{ "cells": [ { "cell_type": "markdown", "metadata": { "id": "OJFRKa5M41ok" }, "source": [ "# El problema\n", "\n", "Vamos a utilizar para ello el IRIS DataSet pero considerando únicamente dos características (nos facilitará la visualización de los resultados)\n", "\n", "## Iris DataSet\n", "\n", "\n", "\n", "Esta es quizás la base de datos más conocida en el mundo del análisis de datos. El conjunto de datos contiene 3 clases de 50 casos cada una, donde cada clase se refiere a un tipo de planta de iris. Una de las clases es linealmente separable de las otras 2; estas últimas NO son linealmente separables entre sí.\n", "\n", "Objetivo: clase de planta de iris.\n", "\n", "\n", "# 0. Carga de Datos\n", "\n", "En este caso los datos nos la facilita la propia API de scikit-learn\n", "\n" ] }, { "cell_type": "code", "execution_count": 32, "metadata": { "id": "pp9tjdSA6rR5" }, "outputs": [], "source": [ "# import some data to play with\n", "from sklearn import datasets\n", "iris = datasets.load_iris()\n", "\n", "# we only take the first two features.\n", "# We could avoid this ugly # slicing by using a two-dim dataset\n", "X = iris.data\n", "y = iris.target\n" ] }, { "cell_type": "markdown", "metadata": { "id": "8tye7Kut6vPB" }, "source": [ "Posteriomente lo cargamos en un DataFrame de Pandas con el nombre de *df_iris*" ] }, { "cell_type": "code", "execution_count": 33, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 204 }, "id": "rxVQ7L5c4B-x", "outputId": "c4cf5183-eefc-4b4c-d2bd-9e5c5c3d91fc" }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
sepal length (cm)sepal width (cm)petal length (cm)petal width (cm)target
05.13.51.40.20
14.93.01.40.20
24.73.21.30.20
34.63.11.50.20
45.03.61.40.20
\n", "
" ], "text/plain": [ " sepal length (cm) sepal width (cm) petal length (cm) petal width (cm) \\\n", "0 5.1 3.5 1.4 0.2 \n", "1 4.9 3.0 1.4 0.2 \n", "2 4.7 3.2 1.3 0.2 \n", "3 4.6 3.1 1.5 0.2 \n", "4 5.0 3.6 1.4 0.2 \n", "\n", " target \n", "0 0 \n", "1 0 \n", "2 0 \n", "3 0 \n", "4 0 " ] }, "execution_count": 33, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# 0. load data in DataFrame\n", "import numpy as np\n", "import pandas as pd\n", "df_iris = pd.DataFrame(data= np.c_[iris['data'], iris['target']],\n", " columns= iris['feature_names'] + ['target'])\n", "\n", "df_iris.target = df_iris.target.astype(int)\n", "\n", "\n", "df_iris.head()" ] }, { "cell_type": "code", "execution_count": 34, "metadata": { "colab": { "base_uri": "https://localhost:8080/" }, "id": "x9kv3zx845B-", "outputId": "1b23768a-fb27-41a1-9e96-ca3f5a97edae" }, "outputs": [ { "data": { "text/plain": [ "array([0, 1, 2])" ] }, "execution_count": 34, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df_iris['target'].unique()" ] }, { "cell_type": "markdown", "metadata": { "id": "Qw1rvr-v5dy5" }, "source": [ "Posteriormente seleccionamos las variables con las que trabajar.\n", "\n", "Es muy importante seleccionar las características cuando se va a resolver un problema mediante kNN ya que muchas variables pueden distorsionar el resultado del algoritmo que está basado en la distancia.\n", "\n", "Para realizar esta selección vamos a utilizar la correlación entre cada una de las características y la variable a predecir.\n", "\n" ] }, { "cell_type": "code", "execution_count": 35, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 376 }, "id": "ec6UrovRcDv4", "outputId": "3bd0eef9-e554-4d5b-d6f2-7950438f929e" }, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 35, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAl4AAAINCAYAAADvO3HfAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAAB85ElEQVR4nO3dd1gUVxcG8HelI0WxAHZFBRtFkWJDwd5bEkWxoKixRmMvsfeCUeyigjX2WLBibwjYUSzYERFFxYKA7H5/EPdzXVYRh51dfH959nnYO3dmzk5ADvfcuSORyWQyEBEREVGOyyN2AEREREQ/CyZeRERERGrCxIuIiIhITZh4EREREakJEy8iIiIiNWHiRURERKQmTLyIiIiI1ISJFxEREZGaMPEiIiIiUhMmXkRERJSrLV68GD4+Pl/t8/LlS/z555+oXr06qlevjnHjxuH9+/eCx8LEi4iIiHKtNWvWYMGCBd/sN3DgQDx69Eje//Tp05g4caLg8egKfkQiIiIikcXHx2PMmDGIjIxE6dKlv9r34sWLOH/+PEJCQmBjYwMAmDRpEnr27IkhQ4bA0tJSsLg44kVERES5TlRUFMzNzbFr1y44ODh8tW9ERAQKFSokT7oAwMXFBRKJBJGRkYLGxREvIiIi0kheXl5f3R4aGqpym6enJzw9PbN0nvj4eFhbWyu06evrI1++fIiLi8vSMbKKiRd9l7Tnd8UOQSt0rPaH2CFohXVbu4odglaQvU4QOwTt8P6t2BFoBaOWQ3P8HNr2uyI5ORn6+vpK7QYGBkhJSRH0XEy8iIiISFjSdEEO87URLSEZGhoiNTVVqT0lJQXGxsaCnotzvIiIiOinZmVlhWfPnim0paam4tWrV4JOrAeYeBEREZHQZFJhXmpSvXp1PH36FA8ePJC3hYWFAQCqVq0q6LmYeBEREZGwpFJhXjkkPT0dCQkJ+PDhAwDAwcEBVatWxeDBg3HlyhWcO3cO48ePR+vWrTniRURERPQj4uLiUKtWLYSEhAAAJBIJAgICUKxYMXTt2hV//PEH6tSpgwkTJgh+bk6uJyIiIkHJ1FgmzIoZM2YovC9WrBhu3ryp0FagQIEsrXD/o5h4ERERkbBysEyo7VhqJCIiIlITjngRERGRsDSs1KhJmHgRERGRsARaQDU3YqmRiIiISE044kVERETCYqlRJSZeREREJCze1agSEy8iIiISlKat46VJOMeLiIiISE044kVERETCYqlRJSZeREREJCyWGlViqZGIiIhITTjiRURERMLiAqoqMfEiIiIiYbHUqBJLjURERERqwhEvIiIiEhbvalSJiRcREREJi6VGlZh4ERERkbA44qUS53gRERERqQlHvIiIiEhQMhmXk1CFiRcREREJi3O8VGKpkYiIiEhNOOJFREREwuLkepW0LvHy8fFB0aJFMWPGjEy329raYvr06Wjbtq3aYnry5AkuXryIZs2aAQA8PT3Rpk0bDBgw4LuO8/btW/z6669YtWoVrKysBI9zypQpKFasGLp16yb4scUWF5+Atl1+x9/T/4JLVXuxw1ELw7xG6DK6G6o3dIORiRFuRkZj9cQVeHz7UZaPMWzZKLx/8x6Lhv6t0G5WwBzewzrDobYTTPKZ4Mm9J/h36Xac2XNK6I+Ro05fvoWArYdwN/YZ8pvmxS9eLvBt4QGJRJJp/4/p6Qjaewo7j0fg2asklLQsCN+WHmjslvE9FZvwEk0Hz1Z5vpa1q2Jy7/Y58lly0umo+1i05zTuxiUiv4kR2te2h2/D6l+5TlIEh0Zgx5koJLx+ixKF8qNHo+poVM1Wod/hi7ex+lA47se/hImhPlxsS+CP1rVQwCyvOj6W4E5HP8KiAxG4G/8S+fMaob17BfjWc8j0Ov0bfgvjNx9XeaxJv3mgpXN5AMCRa/ex4vBF3E94hYKmxmhWtSx6eDpCT1cnxz5LjmOpUSWtS7w00YgRI1C0aFF54pVds2fPRsOGDXMk6QKAAQMGoFmzZqhXrx5KliyZI+cQw5On8eg9eCzevH0ndihqNXjhUJR1KIe109cg+W0yfhnUAeM3TMHg+v3w9vXbr+6bJ08edBvfE66N3XF0S6jCNl09XYxbOxF5zfJi07wNSHz6Aq5N3DFk0XDo6unixI5jOfiphHPp1gMMnLcWjdyqoH/7Brh46z4WbjkEqUwGv1b1Mt1n6fZQBO46jt5tPOFYviQOh1/DiIBN0MkjQQOXKiiUzxRrJ/RR2m/ToXM4cO4q2tR1zumPJbhLd59g0LJ/0ahqefRrXhMX78YiYPfpjOvU2DXTfZaGnMWqA+Ho1cQVjjZFEHrxDkasCkEeiQQNqmYkEwcv3MLwwL1oX6sK+rWoicSkd1i89yz8/t6KjSM7wUBPu379XLofj0FrDqKRQxn0a+SMi/efImB/eMZ18nJS6l+7QnEE92+p0CaTAZO2nsS7lFTUsisOADh76zH+DD6ERg42GNi0Ou7EJWLh/gi8fPcBo9rUVMtnI/XSru/8XOzhw4fYsWMHjh07lmPnMDc3R7NmzbBw4ULMmTMnx86jLlKpFP/uO4w5ASvFDkXtyle1RTWv6pjabSIuHo0EANw4H4VFp1agUZem2LZws8p9S9qVQo9JvWFjXxYpySlK26t5VUfpSmUwosUQxFy5AwC4evoyChUthNa/t9OaxGvpjiOwLWmNab//CgCo6VAeaelSrNp9HD5NasFQX09pn53HI9GkhgP6tPUCALhVLovo+3HYdOgcGrhUgb6eLuzLllDYJ+ruYxw4dxUDf22IqralcvxzCW1ZyDnYFiuEqd2aAABqViqFj+lSrD4YDh/PajDUV/418e/ZKDRxtkWfZu4AADe7koh+/Az/nLgsT7xW7AtDrUqlMbZjffl+paws0HnWRpy4elfeT1ssOxQJ2yIFMLVjRtJe0654xnU6ehk+darA8ItE0sLECBYmRgpt609ew71nrxDUv6V827/ht2CVzwRTO9aFTp48cC9fDInvPmDdiasY2tIdejpaOhWbD8lW6bv+jx4/fhxt27aFg4MD3N3dMXLkSLx+/Vq+PSYmBn5+fnByckKtWrXw559/IiEhQb7dx8cH06ZNw/Dhw+Ho6Ig6depg+fLlkMlk8j5HjhxBhw4d4OTkhCpVqqB9+/Y4c+ZMtj/g0aNH0bZtW9jb26NBgwaYP38+UlNT5dttbW2xefNmdO/eHfb29qhduzaWLVumcIzdu3ejSZMm8niCgoJga2sr/0znz5/Hjh074OnpKd8nISEBAwYMgKOjI1xdXTF9+nSkp6v+Rly9ejVcXV1hYWEhb0tMTMSIESPg6uqKatWqwc/PD/fv3wcALFy4EN26dUNwcDBq1aoFR0dHDBkyBAkJCRg+fDicnJzg4eGBHTt2KJynSZMm2LdvH54+fZrta6opbt25h8lzAtCqSX1MHzdU7HDUyrFOVSS/S8blExflbUmJSbgedg1V61X76r795/0BSR4JRrcehtfPXyltf//mPQ6u3y9Puj6Ju/sEliVzZjRWaKlpHxFx4y68nCsptDeoXhnvP6Tiws37me/38SNMjAwU2vKZGuP12+RM+8tkMkxdswtlihZC5ybaNzqRmvYREbcfw8uxrEJ7fadyeJ+ShgsxsZnv9zEdeb+8TiZGeP3uAwBAKpXBza4E2tWsotCnZOH8AIDHz19Dm6R+TEdETBy8KpdSaK9vXybjOt399r+nz5PeY9GBCPziXgFVShSWt6d9TIeRvi508vz/13E+Y0OkpUvxPiU1s0NpB5lUmFculOXEKzExEf3790e7du0QEhKCgIAAhIeHY9asWQCA+Ph4eHt7o3jx4ti6dSuWLl2Kt2/fokOHDnj//r38OBs2bICRkRG2bduGwYMHY9GiRVixYgUA4Nq1a+jXrx8aNmyIXbt2YcuWLShQoACGDh2qkCxl1YkTJzBo0CD88ssv2LNnD8aPH499+/Zh2LBhCv1mzZqF1q1b499//0W7du0wb948REREAMhI3EaMGIH27dtj165daNeuHebOnSvfd+HChXByckKTJk2wdetWefvWrVvh7OyM3bt3Y9iwYVizZo1SEvS50NBQ1Kv3//LHx48f4evri1u3bmHRokXYvHkzdHR04Ovri48fPwIAIiIiEBERgaCgIMyfPx8HDhxA8+bNUaFCBWzbtg116tTBX3/9hZcvX8qP6+joCDMzMxw/rnrugbawtiqMkH8CMXxgLxgaGoodjloVLVsMzx7GQ5qu+A/T0/txsC5d9Kv7BgyZj3HtR+JB9P1Mt189fRnLRy9WaNPR1UE1r+p4dPPhD8WtLo+fJSLtYzpKWhdQaC9hlfH+QdzzTPfzaVILu09dxOnLt/D2/QfsPX0JZ67cRvOajpn233f2MqLuPsbwzs0VfnFqi8cvXmdcp/8Sok9KFMoHAHgY/zKTvYDOnlWxJ+w6Tkfdx9vkFOw9fwNnrt9HM5cKAIA8eST4s50H6jnYKOwXeikjmbcpUkDpmJrs8YskpKVLUbKQuUJ7iQJmAICHWUgkFx+MRB6JBP0aK5ajO9SshIfPk7Dm2GUkJafgyoN4rD91DbXsisPc+Of6d+1nkeVSY3x8PFJTU1GkSBEULVoURYsWxdKlS+WjOBs3bkThwoXx119/yfeZP38+3NzcsH//fvlk9zJlymDChAmQSCSwsbFBTEwMgoOD4efnBx0dHYwdOxadOnWSH6NLly7w9fXFixcvYG1t/V0fbunSpWjfvj06duwIAChRogQmTpyIrl274vHjxyhWrBgAoE2bNmjVqhUA4I8//sCGDRsQGRkJZ2dnBAYGonHjxujRowcAoHTp0njw4AFWr14NAMiXLx/09PRgaGioMFrVoEEDdO3aFQBQvHhxBAcH49q1a2jfXnnibVxcHOLj41G+/P+H3s+dO4cbN25g3759KFOmDABg8uTJCAwMxKtXrwBklNqmTJkCMzMz2NjYoEKFCtDT00P37t0BAN26dcPmzZvx4MED5M///39Yy5cvj8uXL+O33377ruupaczNTGFuZip2GKLIa5YX79++V2pPfpcMoy/KG19SlXB9TdexvrAuXQQz/aZ+975iePM+Y+TFxEjxF5exoT4A4F0mJVYA6NjAHRei76Pv7DXyttYe1dCteZ1M+wftPQXH8iVRvWIZAaJWvzfvM65D3v+uyyfGBhnv337I/A/ejh6OuHgnFv0W//+PydbuldCtgeo5bg+evcT8HSdQoXhh1KpY+kdDV6s3yRnXIa/Bl9cpo1yt6jp9kvg2GXsib6OLRxWYfTFS6GxjjW517TF/73nM33seAGBXtACme3tmdijtwbsaVcpy4lWhQgU0b94cffr0gbW1NWrUqIG6devKy2vXr19HTEwMnJwUJxmmpKQgJiZG/t7FxUXhDhBHR0esWLECL1++RIUKFWBubo4VK1bg3r17uH//Pm7cuAEAXy3TqXL9+nVcuXJFYaTpU1kzJiZGnnjZ2Cj+VWZiYoK0tDQAQFRUFBo2bKiw3dnZWZ54qVK6tOI/LObm5khJyfwf+0/l2M8Tt5s3b8LMzEyedAFAoUKFMHLkSPn7AgUKwMzMTP7eyMhIITk1MMj4Af/yvBYWFnj+PPO/+EnzSCQSSPIo3jUlyZMnY6ZuJn1lUuX2H9FlTHc07d4C2xdtRfjBMEGPnVOk/12bzO/JQ6Z3oaWmfUT3Kcvx/NUbjO3eCqWLFMLFWw+w8t9jMDbQx4guLRT6X7z1ANEPnmD+4M5Ch682/79OmV+pzAbxUtM+wtd/M54nvcfYDl4oZWWBizGxCNx/HkYGehjxi/KNC3efvkCfhduhp6uDOT2bI08eVf9nNJP8OqkIO4+qDf/ZFhYNqUwG71qVlbZN2XYK/0bcgp+XE1zKFUFs4hssPRiJfiv3YVnvZjDKZI6dVsilZUIhfNf/0blz56Jfv344ceIEzpw5gyFDhqBq1aoIDg6GVCqFm5sbxo8fr7Sfqen/RyV0dRVP+SkR0tHRQXh4OHx9feHh4QFnZ2c0a9YMycnJ6NevX3Y+G6RSKXr27Ik2bdoobStUqJD8a319faXtn+LS1dWFNBuZu46O8m3Askx+UQL//yXw+XZdXV2Vt3J/oqenPDk4TxbKHenp6VnqR5rhl0Ed8OvgjgptZ/eehnnpIkp9DY0N8f6N8khYdugZ6KH/nEGo2bIOdi7Zhg2zggU5rjqY/leiefvFyNb7/0YmTI0NlPY5HH4Ntx4+xbKRvnCrnDHnyblCGZgaG2J60G60rVcd5Yr/f47b4fPXYJbXCLUcbJWOpS1M/xt9effFiM2nuUUmhplcp0t3cCv2OZYOaAs3u4y7o53LFYOpkQFmbD6KtjWqoFzRgvL+4bceYcjy3chrqI+lA9uhaEFzpWNqOlOj/0ZKla5Txh/oJobKv0M+d/jKPbiXL6o02T7+9TtsPx+NHp6O8hJkdRugcvFCaD93G/4Nv4kONStldkjNxxEvlbKceF26dAkhISEYPXo0ypQpg27dumHXrl0YNmwYXrx4gXLlyiEkJATW1tbyRObVq1cYMWIEunfvDjc3NwDA1atXFY574cIFFCtWDObm5ggMDISrqysCAgLk29euXQtAddLyNeXKlcPdu3cVlk44f/48goKCMGHCBBgbG3/zGHZ2drh8+bJC25fvf5SlpSWAjHl0n0bfypYti9evX+PBgwfy+BMTE9GoUSMsXbr0h8738uVLpRE50lyHNhxAZGi4Qlv1Rq5wqOOUMcL12c+GVSlrPL7z4/OwjE2NMXrNeJSvaos1kwOxZ+W/P3xMdSpe2AI6efLgUfwLhfaHTzPelylaWGmfuP9uNHAsr7jUSjW7jJ+VmNhnConXiYvRqFetolavtVS8UD7o5JHgYcIrhfZP78tYK8/FiktMAgA4llGcS+hcLqOCcPfpC3niFRIejb/WHkDJwvmxuF8bWObXzqkBxQuYZVynF0kK7Z/el7HMn9luAID4V29x88kLdK6tPNr19OVbyGSAYynFm1bKWlkgn7EBYp5mPseOtFuWhz1MTEywYcMGzJ49Gw8ePMDNmzexd+9elCpVCvnz54e3tzfevHmDIUOG4MaNG4iOjsaff/6JK1euoFy5cvLjREREYMGCBbh37x62bt2K9evXo2fPngAAa2tr3Lx5ExEREXj8+DG2bduGv//OWNgxO5Pr/fz8cPDgQSxcuBD37t3D2bNnMWrUKCQlJSmMeH3rGAcOHMDq1avx4MED7NixQ54MfpI3b17ExsZm+07BwoULo0iRIoiKipK3ubu7o3Llyhg+fDguX76M27dvY9SoUShQoACqVKnylaN9nVQqRXR0NBwcHLJ9DFKvl88SEXP1jsLr8olLMDY1hqPH/0v7ZhZmqOhaGZdPXPqh8+XRyYORgeNgY18W/v1ma13SBQAG+nqoalcKoRFRConpofBrMDU2RGWb4kr7lLLO+DfhyzseL916AAAoWuj/v1xfv32Ph/EvlJI0bWOgp4uqZYvhyOU7Ctfp8MXbMDUyQOVM7mItbZkxJeLLOx4vxTwBABQpkDGidfLaPYwL3g+HMkWw5s/ftDbpAv67TqWtcOTqfcXrdOUuTI30UbmE6t8n1x5lTCX5MrkCgOIFMxK6C/cUf3fcf/YKr96noIiF9l4zSKXCvHKhLCdeZcuWxcKFC3Hu3Dm0bt0a3t7e0NXVxYoVK5AnTx4UL14c69atQ3JyMry9vdG5c2dIJBIEBQWhQIH//9Xk5eWF27dvo1WrVli6dClGjhwpn/w+cOBAODo6ok+fPmjdujW2bNmCadOmwdDQEFeuXPnuD9e4cWP4+/sjNDQULVq0wNChQ+Hu7q4wovYtderUwcSJE7F+/Xo0a9YM//zzDzp06KBQ5uvQoQNu3bqFli1bZmsuGpBxXc6dOyd/nydPHixevBhFihRBjx490LFjR+jq6iIwMDDT0mhWRUVF4d27dwp3UJL2uXE+CtfOXsHAv/+EV4cGcGnkhr/WT8b7pHc4uG6fvF+xcsVRutL3Tfxu3KUZKrpWwtHNh/E87jnKOdkqvLSFX6t6uBrzGMMWbsSpyzcRsOUQgvaeRM+WdWGor4e37z/gyp2HSEzKWGy2brUKqGJTHKOXbMbmw+dw/noMAncdx9wN++DhZIcqnyVrtx9l/KK0yWTkTNv4NXbB1ftxGBa4F6ei7mHR7jMIOhyBHo1cYKivi7fJKbhyLw6J/5WwPezLoEopK4xZsw+bT1xG+K1HWHXwPOZuPwGPKhnbUtI+YtKGQzA20EfPRi649zQRV+7FyV/xL9+I/Km/n199J1x99AzD1oXiVPQjLNofgaDjV9DD0xGGerp4+yEVVx7EI/GLpUduxyVCX1cHxQuaKR3TwsQInWpVRvCxy1iwLxzhd57g3/Bb6LtyH6zzmaCdq526Pp7gZLJ0QV65kUSWnRpeNn3rcT+a6Pz58yhYsKDCJPelS5di69atOHz4sGDnuXfvHlq2bIkjR45keTQuOyZMmID379/LlwH5XmnP7wockTDOX7gC3wEjsGrhTI14ZFDHan/k+DnymuVF13E94NLIDRKJBDcjb2DNpEA8ufv/kYiJm6aiULHC6FvLL9NjLD61AlHnrik8MmjS5umo6Kp6Xkn7ki1Vbvte67Z2FexYmQkNj8KS7aG4H5eAwvnN8FsDN3RtWhsAEH79LnpOW4lJvdqhVZ2Mtc/evv+AhVsOIjQ8Cq/fJaNYofxoXqsqujStCb3P5qceOHcFwwM2YeesP1C6SM4nX7LXCd/u9AOOXLqDJXvP4v6zlyhsnhe/1XFEl/oZ1yT81iP4/b0VEzs3RCv3jO+Lt8kpCNh9BqGXbuP1uw8oVtAczV0rwMezGvR0dXD+5kP0WrBN5fl6N3XD7/8tviqo919/YsOPOnL1HpYcvID7Ca8yrlONiujikfHvTXjME/gt3YuJv3qgVfX/36E+dfspHLl6H6HjM78JQyaTYf2pa9h69gZiE9+goJkx3MsXQ//GzkpzwoRi1DLn1z1MPrFGkOMY1ekmyHE0CROvb/D398euXbswffp0lChRAjdu3MDYsWPh7e393c9i/JYxY8agYMGCGDx4sKDH/SQxMRHNmjXDxo0bUapUqWwdQ1MTL02jjsQrN8jpxCu3yOnEK9fI4cQrt1BL4nVslSDHMarrK8hxNImW3qeqPv369cO7d+8wfPhwJCYmwtraGt26dZPPSxPSyJEj8csvv6BDhw7fvWZZVixcuBA9e/bMdtJFRESUJVxOQiW1jniR9uOIV9ZwxCtrOOKVNRzxyiKOeGWJWka8jgrzDF2jesIPcoiNI15EREQkrFx6R6IQmHgRERGRsFhqVInLlxMREVGuI5VKsWDBAtSuXRsODg7w9fXFgwcPVPZPSEjAkCFD4OrqCldXVwwaNCjb63N+DRMvIiIiEpYGLKC6ePFibNq0CVOmTME///wDiUQCPz8/lQuyDx48GHFxcVi9ejVWr16Np0+fom/fvj8UQ2aYeBEREZGwZFJhXtmUmpqKVatWYcCAAfDw8ICdnR38/f0RHx+PQ4cOKfVPSkpCeHg4/Pz8ULFiRVSsWBG9evVCVFQUXr4U9tFNTLyIiIhIWCKPeEVHR+Pdu3fy50QDgJmZGSpWrIjw8HCl/gYGBjA2NsbOnTvx9u1bvH37Fv/++y9KlSoFc3NhH+zOyfVERESkkby8vL66PTQ0NNP2T3OzvlwTs3DhwoiLi1Pqb2BggKlTp2LSpElwdnaGRCJBoUKFsG7dOuTJI+wYFUe8iIiISFgij3glJ2c8M/PLZxsbGBggJSVFqb9MJsPNmzfh5OSE9evXIygoCEWLFkW/fv3w9q2w68NxxIuIiIiEJdByEqpGtL7F0NAQQMZcr09fA0BKSgqMjJSfgbl3715s2LABR48ehYmJCYCM5zLXq1cP27ZtQ9euwi32zBEvIiIiylU+lRifPXum0P7s2TNYWVkp9Y+MjETp0qXlSRcAmJubo3Tp0rh//76gsTHxIiIiImGJXGq0s7ODiYkJwsLC5G1JSUm4fv06nJ2dlfpbW1vjwYMHCmXI5ORkPH78GCVLlsx2HJlh4kVERETCEnk5CX19fXTu3Blz5sxBaGgooqOjMXjwYFhZWaFBgwZIT09HQkICPnz4AABo3bo1AOCPP/5AdHS0vL++vj7atm0rxBWRY+JFREREuc7AgQPRvn17jB07Fh07doSOjg4CAwOhr6+PuLg41KpVCyEhIQAy7nbcsGEDZDIZunbtiu7du0NPTw8bN26EmZmZoHFJZDKZTNAjUq6W9vyu2CFohY7V/hA7BK2wbqtwE1ZzM9nrBLFD0A7vhb37LLcyajk0x8+RvGOGIMcxajNSkONoEt7VSERERMLiQ7JVYqmRiIiISE044kVERETC+sEHXOdmTLyIiIhIWEy8VGLiRURERMLifXsqcY4XERERkZpwxIuIiIiExVKjSky8iIiISFhMvFRiqZGIiIhITTjiRURERMLiAqoqMfEiIiIiYbHUqBJLjURERERqwhEvIiIiEhbX8VKJiRcREREJi6VGlVhqJCIiIlITjnjRd+lY7Q+xQ9AKGyPnix2CVjAqUlvsELSCmYGx2CFohaSU92KHoBU+pg7N+ZNwxEslJl5EREQkLC4noRITLyIiIhKUTMrJ9apwjhcRERGRmnDEi4iIiITFOV4qMfEiIiIiYXGOl0osNRIRERGpCUe8iIiISFicXK8SEy8iIiISFud4qcRSIxEREZGacMSLiIiIhMURL5WYeBEREZGwZJzjpQpLjURERERqwhEvIiIiEhZLjSox8SIiIiJhcTkJlZh4ERERkbC4cr1KnONFREREpCYc8SIiIiJhsdSoEhMvIiIiEpSMk+tVYqmRiIiISE044kVERETCYqlRJSZeREREJCze1agSS41EREREasIRLyIiIhIWS40qccSLiIiIhCWVCvP6oRCkWLBgAWrXrg0HBwf4+vriwYMHKvunpaVh7ty5qF27NhwdHdG5c2fcuHHjh2LIDBMvIiIiynUWL16MTZs2YcqUKfjnn38gkUjg5+eH1NTUTPtPmDABW7duxeTJk7Ft2zbky5cPfn5+ePPmjaBxMfEiIiIiYUllwryyKTU1FatWrcKAAQPg4eEBOzs7+Pv7Iz4+HocOHVLq/+jRI2zduhXTp09H3bp1YWNjg2nTpkFfXx/Xrl37kSuhhIkXERERCUsmFeaVTdHR0Xj37h3c3NzkbWZmZqhYsSLCw8OV+p86dQpmZmaoU6eOQv8jR47A3d0923FkhpPriYiISFgCTa738vL66vbQ0NBM258+fQoAsLa2VmgvXLgw4uLilPrfv38fxYsXx8GDB7F8+XLEx8ejYsWKGDlyJGxsbLIZfeY44kVERES5SnJyMgBAX19fod3AwAApKSlK/d++fYuHDx9i8eLFGDJkCJYsWQJdXV14e3vjxYsXgsaWaxIvHx8fjBw5UpBj2draYvv27Sq3jxw5Ej4+PvL3t2/fxrFjx7K8vypPnz5FkyZN8O7du+/eNyv69OmTaW2biIhISDKpVJBXaGjoV1+qGBoaAoDSRPqUlBQYGRkp9dfT08ObN2/g7++PWrVqwd7eHv7+/gCAHTt2CHhlclHiJaRTp06hadOmWe7fu3dvXL169YfPO3bsWPj6+iJv3rw/fKzMjBw5EpMnT8arV69y5PhEREQARJ9c/6nE+OzZM4X2Z8+ewcrKSqm/lZUVdHV1FcqKhoaGKF68OB4/fpztODLDxCsThQoVkmfL6nLu3DlERUWhTZs2OXaOUqVKwd7eHqtXr86xcxAREYnNzs4OJiYmCAsLk7clJSXh+vXrcHZ2Vurv7OyMjx8/KgyifPjwAY8ePULJkiUFjU2QxOv48eNo27YtHBwc4O7ujpEjR+L169fy7TExMfDz84OTkxNq1aqFP//8EwkJCfLtPj4+mDZtGoYPHw5HR0fUqVMHy5cvh0z2/2z3yJEj6NChA5ycnFClShW0b98eZ86cyVJ8bdq0wZQpU+TvDx8+DFtbW+zdu1feNmvWLHh7ewNQLBXKZDIsXrwYderUgaOjI8aMGaNQH/b09ERsbCwCAgIUyo/37t1D9+7dYW9vj1q1amHZsmVfjXHVqlVo1KgRdHX/f7/Do0eP0K9fP1SrVg2urq4YPHgwnj9/DiBj9GrUqFHw9/eHq6srqlWrhsmTJ+Pp06fo06cPHBwc0LBhQxw/flzhPE2aNMHGjRvx4cOHLF07IiKi7ybyiJe+vj46d+6MOXPmIDQ0FNHR0Rg8eDCsrKzQoEEDpKenIyEhQf670NnZGTVq1MCIESMQERGBO3fuYPjw4dDR0UGrVq2EuioABEi8EhMT0b9/f7Rr1w4hISEICAhAeHg4Zs2aBQCIj4+Ht7c3ihcvjq1bt2Lp0qV4+/YtOnTogPfv38uPs2HDBhgZGWHbtm0YPHgwFi1ahBUrVgAArl27hn79+qFhw4bYtWsXtmzZggIFCmDo0KEqF0L7nKenJ06fPi1/f/bsWUgkEpw7d07edvz48Uzvnli+fDlWrlyJ4cOHY/v27TAxMUFISIh8+9atW2FlZQVfX18sXLhQ3r5u3Tq0atUKe/fuhbe3N+bNm4ezZ89mGl9ycjLOnDmDevXqydvevHkDb29vvH//HmvWrMGaNWsQGxuLAQMGyPvs3r0bb968webNmzFq1CisW7cO7du3R+PGjbF9+3aUKVMGI0eOVEhgPTw8kJSUhIiIiG9eNyIiomwReTkJABg4cCDat2+PsWPHomPHjtDR0UFgYCD09fURFxeHWrVqKfw+X7hwIVxcXNC/f3+0b98eb9++RXBwMCwsLH70aij44eUk4uPjkZqaiiJFiqBo0aIoWrQoli5divT0dADAxo0bUbhwYfz111/yfebPnw83Nzfs378fbdu2BQCUKVMGEyZMgEQigY2NDWJiYhAcHAw/Pz/o6Ohg7Nix6NSpk/wYXbp0ga+vL168eKF0u+iX6tWrh4CAAMTFxcHa2hpnzpxBgwYN5EOQjx8/xp07d1C/fn2F/WQyGdauXYsuXbqgefPmAIBRo0YpDF1aWFhAR0cHxsbGyJcvn7y9Y8eOaN26NQCgb9++WLVqFa5du5bpeiBRUVFIS0uDra2tvC0kJEQ+0e/TcadOnYp///1XPuJmZmaGMWPGQEdHByVLlsTcuXPh5uYmP6+3tzeOHj2K58+fo1ChQgAAExMTFCtWDJcvX0atWrW+et2IiIi0lY6ODoYNG4Zhw4YpbStWrBhu3ryp0GZiYoIJEyZgwoQJORrXDydeFSpUQPPmzdGnTx9YW1ujRo0aqFu3Ljw9PQEA169fR0xMDJycnBT2S0lJQUxMjPy9i4sLJBKJ/L2joyNWrFiBly9fokKFCjA3N8eKFStw79493L9/X/78pE8J3tdUrlwZlpaWOH36NGrXro2HDx9izpw5aNu2LeLi4nD8+HGULVtWqY778uVLJCQkoEqVKgrtjo6OCrFnpnTp0grvzczMMr2FFYC87Pp5Vn3z5k2UKlVKIZkrV64chg4dKn9fokQJ6OjoyN8bGRmhePHi8vcGBgYAoHReCwsLecmSiIhIcHxItkqCLKA6d+5c9OvXDydOnMCZM2cwZMgQVK1aFcHBwZBKpXBzc8P48eOV9jM1Nf1/ILqKoXwqj+no6CA8PBy+vr7w8PCAs7MzmjVrhuTkZPTr1y/LMdatWxenT5+Gjo4OKleujEqVKqFo0aIICwvDsWPHlEa7MotFVayZ+TwhUnWcTz4lnNLPHgiqq6urkIhmRk9PT6ktT55vV4/T09MzjY+IiEgIMiZeKv3wHK9Lly5h2rRpKFOmDLp164bly5dj2rRpCAsLw4sXL1CuXDnExMTA2toaJUuWRMmSJWFubo5p06bh1q1b8uN8uRzDhQsXUKxYMZibmyMwMBCurq4ICAhAt27dULNmTfnKs6qSmS95eXnh7NmzOH36tPwRAm5ubjh27BjCwsIynd9lYWEBa2trREZGKrQL/dwmS0tLABnz5T4pW7Ys7t+/r/BwzuvXr8PV1RWxsbE/dL6XL1/KS49ERESkPj+ceJmYmGDDhg2YPXs2Hjx4gJs3b2Lv3r0oVaoU8ufPD29vb7x58wZDhgzBjRs3EB0djT///BNXrlxBuXLl5MeJiIjAggULcO/ePWzduhXr169Hz549AWSsx3Hz5k1ERETg8ePH2LZtG/7++28AyoujqeLm5oaUlBQcOHBAnni5u7vjwIEDMDc3VyonfuLn54f169djy5YtuHfvHubPn48rV64o9MmbNy/u37+f7fKdra0tDAwMEBUVJW9r0aIFzM3NMWzYMERHR+PatWuYMGECypcvj6JFi2brPEBG0vXkyRM4ODhk+xhERERfJfJdjZrshxOvsmXLYuHChTh37hxat24Nb29v6OrqYsWKFciTJw+KFy+OdevWITk5Gd7e3ujcuTMkEgmCgoJQoEAB+XG8vLxw+/ZttGrVCkuXLsXIkSPRsWNHABl3Jjg6OqJPnz5o3bo1tmzZgmnTpsHQ0FApCVLFwMAANWrUAABUrVoVQEbiJZPJ4OnpqbKs16lTJwwbNgxLlixBq1atcPv2bbRv316hj4+PD44dOwZfX9/vvn4AYGxsjBo1aijcZWlkZITAwECkp6ejY8eO6NGjB2xsbLBgwYJsneOTsLAwmJubZ7qOCRERkSCkUmFeuZBEltVaXQ7y8fFB0aJFMWPGDLFDEc3Zs2fxxx9/4OTJk0rPlhKSn58fKleujEGDBmVr//YlWwocUe60MXK+2CFoBaMitcUOQSuYGRiLHYJWSEp5/+1OhI+pPzZdJSve9G0iyHFMF+8T5DiahCvXawh3d3dUqFABO3fuzLFz3LlzB1FRUejWrVuOnYOIiIhUY+KlQaZOnYrAwMAce0j27Nmz8ddff8Hc3DxHjk9ERASAc7y+QpDlJH7U2rVrxQ5BIxQtWhQHDhzIseN/67FFREREQtCAWUwaiyNeRERERGqiESNeRERElIvk0jKhEJh4ERERkbCYeKnEUiMRERGRmnDEi4iIiATFZzWqxsSLiIiIhMXESyWWGomIiIjUhCNeREREJKzc+ZhFQTDxIiIiIkFxjpdqTLyIiIhIWEy8VOIcLyIiIiI14YgXERERCYtzvFRi4kVERESC4hwv1VhqJCIiIlITjngRERGRsFhqVImJFxEREQmKpUbVWGokIiIiUhOOeBEREZGwWGpUiYkXERERCUrGxEsllhqJiIiI1IQjXkRERCQsjnipxMSLiIiIBMVSo2pMvIiIiEhYTLxU4hwvIiIiIjXhiBcREREJiqVG1Zh4ERERkaCYeKnGUiMRERGRmnDEi4iIiATFES/VmHgRERGRsGQSsSPQWEy86Lus29pV7BC0glGR2mKHoBWSn5wUOwStIEt+I3YI2iEtRewIiL6JiRcREREJiqVG1Ti5noiIiAQlk0oEef0IqVSKBQsWoHbt2nBwcICvry8ePHiQpX13794NW1tbPH78+IdiyAwTLyIiIsp1Fi9ejE2bNmHKlCn4559/IJFI4Ofnh9TU1K/uFxsbi4kTJ+ZYXEy8iIiISFAyqTCv7EpNTcWqVaswYMAAeHh4wM7ODv7+/oiPj8ehQ4dU7ieVSjFs2DBUqlQp+yf/BiZeREREJCiZTCLIK7uio6Px7t07uLm5ydvMzMxQsWJFhIeHq9xv6dKlSEtLQ+/evbN97m/h5HoiIiISlFCT6728vL66PTQ0NNP2p0+fAgCsra0V2gsXLoy4uLhM97ly5QpWrVqFrVu3Ij4+PhvRZg1HvIiIiChXSU5OBgDo6+srtBsYGCAlRXnZkffv32Po0KEYOnQoSpUqlaOxccSLiIiIBPWjdyR+ompE61sMDQ0BZMz1+vQ1AKSkpMDIyEip/5QpU1CqVCl06NAhe4F+ByZeREREJCiZTNzzfyoxPnv2DCVKlJC3P3v2DHZ2dkr9t23bBn19fTg5OQEA0tPTAQDNmzdHy5YtMWnSJMFiY+JFREREuYqdnR1MTEwQFhYmT7ySkpJw/fp1dO7cWan/wYMHFd5fvnwZw4YNw/Lly2FjYyNobEy8iIiISFBClRqzS19fH507d8acOXNgYWGBokWLYvbs2bCyskKDBg2Qnp6OxMREmJqawtDQECVLllTY/9Pk/CJFiqBAgQKCxsbJ9URERCQoTVi5fuDAgWjfvj3Gjh2Ljh07QkdHB4GBgdDX10dcXBxq1aqFkJAQgT5x1klkMrErsaRNPoRvEzsErWBSc6DYIWgFPiQ7a/iQ7CziQ7KzRL+MS46f475jA0GOU+qS6sVOtRVLjURERCQoDumoxsSLiIiIBCX2HC9NxjleRERERGrCES8iIiIS1I88ZzG3Y+JFREREghLqWY25ERMvIiIiEpSUI14qcY4XERERkZpwxIuIiIgExTleqjHxIiIiIkFxOQnVWGokIiIiUhOOeBEREZGguHK9aky8iIiISFAsNarGUiMRERGRmnDEi4iIiATFdbxUY+JFREREguJyEqqx1EhERESkJqInXk+ePMHevXuz3H/79u2wtbVVuX3kyJHw8fERIrQsS0tLw5o1a+TvFy5cCE9Pz2wda/r06QrHEtLVq1fRvn17fPz4MUeOT0REBGTc1SjEKzcSPfEaMWIETp48KXYYP2TPnj2YPn36Dx/n4sWLOHr0KDp16iRAVMqqVKmCMmXKYMWKFTlyfCIiIiBjjpcQr9xI9MQrN5AJlJbPnTsXnTp1gp6eniDHy4yvry9WrFiB169f59g5iIjo5yaTSQR55UY/lHjZ2tpi48aN6NixI+zt7dGiRQuEhoYq9Dl69Cjatm0Le3t7NGjQAPPnz0dqaioAwMfHB+fPn8eOHTvkpbmnT59i6NChqFGjBipVqgQPDw/4+/tDKpVmK8b4+HgMHjwYzs7OcHV1RZ8+fXD//n359pEjR2LYsGGYOXMm3N3d4eDggL59+yIhIUHe5+HDh/Dz84OTkxNq1aqFVatWoUGDBti+fTu2b9+OUaNGya9HWFiYfL8VK1bAw8MD9vb28PHxUTjvl65evYrIyEg0bdpUoX3t2rVo1KgR7O3t0bRpU/z7778AgMePH8PW1hbHjx9H27ZtUaVKFbRo0QKXLl3Cli1bUK9ePVStWhV//vknUlJS5Mezs7ODpaUl/vnnn2xdTyIiIsq+Hx7xmjVrFpo3b46dO3fCw8MD/fv3x4ULFwAAJ06cwKBBg/DLL79gz549GD9+PPbt24dhw4YByJgL5eTkhCZNmmDr1q0AgN69eyMxMRGBgYHYv38/evbsiaVLl+LIkSPfHdv79+/h4+OD9PR0rFu3DmvXrkX+/Pnx66+/Ij4+Xt5v3759ePXqFdatW4eAgABERkbC398fAJCcnIxu3bpBKpVi48aNmD9/Pnbs2IFHjx4BAJo2bYrRo0cDAE6dOgUnJycAQGxsLCIjI7Fs2TKsW7cOCQkJGDNmjMpYDx8+jMqVK6NQoULytsDAQMyZMwc9evTAnj170KlTJ4waNQqnT5+W95k0aRKGDh2KnTt3wtDQEL169cK+ffuwdOlSzJgxAwcOHMCWLVsUzlW3bt1sXU8iIqKs4Bwv1X54OYl27drJ5yQNHToU4eHhWLduHapWrYqlS5eiffv26NixIwCgRIkSmDhxIrp27YrHjx+jWLFi0NPTg6GhISwsLPDhwwe0atUKjRo1QtGiRQFkjIotX74cN2/eRP369b8rtr179+Lly5eYO3euvHw3depUhIWFYfPmzRgwYAAAwMTEBJMmTYKenh5sbGzQqlUrHD9+HAAQEhKCxMREbN++Hfny5QMAzJkzBy1btgQAGBoawtTUFAAUkiZdXV3Mnj1bvq1Dhw7yZC4zly5dQvny5RXa1qxZgy5duuDXX38FAHTq1AkfPnxAenq6vE/37t1Ro0YNAEDr1q0xadIkjB8/HiVLloStrS0qVqyIW7duKRzX1tYWwcHBkEqlyJOH1WYiIhJWbp2fJYQfTrxcXFwU3js4OODMmTMAgOvXr+PKlSvYsWOHfPun+VAxMTEoVqyYwr6Ghobo3Lkz9u/fj6CgIDx48ADR0dF49uxZtkqN169fx9u3b5ViTElJQUxMjPx9yZIlFeZVmZqaIi0tTX6M0qVLy5MuICNx+ZRQqVKwYEGFPmZmZvjw4YPK/s+fP4e9vb38fWJiIp49ewYHBweFfj169ACQUWoEgNKlS8u3GRkZAQCKFy8ubzMwMJCXdj+xsLDAx48f8erVK1hYWHz1cxAREZFwfjjx0tVVPMTnoyhSqRQ9e/ZEmzZtlPb7fHTok+TkZHTq1AnJyclo0qQJWrVqhXHjxmX7Lj+pVIrSpUtjyZIlStuMjY3lX+vr66s8ho6OTraSPh0dne/qL5FIFM7zKSaJ5Ot/NXx5/QF8cxTr03k42kVERDkht06MF8IP/+a9evWqwvtLly6hUqVKAIBy5crh7t27KFmypPwVHx+PWbNm4d27d0rHOnnyJKKiorB27VoMHDgQTZs2hYmJCV68eJGtOwfLly+PJ0+ewNTUVH7+okWLYu7cuQgPD8/SMezs7PDgwQO8evVK3nb37l28efNG/v5byVFWWFpaIjExUf7exMQEhQsXVrq+AwcOxJQpU37oXImJidDX11cYxSMiIhIKl5NQ7YcTr6CgIOzevRv37t3DzJkzER0dja5duwIA/Pz8cPDgQSxcuBD37t3D2bNnMWrUKCQlJclHvPLmzYvY2Fg8ffoUVlZWAIBdu3YhNjYWERER6Nu3L9LS0pTKZVnRsmVLmJubo3///rh06RJiYmIwatQoHD9+HOXKlcvSMZo3b478+fNj2LBhiI6OxqVLl+Q3B3xKuD6Nnl27du2r5cSvsbe3R1RUlEJbr169EBQUhJ07d+Lhw4dYv349QkNDv3uu25euX7+uVMIkIiKinPfDiddvv/2G1atXo2XLloiIiEBgYCDs7OwAAI0bN4a/vz9CQ0PRokULDB06FO7u7ggICJDv36FDB9y6dQstW7ZEpUqVMGrUKAQHB6NJkyYYNWoUqlevjubNm+Py5cvfHZupqSnWrVuHAgUKoGfPnmjfvj1iY2MRGBiY5cRLX18fK1euRGpqKn799VcMGDAAbdu2BQD5vDA3Nzc4ODigQ4cOOHr06HfHCQD169fH7du3FUa9OnfujH79+mHBggVo1qwZNm3aBH9/f7i5uWXrHJ+cO3cOXl5eP3QMIiIiVWQCvXIjiewHVv+0tbXF9OnT5YlIbvT48WPcv38ftWrVkrfFx8ejTp06WL9+PZydnQU7V6dOneDp6SmfQJ8Trly5gu7du+PIkSMwNzf/7v0/hG/LgahyH5OaA8UOQSskP9Hup1aoiyz5zbc7EZCW8u0+BP0yLt/u9IPOWLcT5Dg14nLf7xzOrv6GlJQU9OrVC4GBgXj06BGuX7+OcePGoVSpUoKX6/744w9s3LgxW2XVrFqzZg18fX2zlXQRERHRj2Hi9Q02NjaYN28edu/ejebNm6N79+4wNjbG6tWrBX+0T/Xq1VG3bl2sXbtW0ON+cuXKFdy/fx+9evXKkeMTEREBfGTQ1/xQqZF+Piw1Zg1LjVnDUmPWsNSYRSw1Zok6So0nrdoLcpzaT7cKchxN8sPreBERERF9TobcOVolBJYaiYiIiNSEI15EREQkKCknManExIuIiIgEJWWpUSWWGomIiIjUhCNeREREJChOrleNiRcREREJSip2ABqMpUYiIiIiNeGIFxEREQmKpUbVOOJFREREgpIK9PqhGKRSLFiwALVr14aDgwN8fX3x4MEDlf1v376NXr16wdXVFe7u7hg4cCCePHnyg1EoY+JFREREuc7ixYuxadMmTJkyBf/88w8kEgn8/PyQmpqq1Pfly5fo3r078ubNi3Xr1mHFihV4+fIlevbsiZQUYR9FxcSLiIiIBCX2iFdqaipWrVqFAQMGwMPDA3Z2dvD390d8fDwOHTqk1P/w4cNITk7GjBkzUK5cOVSuXBmzZ89GTEwMLly48AORKGPiRURERIKSQSLIK7uio6Px7t07uLm5ydvMzMxQsWJFhIeHK/V3d3fHokWLYGBgoLTt9evX2Y4jM5xcT0RERIKSCjS33svL66vbQ0NDM21/+vQpAMDa2lqhvXDhwoiLi1PqX6xYMRQrVkyhbdmyZTAwMED16tW/J+Rv4ogXERER5SrJyckAAH19fYV2AwODLM3ZCg4OxoYNGzBkyBAUKFBA0Ng44kVERESCEupZjapGtL7F0NAQQMZcr09fA0BKSgqMjIxU7ieTyfD3339jyZIl6N27N7p165at838NR7yIiIhIUDKBXtn1qcT47NkzhfZnz57Bysoq033S0tIwbNgwLF26FMOHD8eQIUN+IALVmHgRERFRrmJnZwcTExOEhYXJ25KSknD9+nU4Oztnus/w4cOxf/9+zJ07Fz169Mix2FhqJCIiIkGJ/axGfX19dO7cGXPmzIGFhQWKFi2K2bNnw8rKCg0aNEB6ejoSExNhamoKQ0NDbN++HSEhIRg+fDhcXFyQkJAgP9anPkLhiBcREREJSiqRCPL6EQMHDkT79u0xduxYdOzYETo6OggMDIS+vj7i4uJQq1YthISEAAD27NkDAJg1axZq1aql8PrURygSmUz2I2VU+sl8CN8mdghawaTmQLFD0ArJT06KHYJWkCW/ETsE7ZAm7ArjuZV+GZccP8dW606CHKd93HpBjqNJWGokIiIiQXFERzUmXkRERCQosed4aTLO8SIiIiJSE454ERERkaCEemRQbsTEi4iIiAQl1Mr1uRETLyIiIhIUJ9erxjleRERERGrCES8iIiISFOd4qcbEi76L7HXCtzsRzAyMxQ5BK3Bh0KyRGJmKHYJWkEnTxQ6B/sPlJFRjqZGIiIhITTjiRURERILi5HrVmHgRERGRoDjHSzWWGomIiIjUhCNeREREJChOrleNiRcREREJiomXaiw1EhEREakJR7yIiIhIUDJOrleJiRcREREJiqVG1Zh4ERERkaCYeKnGOV5EREREasIRLyIiIhIUV65XjYkXERERCYor16vGUiMRERGRmnDEi4iIiATFyfWqMfEiIiIiQTHxUo2lRiIiIiI14YgXERERCYp3NarGxIuIiIgExbsaVWOpkYiIiEhNOOJFREREguLketWYeBEREZGgOMdLNSZeREREJCgpUy+VOMeLiIiISE044kVERESC4hwv1Zh4ERERkaBYaFSNpUYiIiIiNeGIFxEREQmKpUbVmHgRERGRoLhyvWoaW2p88uQJ9u7dm+X+27dvh62trSDnHjlyJHx8fFRuDwsLg62tLR4/fgwAeP/+PdavX5/l/b9mwIABOHToULb2/ZaDBw+ib9++OXJsIiIi+jaNTbxGjBiBkydPinLuMWPGYOHChVnuv2rVKgQGBv7weUNCQvDixQs0aNDgh4+VmYYNGyIpKQm7du3KkeMTEREBGet4CfHKjTQ28RKTqakp8uXLl+X+MtmPf3Okp6dj3rx58PX1/eFjfU337t0xf/58fPz4MUfPQ0REPy+ZQK8fIZVKsWDBAtSuXRsODg7w9fXFgwcPVPZ/+fIl/vzzT1SvXh3Vq1fHuHHj8P79+x+MQlmOJF62trbYuHEjOnbsCHt7e7Ro0QKhoaEKfY4ePYq2bdvC3t4eDRo0wPz585GamgoA8PHxwfnz57Fjxw54enoCAJ4+fYqhQ4eiRo0aqFSpEjw8PODv7w+p9NtT+IKDg+Hi4oL09HQAGYmSq6srevToIe9z584d2Nra4uHDh0qlwoiICPzyyy+wt7dH69atcfPmTfm2hQsXIiAgALGxsQrlx7S0NMycORPu7u5wdHRE37598fz5c5UxHjx4EK9fv0adOnXkbR8/fsTChQvh6ekJBwcHtG3bFidOnACQUe6sWLEizp07h6ZNm6JKlSr47bffcO/ePSxZsgQ1atSAi4sLJk+erJAY1q5dG0lJSThw4MA3rxsREVF2SAV6/YjFixdj06ZNmDJlCv755x9IJBL4+fnJc40vDRw4EI8ePcKaNWuwYMECnD59GhMnTvzBKJTl2IjXrFmz0Lx5c+zcuRMeHh7o378/Lly4AAA4ceIEBg0ahF9++QV79uzB+PHjsW/fPgwbNgxARjLj5OSEJk2aYOvWrQCA3r17IzExEYGBgdi/fz969uyJpUuX4siRI9+MxdPTE69fv8a1a9cAAFFRUXj9+jUuXLiAtLQ0AMCxY8dQrlw5lChRQmHfR48ewdfXFxUqVMCOHTvw+++/Y9GiRfLtvr6+8PX1hZWVFU6dOgVra2sAwMWLF/H69WusX78ey5Ytw6VLlzBr1iyVMR4+fBg1a9aEvr6+vG3atGlYv349hg4dit27d8PDwwN9+/bFnTt3AGSMks2YMQPTpk3D5s2b8eLFC3To0AExMTFYu3YthgwZgnXr1uHYsWPyY+rr66NGjRpZum5ERETaKDU1FatWrcKAAQPg4eEBOzs7+Pv7Iz4+PtN51BcvXsT58+cxffp0VKpUCe7u7pg0aRL+/fdfxMfHCxpbjiVe7dq1Q6dOnVCmTBkMHToU9vb2WLduHQBg6dKlaN++PTp27IgSJUqgVq1amDhxIvbv34/Hjx8jX7580NPTg6GhISwsLPDhwwe0atUKkydPRoUKFVC8eHH4+PigcOHCCqNPqhQrVgzly5fHqVOnAABnzpyBh4cH0tLScOXKFQDA8ePH4eXlpbTv5s2bUbBgQYwfPx42NjZo1KgRfv/9d/n2vHnzwtjYGDo6OihUqBB0dHQAAIUKFcLkyZNRpkwZuLq6omnTpvLELzOXL19G+fLl5e/fvn2LzZs3448//kDTpk1RokQJDBo0CD169MC7d+/k/QYNGgRHR0dUqFABDRs2xLt37zB58mTY2NigQ4cOKFiwIG7fvq1wLltbW1y+fPmb142IiCg7xJ7jFR0djXfv3sHNzU3eZmZmhooVKyI8PFypf0REBAoVKgQbGxt5m4uLCyQSCSIjI7MdR2ZybDkJFxcXhfcODg44c+YMAOD69eu4cuUKduzYId/+qRwWExODYsWKKexraGiIzp07Y//+/QgKCsKDBw8QHR2NZ8+eZanUCGSMep05cwb9+vXD2bNnUb9+fSQlJeHcuXMoX748Lly4IB9x+9ytW7dQsWJFeUIFAFWrVv3m+UqUKKGwj7m5OT58+KCy//Pnz2FhYSF/f+/ePaSlpcHR0VGh3+DBgwFklBoBoHTp0vJtRkZGKFiwIIyMjORtBgYGSElJUTiGhYXFV8ueREREP0KoafGZDYh87stpTJ88ffoUAORVqE8KFy6MuLg4pf7x8fFKffX19ZEvX75M+/+IHEu8dHUVDy2VSpEnTx751z179kSbNm2U9itUqJBSW3JyMjp16oTk5GQ0adIErVq1wrhx49CpU6csx1OvXj0EBgYiMTERkZGRGDduHF68eIGwsDCULl0aBQoUQJUqVTLd98vJ819+tsx8nnRlhUQiUUgi9fT0srTfl7F8usZf8/n/CyIiotwmOTkZABSm7wAZgxGvX7/OtP+XfT/1/3Lw4kflWOJ19epV+cR4ALh06RIqVaoEAChXrhzu3r2LkiVLyrefP38eQUFBmDBhAoyNjRWOdfLkSURFReH06dMoWLAgAODVq1d48eJFlu8odHBwgLm5OZYuXQpzc3OUKVMGNWrUwMqVK5E/f354enpCIlFe8a1ChQrYvn07UlNT5f9Trl69qtAns/2+V+HChZGYmCh/X7JkSejp6eHq1auws7OTt7dv3x6NGzdWmSRmRWJiYqYJLhERkRCEWrle1YjWtxgaGgLImOv16WsASElJUagKfd4/s0n3KSkpSjnJj8qxYY+goCDs3r0b9+7dw8yZMxEdHY2uXbsCAPz8/HDw4EEsXLgQ9+7dw9mzZzFq1CgkJSXJE4K8efMiNjYWT58+hZWVFQBg165diI2NRUREBPr27Yu0tDSVdyd8SSKRwMPDAxs3bpTXfB0cHKCjo4ODBw+ifv36me7XsWNHJCcnY/To0YiJicHRo0cREBCg0MfY2BivX7+Wlwezw8HBAVFRUfL3RkZG6Ny5M/7++2+Ehobi4cOH8Pf3x507d1CvXr1sneOTqKgopRImERGRUMSe4/WpbPjs2TOF9mfPnslzis9ZWVkp9U1NTcWrV69gaWmZ7Tgyk2OJ12+//YbVq1ejZcuWiIiIQGBgoHzkpnHjxvD390doaChatGiBoUOHwt3dXSGh6dChA27duoWWLVuiUqVKGDVqFIKDg9GkSROMGjUK1atXR/Pmzb9rkrinpydSU1PliZeenh6cnZ1hbGysNCftE0tLSwQFBeHp06do06YNZsyYoTC5HshYmLRQoUJo2bIlrl+//r2XCgBQv359REZGKiRuQ4YMQevWrTFhwgS0aNECZ8+exfLlyxUm/32vtLQ0XLhw4Zt1cyIiIm1lZ2cHExMT+XxoAEhKSsL169fh7Oys1L969ep4+vSpwjpfn/bNyrzu7yGRCbH65xdsbW0xffp0tG3bVuhD51ofP35Eo0aNMGzYMDRu3DjHzhMSEoK5c+fiwIEDWZqr9qXkw0tzIKrcp2ibuWKHoBWeXt8qdghaQWJkKnYIWkH27pXYIWgF/ZLCJhKZGVyqgyDH8b+/Kfv7+vtj06ZNmDZtGooWLYrZs2fj8ePH2L17N3R0dJCYmAhTU1MYGhpCJpPB29sbKSkpmDBhAt6/f4/Ro0fD1dUV06dPF+SzfMIZ1hpCV1cXAwcOxJo1a3L0PMHBwRgwYEC2ki4iIqKs0IQFVAcOHIj27dtj7Nix6NixI3R0dBAYGAh9fX3ExcWhVq1aCAkJAZAxHSkgIADFihVD165d8ccff6BOnTqYMGHCD0ahjL99NUirVq1w4MAB7N+/P0dGvfbt24d8+fKhdevWgh+biIhIk+jo6GDYsGGZLhVVrFgxpXVACxQogAULFuR4XDmSeGVlUVPK3OLFi3Ps2E2aNEGTJk1y7PhEREQAIMulD7gWAke8iIiISFBCLSeRGzHxIiIiIkH9yFIQuR0n1xMRERGpCUe8iIiISFAc71KNiRcREREJiqVG1VhqJCIiIlITjngRERGRoHhXo2pMvIiIiEhQXMdLNZYaiYiIiNSEI15EREQkKJYaVWPiRURERIJiqVE1lhqJiIiI1IQjXkRERCQolhpVY+JFREREgpLKWGpUhYkXERERCYppl2qc40VERESkJhzxIiIiIkHxWY2qMfEiIiIiQXE5CdVYaiQiIiJSE454ERERkaC4nIRqTLyIiIhIUJzjpRpLjURERERqwhEvIiIiEhQn16vGxIuIiIgExTleqrHUSERERKQmHPEiIiIiQcn4rEaVmHgRERGRoHhXo2pMvIiIiEhQnOOlGud4EREREakJR7zo+7x/K3YEWiEp5b3YIWiHtBSxI9AKMmm62CFoBUnefGKHQP/hchKqMfEiIiIiQXGOl2osNRIRERGpCUe8iIiISFBcTkI1Jl5EREQkKN7VqBpLjURERERqwhEvIiIiEhTvalSNiRcREREJinc1qsZSIxEREZGaMPEiIiIiQclkMkFeOSklJQUTJ06Eu7s7nJycMHDgQLx48eKr+1y4cAE+Pj6oVq0aateujTFjxuDVq1ffdV4mXkRERCQoKWSCvHLShAkTcPr0aSxcuBBBQUF49OgRBg0apLL/vXv30KNHD9jZ2WHLli3w9/fHlStXMHDgwO86LxMvIiIiEpRMoP9ySnx8PHbu3ImxY8fC2dkZ9vb2mDdvHsLDw3Hp0qVM99m5cycKFy6M0aNHo0yZMnB2dsb48eMRFhaGR48eZfncTLyIiIjopxIZGQkAcHV1lbeVLl0alpaWCA8Pz3Sfli1bYubMmZBIJErbvqfcyLsaiYiISFBSgeZneXl5fXV7aGhoto4bHx+P/Pnzw8DAQKG9cOHCiIuLy3QfGxsbpbYVK1agUKFCsLOzy/K5mXgRERGRoMReTOLx48dfTdoGDRoEfX19pXYDAwOkpKRk6RwzZszA8ePHsWDBAujp6WU5NiZeREREpJGyO6JlaWmJkJAQlduPHz+O1NRUpfaUlBQYGRl99dhpaWn466+/sGPHDowfPx4NGzb8rtiYeBEREZGgxF5AVU9PL9PS4Cc3b97Eq1evkJqaqjDy9ezZM1hZWanc7+3bt+jfvz8iIiIwd+5cNGvW7Ltj4+R6IiIiEpSmLydRrVo1SKVS+SR7ALh79y7i4+Ph7Oyc6T6pqano3bs3rl69ipUrV2Yr6QKYeBEREdFPxtLSEs2aNcPYsWMRFhaGK1eu4M8//4SLiwscHR0BZCRaCQkJ8pLksmXLEBkZiUmTJsHGxgYJCQnyV2ZlS1VYaiQiIiJB5fSq80KYPHkypk2bhv79+wMA6tSpg7Fjx8q3X7x4EV26dEFwcDBcXV2xZ88eyGQyDBkyROlYn/pkhUSmDVeHNEbyrjlih6AVTNv7ix2CVngfvUPsELSDDv9GzgpJ3nxih6AV9AqWyfFzuBTxEOQ4558cF+Q4moSlRiIiIiI14Z9RREREJKicfNyPtmPiRURERILiLCbVmHgRERGRoMRex0uTcY4XERERkZpwxIuIiIgExVKjaky8iIiISFAsNarGUiMRERGRmnDEi4iIiATF5SRU44iXmjx58gR79+4VOwxERkYiIiJC7DCIiCgXk8pkgrxyIyZeajJixAicPHlS7DDg7e2Nhw8fih0GERHRT4mlRiIiIhIUS42qccRLDXx8fHD+/Hns2LEDnp6eePr0KYYOHYoaNWqgUqVK8PDwgL+/P6RSKQBg+/bt8PT0xNSpU+Hs7Iw+ffoAAK5du4ZOnTrBwcEBXl5e2LVrFypWrIiwsDAAGbfvrlixAl5eXnBwcECrVq2wa9cueRy2trYAgFGjRmHkyJFqvgpERPSzYKlRNY54qcHChQvRp08fWFlZ4a+//kL37t1RoEABBAYGwsTEBMeOHcOUKVNQpUoV1K9fHwAQGxuL+Ph47NixAx8+fEB8fDy6du0KLy8vTJw4EbGxsZgwYQLS09Pl5/H398fu3bvx119/wcbGBuHh4ZgwYQLevHmDTp064dSpU6hVqxZGjx6Ntm3binU5iIiIfloc8VKDfPnyQU9PD4aGhjA2NkarVq0wefJkVKhQAcWLF4ePjw8KFy6MmzdvKuzXt29fFC9eHOXKlcM///wDMzMzTJ06FWXLloWHhwfGjRsn7/v+/XusWbMGI0aMQL169VCiRAm0a9cO3bp1Q2BgIACgUKFCAABTU1OYmpqq7wIQEdFPRSbQf7kRR7zUzNDQEJ07d8b+/fsRFBSEBw8eIDo6Gs+ePZOXGj8pVaqU/Ovr16+jUqVK0NPTk7c5OzvLv75z5w5SUlIwYsQIjBo1St7+8eNHpKam4sOHDzA0NMy5D0ZERPSf3FomFAITLzVLTk5Gp06dkJycjCZNmqBVq1YYN24cOnXqpNT380RJR0dHKTH73KfHM8yfPx9lypRR2q6vry9A9ERERN+WW0erhMDES81OnjyJqKgonD59GgULFgQAvHr1Ci9evPjqs63s7OywdetWpKWlyUe9Ll++LN9epkwZ6Orq4smTJ6hXr568PTg4GHfu3MGkSZNy6BMRERFRVnGOl5rkzZsXsbGxyJ8/PwBg165diI2NRUREBPr27Yu0tDSkpqaq3N/b2xtv3rzBuHHjEBMTg7Nnz8qTKYlEAlNTU3To0AHz58/Hzp078ejRI+zYsQOzZ8+WJ3gAYGxsjJiYGLx8+TJnPzAREf20eFejahzxUpMOHTpgxIgR6NevH0aMGIHg4GDMnz8flpaWaNq0KaytrRVGsL5UoEABrFy5EtOmTUOrVq1gZWWFjh07YtasWfIRsFGjRsHCwgILFizAs2fPYGVlhf79+6NXr17y4/j6+mLlypW4e/culixZkuOfm4iIfj4sNaomkX2tvkUa486dO3j9+jWqVasmb7tw4QI6duyIY8eOwdraWi1xJO+ao5bzaDvT9v5ih6AV3kfvEDsE7aDDv5GzQpI3n9ghaAW9gsrzgIVWpqCTIMe5+/yiIMfRJCw1aon4+Hh06dIFO3fuRGxsLC5evIjp06fDxcVFbUkXERFRVshkUkFeuRH/jNISNWvWxJgxY7Bs2TKMGzcOpqam8PT0xNChQ8UOjYiISIGUpUaVmHhpEW9vb3h7e4sdBhEREWUTEy8iIiISFKePq8bEi4iIiATFUqNqnFxPREREpCYc8SIiIiJBsdSoGhMvIiIiElRuXXVeCEy8iIiISFBcuV41zvEiIiIiUhOOeBEREZGgOMdLNSZeREREJCguJ6EaS41EREREasIRLyIiIhIUS42qMfEiIiIiQXE5CdVYaiQiIiJSE454ERERkaBYalSNiRcREREJinc1qsZSIxEREZGaMPEiIiIiQclkMkFeOSklJQUTJ06Eu7s7nJycMHDgQLx48SLL+y9ZsgS2trbffV4mXkRERCQoqUwmyCsnTZgwAadPn8bChQsRFBSER48eYdCgQVna98qVKwgICMjWeZl4ERERkaBkAv2XU+Lj47Fz506MHTsWzs7OsLe3x7x58xAeHo5Lly59dd/3799j2LBhcHZ2zta5mXgRERHRTyUyMhIA4OrqKm8rXbo0LC0tER4e/tV9p06divLly6NVq1bZOjfvaiQiIiJBCVUm9PLy+ur20NDQbB03Pj4e+fPnh4GBgUJ74cKFERcXp3K/Q4cO4fjx49i9ezeOHj2arXMz8SIiIiJBib2O1+PHj7+atA0aNAj6+vpK7QYGBkhJScl0n/j4eIwbNw6zZs1C/vz5sx0bEy8iIiLSSNkd0bK0tERISIjK7cePH0dqaqpSe0pKCoyMjJTaZTIZRo4ciSZNmqBOnTrZiukTJl5EREQkqJycGJ8Venp6sLGxUbn95s2bePXqFVJTUxVGvp49ewYrKyul/k+ePMGZM2dw4cIF7Ny5EwDw8eNHAICTkxN69+6NPn36ZCk2Jl5EREQkKLFLjd9SrVo1SKVSREZGwt3dHQBw9+5dxMfHZ3q3oqWlJQ4ePKjQdvDgQcyZMwc7d+6Eubl5ls/NxIuIiIh+KpaWlmjWrBnGjh2LadOmwcjICOPHj4eLiwscHR0BAKmpqXj9+jXMzc2hr6+PkiVLKhyjQIECAKDU/i1cToKIiIgEpQ0r10+ePBnu7u7o378/evTogTJlymDBggXy7RcvXkStWrVw8eJFQc8rkWn6eCBplORdc8QOQSuYtvcXOwSt8D56h9ghaAcdFieyQpI3n9ghaAW9gmVy/By6+kUFOc7H1FhBjqNJOOJFREREpCYc8SIiIiJSE454EREREakJEy8iIiIiNWHiRURERKQmTLyIiIiI1ISJFxEREZGaMPEiIiIiUhMmXkRERERqwsSLiIiISE2YeBERERGpCRMvIiIiIjVh4kVERESkJky8iIiIiNSEiRcRERGRmuiKHQARCS81NRURERG4fPkynj9/jjx58qBQoUJwdHSEs7Mz8uTh31xfevnyJXR0dGBmZiZ2KESUi0lkMplM7CCIsio+Ph4nTpzINKGoW7cu8ufPL3aIokpKSsLq1auxYcMGJCUloVixYrCwsEB6ejoSExMRFxcHc3NzdO7cGV27doWpqanYIYvmzZs32LRpE44cOYKrV68iPT0dAKCvrw97e3vUr18fbdu2/amvEQBIpVLs378fR44cwZUrV5CQkAAdHR35z52Xlxc8PT2ZzAMYNWoUxowZAxMTE4X2V69eYcyYMVi0aJFIkZEmYeJFWuHRo0dYsGAB9u3bBzMzM5QvX14hobh58ybev3+P5s2bo1+/fihevLjYIavd4cOHMXnyZNjb26NJkybw8PBA3rx5Ffq8evUK586dw/bt23H9+nWMHz8eDRo0EClicUilUixfvhwrVqyApaUl6tatC1tbW1hYWEAqlSIxMRFRUVEICwvD06dP0aNHD/j5+UFHR0fs0NVuz5498Pf3R1JSEmrUqCG/Tunp6Xj58iWioqJw4cIFmJmZYeDAgWjRooXYIatdZGQkHj16BEB14hUTE4N169bh4sWLYoRIGoaJF2m8oKAgrFy5Es2aNUOTJk3g4OCg1Ecmk+HatWvYvn07Dh48CD8/P3Tr1k39wYro999/x8iRI1GyZMks9b9z5w7mzJmDpUuX5nBkmuWXX36BjY0NunfvDltb26/2vXLlCoKCgnD//n1s27ZNTRFqhn79+iEpKQndu3dH7dq1oaenl2m/jx8/4uDBgwgODkb+/PmxZMkSNUcqrgsXLsDb2xsAIJFIkNmvVGNjY/j6+qJ///7qDo80EBMv0nhTpkzBwIEDszz3JjExEQEBAfjrr79yODLSRteuXUPlypW/a5+rV6+iSpUqORSRZjp06NB3j4bu378fjRs3zqGINJ+dnR1OnTqFggULih0KaTAmXkS52IsXL5CSkqLUXqRIERGiIfp5pKamQl9fX+wwSAPxrkbSOq9fv8aDBw8yTSiqV68uQkSa58SJExg1ahQSExMV2mUyGSQSCW7cuCFSZJolNTUVW7Zswe3btzP9fpo+fboIUWmm48eP49atW5leJ5bQ/m/jxo1YsWIFnj59igMHDmDlypUoVKgQrxHJMfEirbJz506MHz8eqampSnMpmFD839SpU2Fvbw9vb28YGBiIHY7GGjVqFA4ePIiKFStydOIrpkyZgnXr1qFgwYJK10kikTCp+M/u3bsxd+5cdO3aFStXrgQA2NjYYM6cOTAwMICfn5/IEZImYKmRtErdunVRp04ddOvWLdOEomjRoiJEpXkcHR2xfft2lClTRuxQNFq1atUwc+ZM1K9fX+xQNJqrqyuGDBmC3377TexQNFqbNm3QpUsXtGnTBk5OTti1axeKFy+OHTt2YMmSJTh48KDYIZIG4IgXaZXXr1/D19cXpUqVEjsUjebm5oaoqCgmXt9gbm6e5btAf2a6urpwcXEROwyNd+/ePTg7Oyu1Ozs74+nTpyJERJqIiRdplYYNG+L48eNMvL5h4sSJaN++PU6dOoVixYpBIpEobGdpKMPvv/+OGTNmYMKECT/l2m9Z1blzZyxZsgRTpkxhSfYrChYsiLt37yp9L124cAGFCxcWKSrSNCw1klZ5/vw5mjVrhrJly6J48eJKCQUnQ2f4NCfHwsIChoaGCtskEglCQ0NFikyzXL58Gb169UJSUlKm2zlnMMP9+/fx22+/4f379yhUqJDSzx2/nzKsWLECGzduxMiRIzFixAjMnz8fsbGx+Pvvv9GtWzf8/vvvYodIGoAjXqRVpk+fjnfv3iE1NRWxsbFih6Oxdu7ciWnTpqFt27Zih6LRxo4di5IlS6J169YwMjISOxyNNXLkSJiZmaF9+/a8Tl/h5+eHN2/eYNiwYUhJSUHv3r2hq6uLDh06oHfv3mKHRxqCI16kVZycnDB//nx4eHiIHYpGc3V1xdatW1k++wZ7e3v8+++/KF26tNihaDR7e3ts3rwZdnZ2YoeiFZKTk3Hnzh3IZDKUKVNG6RFC9HPjiBdplbx586JEiRJih6HxOnTogKCgIIwZM0apLET/V65cOcTHxzPx+obixYsjNTVV7DA0Xnh4uFLbjRs3IJFIoKenBysrK1haWooQGWkSjniRVlm7di0iIiIwdepU/hX5FaNHj8bu3bthbm6OEiVKQFdX8W+s4OBgkSLTLMePH8eUKVPQvXt3lC5dWuk6cUHeDGFhYZg5cyYGDRqU6XXikxAyVKpUCVKpFADk6wx++YePi4sLFi5cmOVHoFHuw8SLtEr37t0REREBmUyGAgUKKP0C4CTfDKNGjfrqdt6EkOFrpTMuyPt/lSpVQnp6OgDFRIJPQlC0e/du+Pv7Y9y4cfJlJS5duoTJkyejY8eOcHBwwIwZM2BnZ4dJkyaJHC2JhYkXaZWAgICvbucyCYo+f15cXFwcrK2tRY5Is3zrBg0uyJvh/PnzX93ONb4yNGjQAH/99Rdq166t0H727FmMHz8eBw8exMWLFzFgwACcOnVKpChJbJzjRVqlf//+kEqlePXqFSwsLAAAFy9eROXKlaGnpydydJrjxYsXGDhwIKpVq4YhQ4YAAFq3bo0KFSrg77//hrm5ucgRaoaiRYvi7NmzSE9PR61atQBkPG6pYcOGLDN+xsXFBTExMXj37h3s7e0BAKtWrULdunW5SO9nEhISMi27Fi5cWL6AqqWlJd68eaPu0EiD5BE7AKLv8eDBAzRs2BArVqyQt/Xu3RutW7dGXFyciJFplqlTp+Ljx49o1aqVvG316tVITk7GrFmzRIxMs+zatQt+fn64ffu2vC0+Ph7du3fH4cOHRYxMs5w8eRJt2rTBkSNH5G0hISFo164dIiIiRIxMs1SuXBkrV66Ul2UBID09HStXrpSXtc+fP49ixYqJFSJpAJYaSav06tULefLkwZQpU1CwYEEAQGJiIkaNGgUDAwMsWLBA5Ag1g6urK4KCgpTmMF27dg1+fn44e/asSJFplubNm8Pb2xve3t4K7evXr8fmzZvx77//ihSZZmnbti3q1q2LgQMHKrT7+/sjLCwMmzZtEikyzXLt2jV069YNZmZmqFy5MqRSKaKiovDmzRusXLkSUqkUXbp0wdixY9GhQwexwyWRcMSLtMqFCxcwYsQIedIFABYWFhg6dCjOnTsnYmSaJT09XX531ed0dXWRkpIiQkSa6dGjR0rzcQCgTp06uH//vvoD0lB3795FmzZtlNrbt2+PmzdvihCRZqpcuTJCQkLQunVr+fIb7dq1w8GDB+Ho6Ahzc3MsX76cSddPjnO8SKvo6uri5cuXSusuJScnixSRZnJ1dcXcuXMxf/58mJqaAgDevn2LBQsWcO7SZ6ytrREWFpbps/UKFSokUlSax8LCAtevX1e6Trdv3+ayCJ/5/fffMXToUKWRwU9sbGxgY2Oj5qhI0zDxIq3i4eGBKVOmwN/fHyVLlgSQMWoxbdq0TEcuflYjR46Et7c36tSpI09S79+/j3z58iEwMFDk6DRHp06dMHXqVDx69AgODg6QSCS4evUqgoKC0K9fP7HD0xht2rTBxIkTkZSUBHt7e/l1mj9/fqYjYT+riIgIGBgYiB0GaTjO8SKtkpiYCF9fX9y8eVP+l3ZSUhIqVaqEJUuWcJTiM2/fvsXevXtx69Yt6OrqomzZsmjRooXSQ7N/doGBgQgKCsKzZ88AZNyB1qtXL3Tu3FnkyDRHeno6pk6dii1btuDjx4+QyWTQ1dWFj48P/vzzT6X19H5W06ZNw7Nnz9CvXz+ULFlSvpQL0eeYeJHWSU9Px9mzZxUSCnd395/+0TjJycnf/QDj7OyTW718+RJ6enp8IsJXvHv3Dvfu3YOuri5KlSrFJP4Lnp6eePLkicp/i7jQLAFMvEgLxMfHf/fzzbKzj7Zr2rQpevbsidatWyNPnq/fN5OWloadO3di1apV2Ldvn5oi1AxBQUHo3LkzdHR0stT/48ePWLt2Lbp3757DkWmWw4cPo379+t+1z8GDB9GwYcMcikjz7dix46vbWZYlgHO8SAv4+vqicePG8PHxQb58+b7a98WLFwgODsahQ4cQEhKingA1xKpVqzBu3DjMnTsXjRo1Qr169VC+fHlYWFhAJpMhMTER165dw7lz57B3715UqFDhp5zvFRcXh6ZNm8LHxwdNmjRBgQIFMu338uVL7Ny5Exs3bkTdunXVG6QG2LVrF9asWYMePXqgdu3aKsuJ6enpCA0NxerVq1GgQIGfOvFiYkVZwREv0njv3r3D7NmzsXPnTri6uiokFFKpFImJiYiKisK5c+dw5swZtGzZEiNGjPhpS0bnzp3D6tWrcebMGXz8+FFhm76+PmrUqIEuXbrA3d1dpAjFd/HiRcybNw8XLlxA5cqVM/1+un37NqpWrYpBgwbJn7v3swkJCYG/vz+SkpJQs2bNTK9TREQEzMzMMGDAALRs2VLskEV35MgR3Lx5U2ER1dTUVFy+fBlBQUEiRkaagokXaY2HDx9izZo1CA0NRXx8vHwehUwmQ5EiReDp6YnOnTujVKlS4gaqIT58+IBr167h+fPnkEgksLS0hK2tLed0febq1as4fPgwLl++LL9OhQsXhoODA+rXr4+KFSuKHaLopFIpDhw4gNDQUJXXydPTM8ul29zM398fy5YtQ+HChZGQkABLS0s8f/4c6enpaNasGebMmSN2iKQBmHiRVnr69CkSEhLkCQXvZiQisdWtWxd+fn7o1KkT6tatiw0bNsDY2Bj9+vWDi4sLBg0aJHaIpAG4cj1pJSsrK1SpUgWVK1dm0kVEGuH58+fw8PAAANjZ2eHKlSvIly8fBg8e/NPNOSXVmHgREREJwNzcHO/evQMAlCxZEnfu3AEAFClSBPHx8WKGRhqEiRcREZEA3N3dMWvWLMTFxaFy5crYt28fEhMTceDAAeTPn1/s8EhDMPEiIiISwLBhw/DixQscOHAAjRo1goGBAWrWrIlZs2aha9euYodHGoKT64lyqbCwMFy7dg0fPnzAlz/m/fv3FykqotzLy8sLW7duhbGxMQwMDPDhwwecPHkS+fPnR79+/RAWFiZ2iKQBuIAqaRWZTIYdO3aoTCimT58uUmSaZfny5Zg3bx5MTU1hamqqsE0ikTDx+s+DBw8wduxY+ffTl/iIlwyJiYmYNWuWyp+70NBQkSITX0hICE6ePAkAiI2NxeTJk5UelB0bG/vNp0nQz4OJF2mVmTNnYs2aNbC1tZU/JJuUrV27FoMGDcLvv/8udigabezYsXj+/DkGDRoEc3NzscPRWOPGjcOlS5fQtGlTXqcvODk5YdOmTfJk9MmTJ9DT05Nvl0gkMDY2xowZM8QKkTQMS42kVdzc3DB8+HC0bdtW7FA0moODA/bu3YtixYqJHYpGc3BwwPr161G5cmWxQ9FoTk5OWL58OapXry52KBrNx8cHixYt4h+F9FUc+yStkpKSAldXV7HD0HjVqlXDxYsXxQ5D4+XPn1+pLETKjIyMuF5eFqxdu5ZJF30TR7xIqwwcOBAuLi7o3Lmz2KFonJ07d8q/vnv3LtatW4dOnTqhVKlSSo9zad26tXqD01ArV65EZGQk5syZg7x584odjsaaNWsWkpKSMHnyZPmjuogoe5h4kcYLCAiQf52QkIBt27ahYcOGKFWqlNKE1Z950ridnV2W+kkkkp960rinp6dC8hAbGwtdXV0UKlRI6fvpZ5403qVLF/nXHz9+xIULF1CoUCGULFlS6ToFBwerOzwircXJ9aTxtm/frvC+cOHCuHTpEi5duqTQ/rPfrRcdHS12CFqhTZs2HLXJgqJFiyq8L1mypEiREOUuHPGiXEMqlfKW7f906dIFixYtUlpK4sWLF+jRo4dCWfJnFh4eDkdHR4W70ICMuYTHjh1Do0aNRIpMszx58gRWVlZKP18fP37E9evXYW9vL1JkRNqHI16kVby8vLBt2zbky5dPoT0+Ph4tW7b8qRcoPH78OK5evQoAOH/+PJYsWQJjY2OFPg8ePEBsbKwY4WmkLl264PTp07CwsFBov3PnDoYNG8bE6z9eXl6ZXqfHjx/Dx8cHly9fFikyIu3DxIs03pcLFE6aNCnTBQp/9vJR0aJFMWnSJMhkMkgkEoSEhCiMUHxaT2j48OEiRim+NWvWYObMmQAyFuStWbNmpv1+9lGc9evXY9WqVQAyrlO7du2URrySkpJQpEgRMcIj0lpMvEjjZXWBwk+/TH9WZcuWlU8G9/T0xNatW5VGKAjo3Lkz8uXLB6lUitGjR2PUqFEKJdlP309ubm4iRim+tm3b4uXLl5DJZFi0aBEaN26sdOdn3rx50bBhQ5EiJNJOnONFWsXHxwcBAQFcPZsEsWPHDjRr1gz6+vpih6LRAgIC0KNHDxgZGYkdCpHWY+JFlEt8fvv/t/D2/wxfu8lAX18flpaWcHR0VFoH7Wfw5MmTLPdluZEo61hqJK1iZ2enci6Xnp4erKys0KpVK/Tt2/enm/P1+e3/KSkpCAkJQYUKFeDo6AhdXV1cvXoVV69exS+//CJilJpl8eLFePz4MaRSqbzc+ObNG0gkEnlpu3Tp0li9ejWsrKzEDFXtvlzv7Gt+5nXhiL4XR7xIqwQFBWHevHnw9vZGtWrVAACXL1/GunXr0KFDB5ibmyM4OBjdu3eHn5+fyNGKZ+zYsTAxMcHIkSMV2ufPn4+YmBgsXLhQpMg0y+rVq7F9+3bMnTsX5cuXB5Cx6v+n54HWr18fY8eOhampKebOnStytOp1/vx5+dc3b95EQEAA+vbtCycnJ+jp6eHKlStYtGgR+vbtiw4dOogYKZGWkRFpkV9++UW2adMmpfbt27fLOnToIJPJZLLDhw/LGjZsqO7QNIqjo6Ps3r17Su337t2TOTg4qD0eTVW7dm1ZZGSkUvvFixdlderUkclkMllUVJTM1dVV3aFplDZt2sgOHjyo1H7kyBFZ48aNRYiISHtxtUnSKtHR0ZnebVatWjVERUUBACpWrIi4uDh1h6ZRzMzMcP36daX2iIgIFChQQISINNObN29gYmKi1G5oaIjXr18DyLiWKSkp6g5No8TExKBs2bJK7SVKlPjpf9aIvhcTL9IqxYoVw9GjR5Xajxw5Ip+D8/Dhw59+GYXffvsNf/31FxYuXIijR4/iyJEjmD17NqZMmQJfX1+xw9MYzs7OmD17Nt68eSNvS0pKwrx58+Dk5AQAOHjwIEqXLi1WiBrB1tYWwcHB8nlvQMaq9cuWLUOVKlVEjIxI+3COF2mV3bt3Y+TIkWjcuDGcnJwglUpx+fJlHDhwABMnTkTVqlXRo0cP1K9fH6NHjxY7XFEtW7YM69atQ0JCAgDA2toafn5+8Pb2FjkyzfHo0SN07doVL1++ROnSpSGTyXD//n3kz58fK1euRFxcHHr16gV/f/+fer2qiIgI9OjRA4UKFULFihUhk8lw9epVJCcnIygoKMsPaCciJl6khY4ePYpVq1YhKioKurq6sLW1Ra9evVC7dm2Eh4fj1KlT6N+/v9Lz935WL1++BADkz59f5Eg004cPH7B3717cuHEDOjo6sLOzk6/tFRsbi5SUFJQpU0bsMEX36NEjbN68Gbdv3wYAVKhQAR07dkThwoVFjoxIuzDxIsoldu7ciaZNm0JfX/+bD8Fu3bq1WmIiIiJFTLxI60RGRiIyMhJpaWn48tu3f//+IkUlPjs7O5w+fRoFChT4aulHIpFw3aX/vHjxAv7+/iq/nz49guln1KVLFwQEBMDMzAw+Pj5fXdOLC/ISZR0XUCWtsnz5csybNw/m5uZKz42TSCQ/deIVHR2d6dek2l9//YWIiAi0bt1a4XmNlLEg76eHYhctWvSnW5CYKKdwxIu0Sp06ddCuXTsMGjRI7FA0WufOnVG/fn14eXmhePHiYoejsRwdHbFo0SLUrFlT7FA02qNHj/h9RCQQLidBWuX169ecn5QF9vb22Lp1Kxo0aIAWLVpg/vz5uHr1qthhaRxjY2NYW1uLHYbG+/R9tGDBAvl6eUSUPRzxIq3i6+uLtm3bonnz5mKHohUeP36M48eP4/jx4zh//jzMzc3h5eWFv/76S+zQNMKsWbPw+vVrTJo06ad8EHZW3blzB8ePH8exY8dw8eJFFCxYEJ6enqhfvz5cXFygq8tZK0RZxcSLtMqWLVswe/ZstG3bFmXKlIG+vr7Cdo6GKYuPj8fZs2dx/PhxHDhwAAAyXdX+ZzR8+HDs27cPpqamKFGihNL3EyeNK3v79i1OnjyJAwcO4NChQ8ibN6/Ccx2J6Ov4ZwpplXHjxgEA1qxZo7RNIpEw8fpPSEgIzp07h7CwMDx8+BDW1tZwd3fHrFmz4O7uLnZ4GkNHR4ejp1mUnp6OK1euyL+vLl68CH19fTg6OoodGpFW4YgXUS5kZ2eHPHnyoH79+vj9999RoUIFsUMiLebn54fIyEh8/PgRVapUgZubG9zd3eHo6MgyI9F3YuJFWunJkyeIiYlB9erV8e7dOz74+QuHDx/GuXPncObMGTx8+BCVK1eGm5sb3NzcULVqVaWS2s/s2bNn2Lx5M+7du4fRo0fj/PnzKF++PGxsbMQOTWO4uroiKSkJHh4e8PDwgLu7O0qVKiV2WERaiYkXaZXU1FSMGDEC+/btQ548eXDgwAHMnDkTb968QUBAANdiysSnOV6nT5/GwYMHIZFIcOnSJbHD0ggPHjzAr7/+ChMTE8THx2Pfvn2YPXs2Tp48icDAQFStWlXsEDVGVFQUzpw5g7NnzyIyMhL58uWTj3yxxE+UdVxOgrTKkiVLEB0djaCgIBgYGADIWGE7NjYWs2fPFjk6zRMfH4/Tp0/j5MmTOHXqFPLkyYNatWqJHZbGmDFjBurXr4/Dhw/Ln+3p7++P+vXrY968eSJHp1kqVaoEPz8/rFq1CkePHkXjxo2xb98+jBo1SuzQiLQKi/OkVfbu3YsJEybA1dVV3ubi4oLJkydj2LBhmDRpkojRaY6pU6fizJkzuHv3LgoXLoy6detixowZcHd3Z5nxMxcvXsS6desUVmXX0dFBnz598Ouvv4oYmWaRyWS4evUqTp8+jTNnzuDSpUswNzdHixYt4OnpKXZ4RFqFiRdplfj4eJQoUUKp3draGklJSSJEpJkiIyPRpEkT1KtXD5UqVRI7HI2Vnp4OqVSq1P727Vuu6/UZNzc3JCUloWzZsvD09MTQoUPh4OAgdlhEWomJF2kVGxsbnDlzRmk0Ys+ePShbtqxIUWme7du3ix2CVqhVqxaWLFmCOXPmyNtevnyJ2bNnw83NTcTINEv//v3h6emJokWLih0Kkdbj5HrSKkePHsUff/yBX375BZs3b0bPnj1x9+5dHDx4EP7+/mjUqJHYIZIWiY+PR5cuXfDq1Su8efMGZcqUQWxsLPLly4d169Yx0SAiwTHxIq1z4sQJLFu2DNevX4dUKkW5cuXg5+fHpIuyJTk5GXv27MGNGzfk30+tWrWCiYmJ2KERUS7ExIuIiIhITTjHizTezp07s9yX6wnRt3Tp0iXLffmsRiISGhMv0ngjR47MUr+f/VmN37Oe0vTp03MwEs3GeVtEJCaWGolyCR8fnyz3Xbt2bQ5GQrmBnZ2dwvpmX3Pjxo0cjoYo92DiRURESrZv357lxKtNmzY5HA1R7sHEiyiX+vjxI168eIH09HQAGauPp6am4vLlyz91SZaISExMvIhyobNnz2LYsGF48eKF0jZDQ0NcvHhRhKhImx05cgQ3b96UJ/IA5Il8UFCQiJERaRdOrifKhebNm4fKlSvDx8cH/fv3x5w5c/DkyRMsWLDgp55YT9nj7++PZcuWoXDhwkhISIClpSWeP3+O9PR0NGvWTOzwiLQKEy+iXOjmzZvYsmULbG1tUbFiRRgbG8PHxwfGxsYIDAxE/fr1xQ6RtMi///6LcePGoVOnTqhbty42bNgAY2Nj9OvXD8WLFxc7PCKtwsSLNB6XSfh+Ojo68pXXS5UqhVu3bsHd3R1ubm6YOXOmyNGJi3frfb/nz5/Dw8MDQMb1u3LlCho3bozBgwdjzJgxGDRokMgREmkPJl6k8R4/fix2CFrHzs4Ohw4dQrdu3VC6dGlERkaia9euePr0qdihiW7atGlZTrwog7m5Od69ewcAKFmyJO7cuQMAKFKkCOLj48UMjUjrMPEijcc1p76fn58f+vfvD319fTRr1gwLFixAr169cPPmTbi5uYkdnqjatm0rdghax93dHbNmzcKUKVNQuXJlLF26FN7e3jhw4AAsLCzEDo9Iq/CuRtI6XCYha6KioqCjowM7OzuEh4dj1apVsLa2xsCBA5EvXz6xw9MYvFvv254+fYo+ffqgdevW8Pb2RocOHeRl2BEjRqBbt27iBkikRZh4kVbhMgnf79WrV9DR0YGpqanYoWicb92tN2fOHLFD1CgpKSkwMDDAhw8fcPLkSVhaWqJKlSos3RJ9hzxiB0D0PT4tk7By5UoYGhoiICAAo0ePhomJCWbPni12eBpl5cqVqFOnDtzd3eHi4oIGDRpg8+bNYoelUT7drXfixAlYWlpiw4YNOHXqFKpWrcq79T7j5eWFV69ewcDAAEDGHzkNGjSApaXlT1+6JvpenONFWoXLJGTN8uXLsXjxYvj4+MDR0REymQyRkZGYNm0aZDIZfvvtN7FD1Ai8W0+1kJAQnDx5EgAQGxuLSZMmyROvT2JjYznaRfSdmHiRVuEyCVmzfv16TJgwQWHOW/369WFjY4Ply5cz8foP79ZTzcnJCZs2bcKn2ShPnjyBnp6efLtEIoGxsTF/7oi+ExMv0ipcJiFrkpKS4ODgoNTu7OyMyZMnixCRZuLdeqpZW1sjODgYAODj44OAgACYm5uLHBWR9uMcL9Iqfn5+mDVrFjZs2IBmzZrh2LFj6NWrF4YMGcK5Jp9p2LBhpstw7NmzR15aI2Do0KF48eIFDhw4gEaNGsHAwAA1a9bErFmz0KVLF7HD0xhr166Fubk5njx5gpMnT+LDhw+Z3uBCRN/GuxpJ63CZhG+bPn06NmzYABsbG1SvXh26urq4du0aIiIi4OXlpXCHI1f7591635KWlobhw4dj3759yJMnDw4cOICZM2fizZs3CAgI4B2zRN+BiRdpLS6ToJqPj0+W+/7MC9R6eXlh27ZtSgl7fHw8WrZsibCwMHEC0zB///039u/fjwkTJqBPnz7YtWsX4uLiMHr0aNSoUQOTJk0SO0QircHEi7TOypUrERwcjISEBABAsWLF4Ofnh19//VXkyEgbfH633o4dO9C0adNM79a7desWzp07J0aIGqdhw4aYMGECatSoAScnJ+zatQvFixeXr6t36tQpsUMk0hqcXE9ahcskZN2HDx+wf/9+3L17F76+vrh16xbKli37008a59163y8+Ph4lSpRQare2tkZSUpIIERFpLyZepFW4TELWPH/+HB06dMDz58+RmpqKX375BatWrcLVq1cRFBSEsmXLih2iaHi33vezsbHBmTNnlEaV9+zZ81N/LxFlB+9qJK3ytWUSfvZ1lz43Y8YMlC1bFmfPnpWX0WbOnAk7OzvMmjVL5Og0B+/Wy5oBAwZg6tSpmDJlCtLT07Fjxw788ccfWLx4MXr37i12eERahYkXaRUuk5A1586dw8CBA2FkZCRvMzc3x7Bhw3Dp0iXxAtMwaWlpGDx4MDw9PdG7d28kJCRg/Pjx6Nq1K968eSN2eBqjXr16WLhwIW7cuAEdHR0EBgbi8ePH8Pf3R6NGjcQOj0irsNRIWiVfvnzYsGEDLly4kOkyCaNGjZL3/ZmXSXj37p1C0vW5jx8/qjkazbV48WJER0cjKCgIffr0AQB06dIFo0ePxuzZs3m33mfq1KmDOnXqiB0GkdbjiBdplevXr8PR0RGmpqaIjo7GtWvXAGSUGl+/fo3Hjx/LXz+z6tWrY/369QptaWlpWLRoEapWrSpSVJpn7969GDduHFxdXeVtLi4umDx5Mo4cOSJiZJrn0qVL+PPPP9GiRQu0bt0aY8aMwe3bt8UOi0jrcMSLtMrPvObU9xgxYgQ6deqE8+fPIy0tDRMmTMDdu3fx5s0brFu3TuzwNAbv1suaI0eOoH///rC3t0eNGjUglUpx6dIltG3bFqtXr4azs7PYIRJpDSZepHW4TMK32djY4N9//8XGjRthbW0NqVSKJk2awNvbG8WKFRM7PI3Bu/Wyxt/fHz169MCff/6p0D5z5kzMnj0b//zzj0iREWkfJl6kVbhMQtZZWlrijz/+kL9PTExkcvqFAQMG4I8//sCtW7fkd+vdvXsXBw8ehL+/v9jhaYyHDx+iXbt2Su2//fYbNmzYIEJERNqLc7xIq3CZhKxJSkrCuHHjcPPmTaSnp6Nbt26oWbMmmjRpgkePHokdnsbg3XpZU6lSJZw9e1ap/dq1a7CxsREhIiLtxUcGkVapVasWli9fjooVKyo8uiQ6OhpdunTB+fPnxQ5RI4waNQoRERFYunQp7t69iyFDhmDatGnYu3cv9PT0sHDhQrFDJC2yadMmzJ49G+3atYOLiwt0dXXlo8y//fYbypUrJ+/7+eLGRKSMpUbSKlwmIWuOHz+ORYsWwcbGBqtWrULNmjXRokULlC9fHp07dxY7PI1y6dIlrF27Frdu3YKOjg4qVaqEbt26KSQTP7sJEyYAAIKDg+Wr/n8SGBgo/1oikTDxIvoGJl6kVT4tkzB27Fh5G5dJUPb+/XtYW1sDAM6cOYPu3bsDAIyMjJCeni5maBqFd+tlTXR0tNghEOUaTLxIq3CZhKyxsbHBsWPHYG1tjbi4OPnCl5s3b+acnM/wbj0iUjcmXqRVuExC1gwcOBADBgxAWloamjdvjlKlSmH69OlYv349Fi1aJHZ4GoN36xGRunFyPWk9LpOQuZcvXyI+Ph52dnYAgMuXL8PExIQjXp/x9vZGixYt0LFjR4X2PXv2YNWqVdi+fbtIkRFRbsXEi7RKUlISZs+ejc6dO6Ns2bLo0aMHwsLCUKpUKSxfvhzFixcXO0TSIrxbj4jUjYkXaRUuk0BC+jQa+C0SiQQ3btzI4WiI6GfAOV6kVbhMAgmJd+sRkbpx5XrSKl8uk1CjRg0AXCaBiIi0A0e8SKtwmQQiItJmTLxIq3CZBCIi0macXE9ah8skEBGRtmLiRURERKQmnFxPREREpCZMvIiIiIjUhIkXERERkZow8SIiIiJSEyZeRERERGrCxIuIiIhITZh4EREREanJ/wAl7f3Zr/SR8QAAAABJRU5ErkJggg==", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import seaborn as sns\n", "\n", "sns.set()\n", "sns.heatmap(df_iris.corr(), square=True, annot=True)" ] }, { "cell_type": "markdown", "metadata": { "id": "a-T6SmVPcAMW" }, "source": [ "Existen variables muy correlacionadas entre si que nos pueden dificultar el trabajo por lo que podríamos eliminar alguna de ellas.\n", "\n" ] }, { "cell_type": "markdown", "metadata": { "id": "1OQNL5kPoJRQ" }, "source": [ "Además vamos a dividir el conjunto de datos en train y test en un porcentaje de 2/3 vs 1/3" ] }, { "cell_type": "code", "execution_count": 36, "metadata": { "id": "Gyl0FFU1oPJj" }, "outputs": [], "source": [ "from sklearn.model_selection import train_test_split\n", "\n", "train, test = train_test_split(df_iris[['petal length (cm)','petal width (cm)', 'target']], test_size=0.33)\n", "train.reset_index(inplace = True)\n", "test.reset_index(inplace = True)\n" ] }, { "cell_type": "markdown", "metadata": { "id": "R7NjZheI7Bhr" }, "source": [ "# 1. Parametrización\n", "\n", "Existen diferentes parámetros para método basado en los vecinos más cercanos utilizando scikit-learn\n", "\n", "\n", "\n", "* **KNeighbors/Radius**: KNeihgbors está basado en el estudio de los k vecinos más cercanos para cada punto, mientras que RadiusNeighboors están basados en un conjunto de vecinos que están dentro de un radio. **Nuestra elección es la primera**, la segunda sería útil cuando los datos no estuvieran muestreados de forma uniforme.\n", "* **K/Radio**\n", " * k: Un número k mayor suprime el efecto del ruido pero hace a los límites de clasificación más distintos.\n", " * Radios, un radio fijo es muy adecuado cuando los datos están muy dispersos (sprarse neighboors\n", "* **Pesos** : dos posibles valores, \"uniform\" cada vecino tiene el mismo peso, weights se asigna un peso a cada vecino proporcional a la distancia que esté del elemento referencia. También se puede definir una función por parte del usuario\n", "\n", "Nuestra elección KNeighbors y k y pesos se van a parametrizar para ello se ejecutará [validación cruzada]([https://es.wikipedia.org/wiki/Validaci%C3%B3n_cruzada) y como medida de éxito vamos a utilizar el Accuracy\n" ] }, { "cell_type": "code", "execution_count": 37, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 302 }, "id": "sCfTgNtbfFDG", "outputId": "abfaa912-2d26-4433-c161-d0e9a35f0a3b" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Max Value uniform : 1.0 (1)\n", "Max Value distance : 1.0 (2)\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from sklearn import neighbors\n", "from sklearn.model_selection import KFold\n", "import matplotlib.pyplot as plt\n", "from sklearn.metrics import accuracy_score\n", "import numpy as np\n", "\n", "cv = KFold(n_splits = 5, shuffle = True) # shuffle = False si hay dimensión temporal\n", "\n", "\n", "for i, weights in enumerate(['uniform', 'distance']):\n", " total_scores = []\n", " for n_neighbors in range(1,30):\n", " fold_accuracy = []\n", " knn = neighbors.KNeighborsClassifier(n_neighbors, weights=weights)\n", " for train_fold, test_fold in cv.split(train):\n", " # División train test aleatoria\n", " f_train = train.loc[train_fold]\n", " f_test = train.loc[test_fold]\n", " # entrenamiento y ejecución del modelo\n", " knn.fit( X = f_train.drop(['target'], axis=1),\n", " y = f_train['target'])\n", " y_pred = knn.predict(X = f_test.drop(['target'], axis = 1))\n", " # evaluación del modelo\n", " acc = accuracy_score(f_test['target'], y_pred)\n", " fold_accuracy.append(acc)\n", " total_scores.append(sum(fold_accuracy)/len(fold_accuracy))\n", "\n", " plt.plot(range(1,len(total_scores)+1), total_scores,\n", " marker='o', label=weights)\n", " print ('Max Value ' + weights + \" : \" + str(max(total_scores)) +\" (\" + str(np.argmax(total_scores) + 1) + \")\")\n", " plt.ylabel('Acc')\n", "\n", "\n", "plt.legend()\n", "plt.show()\n", "\n" ] }, { "cell_type": "markdown", "metadata": { "id": "Xj-T0YFK95v7" }, "source": [ "# 2. Construcción y ejecución del modelo\n", "\n" ] }, { "cell_type": "markdown", "metadata": { "id": "YxzQici8RpSe" }, "source": [ "Una vez que hemos identificado la mejor parametrización vamos a pasar a hacer una ejecución del modelo y vamos graficar sus resultados.\n", "\n", "Recordamos que al final del paso 1 hemos dividido en entrenamiento/tuneado y test" ] }, { "cell_type": "markdown", "metadata": { "id": "IKOb2PaoR9jF" }, "source": [ "Posteriormente, vamos a ejecutar el modelo con la mejor parametrización que hayamos obtenido anteriormente" ] }, { "cell_type": "code", "execution_count": 38, "metadata": { "cellView": "both", "colab": { "base_uri": "https://localhost:8080/" }, "id": "CT-gjUX1S9Em", "outputId": "6409799b-5167-4c44-96a4-b6296378685d" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Acc 0.94\n" ] } ], "source": [ "# constructor\n", "n_neighbors = 29\n", "weights = 'distance'\n", "knn = neighbors.KNeighborsClassifier(n_neighbors= n_neighbors, weights=weights)\n", "# fit and predict\n", "knn.fit(X = train[['petal length (cm)','petal width (cm)']], y = train['target'])\n", "y_pred = knn.predict(X = test[['petal length (cm)','petal width (cm)']])\n", "acc = accuracy_score(test['target'], y_pred)\n", "print ('Acc', acc)\n", "\n" ] }, { "cell_type": "markdown", "metadata": { "id": "c3xiYNMZxfn7" }, "source": [ "Graficamos la solución mediante los diagramas de Voronoi" ] }, { "cell_type": "code", "execution_count": 39, "metadata": { "id": "4xQz3Q3S3y4k" }, "outputs": [], "source": [ "import matplotlib.pyplot as plt\n", "from matplotlib.colors import ListedColormap\n", "cmap_light = ListedColormap(['#FFAAAA', '#AAFFAA', '#AAAAFF'])\n", "cmap_bold = ListedColormap(['#FF0000', '#00FF00', '#0000FF'])\n", "h = .05 # step size in the mesh\n", "\n", "\n", "X = train[['petal length (cm)','petal width (cm)']].to_numpy()\n", "y = train['target'].to_numpy()\n", "\n" ] }, { "cell_type": "code", "execution_count": 40, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 284 }, "id": "dAnH1mjxxNQ_", "outputId": "651b9c28-e7ad-4446-c7c3-11377139c6fd" }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "c:\\Users\\marco\\anaconda3\\envs\\saa\\Lib\\site-packages\\sklearn\\base.py:439: UserWarning: X does not have valid feature names, but KNeighborsClassifier was fitted with feature names\n", " warnings.warn(\n" ] }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "\n", "# Plot the decision boundary. For that, we will assign a color to each\n", "# point in the mesh [x_min, x_max]x[y_min, y_max].\n", "x_min, x_max = X[:, 0].min() - 1, X[:, 0].max() + 1\n", "y_min, y_max = X[:, 1].min() - 1, X[:, 1].max() + 1\n", "xx, yy = np.meshgrid(np.arange(x_min, x_max, h),\n", " np.arange(y_min, y_max, h))\n", "Z = knn.predict(np.c_[xx.ravel(), yy.ravel()])\n", "\n", "# Put the result into a color plot\n", "Z = Z.reshape(xx.shape)\n", "plt.figure()\n", "plt.pcolormesh(xx, yy, Z, cmap=cmap_light)\n", "\n", "# Plot also the training points\n", "plt.scatter(X[:, 0], X[:, 1], c=y, cmap=cmap_bold,\n", " edgecolor='k', s=20)\n", "plt.xlim(xx.min(), xx.max())\n", "plt.ylim(yy.min(), yy.max())\n", "plt.title(\"Classification (k = %i, weights = '%s')\"\n", " % (n_neighbors, weights))\n", "\n", "plt.show()\n" ] }, { "cell_type": "markdown", "metadata": { "id": "onIoRvD449nH" }, "source": [ "Y finalmente visualizamos la Matriz de Confusión" ] }, { "cell_type": "code", "execution_count": 41, "metadata": { "id": "NX9gtCEszUi0" }, "outputs": [], "source": [ "from sklearn.metrics import confusion_matrix\n", "from sklearn.utils.multiclass import unique_labels\n", "\n", "def plot_confusion_matrix(y_true, y_pred, classes,\n", " normalize=False,\n", " title=None,\n", " cmap=plt.cm.Blues):\n", " \"\"\"\n", " This function prints and plots the confusion matrix.\n", " Normalization can be applied by setting `normalize=True`.\n", " \"\"\"\n", " if not title:\n", " if normalize:\n", " title = 'Normalized confusion matrix'\n", " else:\n", " title = 'Confusion matrix, without normalization'\n", "\n", " # Compute confusion matrix\n", " cm = confusion_matrix(y_true, y_pred)\n", " # Only use the labels that appear in the data\n", " classes = classes[unique_labels(y_true, y_pred)]\n", " if normalize:\n", " cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n", " print(\"Normalized confusion matrix\")\n", " else:\n", " print('Confusion matrix, without normalization')\n", "\n", " print(cm)\n", "\n", " fig, ax = plt.subplots()\n", " im = ax.imshow(cm, interpolation='nearest', cmap=cmap)\n", " ax.figure.colorbar(im, ax=ax)\n", " # We want to show all ticks...\n", " ax.set(xticks=np.arange(cm.shape[1]),\n", " yticks=np.arange(cm.shape[0]),\n", " # ... and label them with the respective list entries\n", " xticklabels=classes, yticklabels=classes,\n", " title=title,\n", " ylabel='True label',\n", " xlabel='Predicted label')\n", "\n", " # Rotate the tick labels and set their alignment.\n", " plt.setp(ax.get_xticklabels(), rotation=45, ha=\"right\",\n", " rotation_mode=\"anchor\")\n", "\n", " # Loop over data dimensions and create text annotations.\n", " fmt = '.2f' if normalize else 'd'\n", " thresh = cm.max() / 2.\n", " for i in range(cm.shape[0]):\n", " for j in range(cm.shape[1]):\n", " ax.text(j, i, format(cm[i, j], fmt),\n", " ha=\"center\", va=\"center\",\n", " color=\"white\" if cm[i, j] > thresh else \"black\")\n", " fig.tight_layout()\n", " return ax" ] }, { "cell_type": "code", "execution_count": 42, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 378 }, "id": "jNM0HVdLzVz-", "outputId": "b976e8c0-3570-4f19-cc4d-c043500e511f" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Confusion matrix, without normalization\n", "[[16 0 0]\n", " [ 0 19 2]\n", " [ 0 1 12]]\n" ] }, { "data": { "text/plain": [ "" ] }, "execution_count": 42, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhsAAAHQCAYAAAAI4KhXAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAABx0UlEQVR4nO3dd1yV5f/H8dcBBAco4lbcCW7EbTlRc6TmyCyVNM299yBnuUVzDzQzZ6ZmVq7Sco8cuVIx98KNiijz/P7wy/l5Agdw8MDp/exxHsm9zuc+Bw5vruu6r9tgNBqNiIiIiCQRO2sXICIiIrZNYUNERESSlMKGiIiIJCmFDREREUlSChsiIiKSpBQ2REREJEkpbIiIiEiSUtgQERGRJKWwISKSzGnuRUnpFDYkRfH19aVo0aIcP348zvU+Pj4MHjz4DVcVf76+vvj6+pq+9vT0ZMaMGUn+vFevXsXT05O1a9cm+XO9jqCgIFq3bk2JEiWoVKkST548schxBw8ejI+Pj0WOZU1BQUF06tSJa9euvXS7/fv34+npyf79+99QZSLxo7AhKU5UVBRDhgwhPDzc2qVYzHfffUfz5s2tXcYbt3jxYo4cOcKECROYOXMmadKkschxu3btysyZMy1yLGvas2cPf/zxxyu3K1asGN999x3FihVL+qJEEsDB2gWIxJeLiwtnz55l1qxZ9OnTx9rlWESpUqWsXYJVBAcHkzVrVurXr2/R4+bJk8eix0vunJ2d/7PfQ5IyqGVDUpwiRYrQuHFjFixYwIkTJ166bVRUFMuWLaNhw4aULFmS6tWrM3nyZMLCwkzbDB48mDZt2jBixAjKli1LkyZNiIyMxNPTkxUrVjB48GDKlClD+fLl+fLLL3n69CkTJkygYsWKVKhQAT8/P7Pj3bt3j1GjRlGjRg2KFy9O+fLl6datG1evXn1hnc93o/j6+uLp6Rnn4/nuj++//5733nuP4sWLU716dWbMmEFkZKTZcbds2UKjRo0oWbIkTZo04fTp06/1Gu/evZtWrVrh7e1N5cqVGT58OA8ePDCtv3jxIj179uSdd96hVKlS+Pr6cujQIdP6mO6ajRs30rNnT7y9vSlXrhx+fn48fvwYeNbltXbtWq5fv246/xd1B/y72+nkyZO0adOGMmXK4O3tTdu2bTl69KjZe/p8N8rrfh+0bduWNWvWUKdOHYoXL06jRo3Yvn37S18rX19fhg8fzpw5c6hSpQpeXl506NCBO3fusGbNGmrXrm2q8fnvgaioKObPn0+DBg0oWbIkpUqV4qOPPmLv3r0ArF27liFDhgBQs2ZNU/egj48PY8eOpU2bNpQuXZrhw4ebvW4hISH4+PhQt25dU+uf0WikXbt2VKpUiTt37rz0fESSgsKGpEh+fn64ubm9sjtl+PDhjB07Fh8fH+bMmUOrVq1YunQpXbt2NRt0d/DgQS5dusSMGTPo1q0bDg7PGv0mT56Mo6MjM2fO5P3332fJkiU0btyYGzduMGnSJD766CNWr17NkiVLgGcf6p06dWL37t3069ePhQsX0rVrV/bs2cPw4cNf69xGjBjBd999Z3osW7aM3Llzkz17dqpWrQrAvHnzGDZsGJUqVWLu3Lm0atWKgIAAs+fYtm0bPXv2pFChQsycOZN69eoxYMCAVz7/9u3b+eyzz3B1dWXq1KkMGDDAdCyAf/75h6ZNm3LlyhU+//xzJk+ejMFgoE2bNhw4cCDWueTKlYvZs2fz2WefsWbNGubOnQvAzJkzqVatGlmyZIlXN1JISAifffYZGTNmZPr06UydOpUnT57Qvn17Hj16FOc+r/t9cOLECRYuXEjPnj2ZNWsWDg4O9OzZ0yxoxeWXX35hz549jBkzhiFDhrBnzx5at27NkiVLGDRoEH5+fhw9epTRo0eb9pk8eTKzZs2iRYsWLFiwgNGjR3P//n169epFaGgo1atXp0uXLqbXqmvXrqZ9ly1bZgpo77//vlktzs7OjBkzhosXL5pe6+XLl7N7927GjBlD5syZX+t1FrEoo0gK0rp1a2Pr1q2NRqPRuHXrVqOHh4dxypQppvU1atQwDho0yGg0Go1nz541enh4GGfPnm12jHXr1hk9PDyMf/zxh9FoNBoHDRpk9PDwMF68eNFsOw8PD2Pz5s1NX0dERBhLlSpl9PHxMUZERJiWN2jQwNilSxej0Wg0BgUFGX19fY1//vmn2bG++OILY7FixeI8j5jnmj59epznPGLECGPJkiWNx48fNxqNRuPDhw+NXl5exuHDh5ttt2rVKqOHh4cxMDDQaDQajU2bNjU2bdrUbJt58+YZPTw8jGvWrInzuWL2a9y4sdmyTZs2Gd99911jUFCQsVevXsby5csbHz58aPba1KlTx/jBBx8YjUaj8cqVK0YPDw9j//79zY7j6+trbNCggenrQYMGGWvUqGH6et++fUYPDw/jvn37zPZ7/vU6cuSI0cPDw3jw4EHT+kuXLhknTJhgvH79eqzjxvf74NKlS6ZtDhw4YPTw8DBu2rTpha9X69atjSVKlDAGBweblrVr187o4eFhvHz5smnZ6NGjjWXKlDF93bdvX+OiRYvMjrV582ajh4eH8fDhw0aj0Whcs2aN0cPDw3jlyhXTNjVq1DBWr17dGBUV9dLXbdSoUcZixYoZt2/fbixVqpRx2LBhLzwHkaSmlg1JsXx8fGjUqBELFizg5MmTsdbH/JXdsGFDs+Xvvfce9vb2Zk31qVOnjrOf39vb2/RvBwcHMmbMSPHixU0tHwCurq6mv6izZcvGt99+S9myZbl+/Tp79+5l6dKlHD58mIiIiHif4/Lly1mxYgVjxoyhePHiABw5coQnT57g4+NDZGSk6RHTbbB7926ePn3KyZMnqVmzptnx6tWr99Lni9mvVq1aZsvr1KnD5s2byZYtGwcOHKBGjRq4uLiYvTbvvfcex48fN3WTQOyxKNmzZyc0NDTer8PzChUqhJubG126dGHEiBFs27aNLFmyMHDgQHLkyBFr+/h8H7i5uZl9H2TPnh3glVfJFCxYkAwZMpi+zpIlC25ubuTOndu07PnvEwB/f3/atm3LvXv3OHLkCGvXrmX9+vUAr/xeKViwIHZ2L//47t+/Pzly5KBTp05kzZrV1CUjYg0aICop2ueff87evXsZPHgwa9asMVsX0/SdJUsWs+UxoeH5D/5MmTJhMBhiHd/Z2TnWslddMbF+/XqmTJnCjRs3cHV1pXDhwqROnfq1zynG/v37GTNmDB07dqRBgwam5cHBwQB07Ngxzv1u3brFgwcPMBqNuLm5ma3LmjXrS58zZr9MmTK9dJu4muIzZ86M0WgkJCTEtOzfr5WdnV2i54xIly4dy5YtY86cOWzYsIGVK1eSJk0aGjVqhJ+fH05OTrHqhdf7Pvh3vTHfE9HR0S+tKSHfJ8ePH2fUqFEcP36c1KlT89Zbb5ErVy7g1fNqvE5XSNq0aalTpw4BAQFUrFjRYlf6iCSEwoakaBkyZGDkyJF069aNOXPmxFoHcPv2bdzd3U3LIyIiuH//PhkzZrR4PQcPHmTQoEG0bt2a9u3bm/4ynjhxotkAyle5cuUKPXv2pHLlyrGuuEmfPj3wrM8/X758sfbNnDkzrq6u2NnZxRoMGBNUXsTZ2RmDwcC9e/fMloeHh7N3715KlixJhgwZ4hxkePv2bQAyZszIrVu3XnWKcXrRL/fHjx+TLl0609cFChRg0qRJREVFcezYMX788UdWrFiBu7t7rBBmje+DV4kZd+Lp6cnPP/9saqnYvn07mzdvtshz/PPPPyxevJgiRYqwatUqGjZsSNmyZS1ybJH4UjeKpHi1atWiQYMGzJ8/3+yXZPny5QH46aefzLb/5ZdfiIqKokyZMhav5ciRI0RHR9OzZ09T0IiKimLPnj3Aq/9Chme/iLp06YKbmxv+/v6xmsu9vLxIlSoVN2/epESJEqZHqlSp8Pf35+rVqzg5OeHt7c2WLVvM/kretm3bS587Xbp0FClShK1bt5ot37VrFx07diQoKIhy5crx+++/m7UIREVF8csvv1CiRAkcHR1feY4vEtNCcOPGDdOyBw8ecO7cOdPXmzZtomLFity+fRt7e3u8vb0ZOXIk6dOnJygoKNYxrfF98Crnz58nODiYTz75hEKFCpne4x07dgD//33yqq6SF4mMjGTQoEHkypWLFStWULx4cYYMGZLoLiyRhFLLhtiEYcOGsW/fPrO/uN966y2aNGnCzJkzefr0KRUqVODUqVPMnDmTChUqUKVKFYvXUbJkSQBGjx5Ns2bNePjwIUuXLjVdchoaGhpnk/vz+vfvz5UrV5gyZQrnz583CygxYwo+++wzpk2bRkhICBUqVODmzZtMmzYNg8FA4cKFAejbty9t2rShe/futGjRgosXL8Zq/YlLz5496dKlC71796Zp06bcu3cPf39/atSoQZEiRejevTs7duzgk08+oWPHjjg6OrJ06VKuXLnCggULEvrSAc8uAc6RIwczZ87ExcUFOzs75s+fb9YFULp0aaKjo+nWrRsdO3YkXbp0bNy4kUePHvHuu+/GOqY1vg9eJX/+/Dg7OzN37lwcHBxwcHBg8+bNrF69Gvj/MSIxrVi//vorVatWpWDBgq91/Hnz5nHy5EmWLl1KmjRp+OKLL2jWrBmTJ09+7auiRCxJYUNsgqurKyNHjqR79+5my8eMGUPevHlZs2YNCxcuJGvWrPj6+tKtW7cE/9X4MhUqVGD48OEsWrSITZs2kTlzZipUqMDMmTPp1q0bhw4dolq1ai89xu+//w5gdqljjCZNmjB+/Hh69+5NlixZWL58OQsWLCBDhgxUqlSJvn37mgZuli1bloCAAKZMmUL37t1xd3dn7NixdO7c+aXPX6NGDebNm2e6DDhjxozUq1ePXr16Ac8GaC5fvpwpU6YwdOhQDAYDJUuWNA2MTQx7e3umT5/O2LFj6du3L5kzZ6ZNmzacP3+eCxcuAM/GnSxYsIBp06bh5+fHkydPKFSoEDNmzKBixYpxHvdNfx+8iouLC7Nnz2bixIn06tXL1KK0dOlSOnTowMGDB/Hx8aFChQq8/fbb+Pv7s3fvXubPn//KY58+fZo5c+bw0Ucfmd6PwoUL07ZtWxYuXEjt2rWpVKlSUp+iiBmDMbGjtUREREReQmM2REREJEkpbIiIiEiSUtgQERGRJKWwISIiIklKYUNERESSlMKGiIiIJCmFDREREUlSmtQrBYqIiuba3cev3jAFcrCzI1emtFy7G0rka0ztndLkyeLy6o1SqOdvY6fJe1IWW3/vDEAc91lMMpGRUVwJup+gfXNnz4iDg72FK7I+hY0U6Nrdx3j1XmftMpKEVz43dox9j5ZT/uDoxXuv3iGFufmtr7VLSDIGwMkBwiJt8xeWLbP1987R3jxQJbUrQfcp2nBkgvb9+6eR5Hd/9V19UxqFDREREUszaJTC8xQ2RERELO1N9tukAAobIiIiFmVIRMuGbYYUtfOIiIhIklLLhoiIiCUl5vIX22zYUNgQERGxOA0QNaOwISIiYmkaIGpGYUNERMTS1LJhRq+GiIiIJCm1bIiIiFiaulHMKGyIiIhYlObZ+DeFDREREUtTy4YZjdkQERGRJKWWDREREUsykPBuFBttEFHYEBERsTR1o5hR2BAREbEoDRD9N43ZEBERkSSllg0RERFL0wyiZhQ2RERELM3ONrtDEkphQ0RExNKSQcvG7Nmz2bt3L0uWLAHA19eXAwcOxLnthAkTaNy4cZzrfHx8uHbtmtmyhg0bMnny5NeuRWFDRETE0qx8Nco333zD9OnTKVeunGnZjBkziIiIMNvu888/5/Lly9SqVSvO44SEhHD9+nXmzZtHsWLFTMtTp04dr3oUNkRERGzEzZs38fPz49ChQ+TPn99snaurq9nXP//8M7t27WLt2rU4OzvHebzAwECMRiOlS5cmffr0Ca7L+u08IiIitsTwv0tfE/RIXIvIyZMnyZAhA+vXr8fLy+uF24WGhjJx4kTatGmDp6fnC7c7c+YMWbJkSVTQALVsiIiIWJ6VulF8fHzw8fF55XYrV67k8ePHdOnS5aXbBQYGkjZtWnr06MGRI0dwc3OjadOmfPLJJ9jZvX57hcKGiIiIpSVigOj169fx9fV94fqtW7cm+NgAUVFRLFmyhJYtW+Li4vLSbc+ePcujR4+oX78+3bt35+DBg0yePJkHDx7Qq1ev135OhQ0REZH/kAMHDnD9+nU+/PDDV267aNEiwsLCTGM6PD09efz4MXPmzKFHjx6v3bqhsCEiImJpiehGyZkzZ6JbL17mt99+o2TJkuTOnfuV26ZKlYpUqVKZLfPw8CA0NJQHDx6QMWPG13pODRAVERGxtIQOEH0DDh06RMWKFV+5XXR0ND4+PsyZM8ds+fHjx8mcOfNrBw1Q2BAREbEww/+uSEnAI4lvxBYVFcU///yDh4dHnOsfPXrEvXv3ALCzs6NOnTosWLCAjRs3cvnyZb777jsWLFgQr/EaoG4UERGR/4zg4GAiIiJizbkRY8yYMRw4cIBt27YB0K9fP9KnT4+/vz9BQUG4u7vj5+f3WuM9nqewISIiYmnJYLry8ePHx1qWKVMmzpw589r7ODg40KVLl1deIvsqChsiIiKWZCDhA0Rt9P5tChsiIiKWlgxaNpITvRoiIiKSpNSyISIiYmlq2TCjsCEiImJRhoSP2bDRQRsKGyIiIpamlg0zChsiIiKWZqW7viZXil4iIiKSpNSyISIiYmnqRjGjsCEiImJp6kYxo7AhIiJiSQYDhgTPIGqbIUXtPCIiIpKk1LIhIiJiYQlu2bBRChsiIiKWpqxhRmFDRETEwtSyYU5jNsTqcrml5VJACyoXyWa2PEfGNAR0q8yFeR9yZUELfhxai5J5M1qpSnmZLZs38XaFsqRNm5ZCBfMyacI4jEajtcuS17Rl8ybKli1LxvRp8dD7J0lAYUOsKnfmdKwbWgvXdI5my9M42rNxeB288rnRe+E+Ppu1C+fUqVg3tBbZXNNYqVqJy949e/igSSM8Cxdh7dq1tGzly4hhfkwcP9bapclr2LtnD82aNKJIkSJ8973eP0sx/O+KlPg+bJW6Uf4lNDSUH374gVatWlm7FJtmMEDLqgX5smWZONd/WDk/bi5OlOu/npvBTwA4cv4u28e8R+Ui2Viz9+IbrFZeZuyXoyjpVYpFi5fg5AA1atUlIiKCyRPH07N3X9KkUThMzsZ+OQovr1IsWbKEsEioXUfvnyXYcnBICLVs/MvXX3/NwoULrV2GzSueJyNTPq3A8p3n6DhnV6z11YrlYP3+y6agAXDrwVOKdF+joJGMhIWFsWP7H7zfpKnZ8iZNPyAkJITdu3ZaqTJ5HXr/koaBhLds2GpEUdj4F/VTvhlX7zzGu+86/JYe4klYVKz1+bI6E3jjAX7NvTgzqxl3vm3FL8PepWhu1zdfrLzQhfPnCQ8Pp1AhD7PlBd96C4CzgYHWKEtek96/JGJI5MMG2WTY2L59O02bNsXLy4tKlSoxePBgHjx4AMC5c+fo0KED3t7eVK5cmX79+nH79m0AZsyYwcyZM7l27Rqenp5cvXoVgHXr1tGoUSNKliyJj48Pc+fOJTo62vR869at47333qNEiRJUqVKFMWPGEB4eblq/Zs0aGjduTMmSJSlVqhS+vr6cPHnyDb4iyc/9x+Fcvxf6wvUO9nZ0rVeEKkWz0yNgH5/O2IGbsxO/fP4uOTKqWTe5CA4OBiB9+vRmy11cXAB49Ojhmy5J4iHm/XPR+ydJzObCxr179+jevTvNmjVjw4YNzJw5kz///JOJEydy8+ZNWrZsSe7cuVm9ejVz584lJCSEjz76iNDQUNq1a0e7du3Inj07u3btIkeOHHzzzTcMGzaMFi1asH79evr06cPChQuZOHEiAKdPn+bzzz+nR48ebN68mbFjx/Ljjz+yYMECAH799VdGjBhB27Zt2bhxI4sXL+bp06f4+flZ82VKMZqN38qWv67x059XaD5xK+lSO9Dx3cLWLkv+JyZ0v6h/2mBncx8xNkXvX9LRAFFzNjdA9ObNm4SHh5MzZ05y5cpFrly5mDt3LlFRUaxYsYKsWbMyfPhw0/ZfffUVFStWZNOmTTRt2pS0adNib29PlixZMBqNBAQE0Lp1a9OA0Xz58hEcHMyECRPo1q0bV69exWAw4O7uTs6cOcmZMycLFy7E2dkZAFdXV7788ksaN24MQK5cuWjevDkjRoxI8Dk62Nnhlc8t4S9SMlMwu4vp/6FhkablJy7f560c5n9xXb4dwtuFs6bY87e1j5KMrq4APHr40HRuBiDk0SMAXNNnsLlztiUx71/Iw2ctGDHvla29f9Y4B1sODglhc2GjSJEiNGjQgM6dO5MjRw7efvttqlevjo+PD3///Tfnzp3D29vbbJ+wsDDOnTsX61j37t3jzp07lCljfsVEuXLliIiI4Pz581SpUgVvb2+aNWtGvnz5ePvtt6lZsybFixc3bevm5sbs2bO5dOkSFy5c4NSpU2bdMPGVK1Nadox9L8H7J1fTO1Qy+7q2Vy5qe+WKtV0hsMnzT4mKeBbE3t6eSxf+wfF/nyaODnDl4j8AlCxRFCeb+5SxHTHv38ULz96vmPdQ719iJaaVwjZDik1+G/n7+9OtWzd27NjBnj176Nu3L6VLl8bR0ZGKFSvG2aoQ00f5vBcNFo2Kejag0cHBAScnJ7799lv+/vtvdu3axa5du1i5ciWNGzdm3Lhx/PLLLwwcOJAGDRpQsmRJPvjgAwIDAxk9enSCz+/a3VBaTvkjwfsnN6XyuzG9QyV6BuwlNCySBd2rsPvUTbzyu/Hx5N95EBoBPJuTY3GvqszZdJrvd1+wctUJ8+to2wpJBofUVK5SlTVr1tKzd3+cUhkIj4TvVq3G1dUVr9Llea6xSpKZ59+//v37ExFlwIjtvX+O9jZ7M9UUw+bCxl9//cWGDRsYOnQoBQoUoG3btqxfv54BAwbQtGlT9uzZQ44cOXB0fDaJVHBwMIMGDeLTTz+lYsWKZmk0U6ZMZMqUiUOHDlGrVi3T8oMHD5IqVSry5MnD9u3bOX78ON27d6do0aJ07NiROXPmMHfuXMaNG8fcuXP54IMPGDVqlGn/rVu3As/CTELSb2R0NEcv3kvoS5TsuKRJBcC5oEc8evIsWEz/+STzulbmy1ZlmPDDcVLZGxjewptrd0MZv+YoIU9T5iegLV7rNHjo59SvU4uWH3/IZ+3bsWPXHqb4T+LLcRNInSaNTZ6zLYl5/z788EN827Rj717be/+MvPn2AnWjmLO5sOHs7Mzy5ctJlSoVH374IU+fPuWXX34hX758dOnShc2bN9O3b1+6deuGwWBg0qRJ/P333xQqVAiAtGnT8uDBAy5cuIC7uzvt2rVj2rRpuLu7U7lyZY4dO8bMmTNp0aIFLi4uODg4MGvWLJydnalZsybBwcH8/vvvpq6aHDlycPjwYU6ePImLiwvbtm1j6dKlAISHh+Pk5GS11yo5u3H/Ce+O3MSoj0szr8s7REUb+ePEDYYsOZhig4atql7DhxWr1vDlqBE0btyYnLlyMXbCJHr36Wft0uQ1VK/hw8pVa/hy9AiaN9P7ZzHKGmYMRhucWOL3339n5syZnD9/Hjs7OypWrMigQYPIkycPf//9N/7+/hw+fBh7e3tKlSrFwIED8fB4dp351atX+eyzz7h27RpLly7Fy8uLb7/9lqVLl3L9+nWyZ8/Ohx9+SPv27bG3twdg7dq1fP3111y5coXUqVNTrVo1Bg8ejJubG1euXGH48OH89ddfODo6UrhwYVq0aEGfPn1YsmQJ5cuXj/f5Xbz1CK/e6yz5kiUbXvnc2DH2PaoO/cWmWm9i3PzW19olJBkD4OQAYZG22YJjy2z9vXO0B7s3+Mv/4q0Qyg78OUH7HpzYgHxZnS1ckfXZZNiwdQobKZfChiRHtv7eKWxYn811o4iIiFibxmyYU9gQERGxIIMh4WHDVjOKwoaIiIil2WhoSCjNRSsiImJhyWG68tmzZ+Praz5ObMiQIXh6epo9qlat+tLjbNy4kfr161OiRAkaNmzIjh074l2LWjZERERszDfffMP06dMpV66c2fIzZ87QuXNnWrdubVoWc2VlXPbt28eAAQMYPHgwlSpVYvXq1XTr1o1169ZRsGDB165HLRsiIiIWZq2WjZs3b/LZZ58xbdo08ufPb7YuKiqKf/75hxIlSpAlSxbTw83txfeaCggIoHbt2rRu3ZqCBQsyaNAgihUrxuLFi+NVl8KGiIiIRSUsaDwLG4kLHCdPniRDhgysX78eLy8vs3UXL14kLCzstVskoqOjOXz4MBUrVjRbXqFCBQ4ePBivutSNIiIiYmHWuvTVx8cHHx+fONcFBgZiMBhYvHgxO3bswM7OjmrVqtG7d+847w/28OFDQkNDyZ49u9nyrFmzcuPGjXjVpbAhIiKSjFy/fj3WwM7nxdxfK77Onj2LnZ0duXLlYu7cuVy6dIkJEyYQGBjI4sWLsbMz7+x4+vQpgOleYjGcnJwICwuL13MrbIiIiFhaMrz0tUePHrRt25b06dMD4OHhQZYsWWjRogXHjx+P1e0Sc++u8PBws+VhYWGkSZMmXs+tsCEiImJJiZjUCwPkzJkzwa0XLz20wWAKGjFi7gsWFBQUK2y4urqSNm1abt26Zbb81q1bsbpWXkUDREVERCwsOcyz8W/9+vWjffv2ZsuOHz8OwFtvvRXnOZQuXZoDBw6YLd+/fz9lypSJ13MrbIiIiPwHNGjQgN27dzNnzhwuX77M9u3bGTp0KA0aNDBdofLo0SPu3fv/m2B++umn/PLLLyxatIhz584xceJETp06RZs2beL13AobIiIiFpYcWzZq1KjBtGnT2LJlCw0bNsTPz493332XsWPHmrYZM2YMH3zwgenrypUrM3bsWFasWEGTJk3Yt28fc+fOjdeEXqAxGyIiIpaXDAaIjh8/PtayOnXqUKdOnXjt07hxYxo3bpyoWhQ2RERELMhAIu76atlSkg11o4iIiEiSUsuGiIiIhVlrBtHkSmFDRETEwhQ2zClsiIiIWFRiriyxzZCisCEiImJJibl5q21mDQ0QFRERkaSllg0REREL05gNcwobIiIiFqawYU5hQ0RExMKUNcxpzIaIiIgkKbVsiIiIWJCmK49NYUNERMTC1I1iTmFDRETEwjRA1JzGbIiIiEiSUsuGiIiIhalhw5zChoiIiCUZwM4ugWnDRkOKwoaIiIiFqWXDnMKGiIiIhWmAqDkNEBUREZEkpZYNERERC3o2qVfC97VFChsiIiIWpm4UcwobIiIiFmVIRNiwzZCiMRsiIiKSpNSyISIiYkmGRFz6apsNGwobIiIilqYxG+YUNkRERCxMWcOcxmyIiIhIklLLhoiIiIWpG8WcwoaIiIiFKWuYU9gQERGxoGcziCYsbdhqRlHYEBERsTC1bJjTAFEREREbNHv2bHx9fc2Wbdu2jWbNmuHt7Y2Pjw8TJkzg6dOnLz2Oj48Pnp6eZo/+/fvHqxa1bIiIiFiYtQeIfvPNN0yfPp1y5cqZlh08eJDu3bvTu3dv6tSpw6VLlxg+fDjBwcGMGzcuzuOEhIRw/fp15s2bR7FixUzLU6dOHa96FDZSoDxZXLj5re+rN0yBYn48fx39HkarVpI0Mpbrbu0Skkypwu7sXTGY6r7j+ev0VWuXkyQCt/pbu4QkkcreQPYMjtx/HE5ElO395OXI4Iid/Rv85W/FGURv3ryJn58fhw4dIn/+/GbrVq5cScWKFenYsSMAefPmpU+fPgwdOpRRo0bh6OgY63iBgYEYjUZKly5N+vTpE1yXwoaIiIiFWatl4+TJk2TIkIH169cza9Ysrl27ZlrXrl077Oxij56IjIwkJCQENze3WOvOnDlDlixZEhU0QGFDREQkWbl+/XqssRbP27p16wvX+fj44OPjE+e6okWLmn0dHh7OokWLKFasWJxBA561bKRNm5YePXpw5MgR3NzcaNq0KZ988kmcweVFFDZEREQsLLlfjRIZGcnAgQP5559/WLZs2Qu3O3v2LI8ePaJ+/fp0796dgwcPMnnyZB48eECvXr1e+/kUNkRERCwsMd0oOXPmfGnrRWKFhITQu3dv9u/fz/Tp0/Hy8nrhtosWLSIsLAxnZ2cAPD09efz4MXPmzKFHjx6v3bqhsCEiImJBzyb1Svi+SenWrVt06NCBq1evEhAQQMWKFV+6fapUqUiVKpXZMg8PD0JDQ3nw4AEZM2Z8refVPBsiIiL/AQ8ePKBNmzbcu3eP5cuXvzJoREdH4+Pjw5w5c8yWHz9+nMyZM7920AC1bIiIiFiYIRHdKEnXtjFu3DiuXLnCggULcHNz4/bt26Z1bm5u2Nvb8+jRIyIiInBzc8POzo46deqwYMEC8uXLR7Fixdi7dy8LFizAz88vXs+tsCEiImJh1p7U69+io6PZsGEDERERtGnTJtb6rVu34u7uzpgxYzhw4ADbtm0DoF+/fqRPnx5/f3+CgoJwd3fHz8+PDz/8MF7Pr7AhIiJiSVac1Ot548ePN/3bzs6OY8eOxWsfAAcHB7p06UKXLl0SVYvGbIiIiEiSUsuGiIiIhSW3bhRrU9gQERGxMGUNcwobIiIiFqaWDXMKGyIiIhamrGFOA0RFREQkSallQ0RExIIMgF0CmzZstUFEYUNERMTC1I1iTmFDRETEwjRA1JzGbIiIiEiSUsuGiIiIJRnALhlMV56cvFbYGDJkyGsf0GAwMHbs2AQXJCIiktKpG8Xca4WN/fv3v/YB9QKLiMh/mYGEDxC11d+grxU2Ym41KyIiIhJfCR6zER0dTWBgILdu3aJ06dJERkbi6upqwdJERERSJoPNtlEkTILCxo8//oi/vz+3bt3CYDCwevVqZsyYQapUqfD398fR0dHSdYqIiKQYCR4gaqPifenrhg0bGDRoEBUrVmTq1KkYjUYA3n33XXbs2MHs2bMtXqSIiEjKYcBgSNjDVkdtxLtlY+7cuXz00UeMHDmSqKgo0/KmTZty9+5dVq1aRe/evS1Zo4iISIqiayXMxbtl48KFC9SuXTvOdV5eXty8eTPRRYmIiIjtiHfYyJQpE+fOnYtz3blz58iUKVOiixIREUnJ7AyGBD1sVbzDRv369Zk+fTqbNm0iPDwceDa3xokTJ5g9ezZ169a1eJEiIiIphcGQuIctiveYjd69exMYGEjv3r2xs3uWVXx9fQkNDaVs2bL06tXL4kWKiIikJJrg0ly8w4ajoyMLFixg9+7d7N27lwcPHuDi4kL58uWpVq2aXmARERExk+BJvd555x1Kly7No0ePcHV11dwaIiIi/6O/u80lKGzs2bOHGTNmcPToUYxGI/b29pQqVYrevXtTtmxZS9coIiKSotjyYM+ESNCkXu3atSMsLIzu3bszcuRIOnfuTHBwMG3btmXfvn1JUaeIiEiKYUjgw1bFu2Vjzpw5vPfee/j7+5st79atG127dmXSpEmsWbPGYgWKiIhIyhbvlo1Lly7RpEmTWMsNBgMtW7bk7NmzFilMREQkpUr4dOW2Kd5ho2DBgvz9999xrrtx4wZ58uRJdFEiIiIplYFnN2JLyMNW48ZrdaNcv37d9O927doxfPhw7OzsqFevHlmyZOHBgwfs3LmTGTNmMGbMmCQrVkREJCWw5VaKhHitsOHj42P2whmNRiZPnhxr3IbRaKRTp06cOnXKslWKiIhIivVaYWPs2LFKaSIiIq9JvzLNvVbYaNq0aVLXISIiYhsMiehGsWBImT17Nnv37mXJkiWmZadOnWLMmDGcOHECV1dXfH19ad++/UuPs3HjRmbMmMGVK1fIly8fAwYMoGrVqvGqJUGTegUFBXH48GHTjdgAoqOjefLkCQcPHmTq1KkJOayIiIhNsLNyy8Y333zD9OnTKVeunGnZ/fv3+fTTT6lVqxajRo3ir7/+YtSoUbi6utKsWbM4j7Nv3z4GDBjA4MGDqVSpEqtXr6Zbt26sW7eOggULvnY98Q4bGzduZMCAAURGRpqSm9FoNP27QIEC8T2kiIiIDUnMZayJSyk3b97Ez8+PQ4cOkT9/frN1q1atwtHRkZEjR+Lg4EDBggW5dOkSAQEBLwwbAQEB1K5dm9atWwMwaNAgjhw5wuLFixk9evRr1xXvS1/nzZtH0aJFWbt2LU2bNqVRo0b88ssvDBgwAAcHB4YOHRrfQ4qIiIgFnDx5kgwZMrB+/Xq8vLzM1h08eJBy5crh4PD/7QwVK1bkwoUL3L17N9axoqOjOXz4MBUrVjRbXqFCBQ4ePBivuuLdsnHhwgUmT55M0aJFqVSpEgsWLKBgwYIULFiQu3fvMnfuXN555534HlZERMQmJGbqcQPPppvw9fV94TZbt2594TofHx98fHziXBcUFISHh4fZsqxZs8L/njNTpkxm6x4+fEhoaCjZs2ePtc+NGzdedhqxxLtlw87ODldXVwDy5cvH+fPniY6OBqBKlSr8888/8T2kiIiITbEzGBL0SEpPnz6NdYd2JycnAMLCwuLcHohzn7i2f5l4t2wUKFCAQ4cOUa5cOfLly0dERASnTp2iWLFiPHz40GzQqIiIyH9RYnJDzpw5X9p6kVCpU6eO9Ts6JjSkTZs21vYxQSSufdKkSROv5453y8ZHH33E9OnTmTJlCs7OzlSoUIGhQ4eyZMkS/P39KVasWHwPKQLAls2beLtCWdKmTUuhgnmZNGEcRqPR2mXJS7hnc+XGjolUKVPIbHm9KsXZuaQ/9/dN5dzmL5k8oBnOaZ2sVKW8iNFoZNniBfi8XebZ57mXJyOH9ufRw4fWLk2SQPbs2bl165bZspivs2XLFmt7V1dX0qZNG+c+/+5aeZV4h43mzZvj5+dHREQEAKNHjyYsLIwxY8YQGRmJn59ffA8pwt49e/igSSM8Cxdh7dq1tGzly4hhfkwcP9bapckL5MmRkZ/ndMfVxfwvomplPVj9VUdCQsNoPehr+k1czTul32LT/J7Y28f7I0eS0NwZU/h8QC9q1qnHunXr6NqrH+tWr6RjmxYK+omUHG/EVq5cOQ4dOkRUVJRp2d69e8mfP3+s8Rox51C6dGkOHDhgtnz//v2UKVMmXs+doHk2WrVqZfp3njx52LhxI/fv38fNzS0hh3ujPD09GTdunEUmKpsxYwY//PAD27Zts0Bl/21jvxxFSa9SLFq8BCcHqFGrLhEREUyeOJ6evfvGu8lOko7BYKB1wwqM6xP77s8A7T+ozKnzQTTqNpuIyGcfarsP/8PfP4/ik0YVWfTDnjdZrrxAdHQ0s76aRKs2n+E34kuyZ3CkeLmqpHd1o2u7Vhz76zBe3vH7hSL/LznOINqsWTMWLFiAn58fn332GceOHWPx4sWMGjXKtM2jR4+IiIgw/T7/9NNP6dixI0WLFqVq1aqsWbPGNDFYfLzWnxnXr19/6ePGjRs8ffrU9HVytmvXLurXr2/tMuQ5YWFh7Nj+B+83MQ+ATZp+QEhICLt37bRSZRKXEoVyMn1oC5b9vJ/2wxbHWp8vVyZ+23vKFDQAbt8P4cyFIOpVLf4mS5WXePToIU2bf8z7H7QwW16g4LMusUsXz1ujLJtgMCR8gGhShpRMmTKxYMECLly4QJMmTZg5cyYDBw6kSZP//8NhzJgxfPDBB6avK1euzNixY1mxYgVNmjRh3759zJ07N14TekECb8T2Ksn5RmxZsmSxdgnyLxfOnyc8PJxChcwvySr41lsAnA0MpFbtd61RmsThStB9ijcaxbVbwbHGagAEPwwlb07zJlkHBzvcs2fEMVWCGlMlCWTI4MroCbFne9748zoAPIto/F1KN378+FjLSpYsyXfffRevfRo3bkzjxo0TVYtVb8Q2ePBgzp07x/fff29aFhQURI0aNfj6669xcnLC39+f48eP4+bmRo0aNejXrx/Ozs7AsxBUq1Ytdu3axd27d5k2bRrZs2fniy++4K+//iI6OprSpUszcOBAPD09gdjdKD///DMBAQFcuHCBLFmy0Lp1az799FMAgoODmTZtGtu2beP+/fsUK1aMfv36UbZs2TjP51Xbz5gxg927d5MzZ07++OMP3n//fUaMGGHx1zWlCQ4OBiB9+vRmy11cXIBnf4FJ8nH/YSj3H4a+cP3P24/xaZN36Ne2Fot/3Ecap1SM6NaA9OlS8zg0fpfLyZv15/69zJ3uT536jfAsXNTa5aRoybEbxZqseiO2Jk2a8Mknn3Dp0iXy5s0LwPr168mWLRsZM2bkww8/pHPnzowZM4Y7d+4wceJE2rVrx3fffWcKPytWrGDevHm4uLjg6enJRx99hKenJ2vWrCEyMpIJEybQvXt3fv3111jPv2nTJgYMGECfPn149913+fvvvxkyZAjOzs40bdqUdu3aERERwYQJE8iSJQtLly6lbdu2rFixghIlSpgdKyoq6rW2P3LkCCVKlODHH380G6QTH4mZMCY5Mv5vnhaDwWA6r+fP0c7OzmbOt1Rhd2uXYFFv5cli+v/jJ8+CxI6DZ7G3s2NEt4Z82asxEZFRrN/2FzsP/0MB98wp+jVIZW8r34nmHOwM7Ny5k9Yfvk/e/AWYOmuebZ2rFU5Fd0o3Z9U2zfLly5M7d25++uknunfvDsBPP/3E+++/z8KFC6lUqRJdu3YFnk0g5u/vT61atThw4AAVKlQAoFq1arz99tumY16+fJl33nkHd3d3HBwcGDt2rGniMTs78yEq33zzDfXq1aNjx46m53j8+DFp0qRh165dnDx5kp9++sk049rw4cM5evQoCxcu5KuvvjI7Vny279mzp+mv9oRysqHW6KyZXQF4+vghjv87L0cHePzoEQCZMmawmfPdu2KwtUtIErOHtzT9e9GYtmbrUjnY0+zd/x9oaKuvQUq2cuVK2rZti6enJ5s3b473ZY0Sm667MmfVj3CDwUDjxo1NYePUqVMEBgYyffp0unfvzqVLl/D29o6137lz50xhI6ZFJEafPn1Mg1kqVqxIlSpVqFevXqygAXDmzBnq1atntqx58+bAs5vPuLi4mE3tajAYKFu2LDt3xh6wGBgY+FrbZ8qUKdFBAyAsMtGHSDbc8xbE3t6e04H/EB75LGiER8Lfp5/NRvuWZ1GbOd/qvrH7Q1My7yJ5mD28JV1HL+fxkzAWj/uUsfM3cOvuI/Yfu2Dazt7OwIZ5vfhl+zGmL025V2/9MLevtUuwuNnT/flyhB9Vq1bl66WrIU16gh7Y1uSMmV1S4fCGb8Oqlg1zVv97MWZE7LFjx9i4cSPe3t7kz5+f6OhoGjZsSOfOnWPt8/wltqlTpzZb16pVK+rWrcv27dvZu3cvU6ZMYcaMGaxbt47MmTObbevg4PDCb4jn72T7vOjoaLOb2MR3+3/XmxDG/z1shVPq1FSuUpV1P6ylT9/+gAEjsHbtalxdXSlbrrzNnO9fp69auwSLckn37JLkfy7f5tHjJwB45M2Gb8OKFG00ksjIZ11k7Zq+Q3rn1CxcsztFvwYRUbbynfjM0m8C+GL4UBo1+YDvVy7j3hPbO0fAtj4wUyirt/TkypWL8uXLs2nTJjZs2GC6BKdQoUKcPXuWvHnzmh5RUVGMGzfuhTeAuXPnDqNHjyYiIoKmTZsyadIk1q9fz+3bt2NNSgJQsGBBjh8/brZs7NixdO3aFU9PTx4+fEhgYKDZ+kOHDvHW/66SeF58txdzg4d+zp8H9tPq4w/ZuHEjI0cMY6r/JAYMHqo5NlKYH347QtZMLiwY/QnVy3vQo1UNpg5uzqpNB9l95Jy1y5P/uXUziNGfD8Q9dx7adezK4cOHOfTnfg7/73H3zm1rl5ii2RkS9rBViQobjx494ty5c4SHhyd4sCM8G4C6cuVK7t+/b5oDo127dpw6dYrhw4fzzz//cPToUfr378+FCxfIly9fnMdxdXXljz/+4PPPP+fUqVNcuXKF5cuXkypVKooXj319f8eOHdmwYQPffvstly9f5pdffmHlypXUrl2bd955B09PT/r168f+/fs5d+4co0aNIjAwkDZt2sQ6Vny3F3PVa/iwYtUaAs+coXHjxqxcsYyxEybRt98Aa5cm8XT+6h2a9pzHW3mzsuarznT5qBoTFmym/bBvrV2aPOf33zbx9MkTrl65TON6PlSqVIkGtavSuG41GtetxrYtG61dYoplIOFhw1bzRoK6Ufbv38/kyZM5ceIEBoOB77//noCAALJnz87gwfEf/FWnTh1Gjx5NrVq1TOMZSpUqxYIFC5g2bRpNmzYlTZo0VKxYkUGDBsW6A53pZBwcCAgIYMKECbRt25YnT55QpEgR5s+fT548eWJt7+PjwxdffEFAQAATJ04kV65cDB061NS6smjRIiZMmECPHj0IDw+nWLFifPPNN5QqVSrO547P9hLb+42b0LhxE5wcno1JUctn8rfz0FnSeD8b3P38VSbb9p9mW6vT1ipLXkOLVm1p0aot8Owqm+wZHAl6EG6b3ShWoDEb5gzGeE6Av3fvXjp06IC3tzfVq1dn8uTJrF69mt27d/PVV18xYMAA0zwVkjSijRCe8IakZM0ANh02Mpbrbu0Skkypwu7sXTGYSh+PT9HjMl4mcKu/tUtIErYeNnJkcMThDV7Ke/dxOGO3XXj1hnEY6pOfTOni/oM6JYt3N8pXX31FzZo1WbJkCW3atDHdrKdjx4589tlnZhN0iYiI/BdpzIa5eIeNU6dO0axZMyB2M9E777zDtWvXLFOZiIhISmR4NoNoQh62Omgj3mM2XFxcuH077lHKN27csMgcEiIiIimZncZsmIl3y0bNmjWZOnWq2SWjBoOBoKAg5s6dS/Xq1S1Zn4iIiKRw8W7Z6NevH0ePHuXDDz80TZLVt29fgoKCyJEjB3372t4MeyIiIq/LQMLnlbDV9pB4h40MGTLw/fffs27dOvbt20dwcDAuLi74+vqaLlEVERH5L1MvirkEzbPh6OjIhx9+yIcffmjpekRERFI4QyLGbNhmSol32Fi3bt0rt2ncuHECShEREbENatkwF++w8aIZQg0GA/b29tjb2ytsiIiIiEm8w8bWrVtjLQsNDeXQoUPMnz+fWbNmWaQwERGRlCjm3igJ3dcWxTts5MqVK87lhQoVIiIigi+++ILly5cnujAREZGUSvNsmLPoLeY9PDw4efKkJQ8pIiKS4iR4BlEbZbGwER4ezqpVq8iUKZOlDikiIiI2IN7dKD4+PrHuiRIdHc39+/cJCwtj0KBBFitOREQkxUnMTdVstHUj3mGjQoUKcS53dnamRo0avP3224kuSkREJCUz2GpqSKB4h42GDRtSqlQp0qZNmxT1iIiIpGi6GiW2eI/ZGDhwYJyXv4qIiIjEJd4tG46Ojjg5OSVFLSIiIjYhwWM2bFS8w0anTp0YPnw4p0+fplChQqY7vz6vXLlyFilOREQkJfr3hRT/dfEOGyNGjABg9uzZgPkLajQaMRgMnDp1ykLliYiIpDxq2TAX77Dx7bffJkUdIiIiYqNeK2zUrFmTWbNmUbhwYcqXL5/UNYmIiKRo6kUx91ph49q1a4SHhyd1LSIiIines0tfE5Y2bDWjxLsbRURERF7CSjOI7t+/n08++STOde7u7nFOW/HDDz8wePDgWMu3bNlC3rx5E17MvyhsiIiIWJg1ulG8vb3ZtWuX2bLAwEA6duxI586d49znzJkzlC9fnilTppgtd3Nzs2htrx02unXrhqOj4yu3MxgM/Pbbb4kqSkREROLH0dGRLFmymL6OiIhg3LhxvPvuuzRv3jzOfQIDAylcuLDZfknhtcNG0aJFLZ50REREbJFdMhh9sWzZMm7cuMHXX3/9wm3OnDlDnTp1kryWeLVslCxZMilrERERsQmJ6Ua5fv06vr6+L1z/OrcMCQsLY+7cubRp04asWbPGuc29e/e4c+cOf/75J0uWLCE4OBgvLy/69+9P/vz5E1x/XOJ9bxQRERF5sZgbsSXkYan2kB9//JGwsLCXhpbAwEAA7O3tmTBhAlOnTiU0NJSWLVty584dC1XyjAaIioiIJCM5c+ZM9A1P161bx7vvvkvGjBlfuE3FihU5cOAAGTJkMC2bNWsWNWrUYO3atXTs2DFRNTzvtVo2mjRp8tKCRURE5P/ZGQwJeljCvXv3OHLkCPXr13/lts8HDYC0adPi7u7OzZs3LVJLjNcKG+PGjSN37twWfWIRERFbZTAk7GEJhw8fxmAwvHLG7+XLl1OhQgWePn1qWhYSEsLFixd56623LFPM/2jMhoiIiEUlrFXjWctG4hPH6dOnyZ07N2nSpDFbHhUVxe3bt03hokaNGhiNRgYOHMjZs2c5fvw4PXr0wM3NjSZNmiS6jucpbIiIiNiQO3fu4OrqGmv5jRs3qFy5Mhs2bAAgR44cLF68mMePH/Pxxx/Ttm1bXFxc+Pbbb0mdOrVFa9IAUREREQsykPAuEUv0pIwcOTLO5e7u7pw5c8ZsWZEiRVi4cKEFnvXlFDZEREQsTN0G5hQ2RERELMnw7NYdCd3XFilsiIiIWJiNZoYEU0uPiIiIJCm1bIiIiFiYpSboshUKGyIiIhamqGFOYUNERMSCrH3pa3KkMRsiIiKSpNSyISIiYmEJvvTVRilsiIiIWJi6DcwpbIiIiFiYWjbMKXyJiIhIklLLhoiIiIWpXcOcwWg0Gq1dhMRPtBHCo6xdRdIwAE4OEBYJtviN+Tgs0tolJBl7O3BN40Dwk0iioq1dTdL4YMEBa5eQJAplScf8ll50XH6Us7cfW7sci1vWtjQ5M1j2lukvExIWyabTtxO0b93CWXB2sr12ANs7IxERESsykPAxCrbaIqKwISIiYmEaIGpOA0RFREQkSallQ0RExMLUrmFOYUNERMSSDAm/N4qtphSFDREREQuzs9XUkEAasyEiIiJJSi0bIiIiFqaLUcwpbIiIiFiQ4X//JXRfW6SwISIiYmFq2TCnMRsiIiKSpNSyISIiYmG6GsWcwoaIiIiFqRvFnMKGiIiIhSlsmNOYDREREUlSatkQERGxMFu9hDWhFDZEREQsyADYJTBr2GpEUdgQERGxMLVsmNOYDREREQszGBL2SKxr167h6ekZ6/H999/Huf39+/fp168f5cqVo1y5cgwbNozQ0NDEF/IvatkQERGxEWfOnMHJyYnffvsNw3PpxcXFJc7te/bsSVhYGN988w0PHz7Ez8+PUaNGMWHCBIvWpbAhIiJiYdbqRgkMDCR//vxkzZr1ldseOXKEAwcOsGHDBgoWLAjA6NGj+eyzz+jbty/ZsmWzWF3qRhEREbEwO0PCHol15swZ3nrrrdfa9uDBg2TJksUUNADKly+PwWDg0KFDiS/mOWrZEBERsbDEtGxcv34dX1/fF67funXrC9cFBgaSJUsWWrZsycWLF8mbNy9du3alSpUqsba9efMmOXLkMFvm6OiIq6srN27cSHD9cVHLhoiIiA0IDw/n4sWLhISE0Lt3b+bPn0+JEiXo0KEDe/fujbX9kydPcHR0jLXcycmJsLAwi9amlg0RERELSsyVJQYD5MyZ86WtFy/i6OjIn3/+iYODgylEFC9enHPnzrFw4UIqVapktn3q1KkJDw+PdZywsDDSpk2bsBN4AbVsiIiIWJghgY/ESps2bazWCg8PD27evBlr2+zZs3Pr1i2zZeHh4QQHB1t0cCgobIiIiFicncGQoEdinD59Gm9vbw4ePGi2/MSJE3EOGi1XrhxBQUFcunTJtGz//v0AlC5dOlG1/JvChoiIiA3w8PCgUKFCjBo1ioMHD3Lu3DnGjRvHX3/9RefOnYmKiuL27ds8ffoUAC8vL0qXLk2fPn04duwY+/btY8SIETRu3FgtGyIiIsmdNbpR7OzsmDt3LiVKlKB37940adKEo0ePsmjRIjw9Pblx4waVK1dmw4YNz2o0GJg5cybu7u60adOG3r17U7VqVUaOHJnISmLTAFERERFLs9KtUdzc3Bg7dmyc69zd3Tlz5ozZskyZMjF9+vQkr0thQ0RExKIMiZhnwzZv4KawISIiYmGWuKmaLdGYDREREUlSatkQERGxMDVsmFPYEBERsTSlDTMKGyIiIhb07DLWhKUNW80oGrMhIiIiSUotGyIiIhamq1HMKWyIiIhYmLKGOYUNERERS1PaMKMxGyIiIpKk1LIhIiJiYQmfrtw2KWyIiIhYkiERA0RtNKMobIiIiFiYjWaGBFPYEBERsTSlDTMaICoiIiJJSi0bIiIiFqYBouYUNkRERCzIQMIHiNpqRFE3iiQbWzZv4u0KZUmbNi2FCuZl0oRxGI1Ga5cl8XDlyhXy5sjMrh3brV2KvEIWZ0d+6lwer1zpzZZ75UrPV82Ksb5TedZ8VpZR73mSM0NqK1WZchkS+LBVChuSLOzds4cPmjTCs3AR1q5dS8tWvowY5sfE8WOtXZq8psuXL1G7dm0ePnhg7VLkFbK5ODG5SVGcncwbtwtkTsvkJkV58DSSMZsDmfbHBXJmSM2M5sVJn1oN4ZJwKTpsrF27Fk9Pzzd6jBkzZuDj45Oo55TYxn45ipJepVi0eAl169Zl1Ogv6dNvAJMnjufJkyfWLk9eIjo6muVLvqH62+W5ffu2tcuRlzAAdYtmZd7HJcmQJlWs9fWKZuXS/SeM/OUM+y8Gs+Ofuwxa9zcZ0qSibtGsb77glExNG2ZSdNioX78+u3bteqPHaNeuHatXr07Uc4q5sLAwdmz/g/ebNDVb3qTpB4SEhLB7104rVSav4+SJYwzo3Z2PW/myZMkSa5cjL1Egc1r61CjAllO3GbflbKz1F+6GsubIDZ7vvLwXGsHj8Eh1pcSTIYH/2aoU3S6WOnVqUqdO3A9AfI+RLl060qVLl6jnFHMXzp8nPDycQoU8zJYXfOstAM4GBlKr9rvWKE1eg7t7Hv48dprcud35a3/iwr8krVuPwmm1+DB3QsJjjdUA2HDyFmdvPzZbVso9PelTp+Li3dA3VaZN0C3mzSX7lo3BgwfTvHlzs2VBQUEUKVKE77//3qwLxNPTk6lTp1KjRg3eeecdzp8/z5MnTxgxYgQVKlSgdOnS+Pn50a9fPwYPHgzE7kbx9PRk1apVfPrpp5QsWZIqVaowb9480/p/d6Pcu3ePQYMGUaFCBcqUKUOHDh24ePEiAEajkQULFlCvXj2KFy9OmTJl6NSpE1euXEmKlyrFCg4OBiB9evMPPxcXFwAePXr4pkuSeMjo5kbOXO7WLkNew6OwSO6EhL/29hnSONC/ZkFuPQpj06lbSViZ2Lpk37LRpEkTPvnkEy5dukTevHkBWL9+PdmyZcMQR3T87rvvCAgIICoqigIFCtCzZ0/+/vtvpk6dSubMmZk1axabN2+mcePGL3zOiRMnMmzYMIYPH86PP/7IlClTKFOmDGXLljXbLjIyknbt2mEwGJg1axYZM2Zk0qRJtGvXji1btrB06VLmzZvHhAkT8PT05OrVqwwbNozx48cza9asBL8mtta1Z4yOBsBg+P9GxOfP0c7OzmbO1z7Zx/uEs3/uTbK3s81zLZTFdlo13V1Tm/7/NOLZz2CejGlM6zOkcaB3jQK4pXNkytZz5HZNE+dxUoJU9m/+E8RWPrMsJdmHjfLly5M7d25++uknunfvDsBPP/3E+++/j51d7E+z999/nxIlSgDPLsPbvHkzCxYs4O233waeBYnDhw+/9DmbNGnC+++/D0Dv3r1Zvnw5hw4dihU29u3bx6lTp9i4cSMFChQA4IsvvmDhwoUEBweTJ08exo8fb2oJyZUrF/Xq1eOXX35JxCvyjFOyf+deX9bMrgA8ffwQx/+dl6MDPH70CIBMGTPYzPk6OdjIibyCs5M9rmls71znt/SydgkW17/WW6Z/f17PI85t/OrGvVxeQmnDTLL/NDAYDDRu3NgUNk6dOkVgYCDTp0/nyJEjsbaPaf0A+PvvvwHw9vY2LXNycjKFkRcpWLCg2dfOzs5ERETE2u7MmTOkT5/eFDQAsmTJYuqi8fHx4ejRo0yfPp1Lly5x7tw5zp49S7Zs2V7jzF8uLDLRh0g23PMWxN7entOB/xAe+SxohEfC36f/AeAtz6I2c75PImzkROLw/B+PIWFRBD+xvXMd+MNJa5dgMR5Z09G/1ltM/u0fnkZE83k9D77cGEgaR3u6Vs3Hk/Aopv9xgesPnlq71EQb06gwWZyd3uhz2vJgz4RI9mEDnrU0zJw5k2PHjrFx40a8vb3Jnz9/nGHj+cGe9vb2wLNL8+LD0dEx1rK4JpdycHCIsysnRkBAADNmzKBp06aUL18eX19ftm7dmuiWDeP/HrbCKXVqKlepyrof1tKnb3/AgBFYu3Y1rq6ulC1X3mbONyp+34opy3MNjVHRtnmu/x48mZKldXz2+Xg1+Cmh4VHAs8+VrlXzEfQgjIHr/ubO49cf35GcRUTZyidIypUielVz5cpF+fLl2bRpExs2bKBJkyavtZ+npycGg4G//vrLtCwiIsLU4pFYb731Fg8ePODSpUumZffu3aNcuXIcOnSIOXPm0L17d0aOHEmLFi0oVaoUFy9e1KyYcRg89HP+PLCfVh9/yMaNGxk5YhhT/ScxYPBQ0qRJuX3FIilJmwruONgZ+Gb/FbK4OFIku7PpkTPDm20ZSOkMhoQ9bFWKaNkAaNq0KaNHjyYyMpL69eu/1j65c+emXr16fPHFF4wePZqsWbMSEBDAjRs3Xtoi8boqVapE8eLFGThwIEOHDiVt2rRMnjyZTJkyUaJECXLkyMHu3bvx8fHBzs6OH3/8kS1btpA5c+ZEP7etqV7DhxWr1vDlqBE0btyYnLlyMXbCJHr36Wft0kT+M/K4pQVg1HuxJzrc9PctJvz6z5suKcWy4dyQICkmbNSpU4fRo0dTq1Yt0yWRr+OLL77gyy+/pEePHhiNRho0aECpUqVIlSr27HnxZWdnx+zZsxk/fjzt27cHoEKFCixcuBBHR0cmTpzI6NGjadasGenSpcPLy4tRo0YxcuRIrl69iru7Lhd83vuNm9C4cROcHJ6NSVH7T8pTvXp17odG2GQXiq05eu0hNabtAf7/KpuOy4/aVFeRVSltmDEYbbhNPywsjJ07d1KxYkWcnZ1Ny+vUqUOjRo3o1q2bFatLuGgj/K+L1eYYwKbDxmNbGekaB3s7cE3jQPCTSJsNGx8sOGDtEpJEoSzpmN/Sy2bDxrK2pd/oDKgRkdFcvBuWoH3zZXIilUOKGOEQLymmZSMhHB0dGT16NOXKlaNr167Y29uzevVqrl+/Tt26da1dnoiIiEUFBwczZcoU/vjjD0JCQvD09KRfv36xpm6I8cMPP5iuoHzeli1bzK7uTCybDhsGg4F58+YxadIkWrRoQVRUFEWLFuXrr7+OdXmriIiIRSRmsGciu1/69u3L3bt3mTJlCm5ubixfvpz27duzdu3aOH/vnTlzhvLlyzNlyhSz5W5ubokr5F9sOmwAFClShK+//traZYiIyH+INYZsXLp0id27d7NixQpKly4NgJ+fHzt27ODnn3+mV69esfYJDAykcOHCZMmSJUlrs72OIREREWuzwi3mM2bMyPz58ylevPj/l2EwYDQaefDgQZz7nDlzhrfeeivOdZZk8y0bIiIiKcn169fx9fV94fqtW7fGuTx9+vRUq1bNbNnGjRu5fPkylStXjrX9vXv3uHPnDn/++SdLliwhODgYLy8v+vfvT/78+RN3Ev+ilg0RERELMyTwP0s6dOgQQ4cOpWbNmmZ3K48RGBgIPJtte8KECUydOpXQ0FBatmzJnTt3LFqLWjZEREQsLDHzRubMmfOFrRev67fffqN///54eXnFGvwZo2LFihw4cIAMGTKYls2aNYsaNWqwdu1aOnbsmKganqeWDREREQuzwpANk6VLl9KjRw+qVq1KQECA2T3D/u35oAGQNm1a3N3duXnzpoWqeUZhQ0RExEYsX76cL774glatWvHVV1/FeWPR57etUKECT5/+/519Q0JCuHjxosUHjSpsiIiIWJoVmjYuXLjA2LFjqV27Np06deLu3bvcvn2b27dv8+jRI6Kiorh9+7YpXNSoUQOj0cjAgQM5e/Ysx48fp0ePHri5ub32DU9fl8KGiIiIRSV0eGjiEsfmzZuJiIjg119/pXLlymaPMWPGcOPGDSpXrsyGDRsAyJEjB4sXL+bx48d8/PHHtG3bFhcXF7799tuXdr0khAaIioiIWJCBhA8QTUzjRufOnencufNLtzlz5ozZ10WKFGHhwoWJeNbXo7AhIiJiYbrpqzl1o4iIiEiSUsuGiIiIhSVmng1bpLAhIiJicUobz1PYEBERsTC1bJjTmA0RERFJUmrZEBERsTA1bJhT2BAREbEkQyK6UWw0pShsiIiIWJilbxef0mnMhoiIiCQptWyIiIhYmho2zChsiIiIWJiyhjmFDREREQuy1o3YkjOFDREREQvTAFFzGiAqIiIiSUotGyIiIpamhg0zChsiIiIWpqxhTmFDRETEwnQjNnMasyEiIiJJSi0bIiIiFqarUcwpbIiIiFiYulHMqRtFREREkpTChoiIiCQpdaOIiIhYmLpRzClsiIiIWJCBhA8QtdWMorAhIiJiYWrZMKcxGyIiIpKk1LIhIiJiYWrYMKewISIiYkkGEp42bDSlKGyIiIhYmGYQNaewISIiYmEaIGpOA0RFRERsRHR0NNOnT6dKlSp4eXnRrl07Ll269MLt79+/T79+/ShXrhzlypVj2LBhhIaGWrwuhQ0RERELMyTwkVizZ89m5cqVfPnll3z33XcYDAY6dOhAeHh4nNv37NmTK1eu8M033zB9+nR2797NqFGjLFCJOYUNERERS7NC2ggPD+frr7+mR48eVKtWjcKFCzN16lRu3rzJr7/+Gmv7I0eOcODAAcaNG0exYsWoVKkSo0eP5scff+TmzZuJK+ZfFDZEREQsypDg/xKTOE6fPs3jx4+pWLGiaVn69OkpWrQof/75Z6ztDx48SJYsWShYsKBpWfny5TEYDBw6dCjBdcRFYUNERMQGBAUFAZAjRw6z5VmzZuXGjRuxtr9582asbR0dHXF1dY1z+8TQ1SgpkAFwtLd2FUkjJtM72oPRqpUkjVRpbPSNe46Lk+2e47K2pa1dQpJIZf/sJ29Mo8JERNneT15WF8c3+nwGwCmBv10NwPXr1/H19X3hNlu3bo1z+ZMnT4BngeF5Tk5OPHjwIM7t/71tzPZhYWHxqPrVFDZSIIPBZud9MbHZc/wPXA9nb2e755gzQ2prl5Cksjg7WbsEm5DYz6/bt28naL/UqZ99f4aHh5v+DRAWFkaaNGni3D6ugaNhYWGkTZs2QTW8iMKGiIhIMuLl5fXC1ouXiekSuXXrFnny5DEtv3XrFoULF461ffbs2fntt9/MloWHhxMcHEy2bNni/fwvozEbIiIiNqBw4cI4Ozuzf/9+07KHDx/y999/U7Zs2VjblytXjqCgILN5OGL2LV3asl2GatkQERGxAY6OjrRu3ZrJkyfj5uZGrly5mDRpEtmzZ6d27dpERUVx7949XFxcSJ06NV5eXpQuXZo+ffowcuRIQkNDGTFiBI0bN7Z4y4bBaDTa3mggERGR/6CoqCimTJnC2rVrefr0KeXKlWP48OG4u7tz9epVatasybhx42jatCkAd+/eZdSoUezcuRMnJyfq1q3LkCFDcHKy7PgdhQ0RERFJUhqzISIiIklKYUNERESSlMKGiIiIJCmFDREREUlSChsiIiKSpBQ2REREJEkpbIiIiEiSUtgQERGRJKWwISIiIklKYUNELCJmMuKHDx9auRIRSW4UNkTEIgwGA7/++iuff/45169ft3Y5IpKMKGxIshDzV3FERATh4eFWrkYS4uzZswQEBFC9enWyZ89u7XLkNURHR8e5PCoq6g1XIrZOt5gXqzMajRgMBnbv3s0PP/zA48ePadiwIfXr17d2afKaAgMDGT58OEajkWrVqlm7HHkN0dHR2Nk9+3vzxx9/5MqVKzg6OtK2bVscHR2JiorC3t7eylWKrVDLhlidwWBg8+bNdOnShZCQECIjI+nbty8bNmywdmnymm7cuEGaNGk4fvw4Z86cwc7O7oV/NYv1GY1GU9CYMGECY8eOZcuWLXz33Xf07NmT8PBw7O3t1cIhFqOwIVb3999/88UXXzBs2DDmzp1Lr169AOjbty9r1qyxcnXyOqpVq0afPn3w9vZmwIABHDlyRIEjGTMYDAAsWrSI9evXExAQwPr166lVqxZ//PEHPXr0ICwsDHt7eyIjI61crdgChQ2xulOnTlG4cGGaN29OcHAwc+fO5cMPP6Rdu3b4+fnx66+/WrtEeU7M+Jrjx4+zfv16ZsyYwd69eylcuDCjR4/Gw8ODgQMHKnAkY0ajkdDQUPbv30///v0pWbIkJ0+eZPfu3bRo0YKrV6/St29foqKicHBQb7sknsKGWEXMLyx4dqnkvXv3uHv3LitXriRjxox07drVNGajR48ejB8/Xk26yURMt1e7du34/vvvWbZsGUOGDKF37964u7szaNAg8uTJw5AhQzh48KCpuV6s6/mfOYPBgL29PVevXuXevXvcvHmTZcuW0bBhQz7//HNKlizJ1q1b8fHx4cqVK1asWmyFPgXkjYr5wAsLCwOejXp/7733GDx4MEajkYMHD1KmTBmyZ89OmjRpqFatGqNHj6Z58+YarGZFz/+iOn36NBMmTGDIkCEsWLCAffv28dlnn3H16lWGDh3KW2+9Rffu3XF1dWXMmDGEhYWZ7S9vXnR0tKnr5PHjx9y6dQsnJyc6duxI1apVOX78OE+fPqV8+fKkSpWKXLlyUbNmTWrXrk3OnDmtXL3YArWPyRsTc9XJjh07WLFiBbdv36ZDhw7UqVOHrFmzsnfvXg4ePIifnx/wbIT8w4cPqV+/Ps7Ozlau/r9p2bJlFChQgEqVKpmuXrhw4QIZM2akRo0apgD4wQcfEBYWxpIlSzh58iTe3t7079+fXLly4eTkZOWz+G97fjDonDlz2LFjBxcuXKBdu3a0bNkSZ2dnZs+eTXh4ON7e3gAcPXqU0qVL06VLFwBdmSKJprAhb0zMpE/9+/enTZs2FChQgKJFiwIQGRlJtmzZyJAhA2PGjMHR0ZGDBw/y7bffKmhYSVBQEBs3buTKlSv4+/tTtmxZAM6dO8f9+/fJmDEjAOHh4aROnRpfX1+mT5/OkSNH8PLyMm0v1hXTojFr1ixWrFhBnz59cHJyInXq1KafLQcHB65du8b8+fPZs2cPDx48oEOHDsCzsKKgIYmlbhR5Y27fvs2cOXMYPHgwffv2pXv37ly7do0vvviCIUOGkDVrVrp3706GDBlwdHRk6dKlFC5c2Npl/2dlz56dXr16UaxYMQYNGsSff/4JQNWqVbl58ybz588HMM3J8PjxYwoXLkz+/PmtWbYAJ06cMP3baDRy69Yttm7dyuDBg2nWrBkNGjSgVq1aAOzbt4+mTZsSGRnJtm3bSJ06NatWrcLBwYGoqChTWBFJDLVsyBsTGhrK48ePKVmyJNevX2fcuHFcuXKFJ0+eYGdnR+fOnfn2229p3rw5kZGRGgVvRTFdXuXKlcNgMLBgwQIGDx7MuHHjKF++PJ988glz587FaDTy6aefcv/+fVauXMnNmzcpVKiQtcv/T5s8eTKnTp1iwYIFwLOWjYiICIKCgnB1dQWejeGIabHo3bs37dq1Y/Xq1YSGhpIuXToMBoN+BsWi9J0kb0zu3LnJmDEjHTt2JCQkhFKlStG6dWsaN27MokWL2Llzp+mvKH3IWZfBYDCN0YjpDpk/fz5Dhgxh6tSp9OrVC4PBwIwZM/jmm29wc3Pj8ePHzJ49WwMKraxBgwb07t0bg8HA1atXcXd3J3369ADs3r2bypUrY2dnZ5o/4+233+bu3bvY2dmZulWMRqN+BsWi9N0kSSLmL+Nbt24RFhZmamKfNGkSv//+O25ubtStW9f0gXbjxg3Spk1LWFgYqVOntnL1/20x711oaChPnz4lQ4YMlC1bFhcXFyZOnEifPn2YOnUqAwcOpFGjRuzfv5/cuXNTpEgRcuTIYe3y//Niuh43btzI0KFDmTZtGlWrVqVjx45MmzbNFPif/9l76623zI6hrhOxNINR16SJhcX8svrtt9+YPXs2t2/fJjo6msKFCzNy5Ehy585NaGgoP//8M25ubuzbt49169axbNkyPD09rV3+f1rMe7dt2zYWLVrE1atXcXZ2pnz58vTt25egoCDGjh3LxYsXGT9+POXKlbN2yfI/Me9djH379rFo0SLOnTvH+PHj8fb2ZsqUKXz77bdUq1aN7Nmzc+bMGYKDg/nhhx/UkiFJSmFDksSePXvo1q0bvXv3pkCBAoSGhjJlyhQcHR0JCAggJCSEMWPGcOnSJXLlyoWfn58GgyYTO3fupEePHrRv356CBQty+PBh9uzZQ44cOZgxYwaXL19m2rRpHDhwgICAAEqXLm3tkv/znr+p2qNHj3BxcQHg5MmTzJgxgzNnzjBlyhS8vb359ddfWbJkCRkzZiRz5swMGTLENBhUV51IUlHYkESL+YsqKioKOzs7oqKi+PLLL0mdOjWDBw82bXflyhU+/fRTPD09mTVrFo8ePTINSNPlrdZnNBoJCwujX79+FC5cmB49epjW/fDDD8ybN49atWrRv39/9u7dy/fff0+vXr3ImzevFauW51s0Fi1axK5du3B0dOTjjz+mSpUqnD59mmnTpnHmzBnGjx9PhQoVzMIJoMGgkuR06askSlhYGPPmzePevXvY29tjMBhwcHDg9OnThISEmLaLjo4md+7cdO7cmaNHj3Lx4kVcXFzIli2bgkYyYTAYTJN2ZciQAYCIiAgAmjRpgre3N7t27SIqKopKlSoxduxYBQ0re35m0OnTpzN9+nRy5crFuXPnGDlyJGvXrqVw4cL06tWLwoUL4+fnx++//24WNDQYVN4EhQ1JlMjISLZv306PHj1YuXIlc+bMITQ0lKxZs3L79m1u375tNoNhunTpSJcunWl0vFjXvxs2Y5rRAwMDAUiVKhXh4eEAlChRgujoaJ48eQKggbxW9vzP1bVr1zhx4gQLFixg9OjRbNmyBQ8PD+bPn8+aNWtMgSNLliysXbvW7DgaDCpvgsKGJErq1Knp0KEDN2/eZOTIkaRJk4a0adPSuHFjtm/fzjfffMPt27dN2x87dozs2bPj6OhoxaoF/r/5/fDhw8yfP5/Jkydz9uxZ2rVrx88//0xAQACA6b06deoUuXPn1ntnZR999BG7du0yhYRvv/2Wli1bEhQURNasWU3bzZgxg/z58xMQEMAPP/yAp6cn48aNY9q0adYqXf7D1HYmCRYzKVCOHDkICQkhe/bs/PbbbzRo0AAfHx+GDx/O6NGjOXnyJBkyZMDBwYE//viDZcuWqeskGYiZPn7QoEG8/fbbBAUFUaxYMcqXL0/z5s2ZPXs2p0+fJl++fNy4cYPNmzezfPlyhQ0runPnDlWqVKF8+fKmZQ0aNGDFihUEBgZy+vRpcuXKhZ2dHalSpWLmzJn07NmTMWPG4ObmRvXq1QFijdkQSWoaICqJdu3aNe7evcutW7eYMWMGadKkYfbs2bi5ubFjxw5+/vln7t27R548eWjZsmWsa/rlzXj69KlZ18eVK1do164dbdu2pVWrVoSEhODo6IijoyNHjx413SsjVapUZM+enR49euDh4WHFM5DnzZo1i6xZs9K8eXMePnxIs2bNcHR0ZNSoUZQpU8bU8hEZGcnUqVPp27evrjYRq1HYkHiLaX6/ceMGkZGRPHz4kGLFigGwZcsWZs2aRdq0aZk5cyaZMmXiwYMHZMiQQZfWWdHYsWNNYS/mL9qDBw8yZMgQFi1ahLu7u2nbW7duMWTIELp06ULp0qWxs7MjPDxcLRpW9vzPT3h4OAMHDmTTpk1MmjSJhg0bEhwcTNOmTXF2dmb48OFmgSOuY4i8SWpHk3h5fsKuzp0706FDBz7++GM+/fRTtm/fTu3atenUqRNPnz7ls88+w8/Pj4ULF5oNZpM3L02aNKbgEHOFib29PQ8ePODChQtm2xqNRk6cOMHZs2dN71mqVKneeM3y/56/8+revXsJCwtj1KhRfPjhhwwYMID169fj6urK2rVrefz4MWPGjGHfvn0vHAAs8qbp01/ixWAwsGfPHvr370+LFi1YvXo1X331FXv37uXRo0cA1KtXj44dO5ItWzb27NlDnTp1MBgMGvVuBTG/bPr06UPRokXZvn07S5Ys4dGjR+TOnZusWbOydu1azp8/b9onW7ZsFC9e3DQxFOiKBWt6/vLWAwcOMHnyZObOnYujoyP9+/fngw8+YODAgabAsWbNGs6fP8+6dev0vkmyoQGiEm87duygUaNGtGzZksuXLzNhwgQ+/vhjihQpQkBAAB07dqRu3brUrl2bp0+fajCoFf37l83evXv55ptvsLOzo23btgwePJhu3boB8N5771GwYEHWrFnD33//zahRo6xRsjzn+RbBr7/+moMHD3Lu3DnOnz+PnZ0dnTp1YtCgQRgMBgYOHAhAo0aN2L17N2nSpLFm6SJmFDYkXqKiojh58iRvv/02UVFRtG7dmurVqzNixAhOnjzJV199RZkyZShTpgwODg4KGsnM4MGDMRgMjB8/nqioKNq3b8/cuXPx9/fHz8+PjBkzArBw4UKzcRxiHTFhce7cuSxcuJDhw4fTunVrNm7cyN69e4mKiqJbt24MGDDAFDgyZsxIlSpVAI3RkORDYUNe6t+DQR0cHPDx8WH16tUsXbqUOnXqMHToUIxGI0+fPiVnzpxkypTJ2mUL///enThxggsXLnD58mW6devGoEGDAJg0aRIA7du35+uvv+b27ds8ffqUHDly4ObmZs3S//Ni3juj0cijR4/4448/6N+/Pw0bNgSe3RZ+zpw5rFu3DgcHB7p06ULv3r1xd3enUqVKpuMoaEhyobAhL/T8YNCvvvqKu3fv0qhRI8qUKUNUVBQZMmTgww8/NE11vHPnTtKkSaPZQZOBmPdu06ZNjBgxwhQCr169iru7u1ngsLe3p3HjxhQsWNDKVQvEvqlaqlSpCA4OJjIyEng2hXyqVKno0qULBw8eZNmyZRgMBjp27EjHjh0B3etEkh99N8oLGQwGtm/fTv/+/Rk0aBB58uShUKFCZM2alePHj3P06FG6du1KgQIFiI6O5sSJEyxevFh/FScDMTODjhgxAj8/P+rWrUtYWBhRUVFs2bIFb29vBg0aRKpUqRg/fjypUqXi448/1hVDVvZ80Bg3bhx3795l/PjxuLq6snPnTlq1akWqVKlMYcLDw4OrV6+yd+9esmXLRsuWLXWvE0mW9MkiLxQREcHGjRvp1KkTH3/8MaVKleKPP/6gffv2rF+/nqdPn9K+fXvSpElDqVKlWLlyJUWKFLF22f9pz1/qePnyZSpUqECjRo0ICwtj4cKFtG7dmp49e9KsWTMuXLhA37596dy5MxUrVlTQsLLnB4MeO3aMPXv20LBhQxwcHOjXrx/79u1j2LBhwLMwGR0dzYULF2jbti25c+dm7dq1REVF6QoUSZb06SIvZDQauXjxIhcvXuTw4cO0adOG77//HgcHBz766COOHTtG+fLlmTFjBj179qRAgQLWLvk/LabrZP/+/fz555/cvXuX/fv3M3XqVOrVq8f+/fupXr06u3fvJjIykh9//BGA3r17qwvFij766CMOHDhgCgmLFy+mQ4cOZMiQwTT+omjRogwbNoxffvmFpk2b0qVLFz766CPOnTvHxx9/TMOGDYmMjDTdJE8kuVHYkBdydHSkVatW/Prrr7Rv356sWbPSvXt35s2bR82aNSlSpIgme0omYoLGgQMH+PTTT7l37x7t27enbNmyHDhwgLp16/Lll1/Sp08fMmXKRJEiRcidO7e1y/7Pu3XrFpUqVaJUqVKmZXXr1sXFxYWDBw+yc+dOjEYj6dKlo0mTJvzwww94e3uTJ08eKleuzKZNmwDYtm0b2bJl08+jJFuarlxe6dKlS0RERJjd0+Srr77it99+Y/Hixbr6JJk4ceIE69evx9nZmZ49e5qWR0dHYzQaCQsLIywsjGXLlrFs2TK+++478uTJY8WK5XmzZ88mZ86cNG7cmLt379KsWTPSp0/PyJEj8fb2jtU9cuLECe7cucOuXbtYt24dy5Ytw9PT00rVi7ycRhHJK+XNm5eQkBAmTZpEUFAQkZGR7Nu3j0WLFiloJBNGo5FJkyaxf/9+ypQpQ1hYGE5OThiNRqKioliwYAGrVq3C1dWVBw8esHDhQgWNZGTbtm0EBwczffp0UqdOTd26dVm1ahXNmjXjyy+/ZPjw4abWj+joaMLCwpg7dy7//PMPWbNmZenSpQoakqypZUNeS1hYGD/99BPr16/H09OTjz76SP38yczTp08ZNGgQBw4coHv37nzwwQc4OTkBEBoayooVK8idOzfFihUjV65cVq5WYty5c4fKlSvTpk0b7OzsWLRoEdOmTaNOnTrcunWL5s2bkzVrVgYOHEi5cuVM+0VERPD06VPs7e1JmzatFc9A5NUUNiReYu7ToBHv1hUzRuPOnTvY2dlx9+5dChUqRHh4OD179uTy5cu0b9+ehg0b6m6tyVjM+zhp0iTOnj1L586dWbVqFevWrTMLHNWqVaNJkyaMHTsWML9EViQlUDeKxIs+4Kwv5hfUtm3bCAgI4N69e9y9e5dq1arRq1cvpk+fTpcuXfj666+xs7PjvffeU+BIJmLeuxgx/y5btiw//PAD0dHRfP7550RHR9O7d2+mTZvGu+++y549e8wmy9PPoaQ0+o4VSWFiJlvr3bs39erVY/78+QwaNIhffvmFwMBAHB0dmT17Nrly5cLf35/Nmzdbu2T5n5hwsW7dOrP3pUaNGpQrVw5/f38cHR3p27cvTZo0oWfPnuzfv5+MGTNib29PVFSUtUoXSRSFDZEUxmg08ttvv+Hr68snn3yCnZ0dAQEBtG7dmvz587NixQqcnJyYNm0apUuXNrusUqzv3r17rFq1il69etG3b1++++47AD7++GOio6M5duwY2bNnp3379vTu3ZsyZcqY9tW9TiSlUtgQSUFOnjzJ1atX+euvv3B3d+fJkye0atWKihUr8vnnnxMUFMSkSZM4deoUadKkYfr06ZpPI5lxc3Nj0aJFLF26lIcPH7Jw4UKaN2/Ow4cPuXXrFuvXrwegYMGCdO7cGQcHB9N9UURSKg0QFUkBjEYjoaGhVK5cmQ4dOmBnZ8fBgwc5ffo0NWvWZNiwYTg4OPDbb78xduxYVq1aRebMma1dtrxCSEgId+7cwd/fn4iICNPcGStWrMDb29va5YlYjAaIiqQABoOBdOnS0aFDB3bt2kW5cuU4evQoefPmpVOnTqYbbx07dozMmTNrJskUwtnZGWdnZ2bMmMFff/3F4cOH2blzJyVLlrR2aSIWpZYNkWTq31cuAOzdu5dhw4YxZswYzpw5Q0BAADlz5iRfvnyEhISwf/9+li5dSuHCha1UtcRXXO8zQFRUlMZoiM1Q2BBJxo4dO0ZwcDBVq1Y1LRswYADnzp1j7dq1bNiwgf3793P+/HnTZGvPTysvKc+LwodISqawIZJMPX78mC5dunDgwAGaNm1K5cqVqV+/PqdPn2bYsGF069aN6tWrA8/+Crazs9MvKRFJlnQ1ikgylS5dOmbNmsX8+fM5f/4806ZNo3Xr1ty/f5/w8HA2btxo2tbe3l5BQ0SSLbVsiKQAwcHBXL58mWnTpmE0GgkKCuL8+fMEBARQpUoVa5cnIvJSChsiKczOnTs5fvw4K1euZMWKFbqpmogkewobIinEvwcOhoaG6m6fIpIiKGyIpFC6akFEUgoNEBVJoRQ0RCSlUNgQERGRJKWwISIiIklKYUNERESSlMKGiIiIJCmFDREREUlSChsiIiKSpBQ2RCRBNEWPiLwuhQ0RK/D19cXT09PsUbx4capXr86oUaN48OBBkj332rVr8fT05OrVqwDMmDEDT0/P194/KCiITp06ce3atUTXcvXqVTw9PVm7du0Ltxk8eDA+Pj7xOm5C9onL69QnIq/mYO0CRP6rihYtyogRI0xfR0REcPLkSaZMmcKpU6dYsWLFG5m4q3nz5vG6mduePXv4448/GDZsWBJWJSK2RGFDxEqcnZ0pVaqU2bJy5crx+PFjpk+fztGjR2OtTwrZs2cne/bsSf48IvLfpW4UkWSmePHiAFy/fh141uXSv39/evbsSenSpenYsSMAYWFhTJw4kWrVqlG8eHEaNmzIhg0bzI4VHR3N7NmzqV69Ol5eXnTt2jVWF01c3Si//PILTZs2xcvLi+rVqzNp0iTCw8NZu3YtQ4YMAaBmzZoMHjzYtM/333/Pe++9Z+oOmjFjBpGRkWbH3bJlC40aNaJkyZI0adKE06dPx/v1efr0Kf7+/rz77rsUL16c0qVL8+mnn3Lq1KlY23733XdUr16dkiVL0qZNG/7++2+z9devX6dv376UL18eLy+vOLcRkcRT2BBJZi5cuABA7ty5Tcs2btxIqlSpmDVrFp988glGo5Fu3bqxcuVKPv30U+bMmYO3tzd9+vRh3bp1pv0mTZrErFmzaNasGTNnziRjxoz4+/u/9PlXrlxJ3759KVKkCDNnzqRTp04sX76ckSNHUr16dbp06QLAzJkz6dq1KwDz5s1j2LBhVKpUiblz59KqVSsCAgIYPny46bjbtm2jZ8+eFCpUiJkzZ1KvXj0GDBgQ79dn4MCBrF69mo4dO/L1118zePBgAgMD6dOnj9mg1aCgIGbMmEHv3r2ZMmUKDx484JNPPuHevXsA3Lt3j48++oiTJ08ybNgw/P39iY6OplWrVpw7dy7edYnIi6kbRcRKjEaj2V/+Dx484MCBA8yZM4dSpUqZWjgA7Ozs+OKLL0y3lN+9ezc7d+5k6tSp1K9fH4AqVarw5MkTJk+eTIMGDQgNDWXJkiV88skn9OjRw7TNzZs32blzZ5w1RUdHM2PGDGrXrs2YMWNMy8PCwvjhhx9wdnYmT548ABQpUgR3d3cePXrEnDlzaNGiBZ9//jkAlStXxtXVlc8//5xPP/2UQoUKMWvWLIoVK2YKO1WrVgV4Zfh5Xnh4OI8fP2bYsGGm8y5fvjyPHz9m/Pjx3L59m6xZswIQFRXFzJkzTV1RXl5e1KpVi2+++Ya+ffuyePFigoODWbFiBbly5TLVVL9+faZNm8b06dNfuy4ReTm1bIhYyZ9//kmxYsVMj7fffpu+fftSrFgxpkyZYjY41N3d3RQ0APbu3YvBYKBatWpERkaaHj4+Pty+fZuzZ8/y119/ERERQc2aNc2et169ei+s6cKFC9y5c4datWqZLW/bti0//vgjjo6OsfY5cuQIT548wcfHJ1Yt8CwYPX36lJMnT8arlrg4OjqycOFC6tevz61bt/jzzz/57rvv+P3334Fng2xj5MyZ02zMS5YsWShVqhR79uwBnr2GRYoUIVu2bKaa7ezsqFq1qmkbEbEMtWyIWEmxYsUYNWoU8Ox28U5OTuTIkQNnZ+dY22bOnNns6+DgYIxGI6VLl47z2Ldu3eLhw4cAuLm5ma3LkiXLC2sKDg4GIFOmTK99HjH7xIwliauWBw8eYDQaY9US0woRHzt37mTs2LGcP3+edOnS4enpSbp06QDzuT/+/ZrBs/O6ceOGqe5Lly5RrFixOJ/nyZMn8a5NROKmsCFiJenSpaNEiRIJ2tfFxYW0adPy7bffxrk+b968HDt2DIC7d+9SoEAB07qYcBCX9OnTA5jGNTy/z8mTJ+O8OiZmn8mTJ5MvX75Y6zNnzoyrqyt2dnbcuXMn1nHj4/Lly3Tr1o2aNWsyb948U5fOsmXLYnUNxYSt592+fdsUeFxcXChfvjwDBw6M87niasURkYRRN4pIClS+fHlCQ0MxGo2UKFHC9Dh79iyzZs0iMjISb29vUqdOzaZNm8z2jelyiEuBAgXImDEjW7duNVv+008/0aFDB8LCwrCzM//Y8PLyIlWqVNy8edOsllSpUuHv78/Vq1dxcnLC29ubLVu2mLU+bNu2LV7nfeLECcLCwujUqZMpaACmoPH8sS9dusSlS5dMX9+4cYMjR45QoUIF4NlreOHCBfLnz29W9/r16/n++++xt7ePV20i8mJq2RBJgapVq0a5cuXo2rUrXbt2pWDBghw7dowZM2ZQuXJl01/vXbt25auvviJNmjRUrFiR7du3vzRs2Nvb06NHD0aPHs3IkSOpXbs2Fy9e5KuvvuLjjz/Gzc3N1JLx66+/UrVqVQoWLMhnn33GtGnTCAkJoUKFCty8eZNp06ZhMBgoXLgwAH379qVNmzZ0796dFi1acPHiRebMmROv8y5WrBgODg5MmjSJdu3amS7H/eOPPwAIDQ01bevk5ETXrl3p06cPUVFRTJs2DVdXV9q0aQP8/ziUtm3b0q5dOzJmzMiGDRtYtWqV6fJeEbEMhQ2RFMjOzo758+czbdo05s2bx927d8mWLRtt27alW7dupu06depE2rRpWbx4MYsXL8bb25tBgwYxcuTIFx67VatWpE2bloULF7J69WqyZctGu3btTGMyKlSowNtvv42/vz979+5l/vz59O7dmyxZsrB8+XIWLFhAhgwZqFSpEn379sXFxQWAsmXLEhAQwJQpU+jevTvu7u6MHTuWzp07v/Z5582bF39/f2bOnEmXLl3IkCEDpUqVYsmSJfj6+nLw4EHTnCGenp689957jBw5kkePHlGpUiWGDh1qCmLZsmVj5cqV+Pv7M3LkSMLCwsiXLx9jxozhgw8+iO9bIiIvYTDqbkoiIiKShDRmQ0RERJKUwoaIiIgkKYUNERERSVIKGyIiIpKkFDZEREQkSSlsiIiISJJS2BAREZEkpbAhIiIiSUphQ0RERJKUwoaIiIgkKYUNERERSVIKGyIiIpKk/g8iL9XgJDQFVgAAAABJRU5ErkJggg==", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plot_confusion_matrix(test['target'], y_pred, classes=iris.target_names, normalize=False,\n", " title='Normalized confusion matrix')\n" ] } ], "metadata": { "colab": { "provenance": [] }, "kernelspec": { "display_name": "Python 3", "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": "3.11.5" } }, "nbformat": 4, "nbformat_minor": 0 }