30KRJE/z0008Fnl+3bp1Rf9L1rC5dukh3FTK3zMxMBAUFoXfv3jhw4IDR7f34449YvXp1icuNHDlSr7t06dquqY0bNw7bt2/Xe72LFy8iJCTEqNpjxozBlStX9FrH0P7qKywsDF26dDHq1/3ynDt3Du+//36Jy61YseK5v7aWf/27d+/Cz88PAQEBOHPmjNF9sxZ//fUX5s6dC+DpPRn8/PxM1nZJ77PMzEyMHj1aupsVFSbbv8Mu62xsbLBw4UK0bNlSrz+st1YXL15EWlpagb9TNcbAgQN1Wu7YsWNmadcaaLVazJ49G6tWrTKqnTVr1pioR6a3efNmHDlyxCQ/gvPqq69i2bJlJS4XFxeHhg0bFrt+XFwcqlWrhm+//dboflmTK1euFPjlNFMq6X3m6OgIPz8/LF26FDNmzDBLH+SOgW2lKlSogBEjRmDq1KnYtGlToT/onzlzJho1aoRRo0YVetylSxf4+fnhxIkTSE9Px+jRo3H69GmcP38etra2WLVqlfTrSD/88AMuXbqE3NxcjBgxQvqd4sOHD2PVqlVQqVSoUKECZsyYgRYtWmD58uWIj49HSkoKmjRpgsjIyAL9OnToEFasWAGtVosXX3wRISEhcHR0xKxZs5CcnIyAgABs3ry5wK9BBQcH4+WXX8apU6fw8OFDBAQESDOZ57X32muvYfny5Xj48CHmzp2LLl26oG/fvjh+/Dju3r0r3Ws5b+Y5bNgwrF69Gr/++is2bdoEOzs7ODg4YN68eYX+x6xLu88qbrnNmzdj/fr1sLGxQbVq1TBnzhzUq1cPycnJmDlzJlJSUlCzZs0C93K/evUqwsPD8ejRI2g0GgQHBz/396P37t2L2rVrS/vy8uXLmDdvHh49egSFQoGRI0eiT58+iIuLQ3h4OCpWrIisrCxs27atwPupS5cuWLp0KZ48eYLFixfD3d0dly9fhlqtxscff4yWLVsa1N+oqCh8+eWXiI6OhkKhQL9+/TBu3Dj06dOnwDiK6vegQYMghMCYMWMQFhaGVq1aFdhPN27cwL1795CamgoPDw+Eh4fD0dGx2O3wySefYNeuXZg5cyYcHR3x999/4969e2jSpAkWLlyIHTt2ICEhAYsWLYJSqUT37t2lmnnrh4aGYsmSJcjIyEBwcDDWr19v9Pshv7i4OERGRqJmzZq4du0aKlSogIiICDRo0ADXr1/HvHnzkJWVJY17yZIlcHBwwCuvvIKuXbvi0qVLiIyMRMWKFZ+7X+Li4p67n2vWrIlly5YhIyMDISEh6NOnD548eYJJkybh2rVryMnJwfz589GqVSvk5uYiMjISf/zxBzQaDV5++WWEhobC0dERP/zww3OPs7z3Wf369RESEoIbN27AxsYGzZo1w7x582BjY4OePXsiMjISo0aNQrVq1Qq958s9g//qm8wm7048Go1GDB48WERERAghhFi7dq2YMWOGEKLwXWHyP+7cubNYsGCBEEKI3bt3Cw8PD+lHAd577z2xatUqabmwsDAhhBD37t0TXl5e4p9//hHXr18Xfn5+4sGDB0IIIf755x/Rrl07kZWVJZYtWyZ69OghVCpVoX5fuXJFtG3bViQlJQkhhIiNjRXt2rUTGRkZ4sSJE6J3797PHe+QIUPEmDFjRG5urkhPTxc9evQQhw8fLra9/Hdi69y5s7SN7t27J1599VVpnby7dKnVatGsWTPpnsVRUVFi06ZNhfqia7v5FbVcbGys6Natm3SXsG3btomePXsKrVYr3nvvPbF48WIhhBCJiYnC09NTbNu2TahUKtGrVy+RkJAghBDi8ePHomfPngXuYpRn4sSJYtu2bUIIIVQqlejatavYv3+/1I8OHTqI06dPixMnTggPDw9x69at527/zp07i7/++kucOHFCNG3aVLov8Lp168TgwYOFEMLg/k6ePFmEhYWJkJAQERoaWqh2cf0Woui7rC1btkx07NhRpKamCo1GIyZPniwiIiJK3A5578EZM2aIAQMGiJycHJGbmyv69Okjtm7dKoR4+n7cu3dvoZr519+2bZsYO3ZskdtT3/fDs3U8PDzEH3/8IYQQ4ocffhB9+/YVQjy9j/OOHTuEEELk5uYKPz8/sW/fPmlbRUVFSdu1qP1S3H7OP6685eLj44UQQnzzzTdi6NChQgghli9fLiIiIqS+f/755yIsLKzY4yzvfRYVFSVGjhwphBBCrVaL2bNni8TERGn848aNk/YFFcQZthWzsbHBZ599hj59+qB9+/Z6revj4wPg6V2SqlWrJv0oQJ06dQr84H3ePcOrV6+Odu3a4fjx41AqlUhJScHw4cOl5RQKBZKSkgAAnp6ez71Bw4kTJ9CmTRu4u7sDALy8vFC1alUkJCSU+OMAAwYMgJ2dHezs7ODr64uYmBjUr1+/yPae1bVrV2kczs7OSE9Pl9YDnv6wv6+vL4KCgtCpUye0b9++xJ961KXd4pb7/fff0atXL1StWhUAEBgYiPDwcNy6dQuxsbHSab+6deuidevWAIDExEQkJSVh1qxZUtvZ2dm4cOECPD09C9S8du0ahg4dKq2Xk5Mj7ffq1avDx8cHv//+O1q3bo0aNWqgVq1aJY63Zs2aaNq0KQDg5ZdfRlRUFAAY3N+8+7dXqFDhud95F9fvFi1aFNtXX19faRbWv39/LFiwAP369St2O+TXoUMH6UxD48aNCxwXxtL3/fDse8rDw0M6o9CvXz/MmzcPDx8+xLRp03Ds2DGsWbMGiYmJSElJwZMnT6T18tYpbr80aNCgyP38LHd3dzRv3lzq07Zt2wAAR44cQUZGBmJjYwEAKpUKzs7OOh1nLVu2xOLFixEcHIy2bdti2LBhqFu3rvS6vnewKk8Y2FauRo0a+PjjjzFjxowCpxIVCgVEvt+8UalUBdbLf8qzuLtW5b9TjFarha2tLTQaDby8vLBkyRLptbt378LV1RUHDx5ExYoVn9uWVqstFMxCCKjV6hLvnJX/A4D4v9vqFdfesxwcHKR/P7tt8kRGRuKff/5BbGwsVq9ejejoaCxdurTYfunSblHLPe9uPnn9f7atvPFrNBo4OTkhOjpaeu3+/ftwcnIq1Fb+NjQaTbHbqqh99qz8X1Xkb9/Q/ubdICE3NxcpKSmFgqmkfhcn/52VtFotbGxs9GqvqLGagr7vh2c9765RSqUSkydPhkajQc+ePdGpUyfcvXu3QL/z9nNx+yU+Pl7nsec/bvMvp9VqMWvWLCmMs7KypBthlHScubu74+DBg4iLi8OJEycwYsQIzJs3D126dJFqGnpXuLKOV4nLgK+vLzp27IjvvvtOeq5KlSrSTDM5ORknT540qO28T9Z37tzB8ePH4eXlBS8vLxw7dgxXr14FABw9ehRvvvlmiVdvenl5ISYmRrojU953eHmf0Iuzc+dOaLVapKenY+/evejSpYtR7eVRKpVQq9V48OABvL29UblyZQwfPhwffvih2a+Q79ChA/bs2SNdpb5t2zZUrlwZdevWRYcOHbB582YAT7d9XFwcAKBevXrSXZiA/12N/LyzCvXq1ZPOetSvXx+2trbSFfjJycnYv38/2rZta7Kx6NtflUqFyZMn44MPPsCECRMwadKkQh8sjen3L7/8goyMDGi1WmzZsgWdO3c2yXbIe8+YWnHvh2ddunRJuu3o5s2b0aJFC1SqVAkxMTEYP368dEezs2fPQqPRFFpfn/dRfrqOvX379ti4cSNyc3Oh1WoxZ84cfPHFFzodZz/88ANCQkLQvn17TJs2De3bt8eFCxek12/dulXoe316ijNsmQgNDcWpU6ekx8HBwZg6dSp69OiB2rVro02bNga1m5OTg759+0KlUiE0NFQ6UObNm4fJkydL9ydetWoVXnzxxWLbatiwIcLCwjBhwgRoNBpUqFAB//3vf587O3xWdnY2+vfvj6ysLAwaNAheXl4AYHB7eXx9fREcHIzly5fj3XffxfDhw1GhQgUolUrMnz9f53YM0a5dOwwfPhzDhg2DVqtF1apV8dVXX8HGxgZhYWEICQlBz5494ebmJn1lYW9vj5UrVyI8PBxr166FWq3GBx98gJYtWxZqv0ePHjh48CD69esHOzs7rFy5EvPnz8fy5cuh0Wgwfvx4tGnTRgpXYxjS34ULF6JatWrS3bcOHTqExYsXY/r06VK7xfW7JNWqVcOYMWPw8OFDvPHGG3jnnXdMsh26dOmCL774AiqVCn379jVgaz1fce+H541tyZIluH37NqpWrYpFixYBACZNmoTx48ejYsWKcHR0xBtvvCF9aMuvuP1S3Hbw9PTEl19+iQkTJiA4OLjI5d577z0sXLgQffv2hUajQdOmTaUL+Uo6zvr06YOTJ0+iV69eeOGFF1CjRg2pVm5uLuLj4426JXJZxt8Sp1IXHByMwYMHw9fXt7S7IisajQaBgYFYvXq1dKV4eZH/av6yJv/V7OXN9u3bcfnyZf5ZVxF4SpxIppRKJT755BOz3PubyNKysrKwa9cuTJw4sbS7YrU4wyYiIpIBzrCJiIhkgIFNREQkAwxsIiIiGWBgExERyQADm4iISAYY2ERERDLw/wFW9Dwnq9QUoQAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_dendrogram(model_agg, truncate_mode='level', p=3)\n",
    "plt.title('Hierarchical Clustering Dendrogram')\n",
    "plt.xlabel('Number of points in node (or index of point if no parenthesis)')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Отримання міток кластерів"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "k_best = visualizer.elbow_value_\n",
    "\n",
    "with warnings.catch_warnings():\n",
    "    warnings.simplefilter('ignore')\n",
    "\n",
    "    model_kmn = KMeans(n_clusters=k_best, random_state=42).fit(X)\n",
    "    model_agg = AgglomerativeClustering(n_clusters=k_best).fit(X)\n",
    "\n",
    "labels_kmn = pd.Series(model_kmn.labels_, name='k-means')\n",
    "labels_agg = pd.Series(model_agg.labels_, name='h-clust')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>k-means</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>h-clust</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>35</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "k-means  0  1   2  3  4\n",
       "h-clust                \n",
       "0        0  0  35  0  0\n",
       "1        8  0   0  0  0\n",
       "2        0  9   0  0  0\n",
       "3        0  0   0  0  8\n",
       "4        0  0   0  4  0"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(labels_agg, labels_kmn)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>k-means</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>label</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>BREAST</th>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CNS</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COLON</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>K562A-repro</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>K562B-repro</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LEUKEMIA</th>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MCF7A-repro</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MCF7D-repro</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MELANOMA</th>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NSCLC</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>OVARIAN</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PROSTATE</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RENAL</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>UNKNOWN</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "k-means      0  1  2  3  4\n",
       "label                     \n",
       "BREAST       0  2  3  2  0\n",
       "CNS          0  0  5  0  0\n",
       "COLON        0  0  0  0  7\n",
       "K562A-repro  1  0  0  0  0\n",
       "K562B-repro  1  0  0  0  0\n",
       "LEUKEMIA     6  0  0  0  0\n",
       "MCF7A-repro  0  0  0  1  0\n",
       "MCF7D-repro  0  0  0  1  0\n",
       "MELANOMA     0  7  1  0  0\n",
       "NSCLC        0  0  8  0  1\n",
       "OVARIAN      0  0  6  0  0\n",
       "PROSTATE     0  0  2  0  0\n",
       "RENAL        0  0  9  0  0\n",
       "UNKNOWN      0  0  1  0  0"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(target['label'], labels_kmn)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Візуалізація результатів кластеризації"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x360 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(ncols=2, nrows=1, figsize=(10, 5), sharey=True)\n",
    "\n",
    "for i, s in enumerate([target['label'], labels_kmn]):\n",
    "    ax = axes[i]\n",
    "\n",
    "    sns.scatterplot(x=X[:, 0],\n",
    "                    y=X[:, 1],\n",
    "                    hue=s,\n",
    "                    style=s,\n",
    "                    edgecolor='black',\n",
    "                    linewidth=0.5,\n",
    "                    s=60,\n",
    "                    palette='tab20',\n",
    "                    legend=False,\n",
    "                    ax=ax)\n",
    "\n",
    "    ax.set(title=s.name)"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python 3 (Spyder)",
   "language": "python3",
   "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": "3.11.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}