diff --git a/Workshop3/checkpoint-3.ipynb b/Workshop3/checkpoint-3.ipynb
deleted file mode 100644
index dc559baa8d71a7737f8e1b27665a776e41dae948..0000000000000000000000000000000000000000
--- a/Workshop3/checkpoint-3.ipynb
+++ /dev/null
@@ -1,350 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "code",
-   "execution_count": 42,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "import pandas as pd\n",
-    "import numpy as np\n",
-    "import matplotlib.pyplot as plt\n",
-    "from scipy import stats"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "df = pd.read_csv('HospitalAdmissionsData.csv')"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "Index(['ID', 'AdmissionLengthDays', 'Death_1', 'Admission_Type',\n",
-       "       'Insurance_Type', 'EnglishLanguage_1', 'Religion_Type', 'Married_1',\n",
-       "       'Race', 'Dx'],\n",
-       "      dtype='object')"
-      ]
-     },
-     "execution_count": 3,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "df.columns"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 18,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "object     5\n",
-       "int64      4\n",
-       "float64    1\n",
-       "dtype: int64"
-      ]
-     },
-     "execution_count": 18,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "df.dtypes.value_counts()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 15,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "array(['private', 'medicare', 'government', 'medicaid', 'self pay'],\n",
-       "      dtype=object)"
-      ]
-     },
-     "execution_count": 15,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "df[\"Insurance_Type\"].unique()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 19,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>ID</th>\n",
-       "      <th>AdmissionLengthDays</th>\n",
-       "      <th>Death_1</th>\n",
-       "      <th>EnglishLanguage_1</th>\n",
-       "      <th>Married_1</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>count</th>\n",
-       "      <td>58863.000000</td>\n",
-       "      <td>58863.000000</td>\n",
-       "      <td>58863.000000</td>\n",
-       "      <td>58863.000000</td>\n",
-       "      <td>58863.000000</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>mean</th>\n",
-       "      <td>29508.211984</td>\n",
-       "      <td>10.138978</td>\n",
-       "      <td>0.099417</td>\n",
-       "      <td>0.571072</td>\n",
-       "      <td>0.410665</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>std</th>\n",
-       "      <td>17026.189024</td>\n",
-       "      <td>12.465611</td>\n",
-       "      <td>0.299224</td>\n",
-       "      <td>0.494927</td>\n",
-       "      <td>0.491959</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>min</th>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>-0.945139</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>0.000000</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>25%</th>\n",
-       "      <td>14762.500000</td>\n",
-       "      <td>3.743056</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>0.000000</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>50%</th>\n",
-       "      <td>29523.000000</td>\n",
-       "      <td>6.465972</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>0.000000</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>75%</th>\n",
-       "      <td>44254.500000</td>\n",
-       "      <td>11.798264</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>1.000000</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>max</th>\n",
-       "      <td>58976.000000</td>\n",
-       "      <td>294.660417</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>1.000000</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "                 ID  AdmissionLengthDays       Death_1  EnglishLanguage_1  \\\n",
-       "count  58863.000000         58863.000000  58863.000000       58863.000000   \n",
-       "mean   29508.211984            10.138978      0.099417           0.571072   \n",
-       "std    17026.189024            12.465611      0.299224           0.494927   \n",
-       "min        1.000000            -0.945139      0.000000           0.000000   \n",
-       "25%    14762.500000             3.743056      0.000000           0.000000   \n",
-       "50%    29523.000000             6.465972      0.000000           1.000000   \n",
-       "75%    44254.500000            11.798264      0.000000           1.000000   \n",
-       "max    58976.000000           294.660417      1.000000           1.000000   \n",
-       "\n",
-       "          Married_1  \n",
-       "count  58863.000000  \n",
-       "mean       0.410665  \n",
-       "std        0.491959  \n",
-       "min        0.000000  \n",
-       "25%        0.000000  \n",
-       "50%        0.000000  \n",
-       "75%        1.000000  \n",
-       "max        1.000000  "
-      ]
-     },
-     "execution_count": 19,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "df.describe()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 25,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "0    emergency\n",
-      "dtype: object\n",
-      "0    medicare\n",
-      "dtype: object\n",
-      "0    catholic\n",
-      "dtype: object\n",
-      "0    white\n",
-      "dtype: object\n",
-      "0    newborn\n",
-      "dtype: object\n"
-     ]
-    }
-   ],
-   "source": [
-    "print(df[\"Admission_Type\"].mode())\n",
-    "print(df[\"Insurance_Type\"].mode())\n",
-    "print(df[\"Religion_Type\"].mode())\n",
-    "print(df[\"Race\"].mode())\n",
-    "print(df[\"Dx\"].mode())"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 40,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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\n",
-      "text/plain": [
-       "<Figure size 432x288 with 1 Axes>"
-      ]
-     },
-     "metadata": {
-      "needs_background": "light"
-     },
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "fig, ax = plt.subplots()\n",
-    "df[\"AdmissionLengthDays\"].hist(ax=ax, bins=100, bottom=0.1)\n",
-    "ax.set_yscale('log')"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 74,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAggAAAILCAYAAACNaGTNAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdd3SVVd7F8e+56T0kgUASWigBQkhCF8GuoCIqSLOMin2sWFBn1FHHLgKKvWOhCSgq6GDDAtITQuihhk4S0nvyvH8E53UEFSTJc8v+rMVSzI1341q57nvPOb9jLMtCRERE5NccdgcQERER56OCICIiIkdQQRAREZEjqCCIiIjIEVQQRERE5AgqCCIiInIEFQQRN2SMucwYs6ARnuc0Y8yuhn4eEWl8KggiLsoY098Ys9gYU2CMyTPGLDLG9AKwLOtDy7LOcYKM/zbGrDHGVBtjHrY7j4gcOxUEERdkjAkFPgcmAxFALPAIUGFnrqPIAsYB8+wOIiLHRwVBxDV1BLAsa5plWTWWZZVZlrXAsqwMAGPMVcaYn355sDHmHGPMxsOfNrxsjPneGHPtrx9rjBlvjDlkjNlmjDn3V997tTFmvTGmyBiz1Rhzw7GGtCxrimVZXwBF9fYnF5FGoYIg4po2ATXGmCnGmHONMU1+74HGmChgFnA/EAlsBPr95mF9Dv/zKOAZ4C1jjDn8tQPAYCAUuBqYaIzpXp9/GBFxPioIIi7IsqxCoD9gAW8AB40xnxpjoo/y8POAtZZlzbEsqxp4Adj3m8fssCzrDcuyaoApQAsg+vBzzbMsa4tV53tgATCgYf5kIuIsVBBEXJRlWesty7rKsqw4oCsQA0w6ykNjgOxffZ8F/Pbkwb5ffb308N8GAxz+hGLJ4Y2Q+dQVjqj6+5OIiDNSQRBxA5ZlbQDepa4o/NZeIO6X3xxeOog7yuOOYIzxA2YD44Foy7LCgfmA+cNvFBGXp4Ig4oKMMZ2MMXcZY+IO/74lMBpYcpSHzwOSjDEXGWO8gZuB5sf4VL6AH3AQqD68efGYj08aY3yMMf7UvdZ4G2P8jTFex/r9ImIfFQQR11RE3cbCpcaYEuqKQSZw128faFlWDjCcus2HuUAXYAXHcCTSsqwi4DZgJnAIuBT49DhyvgGUUVde/nn47684ju8XEZuYuuVIEfEUxhgHdXsQLrMs6zu784iIc9InCCIewBgz0BgTfnhPwT+o20NwtOUIERFABUHEU5wEbAFygAuAiyzLKrM3kog4My0xiIiIyBH0CYKIiIgcQQVBREREjuB9PA+Oioqy2rRp00BRREREpDGtXLkyx7Kspkf72nEVhDZt2rBixYr6SSUiIiK2Msbs+L2vaYlBREREjqCCICIiIkdQQRAREZEjHNcehIZQXlXDrkNlhAf6EB7gg7eXOouINKySimrySirx83YQ4OtFoK83Xg5dUCnya7YXhHV7Cxn68uL//j7E35vwQB9ahAWQ0jKc1JbhpLQKp0VYgI0pRcQV7TpUyrJteSzfnsf2nFL2F5VzoLCC4orqIx7r5+2gaYgfXVqE0iUm9L9/jQ0PoO6GbBHPclyTFHv27GnV9ymG3OIKfsrKIb+0ikOlleSXVpFfWsn23FLW7SmksqYWgOah/pzeqRkXpcTQq00EDrV9EfmNqppavt1wgP9k7mPptjx259dNkw7196ZDdAjRoX40C/EnOtSfiCAfKmssyiqrKa2soayyht35ZazbW8i2nBJ+eWmMbxrEeV1bcF5SCzq3CFFZELdijFlpWVbPo37N7oLwRyqqa1i/t4j0nYdYvuMQ3204QGllDbHhAQxJiWFoaiwdokMaLY+IOKesA8XMXJHNnFW7yCmuJCLIlz5tI+p+xUeSEB1yXG8qSiur2bCviIzsfL5av5+ft+RSa0GbyEAGd4vhbye1plmofwP+iUQah8sWhN8qrazmq3X7+ThtNz9uzqGm1uKszs24/cyOJMWF2ZZLRBqfZVn8uDmHyd9uZvn2Q3g7DGd2bsbIXi05pUPTet3PlFtcwYJ1+5m/Zi+LsnLwdjgY1iOOG06Jp01UUL09j0hjc5uC8GsHiyqYtmwnb/20jYKyKs7o1IzbzuxASstwu6OJSANbsT2PZ/+zkaXb8ogND+DKfq25ODWOpiF+Df7cO3JLeP2HrXy0chfVNbWcm9SC287oQEJzfZoprsctC8IvisqrmLJ4O2/+tI380irO6RLNw0MSiQnXpkYRd7NhXyHPfLmRbzccICrYj9vObM/IXi3x8/Zq9CwHisp5Z9F2Pvh5B6VVNVx5UhvGnt2BEH+fRs8i8le5dUH4RXFFNe/8tI2XFmbhZQx3nZPAlf3a6OiSiBuoqK7hxW+zeGXhFoL8vLnx1HZc2a81gb62H8TiUEklzy7YyLRlO4kK9uOf53XmwpQYbWYUl+ARBeEX2XmlPPBJJt9vOki3uDCeuDiJrrHanyDiqtJ2HmLcrAw2HyhmaPdYHhrchfBAX7tjHWF1dj4Pzs0kY1cBfdpG8OwlybSKDLQ7lsgf8qiCAHWblz7P2Msjn63jUGkld5zZgZtPb6+jkSIupLyqhglfbeLNH7cSHerPE0OTOD2hmd2x/lBNrcWM5dk8+cV6sODxoUkMSY6xO5bI7/K4gvCLgtIqHvo0k7npexjQIYpJI1OIDG74TUwicmKy80q56cOVZO4u5NI+rbj/3E4utbafnVfK7dPTWLUznxE943h4SKJTLIeI/JbHFgSo+zRh2rJsHv5sLRGBvky+NJVebSLsjiUiv+O7jQe4Y3o6tZbFxBEpnNUl2u5If0lVTS2Tvt7Eywu3EB8VxOTR3ekSE2p3LJH/8UcFwe0vPjDGcGmfVnz89374+zgY9foSXvt+C8dTjESk4dXWWjz/9WbGvLucmPAAPr+1v8uWAwAfLwf3DOzEh9f0oai8mqGvLOLLzL12xxI5Zm5fEH6RGBPGZ7f2Z2BiNE9+sYH7Zq+h6vAYZxGxV0lFNde9t4KJX2/i4pRY5tzUj9aR7jGAqF/7KObdNoDOLUK58YNVvKo3KOIiPKYgAIT4+/DSpd259Yz2zFiRzTVTVlBUXmV3LBGPllNcweg3lrBw00EevTCR50YkE+Db+HMNGlLTED+mXdeXwd1a8NThNyiV1XqDIs7NowoC1C053HVOAk8NTWJRVg4jXlvCvoJyu2OJeKQduSUMe2Uxm/YX8foVPfjbSW3cdn6Av48XL4xK/e8blCvfXkZBqd6giPPyuILwi1G9W/HWlT3ZmVvCxS8vYuO+IrsjiXiUNbsKGPbKYgrLqph6XV/O7Oy6+w2OlcNR9wZlwohkVu44xKg3lpBbXGF3LJGj8tiCAHBaQjNm3ngSNbUWo17/mXV7Cu2OJOIRFmXlMOr1n/Hz9mLWTf3o3qqJ3ZEa1dDucbx1VU+25RQz8vUlHCjUp5jifDy6IEDd5sWZN5yEv48Xl765hMzdBXZHEnFrP23OYcy7y2kZEcicv/ejXdNguyPZYkCHprx7dW/25pcx4rWf2Z1fZnckkf/h8QUBoE1UEDOuP4kgX28ufWMJGbvy7Y4k4pYWZeVwzZTltI0KYup1fYkO9bc7kq36xkfy/rV9yC2pZMSrP7Mjt8TuSCL/pYJwWKvIQKZf35ewQB8ue3Mp6dkqCSL1afFvykFEkPPdp2CH7q2aMO26vpRWVjPiNZUEcR4qCL/SMiKQ6defRESQL1e8uVSfJIjUk8VZOYyZspzWEUF8eG0flYPf6BobxvTrT6KyupbL3lyqk1XiFFQQfiM2POC/nyRc9c5ythwstjuSiEtbuSOPMVOW0yoikA+v66P7UH5HQvMQpozpTX5pFZe/tZS8kkq7I4mHU0E4ihZhAbx/TR8cBq54cyl7tHlI5C/ZtL+IMe+uoEVYAFOv60uUysEf6hYXzptX9iQ7r5Qr315GoQa5iY1UEH5H26gg3r26N0Xl1VyhNi9y3Hbnl/G3t5bh5+3gvTG9VQ6OUd/4SF65vDvr9xZy7bsrKKussTuSeCgVhD/QNTaMN6/sya5DZVz9zjKKK6rtjiTiEg6VVPK3t5ZSUlnNlDG9aRkRaHckl3JGp2gmjExh+Y48bpm6imrdGyM2UEH4E33iI3nx0u5k7inkpg9W6oInkT9RWlnN1e8uJ/tQGW/+rSedW+iK479iSHIMjw5J5JsNB3j083W64EkanQrCMTi7SzRPXpzEj5tzeGhupn5QRX5HTa3FrVPTyNiVz+TRqfSJj7Q7kku74qQ2XDegLe/9vIO3ftpmdxzxMN52B3AVI3q1ZHtuCS8v3EJ8VDDXnRJvdyQRp/P4vPV8s+EA/76oKwMTm9sdxy3cf25nsvPKeHz+euKaBDKoq/67SuPQJwjH4e5zEjg/qQVPfLGe/6zdZ3ccEacydelO3l60jatPbsMVfVvbHcdtOByGiSNTSI4L544ZaaTtPGR3JPEQKgjHweEwPDcimeS4cG6fnsaaXbq3QQTqRig/NDeT0xKa8sD5XeyO43YCfL1488qeNA3x49opK8jOK7U7kngAFYTj5O/jxRt/60lkkB/XTFmuGQni8bYcLOamD1YS3zSIyaNT8XIYuyO5pahgP965qjeVNbVc994KSit1qkoalgrCX9A0xI93ru5FaWUNN36wkvIqnVMWz5RfWsm1U1bg7eXgrSt7EeLvY3ckt9a+WTCTR6eycX8R93yUoQ3T0qBUEP6ijtEhTByZQsauAv75sU42iOepqbW4dVoauw+V8foVPTTroJGcltCMewd1Yt6avby8cIvdccSNqSCcgLO7RHP7mR2YvWoXUxZvtzuOSKN6bsFGftycwyMXJtKzTYTdcTzKDafEMyQ5hvELNvLthv12xxE3pYJwgm4/swNndY7m3/PWs2Rrrt1xRBrFl5l1715H927J6N6t7I7jcYwxPD2sG11ahHL7tHSyDuhSOal/KggnqO4IUjKtIwO5+cNV2rQobi/rQBF3zVxNcstwHh6SaHccjxXg68Xrf+uJr7eD699fQZEudpJ6poJQD0L8fXj9ip5UVNdq06K4taLyKq5/fyX+Pl68enl3/Ly97I7k0WLDA3jpsu5szynhvtlrtBdK6pUKQj1p3yyYCSOSydhVwGPz1tkdR6TeWZbF3R+tZkduKS9d1p0WYQF2RxLqbn+8Z2DdpsV3tRdK6pEKQj06J7E5N5wSzwdLdjI3fbfdcUTq1Zs/buM/a/dz/7md6Ks7FpzKDafEc1bnaB6ft56VOzRpUeqHCkI9u3tgAr3aNOH+OWvIOlBkdxyRerFyRx5Pf7mBQYnNuaZ/W7vjyG84HIbnhifTItyfW6auIre4wu5I4gZUEOqZj5eDyaO7E+DjxU0frNK0M3F5eSWV3DI1jZjwAJ4Z3g1jNCnRGYUF+vDKZT3ILankjhnp1NRqP4KcGBWEBtA8zJ/nR6WSdbCYBzRESVxYba3FnTPTyS2u5OXLuhOqSYlOrWtsGI8OSeTHzTlM/naz3XHExakgNJD+HaK4/cwOzEnbzfTl2XbHEflLXv1hCws3HuTBC7rQNTbM7jhyDEb2asnQ1Fhe+GazZrPICVFBaEC3ntGB/u2jePjTtWzar/0I4lqWbcvjuQWbGNytBZf30TAkV2GM4d8XdaVNZBC3T0/TfgT5y1QQGpCXwzBhZDIh/t7cMnWV5iOIyzhUUslt09JoFRHIk0OTtO/AxQT5eTP50lQOlVZx90erqdV+BPkLVBAaWLMQfyaMSGHT/mIe/VzzEcT5WZbFuNkZ5JVUMnl0qm5odFGJMWE8cH5nvtt4kLd+2mZ3HHFBKgiN4JSOTbnh1HimLt3J/DV77Y4j8ofeX7KDr9bt595zO2nfgYu7om9rBiZG8/SXG0jPzrc7jrgYFYRGcvc5CSS3DOfe2Rlk55XaHUfkqNbvLeSxees5o1Mzxpzcxu44coKMMTwzLJnoUH9unbaKQt3XIMdBBaGR+Hg5mDwqFSy4fXoaVTW1dkcS+R9llTXcOi2NsAAfnr1E8w7cRVigDy+MTmVPfjkPfpJpdxxxISoIjahVZCBPDE1i1c58XvhGZ5TFuTz6+Vq2HCxm0sgUIoP97I4j9ahH6ybcfmYH5qbv4eO0XXbHERehgtDILkiO4ZIecbz0XRbLtuXZHUcEgC/W7GXasmxuOrUdJ7ePsjuONICbT29P7zYRPPjJWnbmaplT/pwKgg0eHpJIy4hAxs5Ip6BMa4Jir70FZdw3Zw3JcWGMPbuj3XGkgXg5DBNHpWAM3KZlTjkGKgg2CPbz5vlRqewrLOefH+sOd7FPba3FnTNWU1VTy6RRqfh46SXBncWGB/Dk0CTSs7XMKX9OrwY2SWkZztizOvB5xl7mrNLV0GKPN37cys9bc3n4gkTaRgXZHUcaweBu/7/MuVSjmOUPqCDY6KbT6tYEH5qbyY7cErvjiIfJ3F3A+AUbObdrc4b3jLM7jjSih4ck0krLnPInVBBs9MuaoMNhuH16OtVaE5RGUlZZw23T04gM8tMoZQ8U7OfNxJEp7C+q4F9zdfRRjk4FwWax4QE8fnHdmuCL32XZHUc8xGPz1rEtp4QJI5IJD/S1O47YILVVE247owOfpO9hbrqWOeVIKghOYEhyDBelxDD52yzSdh6yO464uW/W7+fDpTu5bkA8/XSk0aPdfHo7urcK54FPMtmdX2Z3HHEyKghO4pELu9I81J+xM9Ipqai2O464qZziCu6dnUGn5iHcdY6ONHo6by8HE0emUFtrcdfMdGp066P8igqCkwgL8OG5EcnsyCvlsXm69VHqn2VZ3Dc7g8Lyap4flYqft5fdkcQJtI4M4l9DElmyNY83f9xqdxxxIioITqRvfCQ3nNKOacuy+WrdfrvjiJuZvjybr9cfYNzABBKah9gdR5zI8B5xDEpszvgFG1m7p8DuOOIkVBCczJ1nd6RLi1DunZ3BgaJyu+OIm9iWU8Kjn63j5PaRjDm5rd1xxMkYY3hyaBJNAn0ZOyOd8qoauyOJE1BBcDK+3g6eH5VCSUU1987K0JRFOWHVNbWMnZGOj5dh/PBkHA4daZQjNQny5dnhyWzaX8wzX260O444ARUEJ9QhOoT7zu3EdxsPMnXZTrvjiIt78bss0rPzeWJoEi3CAuyOI07s1I5NufKk1ry9aBs/bc6xO47YTAXBSV15UhsGdIjisc/Xs/Vgsd1xxEWl7TzE5G+zuDg1lsHdYuyOIy7gvnM7065pEHd/tJr80kq744iNVBCclMNhePaSZHy9HYyduVo3r8lxK62s5s6Zq2ke6s8jFybaHUdcRICvF5NGppJTXMGDc9faHUdspILgxJqH+fP4xV1ZnZ3PS5qyKMfpsXnr2Z5bwvjhyYT6+9gdR1xIUlwYd5zVgc9Wa8qiJ1NBcHKDu8VwcWqspizKcfl2w36mHp6WeFK7SLvjiAu68dR29GjdRFMWPZgKggt45MJEmof6c+fM1ZRWasqi/LHc4grGzVqjaYlyQry9HEwc8f9TFms1ZdHjqCC4gFD/uimL23NLeGzeervjiBOzLIv756yhsKyKSaNSNC1RTkiryED+dcHhKYs/acqip1FBcBF94yO5/pR4pi7dyTfrNWVRjm7mimwWrNvPuEEJdGoeancccQPDe8ZxTpdoxv9nE+v3FtodRxqRCoILufPsjnRuEcq4WRkcLKqwO444me05JTzy2Tr6tdO0RKk/xhieGtaNsEAf7piuKYueRAXBhfh5e/H8qBSKKqq5b7amLMr/q66p5Y4Z6Xg7DM+N0LREqV8RQb48c0k3Nu4vYvx/NGXRU6gguJiO0SHcf24nvtlwgGnLsu2OI05C0xKloZ2e0Iwr+rbmzZ+2sShLUxY9gQqCC/plyuK/P1+nKYvCqsPTEodqWqI0sH+c15n4pkHcNVNTFj2BCoILcjjqLt3x83Ewdka6pix6sJKKasbOSKd5qD8Pa1qiNLAAXy9eGJVKbkkF989Zo2VON6eC4KKiQ/158uIkVu8qYNLXm+yOIzZ59LN17MwrZcIITUuUxtE1Noy7zkngi8x9fLRyl91xpAGpILiwc5NaMLJnS15euIWlW3PtjiON7Is1e5mxIpsbT21Hn3hNS5TGc92AePrGR/DIp2vZnlNidxxpICoILu6hC7rQOiKQsTPSKSitsjuONJI9+WXcN2cNyXFh3Hm2piVK4/JyGCaMSMHLYbhDy5xuSwXBxQX5efP8qFQOFFXwj0+0JugJamot7pxZ96L8/KhUfLz0YyyNLyY8gCeGJpGenc/kbzbbHUcagF5Z3EByy3DGnt2ReRl7mb1KN6+5u9d+2MKSrXk8PCSRNlFBdscRDza4WwzDusfx4ndZLNuWZ3ccqWcqCG7ixlPb0bttBP+am6k1QTe2OjufCQs2cX5SC4b3iLM7jgiPXJhIy4hA7piepmVON6OC4Ca8HIaJI+vWBG+bnkZltdYE3U1JRTV3zEinWYgfT1ychDGalij2C/bz5oXDy5z3zdGEV3eiguBGYsMDeOaSbmTsKmD8Ao1DdTcPzV3LjtwSJoxMISxQRxrFeSS3DOeegXVHHzXh1X2oILiZQV1bcFmfVrz+w1YWbjxgdxypJ3NW7WL2ql3cekYH+upIozih6wbEM6BDFI9+vpbN+4vsjiP1QAXBDT04uAsJ0SHc/dFqDhSV2x1HTtDWg8U88EkmvdtGcOsZ7e2OI3JUjsMXhQX5enPrtDTd+ugGVBDckL+PF5MvTaW4opo7Z6ymtlZrgq6qorqGW6el4evt4PlRKXjrSKM4sWYh/owfkcyGfUU8MX+93XHkBOnVxk11jA7hXxck8lNWDq/9sNXuOPIXPfXFBtbuKWT8Jcm6pVFcwukJzbi2f1ve+3kH89fstTuOnAAVBDc2qldLzk9qwfgFG1mxXWeUXc1X6/bzzqLtXH1yG87qEm13HJFjNm5QJ1JahnPvrAx25OrYtatSQXBjxhieHJZEXJMAbpmaRm5xhd2R5BjtOlTKPbNW0zU2lPvO7WR3HJHj4uvt4MVLU3E4DDdPXaX9CC5KBcHNhfr78NKl3ckrreSOGenaj+ACKqpruHlqGjU1Fi+O7o6ft5fdkUSOW1yTQJ4bnkzm7kLtR3BRKggeoGtsGA9fkMiPm3N46bssu+PIn3hi3npWZ+fz7PBkjVIWl3ZWl2iuG1C3H2FehvYjuBoVBA8xundLLkqJYeLXm1i8JcfuOPI7Plu9hyk/7+Da/m0Z1LW53XFETti4QZ1IbRXOvbMzNAbexaggeAhjDI9fnETbqCBum5bOgULNR3A2WQeKuW92Bj1aN+Fe7TsQN+Hj5eDFS7vj7WW48YOVlFZW2x1JjpEKggcJ8vPmlct7UFJRzS1T03SHuxMprazm7x+uxN/Hi5cu7a4rnMWtxIYH8MKoVDbuL+L+ObqW3lXoVcjDdIwO4alhSSzbnsfj87RxyBlYlsX9c9aw+UAxz49KpXmYv92RROrdKR2bcvc5CcxN38M7i7bbHUeOgQqCB7owJZZr+7fl3cXbmbVyl91xPN6bP25jbvoe7jq7I/07RNkdR6TB3HRqO87uEs3j89ezdGuu3XHkT6ggeKj7zu1Ev3aR/OPjNazZVWB3HI/1w6aDPPnFes5Las7Np+ueBXFvv9zX0DoikJunprGvQHuhnJkKgofy9nIweXQqTYP9uOH9FRqiZIMduSXcOi2NDs1CePaSZIwxdkcSaXCh/j68dkUPSiuruenDlVRUa4iSs1JB8GCRwX68enkPckoqtWmxkZVUVHP9eysBeP1vPQjy87Y5kUjj6RAdwvjhyaTtzNemRSemguDhkuLCePLiJH7emsvDn67VD2ojqK21uPuj1Ww+UMTk0am0jtQwJPE85yW14I6zOjBn1W5dKOek9LZFGNYjjs0Hinn1+y3ENw3mmv5t7Y7k1iZ+vYkvMvfxj/M6cUrHpnbHEbHN7Wd2IOtAMU9/uYF2TYM5W5eSORV9giAAjBuYwMDEaB6bt45v1u+3O47b+mhFNpO/zWJkz5ZcNyDe7jgitjLGMH54MkmxYdw+PY31ewvtjiS/ooIgQN3u4okjU0iMCeXWaWms26Mf1Pq2OCuH++es4eT2kTx2cVdtShQB/H28eONvPQnx9+baKSs4WKQN085CBUH+K9DXm7eu7EWovw/XTlmuccz1KOtAETd+sJK2UUG8fFkPTUoU+ZXoUH/e/FsvcksquPa9FRrH7CT0KiX/IzrUnzev7Mmh0irGTFlOcYV+UE9UTnEFV7+7HF9vB29f1YuwAB+7I4k4naS4MF4YlcqaXfn8/cNVOlXlBFQQ5AhdY8N46bJU1u8t4ob3V+ic8gkoqajmmikrOFBYwZtX9qJlRKDdkUSc1jmJzXnsoiQWbjzIfbN1/NFuKghyVGd0iuaZYd1YlJXL2Bnp1NTqB/V4VVTXcP37K8jcXcDk0amktAy3O5KI07u0TyvuOKsDs1ft4tn/bLQ7jkfTMUf5XcN6xJFXUsnj89cTHpjJ4xdpY92xqq6p5fZp6SzKymX88GTOSWxudyQRl3H7mR3YX1jBywu30CzEj6tO1tFrO6ggyB+67pR4ckoqeO37rUQF+3Hn2R3tjuT0frmd8cu1+3hocBcu6RFndyQRl2KM4d8XJpJTXMEjn68jNMCHod31c9TYtMQgf+q+QZ0Y3iOOF77ZzOs/bLE7jlOzLIvH563no5W7uO3MDozR0CmRv+SX+2L6to3k7o9WMzd9t92RPI4KgvwpYwxPDk3i/KQWPDF/A699r5JwNJZlMeGrTbz50zau6teGsWd1sDuSiEvz9/Hirat60rNNBGNnpDMvY6/dkTyKlhjkmHh7OXh+VAoYePKLDVjAjae2szuW07Asi6e+2MBrP2xlVK+WPDS4i/ZriNSDQF9v3rmqF1e+vYzbpqfh5YBBXVvYHcsj6BMEOWbeXg6eH5nCBckxPPXFBl5ZqE8SoK4cPPr5Ol77YStX9G3NExcn4XCoHIjUlyA/b94d05vkuDBumZrGgrX77I7kEVQQ5Lh4ezmYOCKZIckxPP3lBl76LsvuSLaqrbV44JNM3lm0nWv6t+XRCxNVDkQaQPDhkpAYG8bfP1zFnFW77I7k9lQQ5E/F7+oAACAASURBVLh5ezmYMCKZi1JiePY/G3n0s3XUeuCchOqaWu6dncGHS3dy02nteOD8zlpWEGlAof4+fHBNb3q3jeDOmat5Q9dENyjtQZC/pK4kpNAkyJe3F21jf2E5z41Ixt/Hy+5ojaKovIpbpqbx/aaD3HFWB24/s4PKgUgjCPH34Z2re3HnjNU8Pn89B4sruG9QJ31y1wBUEOQvczgM/7ogkdjwAB6bV/eD+sYVPQkLdO+7BvbklzHm3eVsPlDMU0OTGNW7ld2RRDyKn7cXL4xOJTLYl9d/2EpOcQVPD+umS9Dqmf5rygm7dkA8k0enkr4zn2GvLmbXoVK7IzWYzN0FXPzyInYfKuPdq3upHIjYxMtheGRIIned3ZE5q3ZzxVtLySnWVdH1SQVB6sUFyTFMGdOb/YXlXDD5J77fdNDuSPVuwdp9jHjtZ7yMYdZN/RjQoandkUQ8mjGGW8/swIQRyaTtzGfI5J9YnZ1vd6x6Z9eFeSoIUm9OahfJp7f0JzrUn6veWcaErza5xSVPldW1PPrZOq5/fyXtmwXzyc0nk9A8xO5YInLY0O5xzL6pH8YYhr/2MzOXZ9sdqV4UV1Rz18zVXP/eSls2gqsgSL1qGxXEx38/mWHd60YzX/XOMnJd+GO/HbklXPLqYt5eVDcd8aMbT6JZqL/dsUTkN7rGhvHZrf3p1aYJ42Zn8I+P11BW6bpX1adn53P+Cz/ycdoukuPCsOOtljme+7Z79uxprVixogHjiDuZsXwnD85dS0SgL08NS+K0hGZ2Rzou8zL2ct/sDIyBZ4cnM1A3Moo4veqaWp79z0Ze+2ErbaOCGD+8Gz1aR9gd65jV1Fq8+v0WJn61iehQfyaOTKF324bLb4xZaVlWz6N+TQVBGlLm7gLumJFO1oFiLkqJ4aELEokI8rU71h/aV1DOo5+vZf6afaS0DGfy6FRaRgTaHUtEjsOirBzGzcpgT0EZ15zclrsHJjj9MeztOSXcNyeDJVvzOL9bC564OImwgIY9FaaCILaqqK7h5e+28PLCLEL8fXhocBcuTIlxurkB1TW1TPl5BxMWbKS61uLWM9pzw6ntdHRKxEUVV1TzxPz1TF26k/ioIJ4cmkSf+Ei7Yx2htLKal77L4o0ftuHjZXh4SCKX9IhrlNdIFQRxChv3FXHv7AzSs/Pp1y6Suwcm0L1VE7tjAbByRx4PfrKWdXsLOS2hKY8O6UqrSH1qIOIOftqcw72zM9idX8ZZnZtxz8BOTrHR2LIsPsvYyxPz1rOvsJyhqbHcd26nRt3npIIgTqOm1uKDJTt4/pvN5JVUcnpCU8ae3ZFuceGNnsWyLH7emstL32WxKCuX6FA/Hr4gkUFdmzvdpxsicmLKKmt4e9E2Xv1+C8UV1QzrHsfYszsSGx7Q6Flqay2+23iAVxZuYcWOQyTGhPLIkER6tmn8vRIqCOJ0SiqqmfLzdl7/YSv5pVWc1TmaMf3b0LdtZIOPTLUsi4UbDzL5282s2plP0xA/rh8Qz6V9WhHkp+GiIu7sUEklr3y/hXcXbwcLBndrwaV9WtGjdZMGf2NQXlXDJ2m7eePHrWw5WEJMmD83n9GeUb1a4WXTqGgVBHFaReVVvLNoO2/+uJXC8mpiwwMY1j2Wod3jaBMVVG/PY1kWa/cU8tnqPXyesZfd+WXEhgdw46nxDO/Z0uk3L4lI/dqdX8arC7fwcdpuiiuq6RgdzOjerbg4NZbwwPrbSF1Ta7Fq5yEWrN3Hx2l7yCmuIDEmlOtPiee8pBa273FSQRCnV1ZZw4J1+5i1chc/ZeVgWZDcMpy+bSPo0boJPVo3ITLY75j/fZZlsb+wgjW7C1i18xBfZu5jW04J3g7DgA5RXJgSy/nd7P/hFBF7lVZW89nqPUxdls3q7HwcBrrFhdOvXST92kXRs02T43oDYVkW+wrLydxdyLcb9vPVuv3kFFfi6+XglI5RjDm5LSe1i3SaZUwVBHEp+wrK+ThtN1+t20fm7kIqa2oBiI8Kol2zYKKCfYkM8iMy2JfwQB9KK2soLq+mqLya4opqsvNKydhdwMGiugFNDlM35XFwtxgGJTaniZMfsxQRe6zdU8B/MvexaEsuq7Pzqa618PVy0CYqkJjwgLpfYf40DfGjprbuhFZldS2V1bXsKyxn0/4iNu4rorC8GoBgP29OS2jKwMTmnJbQlBB/57vITgVBXFZ5VQ1rdhewYvshVu44xK5DpeSWVJJXUnnEGGeHqfuBjA71JykujG6xYSTFhdGlRRgBvlpCEJFjV1xRzfJteSzZmsu2nBL2FJSxJ7+cvJLKoz4+1N+bTs1D6dg8mIToEDpGh5DSKhw/b+d+7VFBELdTW2uRX1ZFfmklAb5ehPj7EOTr5TQf24mIeyqrrCGnuAIfLwe+3g78vOv+6u0wLvn680cFQVu2xSU5HIaIIF+nn8ooIu4lwNfLYyaraoeWiIiIHEEFQURERI6ggiAiIiJHUEEQERGRIxzXKQZjzEFgR8PFERERkUbU2rKspkf7wnEVBBEREfEMWmIQERGRI6ggiIiIyBFUEEREROQIKggiIiJyBBUEEREROYIKgoiIiBxBBUFERESOoIIgIiIiR1BBEBERkSOoIIiIiMgRVBBERETkCCoIIm7IGHOZMWZBIzzPacaYXQ39PCLS+FQQRFyUMaa/MWaxMabAGJNnjFlkjOkFYFnWh5ZlnWNzvmbGmGnGmD2HMy4yxvSxM5OIHDsVBBEXZIwJBT4HJgMRQCzwCFBhZ67fCAaWAz2oyzgFmGeMCbY1lYgcExUEEdfUEcCyrGmWZdVYllVmWdYCy7IyAIwxVxljfvrlwcaYc4wxGw+/k3/ZGPO9MebaXz/WGDPeGHPIGLPNGHPur773amPMemNMkTFmqzHmhmMJaFnWVsuyJliWtfdwxtcBXyChPv9DiEjDUEEQcU2bgBpjzBRjzLnGmCa/90BjTBQwC7gfiAQ2Av1+87A+h/95FPAM8JYxxhz+2gFgMBAKXA1MNMZ0P97AxpgU6gpC1vF+r4g0PhUEERdkWVYh0B+wgDeAg8aYT40x0Ud5+HnAWsuy5liWVQ28AOz7zWN2WJb1hmVZNdQtBbQAog8/1zzLsrZYdb4HFgADjifv4SWR94FHLMsqOJ7vFRF7qCCIuCjLstZblnWVZVlxQFcgBph0lIfGANm/+j4L+O3Jg32/+nrp4b8NBjj8CcWSwxsh86krHFHHmtMYEwB8BiyxLOvJY/0+EbGXCoKIG7AsawPwLnVF4bf2AnG//Obw0kHcUR53BGOMHzAbGA9EW5YVDswHzB9+4/9+/yfAbuCY9i6IiHNQQRBxQcaYTsaYu4wxcYd/3xIYDSw5ysPnAUnGmIuMMd7AzUDzY3wqX8APOAhUH968eEzHJ40xPtTtfSgD/mZZVu0xPqeIOAEVBBHXVETdxsKlxpgS6opBJnDXbx9oWVYOMJy6zYe5QBdgBcdwJNKyrCLgNmAmcAi4FPj0GDP2o25z4zlAvjGm+PCv49q/ICL2MHXLkSLiKYwxDur2IFxmWdZ3ducREeekTxBEPIAxZqAxJvzwnoB/ULeH4GjLESIigAqCiKc4CdgC5AAXABdZllVmbyQRcWZaYhAREZEj6BMEEREROYIKgoiIiBzB+3geHBUVZbVp06aBooiIiEhjWrlyZY5lWU2P9rXjKght2rRhxYoV9ZNKREREbGWM2fF7X9MSg4iIiBxBBUFERESOoIIgIiIiRziuPQgiIqWV1ezJL2dPfhl78svIK63E18uBr7cDXy8Hfj4OmoX40zE6hKhgX+oujxQRV6OCICK/y7IssvPKWLQlh8VbclmyNZeDRX96x9N/RQT50jE6mE7NQzm1Y1P6tY/Ez9urAROLSH1RQRCRI2zeX8TUZTtZsHY/u/PrJjI3C/Hj5HaRdGweQkxYADHhAcSE+xMZ5Ed1bS2V1bVUHP61J7+MjfuK2LS/iI37i5ixPJt3F28nyNeL0zo1Y2Bic05PaEqIv4/Nf1IR+T0qCCICQHlVDfPX7GXasp0s334IHy/DaQnNuOHUePq1i6Jd06A/WC74308F2kYFcXL7qP/+vqK6hsVbclmwdh9frdvPvIy9BPp6MbJXS8ac3JaWEYEN+CcTkb/iuO5i6Nmzp6U5CCLupayyhncWb+P1H7aSX1pF26ggRvduybDucUQG+9X789XUWqTtPMTUZTv5NH0PtZbFeUktuP6UeLrFhdf784nI7zPGrLQsq+dRv6aCIOKZqmtq+WjlLiZ9vYn9hRWc0akZ1w5oy0nxkY22sXBvQRnvLtrO1KU7Kaqo5qzOzXhwcBdaRwY1yvOLeDoVBBH5H1+t28+TX6xn68ESurcK575zO9O7bYRteQrLq3j/5x28/F0WVbUW1w+I5++ntyPQV6ugIg1JBUFEAMgrqeShuZl8nrGX9s2CGTcwgbO7RDvNUcT9heU89cUGPk7bTYswf/55fmfOT2rhNPlE3I0KgojwZeZeHvgkk4KyKm47owM3ntYOHy/nnJW2YnseD81dy7q9hQzu1oLHL04iLEAnHkTq2x8VBH1+J+Lm8ksreXDuWj5bvYfEmFDev6YPnVuE2h3rD/VsE8Fnt/bn1e+3MOGrTaTtzGfSqBR6tbFvGUTE0zjn2wcRqRdr9xQwePJPfJm5lzvP7sgnN5/s9OXgF14Ow82nt2fWjSfh5TCMfO1nJn61ieqaWrujiXgEFQQRN/VJ2m6GvbKY6hqLmTecxG1ndnDaJYU/ktqqCfNvH8DFqXE8/81mLntzKXkllXbHEnF7rvdqISJ/qKqmlkc+W8sdM9LpFhfOZ7f2J7VVE7tjnZBgP2+eG5HMxJHJpGXnc9FLi9i8v8juWCJuTQVBxI3kl1Zy+ZtLeWfRdsac3JYPr+1D05D6H3Zkl4tT45hxfV9KK2sY+vJivt900O5IIm5LBUHETewtKGP4qz+TtjOfiSOTeeiCLi65pPBnUls1Ye4tJxMXEcjV7yxjyuLtdkcScUvu9+oh4oGyDhQx7OXF7CsoZ8qY3lycGmd3pAYVGx7ArBtP4oxO0fzr07U8OX89x3NkW0T+nAqCiItL23mIS179mcoai+k39OWkdpF2R2oUQX7evHZFD67o25rXftjKg3Mzqa1VSRCpL5qDIOLCFm48wE0frKJZqB/vjentcXcYeDkMj16YSJCfN69+v4XSihqeuaQb3m64tCLS2FQQRFzU95sOcv17K2nfLJgpY3q71WbE42GM4d5BCQT5evHcV5soq6rh+VGp+HqrJIicCP0EibignzbncN17K+gQHczU69zrpMJfYYzh1jM78MD5nfkicx/Xv7+Ciuoau2OJuDQVBBEXszgrh2umLCc+KogPrulDeKCv3ZGcxrUD4nni4iQWbjzIbdPSNHVR5ASoIIi4kCVbc7lmygpaRwby4bV9aBKkcvBbl/ZpxUODu/CftfsZNytDGxdF/iLtQRBxEat2HmLMu8uJbRLAh9f2JTLYs5cV/siY/m0pqajmua82Eejnxb8v7Koro0WOkwqCiAvIOlDMmHeX0zTET3sOjtEtZ7SnuLKa177fSpCfN/cN6qSSIHIcVBBEnNz+wnKufHsZ3g7De2N60yzE3+5ILsEYw32DOlFSUVcSwgJ8+Ptp7e2OJeIyVBBEnFhBWRVXvr2M/NJKZtxwksfNOThRxhgeHdKVwrJqnvlyI7HhAVyYEmt3LBGXoIIg4qTKq2q4/r0VbDlYzNtX9aJrbJjdkVySw2F4dng39heWc89HGUSH+tM33jOmTYqcCJ1iEHFCtbUWd81czdJteYwfnsyADk3tjuTS/Ly9eP2KnrSMCOD691aQdUBXRYv8GRUEESc04atNzFuzl3+e11kfideTsEAf3r26N77eXlz59nIOFJXbHUnEqakgiDiZuem7efG7LEb1asm1A9raHcettIwI5O2repJXUsk1766gtLLa7kgiTsv2glBeVcPzX2+mvEpjUUVW7TzEPbMy6NM2gkd1dr9BdIsLZ/LoVDL3FHDPrAxdEy3yO2wvCKt2HGLi15u4f84a/aCKR9udX8b1762keag/r17eQ5cNNaCzukQzbmAn5mXs5eWFW+yOI+KUbH8F6tc+ijvP7sjHabt5e9F2u+OI2KKkopprp6ygoqqGt67sqRHKjeDGU+MZkhzD+AUb+XrdfrvjiDgd2wsCwC2nt2dgYjRPzF/Poqwcu+OINCrLsrhn1mo27itk8qWpdIgOsTuSRzDG8PSwbiTGhHLHjHSdbBD5DacoCA6H4bkRKcRHBXHL1FVk55XaHUmk0bz+w1bmr9nHvYM6cVpCM7vjeJQA37rjj/4+Dq57byUFpVV2RxJxGk5REACC/bx5/W89qa61uP79ldpdLB5hUVYOT3+5gfOTWnD9KfF2x/FIMeEBvHp5D3YdKuW26WnU6PZHEcCJCgJA26ggXhidyoZ9hYzT7mJxc7vzy7h1Whrtmgbz9CXddGLBRj3bRPDwkES+33SQyd9utjuOiFNwqoIAcHpCM+4+J4HPM/YyZfF2u+OINIjyqhpu+mAlldW1vHpFD4L9NPXcbpf2bsXQ7rE8/81mFm48YHccEds5XUEAuOnUdpzZqRmPz19P2s5DdscRqXcPf7qWjF0FPDcimXZNg+2OI9RtWnz8oiQSokO4Y0Y6uw5pL5R4NqcsCHWbFpOJDvXn5g9Xcaik0u5IIvXmoxXZTF+ezc2nt2NgYnO748ivBPh68crlPaipsbj5w1VUVGuAm3gupywIAOGBvrx8WXdyiisZOzOdWm0cEjewaX8RD87N5KT4SO48O8HuOHIUbaOCeHZ4Mqt3FfDY5+vtjiNiG6ctCFA3EvXBC7qwcONBXl6YZXcckRNSWlnN3z9cRbCfD8+PTsHLoU2JzmpQ1+Zcf0o87y/Zwdz03XbHEbGFUxcEgMv7tOLClBgmfLWJxRqiJC7swU/WsuVgMc+PSqFZiL/dceRPjBuYQK82TfjHnDVsyymxO45Io3P6gmCM4YmLk2gbFcTtM9LJKa6wO5LIcZu5IpvZq3Zx2xkdOLl9lN1x5Bh4ezl4flQqPt4Obv5wlS6UE4/j9AUBIMjPmxcv7U5BWRV3zVyt/QjiUjbuK+KhuZn0axfJbWd2sDuOHIeY8ADGX5LMur2FPDlf+xHEs7hEQQDo3CKUBwd34ftNB3nzp612xxE5JmWVNdw8tW7fwaRR2nfgis7qEs2Yk9sy5ecdfJm5z+44Io3GZQoC1O1HGJTYnGe+3Eh6dr7dcUT+1KOfr2PLwWImjdS+A1d237md6BYXxrhZq3VXjHgMlyoIv9y+Fh3qz63TVlFYrotVxHnNX7OXact2csMp7ejfQfsOXJmvt4PJo1OpteC26WlU1dTaHUmkwblUQQAIC/ThhdGp7Mkv5/45a3Rfgzil3fll3Dc7g+SW4dx1Tke740g9aB0ZxJNDk0jbmc8L3+i+BnF/LlcQAHq0bsJd53RkXsZePlqxy+44Iv+juqaWO6anUWvBC6NS8PFyyR8zOYoLkmMY3iOOF7/LYsnWXLvjiDQol33luvGUdpwUH8nDn63VGWVxKpO/zWL59kM8fnFXWkcG2R1H6tnDQxJpExnE2BnpFJRqmVPcl8sWBIfDMGFkMj5eDm6fnkZltdYExX7LtuUx+dvNDO0ey4UpsXbHkQYQ5OfN86NSOFhUwf0f61p6cV8uWxAAWoQF8PSwJDJ2FTDx6012xxEPV1BWxdgZ6bSMCOTRC7vaHUcaULe4cO4emMD8NfuYuSLb7jgiDcKlCwLAoK4tGNWrJa9+v4XFWzSKWezz4CeZ7CssZ9LIFIL9vO2OIw3s+gHx9GsXycOf1h1lFXE3Ll8QAB66oAttI4O4c8Zq8kt1NbQ0vk/SdvPp6j3ccWYHUls1sTuONAKHwzBhRAr+PlrmFPfkFgUh0Neb50elkltSoaOP0uiy80p58JNMerZuwt9Pb293HGlEzcP8eXJoNzJ3FzJJy5ziZtyiIAAkxYVx1zkJfJG5j9mrdD2rNI7qmlrGzkgHYOJIjVL2RIO6Nmdkz5a88v0Wlm3LszuOSL1xm4IAcN2AePq0jeBfczPZmatxqNLwXlm4hRU7DvHoRYm0jAi0O47Y5KELutAqIpCxM9I14VXchlsVBC+HYcLIFBwOw50z06nWOFRpQOnZ+Uz6ZjMXJMdwkY40erQgP28mjkxhX2E5/5q71u44IvXCrQoCQGx4AI9d1JUVOw7x6vdb7I4jbqq0spqxM9KJDvHjsYu6YoyWFjxd91ZNuPWM9nx8eMOqiKtzu4IAcGFKLEOSY5j09WZW69ZHaQCPz1vP9twSxo9IJizAx+444iRuOb09KS3DeeDjNezJL7M7jsgJccuCAPDvC7vSLMSPsTPSKa2stjuOuJFvN+znw6U7ubZ/W/q10y2N8v+8vRxMGplCda3FXTNXU1urE1Xiuty2IIQF+jB+RDLbckt4fN56u+OIm8gprmDcrAw6NQ/h7oEJdscRJ9QmKoiHBnfh5625vL1om91xRP4yty0IAP3aRXFt/7Z8uHQn327Yb3cccXGWZXHf7DUUllUzaVQKft5edkcSJzWyV0vO6hzNM19uZOO+IrvjiPwlbl0QAO4emECn5iGMm7WG3OIKu+OIC5uxPJuv1+9n3KAEOjUPtTuOODFjDE8NSyI0wJvbp6dRUV1jdySR4+b2BcHP24tJo1IoLKvSlEX5y7bnlPDo5+vo1y6SMSe3tTuOuICoYD+euaQbG/YVMWGBpizKX7NyRx5Ltuba8txuXxAAOjUP5Z6BCSxYt5+PVuyyO464mOqaWu6cmY63wzB+eDIOTUuUY3RGp2gu7dOK13/catuLvLiuwvIqbpuWzv1z1tgy18cjCgLANf3b0jc+gkc+W6spi3JcXlm4hVU78/n3RV2JCQ+wO464mH+e15nWEYHcNXO1pizKcXn407XsLShj/PBkvL0a/3/XHlMQHA7DcyPqpiyO1ZRFOUYZu/J5/pvNDEmO4UJNS5S/QFMW5a+Yv2Yvc1bt5pbT29OjtT03xHpMQYC6KYv/vrArKzVlUY5BWWUNd8xIp2mIH/++sKvdccSFpf5qyuLnGZqyKH9sf2E5//h4DclxYdx6ZgfbcnhUQQC4MCWGwd1aMOnrzWTs0pRF+X1PfrGerQdLGD88mbBATUuUE/PLlMV/fpzJvoJyu+OIk7Isi3tmZVBeVcOEkSn42LC08AuPKwjGGB6/KImmIX7cMSOdskodP5IjLdx4gPd+3sE1/dtycntNS5QT5+3lYOLIFCqra7n7I01ZlKN77+cd/LDpIP88vwvtmgbbmsXjCgIcnrI4PJmtB0t4Yr6mLMr/yi2u4J5ZGSREh3CPpiVKPWobFcSDg7vwU1YO7y7ebncccTJZB4p4Yv56Tk9oyuV9WtkdxzMLAsDJ7aO4pn9b3l+yg+82HrA7jjgJy7K4f84aCkqrmDQqBX8fTUuU+jW6d0vO7NSMp77coCmL8l+V1bXcPj2dID9vnr6km1PcEOuxBQHgnoEJJESHMG5WBnkllXbHEScwc0U2C9bVTUvs3ELTEqX+GWN4+pJuhPpryqL8vwlfbWLtnkKeHtaNZiH+dscBPLwg+PvUTVksKK3i3tkZmrLo4bbnlPDIZ5qWKA0vKtiPp4fVTVkc/5+NdscRm/28JZfXftjC6N6tOLtLtN1x/sujCwJA5xahjBuUwFfr9jN9ebbdccQm1TW13DGjblricyM0LVEa3pmdo7m8byve+HEbi7Jy7I4jNikoq+Kumem0iQziwcGd7Y7zPzy+IACMObkt/dtH8ehn69hysNjuOGKDyd9mkZ6dzxNDk2gRpmmJ0jj+eV4X2jUN4q6Zq8kv1TKnJ3rwk0z2F1UwaWQKgb7edsf5HyoI/DJlMRk/Hwd3TE+nslpTFj3Jiu15TP52M0NTYxncLcbuOOJBAny9eH5UKrklFfzjY10m52k+SdvNp6v3cMeZHUhuGW53nCOoIBwWHerPU0O7sWZ3AZO+1s1rnqKwvIrbp6cT1ySQRy5MtDuOeKCusWHcdU4C89fsY9ZKXSbnKbLzSnnwk0x6tm7C309vb3eco1JB+JVBXZszundLXvl+i25e8wCWZfHAx5nsKyxn0qgUQvw1LVHscd2AePrGR/Dwp2vZnlNidxxpYNU1tdw+PQ2AiSNT8HLSPU8qCL/x4OAutI0M4s4Z6RSU6uY1d/bxrz7e697KnstQRAC8HIYJI1Lw9nJw+/Q0LXO6uRe+2cyqnfk8PjSJlhGBdsf5XSoIvxHo682kUSkcKKrgvjk6+uiuduSW8NDctfRuE+G0H++JZ4kJD+DpYUms3lXAc1/p6KO7Wro1lxe/y2JY9ziGJDv3nicVhKPoFhfOuEEJfJG5j2nLdPTR3VTV1E0scxiYOMp5P94TzzOoawsu7dOK177fyo+bD9odR+pZQWkVY2ek0yrCNfY8qSD8jmv7xzOgQxSPfLaWTfs1DtWdTPp603+PNMaG60ijOJcHz+9Ch2bB3DlzNbnFFXbHkXpiWRb3f5zBgaIKnh+VSrCfcx1pPBoVhN/xy9HHEH9vbp2aRnmVxqG6g5825/Dywi2M7NlSRxrFKQX4evHC6FQKyqq4+6PVWuZ0EzOWZzN/zT7uOifBKY80Ho0Kwh9oFuLPcyNS2Li/iMfmrbM7jpygg0UV3DEjnfZNg3l4iPN/vCeeq3OLUP5xbie+23iQtxdttzuOnKCN+4p4+LO1nNw+khtOibc7zjFTQfgTp3ZsyvWnxPPBkp18mbnX7jjyF9XWWtw5M52i8ipevLQ7Ab66pVGc25X92nBW52Y89cV6Mnbl2x1H/qLSympumbqKYD8fJo5Mcakx7ioIx+DucxLoFhfGuFkZZOeV2h1H/oJXf9jCj5tzeHhIIgnNQ+yOI/KnjDE8e0kyTYP9LdjYYgAAIABJREFUuHnqKgrKdOzaFf1r7lqyDhYzaWSK09zSeKxUEI6Br7eDF0d3xwJumbpKZ5RdzModeTy3YBPnd2vBqF4t7Y4jcsyaBPky+dJU9uSXc59unHU5c1bt4qOVu7jl9Pb07xBld5zjpoJwjFpFBvLsJcms3lXAE/PX2x1HjlF+aSW3TUsnNjyAJ4cmYYzrfLwnAtCjdQTjBtYdu35/yQ6748gx2nKwmAc+yaR3mwhuP7OD3XH+EhWE4zCoa3OuPrkN7y7ezhdrtB/B2dXWWoydkc7Bogomj04lVKOUxUVdNyCeMzo147HP15O5u8DuOPInyqtquGVqGn7eDl4YnYq3l2v+r9Y1U9vo/nM7k9wynHGzMtiRq5npzuzlhVl8t/EgD17QxWWOFYkcjcNheG54MpHB/9fefUdHVa5tHP7tlEmvpFBCTYcQSAihqiCoKBaKSEeKBeyI3eOxHHtFRFQsFEE6CnzYFUWpCWkQCCSBQAKk9z5lf38EPR5BpSTZU55rrax1dCbkHtchc8/e7/u8Ou5amURlvaxHMGf/3nSAQ6crefOW3rT1sqx1B38kBeEC6RzseHdSDHZ2CnetTJL5CGZqR1Yxb353hJt6t2dKv05axxHikvm46XhnYgwny+uYtzYVk0nWI5ij1XtPsDYxj3uvDGFoRIDWcS6JFISLEOTjyhvjepF+qpJnNqdrHUf8SX5FPfetSibY350XR8u6A2E94rr48sR1kXx3sID3fs7WOo74k/15Ffx7czqXhfrxwPAwreNcMikIF2l490DuGhLM6oRcVu09oXUccYbeaOKez5Ko0xt5b0osbhYwzlSICzFzUBdu6NWeN749LOc1mJHy2kbmrNyHn5uOtyfEWMUZL1IQLsG8q8O5LNSPpzelk3yiTOs4AnjpywwSj5fx8thoQgJk3oGwPoqi8PKYnoQEuHPfqmTyymQ2i9Z+WxBdUFnPoil98HXTaR2pWUhBuAT2dgoLJsQQ4OnEnBVJFFXJwSpa2rAvj092HGP6wC5mf4yqEJfCzcmB96f0wWBUZS2UGXjnx6YF0f++oQe9rWhBtBSES+TjpuODqX0oq23kns+SMBhliJIWUnLLefzz/Qzo1oYnR0ZqHUeIFtfN3503bulFWl4FT29KlyFKGvk2PZ+3vj/CmNgOVrcgWgpCM+jR3ouXx/Zkz7FSXvoqQ+s4Nqewqp7Zn+4jwMOJdyfH4mihe46FuFBX92jL3UODWZOYy/JdMkSptWXkVzJ3TQq9Onpb5YJoWcHVTEbHBJGaW8HHvx4jvK0Ht8TJSN/W0GAwMmdF05z6DXMGWs29PyHO17yrwjmcX8Vz/3eQbv5uXBbqr3Ukm1Ba08jtyxNxc3Jg8dQ+ODta3wFw8lGrGT05MpLLQv148vP97D5aonUcq6eqKk9vSmff8TJeH9eL7u09tY4kRKuzs1OYPyGGEH937l6ZxNGiaq0jWT290cRdK/dRUNnA4mlxBHpa7jCkvyMFoRk52tuxcFIsnXxdmb1in0xabGFLduSwOiGXu4cGMzK6ndZxhNCMu5MDH90ah4O9HbctS6SiViYttqTnthxk99FSXhnb06oWJf6ZFIRm5uXiyMe39gVg5tIEOaK1hXyTns9/th7kmh6BPHhVuNZxhNBcR19X3p/Sh9yyWu5ZJQumW8ryXTl8uvs4d17ejdExQVrHaVFSEFpAFz833p/ShxOltbKzoQWk5pZz/+pkooO8mT/eOgaSCNEc4rv68vyoKH7JLOaZLbKzobl9m57PM5vTGR4ZyCMjIrSO0+KkILSQ/t3a8MKonvySWcxTsgWp2eSW1jJrWSJ+7k58NC0OF531LQwS4lKM79uJO6/oxordJ1j0k4xjbi7JJ8q4b3UyPYO8eWeibXwwkV0MLeiWvh05VlLDez9lE+jpZBWzubVUUadnxtIEGg1GVt/RD38PJ60jCWGWHr0mgoKKel775jABHk6Mk11VlySnuIZZyxIJ8HDm41tt54OJFIQW9sg14RRXNTD/+0z83J2Y0r+z1pEsUr3eyOxPmxZ+Lp/ZT8YoC/E37OwUXr25F0XVDTy2cT/+Hk4MCbfskwW1UlLdwPQle1FVlWUz4/Fzt50PJnKLoYUpisJLY3oyLCKApzYd4Mv9p7WOZHH0RhP3rkpm19ESXr05mgHBbbSOJITZ0znY8f6UPoQHenDXyiT251VoHcni1DQYmLUskdMV9Xx0a1+6+rlpHalVSUFoBQ5ntj/GdPTmgdUp7MqWGQnny2RSeWR9Gt8dLOC5m3pY/aphIZqTh7MjS2f0xcdVx4yle8mWGQnnrV5v5LZliew/WcGCiTH06eyjdaRWJwWhlbjo7Plkel86tXHljuWJpOWVax3J7KmqytOb0/k8+SQPXxPOtAFdtI4khMUJ8HRm+ax4VBUmfbhb5rOch0aDiTkr9rH7WAlvjOvFNT3aah1JE1IQWpG3q47lM+PxcnVkykd75JLfP3j928O/7ze+a0iw1nGEsFjB/u6svL0fjQYTkz7cQ26pHBH9VwxGE/evTmbb4SJeHN2TUTEdtI6kGSkIray9twur7+iPp4sjUz7ew4GTUhLO5Z0fMnl3WzYT4zvx2LURVncIihCtLaKtJ5/O6kdVvZ5JH+3mVHmd1pHMzm+3NL86kM+/r+/OxHjrOp3xQklB0ECQjyurbu+Pu5MDkz+SkvBHqqry2jcZvPHdEcbEdOD5UVFSDoRoJlEdvPh0Vj/Ka/RM/mgPBZX1WkcyG0aTyqMb0th45pbmzMFdtY6kOSkIGuno68rqO5pKwpSP95B+SkqCqqo8v/XQmSsHHXl9XC+bGEYiRGvq1dGbpTP7UlhZz4TFu+V2A01rDu5dlcS6fXk8MDyUu4eGaB3JLEhB0NBvJcFN58DExbvZe6xU60iaMZlU/vXFAT7+9RjTB3bhxdE9sZNyIESL6NPZl+Wz4impbuDm93dypKBK60iaqWs0cvvyRL7cn8+/RkbKQLs/kIKgsY6+rqy5sz9+Hk5M+XgPXx+wvTkJBqOJRzaksXLPCWZfEczTN3SX2wpCtLA+nX1ZO3sAqgrj3t/FvuNlWkdqdZX1eqZ9softmUW8MrYnt13WTetIZkUKghkI8nFlw+yB9GjvyZyVSXy6+7jWkVpNZX3T+OT1+/KYOzyMR0eESzkQopVEtPVk/eyBeJ/ZWfXzkSKtI7Wawqp6Jn24m+QT5bwzMYbxfW17QeK5SEEwEz5uOj67rX/TxMUvDvD6N4et/oCn3NJaxi7aya7sEl4Z25P7h4dKORCilXVq48q62QPo4ufGbcsSWJuYq3WkFnfgZAU3LdxBdmENH06L4/ro9lpHMktSEMyIi86e96f0YULfjizclsV9q1OoaTBoHatFJJ0oY/SiHRRU1rN8Zry0dyE0FODhzJo7+9OvaxseWZ/GM5vT0VvpMfVb005z8/s7UYD1cwYwNELOqPgrUhDMjIO9HS+N6ckjI8LZmnaK0Yt2cNTKxqN+kXySCYt34+bkwMa7BjEwxE/rSELYPM8zY5lvv6wrS3fmMOWjPZRUN2gdq9mYTCpvfXeEuz9Lons7TzbdM5ge7b20jmXWlAu5jB0XF6cmJia2YBzxRzuyirl3VTKNBhOvj+vFiCjLHvdZ22jgmc3prE3MI76LL+9P7YOvm07rWEKIP/k8OY/HNuzHz92JD6b2IaqDZb+RltU08uiGNL49WMDY2CBeHBOFk4NtHNn8TxRF2aeqatw5H5OCYN5OldcxZ2USqbnl3HlFN+ZdFY7OwfIu/KSfquDeVckcK67h7iEhPDA8FAd7y3sdQtiK/XkV3PlpIsU1jTx0dRizBnezyLkkv2QW8dC6VEprGnl0RASzBneVtU5/IAXBwjUYjDy35SAr95wgPNCD18ZFEx3krXWs86KqKst3HeeFrYfwdnVk/vjecktBCAtRUt3A4xv38+3BAuI6+/D6uF50sZAjj+v1Rl79+jCf7DhGSIA788f3tvgrIS1BCoKV+OFQAU9+foDCqnpuv7wbc4eH4exovpfJMvIr+fcX6ezNKWVouD+vj+tFG3cnrWMJIS6Aqqp8kXKSf29Kx2BUeWJkJFP6dTLrT+GpueU8uiGNjPwqbh3QmceujcRFZ76/K7UkBcGKVNbreenLQ6zam0s3PzeeHxVldp/Iq+r1vP19Jkt25uDh7MAj10QwoW9HmYwohAU7XVHHI+vT+CWzmLjOPjwxMpLYTj5ax/ofBZX1vPJ1BhuTTuLv4cSrN0czNFx2KfwdKQhW6NfMYh7bmEZeWR2Xhfrx0NXh9Oqo7W0Hg9HElrRTvPRlBkXVDUzo24lHrgnHRxYiCmEVVFVlbWIur31zhOLqBq7r2ZZHronQ/LZDvd7Ix78e491tWRiMKrMu68rdQ0Nwd3LQNJclkIJgper1RlbsPs6in7IprWnkmh6BzLs6nLBAj1bPsSEpj8Xbj3K8pJaeHbz4z6goemtcWIQQLaOmwcDi7UdZvP0oBpOJyf06M2twVzr6urZqjsp6PWsTclmyI4eT5XVc0yOQJ66LpHMby1gnYQ6kIFi56gYDn/x6jA+3H6W60cDgED9uievI1T0CW3QrT1lNI6sTcvn412MUVzfQK8iLOUNCuLp7oNxOEMIGFFbW89b3R1iTkIsKXBkewNQBnbk81L9FfwccL6lhyY4c1iXmUtNoJL6rL/cPC2WQmd1utQRSEGxEWU0jy3cdZ21iLifL6/B2dWRU7w6MjulAVAevZtmiVFLdwDfpBXx14DQ7s0swmlQuC/VjzpBgBnRrY9YLl4QQLeNUeR2r9p5g1d5ciqsb6OTrys19grgizL/ZfvecKKnl+0MF/JBRwM7sEhzsFG6Ibs/MwV1ld8IlkIJgY0wmlR3ZxaxJyOXb9AIajSY8nByI6+JDfNc29OvmS7C/O57ODn/7hm4wmjhWXMPB05UcPF1JyolyEnJKManQpY0r1/Vsxw292hPZzrMVX50Qwlw1Gkx8k57Pp7uP/358vberI4NC/LgsxI+Idp509HHB1033t7976hqNZBdVk11UzcFTlfyYUUhmYdNE2ZAAd66Lasvk/p0J9HRulddlzaQg2LCymka2Zxax51gpe4+VklX437HNzo52BHo6E+jhjLerIw0GE3WNRmr1BmobjJwsr6PB0DSPXWdvR1hbd4aGB3BtVDsi23nI1QIhxF8qqW7g16xifsks5pfMIgoq/zu22U1nT0dfVwI8nVFVFaPpv1+nK+o5WV73+3Md7BTiu/oyLDKQ4ZEBsr6gmUlBEL8rrm4gMaeMvLJaCirrKahsoKCynvJaPc6Odrjo7HHVOeCis6etpzM92nvSvb0nwf7uOMrkQyHERVBVleyiGo4WVZNbVkduaS25pbUUVzdgZ6dgryjY2zV9+bk7ERLg/vtX5zauMha5Bf1dQZA9IDbGz93J4s90EEJYFkVRfn/DF5ZDPhIKIYQQ4ixSEIQQQghxFikIQgghhDiLFAQhhBBCnEUKghBCCCHOckHbHBVFKQKOt1wcIYQQQrSizqqq+p/rgQsqCEIIIYSwDXKLQQghhBBnkYIghBBCiLNIQRBCCCHEWaQgCCGEEOIsUhCEEEIIcRYpCEIIIYQ4ixQEIYQQQpxFCoIQQgghziIFQQghhBBnkYIghBBCiLNIQRBCCCHEWaQgCGGFFEWZrCjKt63wc4YoipLX0j9HCNH6pCAIYaEURRmsKMpORVEqFEUpVRRlh6IofQFUVV2pqurVZpBxm6IoRYqiVCqKkqooyk1aZxJCnB8HrQMIIS6coiiewP8Bc4C1gA64DGjQMtc53A8cVFXVoChKP+B7RVHCVFU9rXUwIcTfkysIQlimMABVVVepqmpUVbVOVdVvVVVNA1AUZbqiKL/+9mRFUa5WFOXwmasNixRF+VlRlNv++FxFUV5XFKVMUZRjiqJc+4fvnaEoyiFFUaoURTmqKMqd5xtSVdU0VVUNv/0j4Ah0bIbXL4RoYVIQhLBMRwCjoijLFEW5VlEUn796oqIofsB64HGgDXAYGPinp/U78+/9gFeBjxVFUc48VghcD3gCM4C3FEWJPd+giqL8n6Io9cAe4Ccg8Xy/VwihHSkIQlggVVUrgcE0fSr/EChSFGWzoiiB53j6dUC6qqobz3yaXwDk/+k5x1VV/VBVVSOwDGgHBJ75WVtVVc1Wm/wMfEvT7YzzzXo94HEmxzeqqpou6MUKITQhBUEIC6Wq6iFVVaerqhoERAHtgfnneGp7IPcP36cCf955kP+Hx2vP/E93gDNXKHafWQhZTtMbvd8FZtWrqvoVcI2iKDdeyPcKIbQhBUEIK6CqagawlKai8GengaDf/uHMrYOgczzvLIqiOAEbgNeBQFVVvYEvAeVvv/GvOQDBF/m9QohWJAVBCAukKEqEoijzFEUJOvPPHYGJwO5zPH0r0FNRlFGKojgAdwNtz/NH6QAnoAgwnFm8eF7bJ89kvFZRFBdFURwVRZkCXA78fJ4/WwihISkIQlimKpoWFu5RFKWGpmJwAJj35yeqqloMjKNp8WEJ0J2mhYL/uCVSVdUq4D6atlKWAZOAzeeZUQGeoWmRYxFNWx7Hq6qadJ7fL4TQkNJ0O1IIYSsURbGjaQ3CZFVVt2mdRwhhnuQKghA2QFGUaxRF8T6zpuAJmj7dn+t2hBBCAFIQhLAVA4BsoBi4ARilqmqdtpGEEOZMbjEIIYQQ4ixyBUEIIYQQZ7mgw5r8/PzULl26tFAUIYQQQrSmffv2Fauq6n+uxy6oIHTp0oXERBmjLoQQQlgDRVGO/9VjcotBCCGEEGeRgiCEEEKIs0hBEEIIIcRZpCAIIYQQ4iwXtEhRCHHpjCaVw/lV5JTUkFtay4nSWnLL6iitacBeUbC3++9XoKczwf7uhAS4E+zvThc/V5wc7LV+CUIIGyAFQYhWcLqijl+OFLM9s4gdWcWU1ep/f8zb1ZFOvq4EeDhjUlWMJhWDUcVgMpGYU8amlFO/P9fRXqF/tzYMiwhgWGQgHX1dtXg5QggbcEGTFOPi4lTZ5ijE+alrNLI59SQrdp9g/8kKAAI8nLgs1J/BoW0IC/Sgo68rns6Of/vn1DYaOFpUQ3ZRNQdOVvBDRiFHi2oACA/04Prodkzs1wk/d6cWf01CCOuiKMo+VVXjzvmYFAQhmld2UTUrd59g/b5cKusNhAW6MzY2iCvC/QkP9EBRlEv+GceKa/jhUAHfHixg77FSdA52jO7dgZmDuxLe1qMZXoUQwhZIQRCiFeSW1vLqN4fZknoKR3uFEVHtmNq/M327+DRLKfgrWYXVLNlxjA1JedTrTQwO8WPuVWH06ezTYj9TCGEdpCAI0YLKaxtZ+GMWy3cdx84OZg3uyq0DuxDg4dyqOcpqGlmVcIIlO3IoqmpgVO/2PHptBO28XFo1hxDCckhBEKIFGIwmlu7MYcEPmVQ1GBjXJ4gHrwqnrVfrFoM/q2kw8P7P2Xyw/Sj2isKcIcHccXk3nB1l94MQ4n9JQRCimR0tqmbeulSST5RzeZg/T1wXQURbT61j/Y/c0lpe/iqDrftPE+TjwhvjetGvWxutYwkhzIgUBCGaicmksmxXDq98nYGTgz3P3dSDG3u1b9E1Bpdq99ESHtuQxvHSWmZfEczc4WHoHGRGmhDi7wuCzEEQ4jydLK/jobWp7DpawpBwf14ZG02gp7a3E85H/25t2HrfZTy35SDv/ZTNL5lFzB8fQ0iAu9bRhBBmTD5GCHEedmYXc8M7v5KWV87LY3qyZHpfiygHv3FzcuCVm6N5f0ofTpbVcf07v7Am4YTWsYQQ/2D13hO8+nWGJj9briAI8TdUVWXpzhye33qIrn5uLJ7ah27+lvvJe0RUW2I6efPg2hQe3bCfjPwqnrwuEgd7+awghDkxGE288OUhluzI4fIwf/RGE46t/PdUCoIQf6Feb+TJzw+wISmP4ZGBvDW+Fx7/MPXQEgR6OrNsRvzvv3yyCqtZODEWL1fLf21CWIOKOj33fJbEL5nFzBzUlSeui9CkxEtBEOIcCqvquX1ZIql5Fdw/LJT7h4ViZ2e+CxEvlIO9HU/f0IPwQA+e2nSA0Yt28NGtcRZ9dUQIa3C0qJrblic27UIa05MJ8Z00yyLXFYX4kxMltYx7fxdHCqr5YGof5l4VZlXl4I8mxHdixax+lNfpGfXuDvYeK9U6khA2KyGnlFHv7qC8Vs+KWf00LQcgBUGI/3HodCVj399JRZ2elbf345oebbWO1OL6dWvDprsH4efhxLRP9vDzkSKtIwlhc34+UsTUj/fg5+HEprsHmcXMEikIQpyRmFPK+A92Ya8orLtzALGdbOcsg46+rqy9cwDd/Ny5bVkCX+0/rXUkIWzGV/tPc9uyBLr5ubP2zgFmc4y7FAQhgG0ZhUz5eA9+7k6snzOA0EDbOxHRz92JVXf0JzrIm7s/S2JdYq7WkYSweuv35XH3Z0lEB3mz6o7+ZnVsuxQEYfO2ZRRyx6eJhAS4s3b2AIJ8zKO9a8HLxZFPZ8UzKMSPh9ensWxnjtaRhLBay3bm8NC6VAaF+PHprHi8XMxrJ5EUBGHTfsks4s4V+wgL9GDlLPNq71px1Tnw0a1xXN09kKc3p7Nyz3GtIwlhdVbsPs7Tm9O5unsgH90ah6vO/DYVSkEQNmv30RJuX55INz83VszqJ3MA/sDJwZ6Fk2K5MiKAf31xgI1JeVpHEsJqbEzK46lNB7gyIoCFk2JxcjDPk1alIAiblJhTysylCXT0cWXFbf3wcdNpHcns6BzsWDQ5lgHd2vDQulS+lIWLQlyyL/ef5qF1qQzo1oZFk2PN+uA0800mRAtJyytn+pIEAj2dWXlbP7mt8DecHe35cFocMZ18uG9VMj9mFGgdSQiL9WNGAfetSiamkw8fTovD2dE8rxz8RgqCsCk5xTXMWJKAt6sjn93ejwALOnBJK25ODiyZ0ZfIdp7MXpHEruwSrSMJYXF2ZZcwe0USke08WTKjL25O5rfm4M+kIAibUVzdwK1L9mJSVZbPjKedl4vWkSyGp7Mjy2fG09nXlTs+TeRwfpXWkYSwGIfzq7hjeSKdfV1ZPjMeTws500UKgrAJNQ0GZi5NoKCyno+n95UzBy6Cj5uOpTPjcXG0Z8aSvRRU1msdSQizl19Rz/Qle3HR2bN0ZrxFrXeSgiCsnt5o4u7PkjhwsoKFE2NtakJic+vg7cKSGX2pqNMzfUkCVfV6rSMJYbaq6vXMWJpAZZ2eT6b3pYO3ZV21lIIgrJqqqjyxcT8/HS7ihdE9Gd49UOtIFq9Hey8WTenDkYIq7lqZhN5o0jqSEGZHbzRx18okjhRUsWhKH6I6eGkd6YJJQRBW7f2fj7JuXx73DQtlosYno1mTK8L8eWl0T37JLObxjftRVVXrSEKYDVVVefLz/fySWcyLo6O4Isxf60gXxfyXUQpxkb47WMCr32RwQ6/2zB0eqnUcq3NL347kldex4IdMQgPcufOKYK0jCWEWFm8/ytrEPO67MoTxfS33g4kUBGGVMvIreWB1Mj07ePHazdEoiqJ1JKs0d3go2YXVvPx1BmGBHgyNCNA6khCa2pZRyMtfZzCyZzvmXhWmdZxLIrcYhNUpqW7gtmWJuDs7WMQwEkumKAqvjYsmsq0n961KJquwWutIQmgmq7Ca+1YlE9nWk9fGWf4HEykIwqo0GkzMXrGPoqoGFk+NI1AGIbU4V50DH94ah87BjtuXJ1JRKzsbhO2pqNVzx/JEdA52fGimhy9dKCkIwqo8vfkACTllvDauF706emsdx2Z08Hbh/al9yCur5Z5VSRhkZ4OwIUaTyr2rk8ktq+X9qX0sbjvjX5GCIKzG6r0nWLU3lzlDgrmxV3ut49icvl18+c9NUfySWcwrX2doHUeIVvPK1xlsP1LEf26Kom8XX63jNBvLvwYiBE0HMP17czqDQ/x46OpwrePYrAnxnTh4upIPfzlGTCcfruvZTutIQrSoL/efZvH2o0wb0JkJVraVWq4gCItXWtPInBVJ+Ls7sWBiDPZ2lr0wyNL9a2R3Yjp58/C6VFm0KKxadlE1D69LJaaTN/8a2V3rOM1OCoKwaEaTyv2rkymqauC9KbH4WtCcc2ulc7Bj0eRYnBztmbNiHzUNBq0jCdHsahoMzP50H06O9iyaHIvOwfreTq3vFQmb8tZ3R/gls5jnbupBdJAsSjQX7bxceGdiDNlF1TJpUVgdVVV5fON+souqeWdijNWeDCsFQVisHw4VsHBbFhP6drS6e3/WYFCIH/OuDmdz6imW7czROo4QzWb5ruNsTj3FvKvDGRTip3WcFiMFQVikU+V1zFuXSvd2njxzYw+t44i/MOeKYIZHBvD81kMknSjTOo4QlyzpRBnPbz3I8MgA5lj5eHEpCMLi6I0m7l2VjMGo8u7kWJmUaMbs7BTeGNebtl7O3PtZsgxREhatolbPvZ8lE+jpzBvjemNn5QuipSAIi/P6t4fZd7yMl8b0pKufm9ZxxD/wcnVk4aRYCirreWRDqqxHEBZJVVUe2ZBKQWU9CyfF4uXqqHWkFicFQViUHzMK+ODno0zu14kbZBiSxejd0ZvHro3gm/QCWY8gLNLyXcf5Jr2Ax66NoLeNTGmVgiAsxqnyOuatTSWynSdPXW99e46t3azBXRkWEcCLX2awP69C6zhCnLcDJyt4YeshhkUEMGtwV63jtBopCMIiGIwm7l+dTKPBxLuTYmTdgQVSFIXXx/WijbuOe1YlUVUv6xGE+auq13PPZ0m0cdfx+rheFn9C44WQgiAswoIfs0jIKePFMT3p5u+udRxxkXzcdCyYGENeWR1PfH5A1iMIs6aqKk9+foDcsjoWTIzBx8YGsUlBEGZv77FSFv6YyZjYDtzUu4PWccQl6tvFlwevCmNL6inW78vTOo4Qf2n9vjw2p55i7vAQ17OoAAAY+klEQVRQqzqE6XxJQRBmraJWzwOrk+nk68pzN0VpHUc0k9lXBNO/my9Pb07nWHGN1nGEOMux4hqe3pxO/26+zBkSonUcTUhBEGZLVVUe25hGUXUDCybG4O4kh49aC3s7hbfG90bnYMd9q5rWlghhLhoNTWueHO3teGt8b5s9AE4KgjBbqxNy+epAPg9dHS7nLFihdl4uvDwmmv0nK3jju8NaxxHid298d5i0vApeGRtttecsnA8pCMIsZRVW8eyWdC4L9eP2y7ppHUe0kBFRbZnUrxMf/HyUXzOLtY4jBL9mFvPBz0eZ1K8TI6Laah1HU1IQhNlpMBi5b1UKrjoH3hjXy+rHmdq6p0Z2JyTAnQfXplBS3aB1HGHDSmsaeXBtCiEB7jw1UmatSEEQZufNb49w8HQlr46NJsDTWes4ooW56OxZMCGG8lo9j26Qo6GFNlRV5dENaZTX6lkwIQYXncxakYIgzMrO7GIW/9J0eW9490Ct44hW0r29J4+MCOf7QwWsTsjVOo6wQasTcvnuYAGPjAine3tPreOYBSkIwmyU1zby4JpUuvq58a+RkVrHEa1s5qCuDA7x47ktBzlaVK11HGFDjhZV89yWgwwO8WPmINsZpfxPpCAIs/DbxLLi6gbeHh+Dq062NNoaO7umUcw6BzvmrklBb5Stj6Ll6Y0m5q5JQedgx+uy5ul/SEEQZmFD0km27j/Ng1eH0TPIS+s4QiNtvZx5aUxPUvMqWPBDptZxhA1Y8EMmqXkVvDSmJ229ZM3TH0lBEJrLLa3lmc3pxHf15c7Lg7WOIzR2Xc923NwniHe3ZZGYU6p1HGHFEnJKeXdbFjf3CeK6nu20jmN2pCAITRlNKnPXpKAAb97Sy2Ynlon/9cyNPQjyceWBNSly6qNoEVX1euauSaGDjwtP3yBbGs9FCoLQ1Ps/Z5N4vIznRjW9IQgB4O7kwFvje3GqvI5ntxzUOo6wQs9uOcip8jreuqU3Hs6OWscxS1IQhGb251Xw1ndHGBndjlFySqP4kz6dfbl7aAjr9+Xx9YHTWscRVuTrA6dZvy+Pu4aEEGeDpzSeLykIQhN1jUYeWJOMn7sTL4yKQlHk1oI4233DQokO8uLxjfsprKzXOo6wAoWV9Ty+cT89O3hx//BQreOYNSkIQhMvf3WI7KIaXh/XC29XndZxhJn67TS9Or2Rh9enyZRFcUlUVeXh9WnU6Y28Nb43jvbyFvh35L+OaHU/Hyli2a7jzBjUhcGhflrHEWYu2N+dJ66L5OcjRXy6+7jWcYQFW7H7OD8fKeKJ6yIJCXDXOo7Zk4IgWlVZTSMPr0slNMCdR0dEaB1HWIip/TtzRZg/L2w9RFahTFkUFy67qJoXvjzEFWH+TO3fWes4FkEKgmg1qqryxOf7KattZP6E3jg7ymEo4vwoisJrN0fjqrNn7poUGg0yZVGcv9+mJbo42vPazdGy5uk8SUEQrWZj0km+OpDPg1eF06O9TEsUFybAs2nK4v6TFbzzo0xZFOdvwQ+ZpJ2ZlignxJ4/KQiiVeSW1vL05nTiu/hyx+XdtI4jLNSIqP9OWdx3XKYsin+27/h/pyWOiJJpiRdCCoJocUaTyry1qQC8IdMSxSV6+obutPd2Ye6aVKobDFrHEWasusHA3DWptPeWaYkXQwqCaHGLtx9lb04pz9zYg46+Mi1RXBoPZ0feGt+bvLJa/iNTFsXf+M+Wg+SW1fLWeJmWeDGkIIgWlX6qgje/O8y1UW0ZGyvTEkXz6NvFl9lXBLMmMZdv0vO1jiPM0Dfp+axJzGX2FcH0lWmJF0UKgmgx9Xoj969OwcdVxwuje8rKYdGsHhgeRo/2nk1TFqtkyqL4r8KqpmmJPdp7Mnd4mNZxLJYUBNFiXv4qg6zCal4f1wtfN5mWKJqXzsGO+eN7U9Ng4FGZsijOUFWVR9enUd1gYP743ugc5G3uYsl/OdEifjpcyNKdOcwY1IXLw/y1jiOsVGigB09cF8m2wzJlUTT5dPdxth0u4olrIwgN9NA6jkWTgiCaXWlNIw+vTyMsUKYlipY3bcB/pyxmFlRpHUdoKLOgihe2Nk1LvHVgF63jWDwpCKJZqarKYxvSqKjVM398jExLFC1OURReGxeNm5MD969OocFg1DqS0ECjwcT9q1Nwc3LgtXEyLbE5SEEQzWptYi7fHizg4WvC6d7eU+s4wkYEeDjzythoDp6u5M1vj2gdR2jgje8Oc/B0Ja+MjSbAQ6YlNgcpCKLZHCuu4dktBxkY3IZZg7tqHUfYmKu6BzIxvhOLfznKzuxireOIVrQru4TF248yMb4TV3UP1DqO1ZCCIJpF0+W9ZHQOdrxxSy/sZFqi0MBT10fStY0b89amUl7bqHUc0QoqavXMW5tC1zZuPHV9pNZxrIoUBNEs3vzuCGl5Fbw8Jpp2Xi5axxE2ylXnwPwJvSmubuCxDftl66OVU1WVxzamUVjVwPwJvXHVOWgdyapIQRCXbEdWMR9sz2ZifCdGRLXVOo6wcdFB3jx0dThfp+ezOiFX6ziiBa1OyOWrA/k8fE040UHeWsexOlIQxCUprWnkwbUpdPOTy3vCfNx+WTcGh/jx7JZ0sgpl66M1yiqs4tkt6QwO8eP2y+SE2JYgBUFcNFVVeXRDGmU1et6eECOX94TZsLNTePOWXrjqHLh3VQr1etn6aE3q9UbuXZWCq86BN2XNU4uRgiAu2so9J/juYAGPjAgnqoOX1nGE+B8Bns68dnM0h05X8urXh7WOI5rRK19ncOh0Ja/dHE2Ap2xpbClSEMRFOXS6kuf+7yCXh/kzc5BsaRTmaVhkINMHduGTHcfYllGodRzRDLZlFLJkRw7TB3ZhWKRsaWxJUhDEBatpMHD3Z0l4uTjK5T1h9h67NoKIth7MW5dKfoWc+mjJ8ivqmbculYi2Hjx2rYxxb2lSEMQF+/emdI4V1/D2hN74uTtpHUeIv+XsaM/CSbHU643ctyoZg9GkdSRxEQxGE/etSqZeb+TdybEyxr0VSEEQF2TDvjw2JOVx75WhDAz20zqOEOclJMCd50dFsTenlLd/yNQ6jrgIb/+Qyd6cUl4YHUWwv7vWcWyCFARx3rKLqnlq0wH6dfXl/mGhWscR4oKMiQ1iXJ8gFm7L4tdMGcVsSX7NLGbhtixuiQtidEyQ1nFshhQEcV7q9UbuXpmEs6M9b0+IwV7WHQgL9OxNPQjxd+eBNckUVsp6BEtQWFnPA2uSCfF355kbe2gdx6ZIQRDn5elN6WTkV/HGuF609ZJtRcIyueoceHdyLNUNBu5fnYLRJKOYzZnRpPLAmhSqGwy8OzlWZq20MrMoCAk5pTIz3YytTchlTWIu9wwNYWhEgNZxhLgkYYEePHdTFLuOlvDWd3I0tDmb//0RdmaX8NxNUYQFemgdx+ZoXhB+OlzIuPd38enu41pHEeeQfqqCpzYdYFBIG+ZeFaZ1HCGaxS1xHbklrmk9wvcHC7SOI87h+4MFvPNj07qDW+I6ah3HJmleEC4P9WdYRADPbTlIYk6p1nHEH1TU6ZmzIgkfV52sOxBW57mboojq4MnctSkcL6nROo74g+MlNcxdm0JUB0+euylK6zg2S/OCYGen8Ob43nTwceGulUmycMhMmEwq89amcqq8jncnx8q8A2F1nB3teW9yH+wUhdkrkqhrlPMazEFdo5HZK5KwUxTem9xH5h1oSPOCAODl4sgHU/tQVd80oU8vg0w098H2o3x/qIAnR0bSp7OP1nGEaBEdfV2ZP6E3GfmV/OuLA7IWSmOqqvLkF/vJyK9k/oTedPR11TqSTTOLggAQ0daTl8f2JCGnjBe2HtI6jk376XAhr36TwfXR7Zg+sIvWcYRoUUPDA7h/WCgbkvJYueeE1nFs2so9J9iYdJL7h4UyNFwWRGvNbAoCwE29OzBzUFeW7szhi+STWsexSUeLqrl3VTKRbT159eZoFEXWHQjrd9+VoQwN9+fZLensPSZrobSw91gpz25JZ0i4P/ddKYPYzIFZFQSAx6+LIL6rL49uSCM1t1zrODalsl7PbcsTcbS3Y/G0PrLnWNgMOzuF+RNi6OjryuwV+8gtrdU6kk3JLa1l9op9dPRx5e0JMXIAnJkwu4LgaG/HojOL4m5fniinr7USo0nlgdUpnCipZdHkWIJ85N6fsC1eLo58NC0Og9HE7csTqWkwaB3JJtQ0GLh9eSJ6o4kPb43Dy8VR60jiDLMrCAB+7k58PD3u9//jyOrilvfGt4f5MaOQp2/sQf9ubbSOI4Qmuvm7s3BSLEcKqpi7JgWTTFpsUSaTytw1KRwpqOLdSbFyCJOZMcuCAE2LFhdMjOHAqQrmrZO/qC3pi+STLPopm0n9OjG1f2et4wihqcvD/PnXyO58e7CAt76XSYstaf73R/j2YAFPjuzO5WH+WscRf2K2BQFgWGQgj18bwZf785kvR7S2iF3ZJTy8PpV+XX155gY5CEUIgBmDujA+riPv/JjFphRZMN0SNqWcZMGZSYkzB3XROo44B7NfhXb7Zd3ILKhmwQ+ZdPNzY1RMB60jWY2swiru/DSRzm3cWDw1Dp2DWfdFIVqNoij8Z1QUOSU1PLQuFX8PJwYG+2kdy2rsyi7h4XVpxHf15T+jomS3lJky+3cERVF4YXRP+nX15eH1qXKOezMprKrn1k8S0DnYs2R6X7xcZWGQEH+kc7Bj8dQ4uvq5cefyfWTkV2odySpk5Fdyx6eJdG7jyodT43BykEmJ5srsCwKc+Ys6LY5gf3fu/DSRAycrtI5k0WobDcxamkhpTSOfTI+TaWVC/AUvV0eWzIjH1cmeGUsSOF1Rp3Uki3a6oo4ZSxJwcbRn6cx4+WBi5iyiIEDTFqSlM+LxdtUxfUkCJ0pkn/LF0BtN3PtZMumnKlg4KYboIG+tIwlh1jp4u7BkejxV9QZmLEmgsl6vdSSLVFmvZ8aSBKrqDSydEU8HbxetI4l/YDEFAaCtlzPLZvbFYDIx7ZM9FFc3aB3JophMKg+vS+WHjEKevSmKYZGBWkcSwiJ0b+/J+1P6kFVYzZ3L91Gvl63XF6Jeb+TO5fvIKqzm/Sl96N7eU+tI4jxYVEEACAnw4ONb+5JfWc/MpQlUSZs/L6qq8tSmA3yRcoqHrwmX7YxCXKDBoX68Ni6aXUdLuGtlEo0GOVTufDQaTNy9MoldR0t4bVw0g0NlsaelsLiCANCnsw/vTorl4KlKpi9JoFomnv0tVVV56asMVu45wZwhwdw9NETrSEJYpNExQTw/KoofMwq5f3UyBjl59m8ZjCbuX53MDxmFPD8qitExQVpHEhfAIgsCNM1IWDAxhpTccmYuSaC2UUrCX3nnxywWbz/KtAGdeeSacK3jCGHRpvTvzFPXd+erA/nMW5eKUYa4nZPRpDJvXSpfHcjnqeu7M0WuWlociy0IANf1bMf88b1JPF7KzKUJMpL5HBZvz+bN744wJrYDz9zQQ/YbC9EMZg3uysPXhLMp5RSPb0yTSa9/YjKpPL4xjU1nbmnOGtxV60jiIpj9oKR/ckOv9pjUpnnety1P4ONb++LsKPtqVVVlwQ9ZvPX9EUZGt+PVsdFyQpoQzejuoSHU642882MW9nZ2PD8qCnv5O4bJpPKvTQdYm5jHfVeGyC1NC2bxBQHgpt4dMBhVHlqfyqxlCXwwNQ53J6t4aRdFVVVe/iqDD7YfZWxsEK+M7YmDvUVfLBLCLD14VRhGk8qin7KpbjDwxrheNj2RtNFgYt66VLaknuKuIcHMvSpM60jiEljNu+jYPkEoCjy8Po2Ji3ezZEZf/NydtI7V6kwmlX9vPsCK3SeYNqAzz9zQQ64cCNFCFEXhkREReDg78srXGVTV63lvch9cdLZ3FbOu0ciclfv46XARj46IYM6QYK0jiUtkVVV3TGwQH07rQ2ZhFTe/t5PcUtsapmQwmnhofSordp9g9hXBPHujlAMhWsOcIcG8NKYnPx8pYtone6ios63t1xV1eqZ9soefjxTx0pieUg6shFUVBIArIwJZeVs/ymr1jHlvJwdP2cb89Io6PTOWJrAx6STzrgrj0RHhsiBRiFY0Mb4T75zZWTVx8W4KKuu1jtQqCivrmbh4Nym55SycGMvE+E5aRxLNxOoKAkCfzr6snz0ABzuF8R/sYvuRIq0jtaic4hrGLNrB7qMlvHpzNPcOC5VyIIQGro9uz4fT4sgpqeGGd34lJbdc60gtKjW3nBsW/kpOSQ0f3dqXkdHttI4kmpFVFgSA0EAPNswZSAcfF25dspd3t2Whqta3FWn30RJGLdpBaU0jK2b145a4jlpHEsKmDQkPYMOcgegc7Ljlg11sTMrTOlKL+Dw5j3Ef7MLR3o4NcwZyRZi/1pFEM7PaggDQ3tuFjXcNZGTPdrz2zWHuWplkNVMXVVVl1d4TTP14D37uTnxx9yD6dWujdSwhBBDZzpPN9wwmtpM3D65N5cUvD1nNQCWjSeXFLw8xd00qsZ282XzPYCLbydkK1shqdjH8FVedA+9MjKFXkDcvfXWIrMJqPpjah27+7lpHu2gVdXqe+Hw/W9NOc3mYPwsnxeDpLMemCmFOfN10fDqrH//5v4Ms3n6UQ6creX1cLwI9nbWOdtEKK+uZty6VXzKLmTagaaKko2yhtlrKhVx2j4uLUxMTE1swTsvamVXMPauSadAbefy6SCbFd7K4Vf4JOaU8sDqFgsp6Hrw6jDsvD5bhLEKYudV7T/DMlnScHe15YVRPi7xX/+X+0zz5+X7q9EaevqGHLEa0Eoqi7FNVNe6cj9lSQQA4VV7HI+vT+DWrmEEhbXhlbDRBPq5ax/pHBqOJBT9msfDHTDr6uvL2hBh6d/TWOpYQ4jxlF1Xz4JoUUvMqGB3TgWdu7IGXi/lf+auo0/PM5nQ+Tz5JryAv3hzfm2ALvgIr/pcUhD9RVZXP9p7gxa2HAHhiZNPVBHNd+b8ru4RnNqdzuKCKsbFBPHtTD5ueFCmEpdIbTby7LYt3fswi0MOJ526KYlhkgFn+7lFVlR8zCnnqiwMUVDVwz9AQ7rkyRG4pWBkpCH8ht7SWxzamsSOrhJhO3jx+bSTxXX21jvW7U+V1vPDlIbamnSbIx4Wnru/ONT3aah1LCHGJUnLLmbc2heyiGgaFtOHJ67rTvb35LPQ7dLqSF7Ye4tesYoL93Xjjlt5yxdJKSUH4G6qqsi4xjze+O0xBZQPDIgJ4ZEQE4W09NMtUWa9n6Y4c3vspG5OqMmdIMLOvCJZDqISwInqjiZW7jzP/h0wq6vSMj+vIg1eHEeCh3SLGwqp63vz2CGsTc/F0ceSBYaFM7t9ZrhpYMSkI56Gu0cjSnTks+imL6gYDo3t3YObgrkR18Gq1DKfK6/jk12OsTsilusHAiB5teXJkJB19zX+NhBDi4lTU6lnwYybLd+XgYGfH6NgOTB/YhbDA1vuQkllQxbJdOWzYdxKDycStA7pw75WheLma/xoJcWmkIFyA8tpG3vspm2W7cqjXm4gO8mJifCdu7NUetxa47280qSSfKGPF7uP8X9ppVOCG6Hbcdlm3Vi0nQghtHSuuYdG2LDalnqLRYGJQSBtuHdCFYZGBLbJTyWhS+eFQAct25bAjqwSdgx039mrPPUND6OLn1uw/T5gnKQgXoaJOzxfJJ/lszwkOF1ThprPnmh5tGRTix8CQNrTzcrnoP7vBYGRndgnfpufz3cFCiqsbcNPZMzG+EzMGd6WD98X/2UIIy1Za08iqvSdYsfs4pyvqaeOm44pwf4aGB3B5qP8lfaovr21ke2YxPx0uZPuRIoqrG2nn5cyU/p2Z0LcjbWzwBFxbJwXhEqiqStKJclbtPcEPhwooq206pa2bnxv9g9sQ4u9Oe28X2ns7097bBV9XHY1GU9OXwUS93kheWR1HCqrIyK/iSH4Vh05XUtNoxE1nz5CIAK7uHsiVEQF4yLAjIcQZBqOJ7w4W8HV6PtuPFFFWq8dOgZhOPkS09aCrnxtd2rjRxc+NQE8n7P6wE0IF8ivqySmuIaekhmPFNRw6XUlKbjkmFbxdHbk81J9ro9pyVfdAHGSNgc2SgtBMTCaVjPwqdmYXsyu7hL3HSqm6gNHNns4ORLT1JKKdB0PC/RkY7CcLD4UQ/8hoUknNK+enjEJ+zSomu6jmgo6U9nR2ICTAncEhflwRHkDvjt4yYE0AUhBajKqqlNXqOVVe9/tXWa0enYMdTme+dA52tPVyITzQg0BPJ7Pc7yyEsDxlNY0cK6khp7iG4uqG/3lMVSHA04nObdzo2sYNHzedRimFufu7giDTdi6Boij4uunwddPJgkIhRKvycdPh46YjtpOP1lGElZIbT0IIIYQ4ixQEIYQQQpxFCoIQQgghziIFQQghhBBnuaBdDIqiFAHHWy6OEEIIIVpRZ1VV/c/1wAUVBCGEEELYBrnFIIQQQoizSEEQQgghxFmkIAghhBDiLFIQhBBCCHEWKQhCCCGEOIsUBCGEEEKcRQqCEEIIIc4iBUEIIYQQZ5GCIIQQQoiz/D+FhXpzEVcTjgAAAABJRU5ErkJggg==\n",
-      "text/plain": [
-       "<Figure size 648x648 with 3 Axes>"
-      ]
-     },
-     "metadata": {
-      "needs_background": "light"
-     },
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "# Our data...\n",
-    "x = np.linspace(0, 10, 100)\n",
-    "y1, y2, y3 = np.cos(x), np.cos(x + 1), np.cos(x + 2)\n",
-    "names = ['Signal 1', 'Signal 2', 'Signal 3']\n",
-    "\n",
-    "fig = plt.figure(figsize = (9,9))\n",
-    "sig1 = fig.add_subplot(311)\n",
-    "sig1.plot(x, y1)\n",
-    "sig1.title.set_text(names[0])\n",
-    "sig1.axes.xaxis.set_visible(False)\n",
-    "sig1.axes.yaxis.set_visible(False)\n",
-    "sig2 = fig.add_subplot(312)\n",
-    "sig2.plot(x, y2)\n",
-    "sig2.title.set_text(names[1])\n",
-    "sig2.axes.xaxis.set_visible(False)\n",
-    "sig2.axes.yaxis.set_visible(False)\n",
-    "sig3 = fig.add_subplot(313)\n",
-    "sig3.plot(x, y3)\n",
-    "sig3.title.set_text(names[2])\n",
-    "sig3.axes.xaxis.set_visible(False)\n",
-    "sig3.axes.yaxis.set_visible(False)\n",
-    "plt.show()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  }
- ],
- "metadata": {
-  "kernelspec": {
-   "display_name": "Python 3",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
-   },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.8.3"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 4
-}
diff --git a/Workshop3/checkpoint3.ipynb b/Workshop3/checkpoint3.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..2fb0cf5a7e6d7c59d4ab966e83264a18ac5d1e8a
--- /dev/null
+++ b/Workshop3/checkpoint3.ipynb
@@ -0,0 +1,643 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h1> The Challenge:</h1>\n",
+    "\n",
+    "Based off this dataset with school financial, enrollment, and achievement data, we are interested in what information is a useful indicator of student performance at the state level.\n",
+    "\n",
+    "This question is a bit too big for a checkpoint, however. Instead, we want you to look at smaller questions related to our overall goal. Here's the overview:\n",
+    "\n",
+    "1. Choose a specific test to focus on\n",
+    ">Math/Reading for 4/8 grade\n",
+    "* Pick or create features to use\n",
+    ">Will all the features be useful in predicting test score? Are some more important than others? Should you standardize, bin, or scale the data?\n",
+    "* Explore the data as it relates to that test\n",
+    ">Create 2 well-labeled visualizations (graphs), each with a caption describing the graph and what it tells us about the data\n",
+    "* Create training and testing data\n",
+    ">Do you want to train on all the data? Only data from the last 10 years? Only Michigan data?\n",
+    "* Train a ML model to predict outcome \n",
+    ">Pick if you want to do a regression or classification task. For both cases, defined _exactly_ what you want to predict, and pick any model in sklearn to use (see sklearn <a href=\"https://scikit-learn.org/stable/modules/linear_model.html\">regressors</a> and <a href=\"https://scikit-learn.org/stable/auto_examples/classification/plot_classifier_comparison.html\">classifiers</a>).\n",
+    "* Summarize your findings\n",
+    ">Write a 1 paragraph summary of what you did and make a recommendation about if and how student performance can be predicted\n",
+    "\n",
+    "** Include comments throughout your code! Every cleanup and preprocessing task should be documented.\n",
+    "\n",
+    "\n",
+    "Of course, if you're finding this assignment interesting (and we really hope you do!), you are welcome to do more than the requirements! For example, you may want to see if expenditure affects 4th graders more than 8th graders. Maybe you want to look into the extended version of this dataset and see how factors like sex and race are involved. You can include all your work in this notebook when you turn it in -- just always make sure you explain what you did and interpret your results. Good luck!"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt\n",
+    "import seaborn as sns\n",
+    "\n",
+    "# feel free to import other libraries! "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df = pd.read_csv('states_edu.csv') #dataset import"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Chosen test: **<hit `Enter` to edit>**"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2> Cleanup (optional)</h2>\n",
+    "\n",
+    "_Use this space to rename columns, deal with missing data, etc._"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 227,
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Index(['PRIMARY_KEY', 'STATE', 'YEAR', 'ENROLL', 'TOTAL_REVENUE',\n",
+       "       'FEDERAL_REVENUE', 'STATE_REVENUE', 'LOCAL_REVENUE',\n",
+       "       'TOTAL_EXPENDITURE', 'INSTRUCTION_EXPENDITURE',\n",
+       "       'SUPPORT_SERVICES_EXPENDITURE', 'OTHER_EXPENDITURE',\n",
+       "       'CAPITAL_OUTLAY_EXPENDITURE', 'GRADES_PK_G', 'GRADES_KG_G',\n",
+       "       'GRADES_4_G', 'GRADES_8_G', 'GRADES_12_G', 'GRADES_1_8_G',\n",
+       "       'GRADES_9_12_G', 'GRADES_ALL_G', '4MATH', '8MATH', '4READING',\n",
+       "       '8READING', 'PERCENT_EXPENDITURE', 'MONEY_PER_STUDENT',\n",
+       "       'PERCENT_PER_STUDENT', 'MONEY_PER_STUDENT_AVG',\n",
+       "       'PERCENT_PER_STUDENT_AVG'],\n",
+       "      dtype='object')"
+      ]
+     },
+     "execution_count": 227,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "#renaming columns to make them easily accessible\n",
+    "df.rename({'AVG_MATH_4_SCORE' : '4MATH', 'AVG_READING_4_SCORE' : '4READING', 'AVG_MATH_8_SCORE' : '8MATH', 'AVG_READING_8_SCORE' : '8READING'}, axis = 1, inplace = True)\n",
+    "#keep only documented values\n",
+    "df.dropna(subset=['4MATH', '8MATH', '4READING', '8READING', 'STATE'], inplace=True)\n",
+    "df.columns #for my reference later"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2> Feature Selection </h2>\n",
+    "\n",
+    "_Use this space to modify or create features_"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 240,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "6.004382533196235\n",
+      "1.7261459623412743\n",
+      "12.327728713667957\n",
+      "3.4526777193699165\n"
+     ]
+    }
+   ],
+   "source": [
+    "df['PERCENT_EXPENDITURE'] = 100*df['INSTRUCTION_EXPENDITURE']/df['TOTAL_EXPENDITURE'] # percent of total expenditure directed towards instruction\n",
+    "df['MONEY_PER_STUDENT'] = df['INSTRUCTION_EXPENDITURE']/df['GRADES_ALL_G'] #amount of instruction expenditure per student\n",
+    "df['PERCENT_PER_STUDENT'] = 100*df['PERCENT_EXPENDITURE']/df['GRADES_ALL_G'] #percent of instruction percentage per student\n",
+    "arr = df['STATE'].values[0:51] #list of states\n",
+    "avgdf = pd.DataFrame(columns=['STATE', 'MONEY_PER_STUDENT', 'PERCENT_PER_STUDENT', '8READING', '8MATH', '4READING', '4MATH', 'STUDENT_POP'], index = range (51)) #new df for averages\n",
+    "avgdf['STATE'] = arr #population of new df\n",
+    "for i in range (0, 51):\n",
+    "    avgdf['MONEY_PER_STUDENT'][i] = df.groupby('STATE')['MONEY_PER_STUDENT'].mean()[i]\n",
+    "    avgdf['PERCENT_PER_STUDENT'][i] = df.groupby('STATE')['PERCENT_PER_STUDENT'].mean()[i]\n",
+    "    avgdf['8READING'][i] = df.groupby('STATE')['8READING'].mean()[i]\n",
+    "    avgdf['8MATH'][i] = df.groupby('STATE')['8MATH'].mean()[i]\n",
+    "    avgdf['4READING'][i] = df.groupby('STATE')['4READING'].mean()[i]\n",
+    "    avgdf['4MATH'][i] = df.groupby('STATE')['4MATH'].mean()[i]\n",
+    "    avgdf['STUDENT_POP'][i] = df.groupby('STATE')['GRADES_ALL_G'].mean()[i]\n",
+    "meanmoney = avgdf['MONEY_PER_STUDENT'].mean()\n",
+    "stdmoney = avgdf['MONEY_PER_STUDENT'].std()\n",
+    "maxmoney = avgdf['MONEY_PER_STUDENT'].max()\n",
+    "minmoney = avgdf['MONEY_PER_STUDENT'].min()\n",
+    "print(meanmoney)\n",
+    "print(stdmoney)\n",
+    "print(maxmoney)\n",
+    "print(minmoney)\n",
+    "avgdf = avgdf.dropna()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Final feature list: **<LIST FEATURES HERE\\>**"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Feature selection justification: **<BRIEFLY DESCRIBE WHY YOU PICKED THESE FEATURES\\>**"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2> EDA </h2>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Visualization 1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 176,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0, 0.5, '4th grade reading score')"
+      ]
+     },
+     "execution_count": 176,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "#4 scatter plots for money\n",
+    "avgdf.plot.scatter(x='MONEY_PER_STUDENT',y='4MATH')\n",
+    "plt.xlabel('4th grade money per student ($)')\n",
+    "plt.ylabel('4th grade math score')\n",
+    "avgdf.plot.scatter(x='MONEY_PER_STUDENT',y='8MATH')\n",
+    "plt.xlabel('8th grade money per student ($)')\n",
+    "plt.ylabel('8th grade math score')\n",
+    "avgdf.plot.scatter(x='MONEY_PER_STUDENT',y='8READING')\n",
+    "plt.xlabel('8th grade money per student ($)')\n",
+    "plt.ylabel('8th grade reading score')\n",
+    "avgdf.plot.scatter(x='MONEY_PER_STUDENT',y='4READING')\n",
+    "plt.xlabel('4th grade money per student ($)')\n",
+    "plt.ylabel('4th grade reading score')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "**<CAPTION FOR VIZ 1>**"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Visualization 2"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 177,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0, 0.5, '4th grade math score')"
+      ]
+     },
+     "execution_count": 177,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 432x288 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "#4 scatter plots for percentage\n",
+    "avgdf.plot.scatter(x='PERCENT_PER_STUDENT',y='8READING')\n",
+    "plt.xlabel('8th grade percent of revenue per student (%)')\n",
+    "plt.ylabel('8th grade reading score')\n",
+    "avgdf.plot.scatter(x='PERCENT_PER_STUDENT',y='4READING')\n",
+    "plt.xlabel('4th grade percent of revenue per student (%)')\n",
+    "plt.ylabel('4th grade reading score')\n",
+    "avgdf.plot.scatter(x='PERCENT_PER_STUDENT',y='8MATH')\n",
+    "plt.xlabel('8th grade percent of revenue per student (%)')\n",
+    "plt.ylabel('8th grade math score')\n",
+    "avgdf.plot.scatter(x='PERCENT_PER_STUDENT',y='4MATH')\n",
+    "plt.xlabel('4th grade percent of revenue per student (%)')\n",
+    "plt.ylabel('4th grade math score')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "**<CAPTION FOR VIZ 2>**"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2> Data Creation </h2>\n",
+    "\n",
+    "_Use this space to create train/test data_"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 197,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sklearn.model_selection import train_test_split"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 295,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "X\n",
+    "y = avgdf.loc[X.index]['8MATH']\n",
+    "#based on money per student, can the score be predicted for 8th grade math tests?"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 318,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "X1_train, X1_test, y1_train, y1_test = train_test_split(\n",
+    "     X, y, test_size=.75, random_state=0) #25% of data allocated to prediction\n",
+    "\n",
+    "X2_train, X2_test, y2_train, y2_test = train_test_split(\n",
+    "     X, y, test_size=.65, random_state=0) #35% of data allocated to prediction"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2> Prediction </h2>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Chosen ML task: **<REGRESSION/CLASSIFICATION>**"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 319,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# import your sklearn class here\n",
+    "from sklearn.linear_model import LinearRegression"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 320,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# create your model here\n",
+    "model = LinearRegression()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 321,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "7.884823712911691"
+      ]
+     },
+     "execution_count": 321,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "model.fit(X1_train, y1_train)\n",
+    "model.score(X1_test, y1_test)\n",
+    "np.mean(model.predict(X1_test)-y1_test)\n",
+    "np.mean(np.abs(model.predict(X1_test)-y1_test))\n",
+    "np.mean((model.predict(X1_test)-y1_test)**2)**0.5"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 322,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "y1_pred = model.predict(X1_test)\n",
+    "y2_pred = model.predict(X2_test)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 323,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# for classification:\n",
+    "#from sklearn.metrics import plot_confusion_matrix\n",
+    "\n",
+    "#plot_confusion_matrix(model, X_test, y_test,\n",
+    "                        # cmap=plt.cm.Blues)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 324,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "Text(0.5, 1.0, 'Prediction with 35% trained')"
+      ]
+     },
+     "execution_count": 324,
+     "metadata": {},
+     "output_type": "execute_result"
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 864x432 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtQAAAGECAYAAAAWS7eaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzdeXxV1bn/8c+TEBLDEGQeQkioSBgyQEAZAoIooiKChFqlDj/bYh1KW69YNFXRNq2t3lq9rfVyLxZ7mzoQQMDaShERAZUSQJmCEE1CAJkJiWFKsn5/nJOYQBISkpPx+3698jpnr7323s85J4QnK89ey5xziIiIiIjIxfGr7wBERERERBozJdQiIiIiIjWghFpEREREpAaUUIuIiIiI1IASahERERGRGlBCLSIiIiJSA0qoRUQAM5tvZr/0Ph9lZjsv8jwvm9njtRtd7V3fzOaY2V/rMqb6UpPPsQrnLvl+ERFRQi0ijYaZZZjZSTPLM7MDZvZnM2td29dxzn3onOtbhXjuNrM15xz7Q+fcL2o7pqoqfX0zG2Nm2TU5n5m9b2aHzOyEmX1qZjeX2jfGzIq8n0fx112l9s8ys8NmttXMBpZqH2lmb13gujVO/Kv6OYqI1JQSahFpbG5yzrUGBgNDgZ+f28HMWtR5VE3Xj4Fuzrm2wAzgr2bWrdT+fc651qW+XgXw9vke0Bt4GXjG294C+E/gJzUJyjz0f5iINAj6YSQijZJzbi/wD2AggJk5M3vAzHYBu7xtE81ss5kdN7N1ZhZdfLyZDTKzjWaWa2ZvAEGl9pUZ2TWznma2yDtSe8TM/mBm/fAkisO9I7PHvX3LlAKY2Q/MbLeZHTWzpWbWvdQ+Z2Y/NLNdZnbMzP5oZnbuazWzIO/IfEfv9s/NrMDM2nq3f2lmvy99fTNr5X1/upcaPS6+dksz+4v3tW8zsyGVvM+fOecKijeBAKDnhT4fIAzY5Jw7AazAk1iDJ5Fe6pzLqOhAM5sAPAbc6o37U2/7KjNLMrO1QD7Q28z+n5nt8L6WL8zs3lLnOfdzzDCzh83sMzPLMbM3zKz0535R3y8iIkqoRaRRMrOewA3AplLNk4Ergf5mNhh4BbgX6AD8N7DUzALNrCXwFvB/QHtgATC1guv4A28DmUA40AN43Tm3A/gh8JF3ZLZdOcdeDfwa+DbQzXuO18/pNhHPSHuMt991557HOXcK+DdwlbdptPdcI0ttf3DOMV8D11N2BHmfd/ckbxztgKXAH8p77aVex9tmdgr4BFgFbCi1u7O3/OZLM3vem8gD7AaizKwdcA2wzfuZfQd4rrLrOef+CfwKeMMbd0yp3XfgGSlv430PDuJ5D9sC/w943vvZV+TbwAQgAogG7va+xlr5fhGR5kkJtYg0Nm95R4PX4Ekif1Vq36+dc0edcyeBHwD/7Zz7xDlX6C1FOA0M834FAL93zp11zqXgSVjLcwXQHZjlnPvaOXfKObemgr7nmg684pzb6Jw7DTyKZ0Q7vFSfZ5xzx51zWcD7QGwF5/oAuMpbMhENvOjdDsKTkH9YxZgA1jjn3nHOFeJJEmMq6+ycm4gngb0BeNc5V+TdleaNtxtwNRAH/M57zBEgCVgJ3Ag8DLwA/AyYYmYfmNkSMwutRtwA851z25xzBd7P7u/OuXTn8QGwHBhVyfEvOuf2OeeOAsv45v2ure8XEWmGlFCLSGMz2TnXzjnXyzl3vzd5Lran1PNewH94/3x/3JuE98STHHcH9jrnXKn+mRVcryeQWarsoTq6lz6vcy4POIJnlLvYV6We5wMV3WT5ATAGT+34FuBfeEashwG7nXOHqxHXudcMulDduTeR/AdwnZlN8rZ95Zzb7pwrcs59CTwCJJQ65jXn3GDn3PV4SnNO4/mLwnPATXhGeisdrS5H6c8YM7vezD72ltQcx5P0d6zk+Ire79r6fhGRZkgJtYg0JaUTnj1Akjf5Lv4Kds69BuwHepxTrxxWwTn3AGEVJJyunLbS9uFJ1ADwlkN0APZe6IWUYx3QF5gCfOCc244n5hs5p9yjGvFdjBbAtyq5Xnk14Jfg+UvCfwB9gD3e2up/4xltr+hclbabWSCwEE9S3sVbdvNOeTFUQW19v4hIM6SEWkSaqv8BfmhmV5pHKzO70czaAB8BBcBMM2thZrfgKe0oz3o8CdUz3nMEmVlx7fIBINRbY1uevwH/z8xivcnfr4BPKrshryLOuXwgFXiAbxLodXhqfitKqA8AHcwspLrXAzCzSO8I8CVmFmBm36VUvbb3pr8w7/vbE89MHkvKOdXP8ZRq7AOygL5m1gUYC3xRSezhVvlMHi2BQOAQUGBm1wPjL+KlQu19v4hIM6SEWkSaJOfcBjx1sX8AjuG5Se5u774zwC3e7WPArcCiCs5TiKc84TI8yWC2tz946oO3AV+Z2XklF86594DH8Yyi7sczsvudGrysD/DU8q4vtd0GWF1B7GnAa8AX3jKG7uX1q4QBc/Dc+HcIzxR6tzrnNnr3D8aTbH6NJ7nfCswscwKzvniS3P/yxrQfT+K9zdv30QquvcD7eMTMNpbXwTmX6z3Hm3g+x9vx3GRZbbX1/SIizZOVLQkTEREREZHq0Ai1iIiIiEgNKKEWEREREakBJdQiIiIiIjWghFpEREREpAaUUIuIiIiI1EClK2M1dB07dnTh4eH1HYaIiIiINHGpqamHnXOdytvXqBPq8PBwNmzYUN9hiIiIiEgTZ2aZFe1TyYeIiIiISA0ooRYRERERqQEl1CIiIiIiNdCoa6jLc/bsWbKzszl16lR9hyJ1JCgoiNDQUAICAuo7FBEREWmGmlxCnZ2dTZs2bQgPD8fM6jsc8THnHEeOHCE7O5uIiIj6DkdERESaoSZX8nHq1Ck6dOigZLqZMDM6dOigv0iIiIhIvWlyCTWgZLqZ0ectIiIi9alJJtRNQUZGBn/729+qfdz8+fN58MEHAZgzZw7PPfdctY5v3bp1ta9ZXeHh4Rw+fLjGfUREREQaAiXUDVRlCXVBQUEdRyMiIiIiFVFC7QN/+ctfiI6OJiYmhjvuuAOAzMxMxo0bR3R0NOPGjSMrKwuAu+++m5kzZzJixAh69+5NSkoKALNnz+bDDz8kNjaW559/nvnz5zNt2jRuuukmxo8fz9GjR5k8eTLR0dEMGzaMzz77rNKY0tPTmTBhAnFxcYwaNYq0tDQAvvzyS4YPH87QoUN5/PHHyz02IyODyMhIvv/97zNw4ECmT5/OihUrGDlyJH369GH9+vUAFcZ05MgRxo8fz6BBg7j33ntxzpWc+69//StXXHEFsbGx3HvvvRQWFtbgnRcRERGpe01ulo/Snlq2je37TtTqOft3b8uTNw2ocP+2bdtISkpi7dq1dOzYkaNHjwLw4IMPcuedd3LXXXfxyiuvMHPmTN566y0A9u/fz5o1a0hLS2PSpEkkJCTwzDPP8Nxzz/H2228DnlKOjz76iM8++4z27dvzox/9iEGDBvHWW2+xcuVK7rzzTjZv3lxhXDNmzODll1+mT58+fPLJJ9x///2sXLmSH//4x9x3333ceeed/PGPf6zw+N27d7NgwQLmzp3L0KFD+dvf/saaNWtYunQpv/rVr3jrrbd48skny43pqaeeIj4+nieeeIK///3vzJ07F4AdO3bwxhtvsHbtWgICArj//vtJTk7mzjvvrPbnIiIiIk1XcjIkJkJWFoSFQVISTJ9e31F9o0kn1PVh5cqVJCQk0LFjRwDat28PwEcffcSiRYsAuOOOO3jkkUdKjpk8eTJ+fn7079+fAwcOVHjua6+9tuR8a9asYeHChQBcffXVHDlyhJycnHKPy8vLY926dUybNq2k7fTp0wCsXbu25Dx33HEHP/vZz8o9R0REBFFRUQAMGDCAcePGYWZERUWRkZFRaUyrV68uee033ngjl156KQDvvfceqampDB06FICTJ0/SuXPnCl+/iIiIND/JyTBjBuTne7YzMz3b0HCS6iadUFc2kuwrzrkqzTpRuk9gYGCZ4yvSqlWrSvtVdN2ioiLatWtX4Qh2VeItHaOfn1/Jtp+fX0lNd2UxlXcN5xx33XUXv/71ry94fREREWmeEhO/SaaL5ed72htKQq0a6lo2btw43nzzTY4cOQJQUvIxYsQIXn/9dQCSk5OJj4+v9Dxt2rQhNze3wv2jR48mOTkZgFWrVtGxY0fatm1bbt+2bdsSERHBggULAE8i++mnnwIwcuTIMnHVREUxlW7/xz/+wbFjxwDPe5WSksLBgwcBz3uVmZlZoxhERESkafHedlbl9vqghLqWDRgwgMTERK666ipiYmJ46KGHAHjxxRf585//THR0NP/3f//HCy+8UOl5oqOjadGiBTExMTz//PPn7Z8zZw4bNmwgOjqa2bNn8+qrr1Z6vuTkZObNm0dMTAwDBgxgyZIlALzwwgv88Y9/ZOjQoRWWjFRVRTE9+eSTrF69msGDB7N8+XLCwsIA6N+/P7/85S8ZP3480dHRXHvttezfv79GMYiIiEjT4k0bqtxeH6yyEoOGbsiQIW7Dhg1l2nbs2EG/fv3qKSKpL/rcRUREmqZza6gBgoNh7ty6Lfkws1Tn3JDy9mmEWkREREQarOnTPclzr15g5nms62T6Qpr0TYkiIiIi0vhNn96wEuhzaYRaRERERKQGlFCLiIiIiNSAEmoRERERkRpQQi0iIiIiUgNKqGvZ8ePHeemll3x2/lWrVrFu3bpqH7dhwwZmzpx5wX4jRoy4mLBEREREmi0l1LWssoS6sLCwxuevLKEuXgK8PEOGDOHFF1+84PkvJlkXERERac6UUCcnQ3g4+Pl5Hmu4/Pbs2bNJT08nNjaWWbNmsWrVKsaOHcvtt99OVFQUGRkZDBw4sKT/c889x5w5cwBIT09nwoQJxMXFMWrUKNLS0sqcOyMjg5dffpnnn3+e2NhYPvzwQ+6++24eeughxo4dy89+9jPWr1/PiBEjGDRoECNGjGDnzp2AJxGfOHEi4FnR8J577mHMmDH07t27TKLdunXrkv5jxowhISGByMhIpk+fTvEiQO+88w6RkZHEx8czc+bMkvOKiIiINEfNex7qc5feycz0bMNFT3b4zDPPsHXrVjZv3gx4EtP169ezdetWIiIiyMjIqPDYGTNm8PLLL9OnTx8++eQT7r//flauXFmyPzw8nB/+8Ie0bt2ahx9+GIB58+bx+eefs2LFCvz9/Tlx4gSrV6+mRYsWrFixgscee4yFCxeed620tDTef/99cnNz6du3L/fddx8BAQFl+mzatIlt27bRvXt3Ro4cydq1axkyZAj33nsvq1evJiIigttuu+2i3icRERGRpqJ5J9SJiWXXsQTPdmJirc4efsUVVxAREVFpn7y8PNatW8e0adNK2k6fPl2l80+bNg1/f38AcnJyuOuuu9i1axdmxtmzZ8s95sYbbyQwMJDAwEA6d+7MgQMHCA0NPS/u4rbY2FgyMjJo3bo1vXv3Lnk9t912G3Pnzq1SnCIiIiJNUfNOqLOyqtd+kVq1alXyvEWLFhQVFZVsnzp1CoCioiLatWtXMrJ9sed//PHHGTt2LIsXLyYjI4MxY8aUe0xgYGDJc39//3Lrr8vrU1z2ISIiIiIezbuGOiyseu1V0KZNG3Jzcyvc36VLFw4ePMiRI0c4ffo0b7/9NgBt27YlIiKCBQsWAOCc49NPP632+XNycujRowcA8+fPv+jXUZHIyEi++OKLktKVN954o9avISIiItKYNO+EOikJgoPLtgUHe9ovUocOHRg5ciQDBw5k1qxZ5+0PCAjgiSee4Morr2TixIlERkaW7EtOTmbevHnExMQwYMAAlixZct7xN910E4sXLy65KfFcjzzyCI8++igjR46slVlFznXJJZfw0ksvMWHCBOLj4+nSpQshISG1fh0RERGRxsIa85/whwwZ4jZs2FCmbceOHfTr16/qJ0lO9tRMZ2V5RqaTkmq1fropysvLo3Xr1jjneOCBB+jTpw8//elP6zWman/uIiIiItVgZqnOuSHl7WveI9TgSZ4zMqCoyPOoZPqC/ud//ofY2FgGDBhATk4O9957b32HJCIiIlJvmvdNiXJRfvrTn9b7iLSIiIhIQ+GzEWozCzKz9Wb2qZltM7OnvO3tzexfZrbL+3hpqWMeNbPdZrbTzK7zVWwiIiIiIrXFlyUfp4GrnXMxQCwwwcyGAbOB95xzfYD3vNuYWX/gO8AAYALwkpn5+zA+EREREZEa81lC7TzyvJsB3i8H3Ay86m1/FZjsfX4z8Lpz7rRz7ktgN3CFr+ITEREREakNPr0p0cz8zWwzcBD4l3PuE6CLc24/gPexs7d7D2BPqcOzvW3nnnOGmW0wsw2HDh3yZfgiIiIiIhfk04TaOVfonIsFQoErzGxgJd2tvFOUc865zrkhzrkhnTp1qq1QG6xVq1YxceJEAJYuXcozzzxTYd/jx4/z0ksvVfsac+bM4bnnnjuv/a233mL79u3VPt+F4gTYt28fCQkJ1T63iIiISENTJ9PmOeeOA6vw1EYfMLNuAN7Hg95u2UDPUoeFAvvqIr76cDGLrkyaNInZs2dXuP9iE+qKVJZQl7dUebELxQnQvXt3UlJSahSfiIiISEPgy1k+OplZO+/zS4BrgDRgKXCXt9tdQPFygEuB75hZoJlFAH2A9b6Kr1jylmTCfx+O31N+hP8+nOQtyTU6X0ZGBpGRkdx1111ER0eTkJBAfn4+AOHh4Tz99NPEx8ezYMECli9fzvDhwxk8eDDTpk0jL89Tcv7Pf/6TyMhI4uPjWbRoUcm558+fz4MPPgjAgQMHmDJlCjExMcTExLBu3Tpmz55Neno6sbGxJas0PvvsswwdOpTo6GiefPLJknMlJSXRt29frrnmGnbu3Hne61i3bh1Lly5l1qxZxMbGkp6ezpgxY3jssce46qqreOGFF1i2bBlXXnklgwYN4pprruHAgQPnxXn33Xczc+ZMRowYQe/evUuS6IyMDAYOHFjS/5ZbbmHChAn06dOHRx55pCSOefPmcfnllzNmzBh+8IMflJxXREREpKHw5TzU3YBXvTN1+AFvOufeNrOPgDfN7HtAFjANwDm3zczeBLYDBcADzrnaXzu7lOQtycxYNoP8s56ENzMnkxnLZgAwPeriF3jZuXMn8+bNY+TIkdxzzz289NJLPPzwwwAEBQWxZs0aDh8+zC233MKKFSto1aoVv/nNb/jd737HI488wg9+8ANWrlzJZZddxq233lruNWbOnMlVV13F4sWLKSwsJC8vj2eeeYatW7eyefNmAJYvX86uXbtYv349zjkmTZrE6tWradWqFa+//jqbNm2ioKCAwYMHExcXV+b8I0aMYNKkSUycOLFMacbx48f54IMPADh27Bgff/wxZsb//u//8tvf/pb//M//PC/W/fv3s2bNGtLS0pg0aVK5pR6bN29m06ZNBAYG0rdvX370ox/h7+/PL37xCzZu3EibNm24+uqriYmJubgPRURERMRHfJZQO+c+AwaV034EGFfBMUlAkq9iOlfie4klyXSx/LP5JL6XWKOEumfPnowcORKA7373u7z44oslCXVxgvzxxx+zffv2kn5nzpxh+PDhpKWlERERQZ8+fUqOnzt37nnXWLlyJX/5y18A8Pf3JyQkhGPHjpXps3z5cpYvX86gQZ6PIS8vj127dpGbm8uUKVMIDg4GPCUaVVU6wc/OzubWW29l//79nDlzhoiIiHKPmTx5Mn5+fvTv379kFPtc48aNIyQkBID+/fuTmZnJ4cOHueqqq2jfvj0A06ZN4/PPP69yrCIiIiJ1oVkvPZ6Vk1Wt9qoyswq3W7VqBYBzjmuvvZbNmzezefNmtm/fzrx588o9/mI553j00UdLrrF7926+973v1egaxfED/OhHP+LBBx9ky5Yt/Pd//zenTp0q95jAwMAyMV2oj7+/PwUFBRX2FREREWlImnVCHRYSVq32qsrKyuKjjz4C4LXXXiM+Pv68PsOGDWPt2rXs3r0bgPz8fD7//HMiIyP58ssvSU9PLzm+POPGjeNPf/oT4LnB8cSJE7Rp04bc3NySPtdddx2vvPJKSW323r17OXjwIKNHj2bx4sWcPHmS3Nxcli1bVu41zj3fuXJycujRwzOz4auvvlphv4t1xRVX8MEHH3Ds2DEKCgpYuHBhrV9DREREpKaadUKdNC6J4IDgMm3BAcEkjatZ1Um/fv149dVXiY6O5ujRo9x3333n9enUqRPz58/ntttuIzo6mmHDhpGWlkZQUBBz587lxhtvJD4+nl69epV7jRdeeIH333+fqKgo4uLi2LZtGx06dGDkyJEMHDiQWbNmMX78eG6//XaGDx9OVFQUCQkJ5ObmMnjwYG699VZiY2OZOnUqo0aNKvca3/nOd3j22WcZNGhQSYJf2pw5c5g2bRqjRo2iY8eONXrPytOjRw8ee+wxrrzySq655hr69+9fUhYiIiIi0lBYY/6z+pAhQ9yGDRvKtO3YsYN+/fpV+RzJW5JJfC+RrJwswkLCSBqXVKP66YyMDCZOnMjWrVsv+hzyjby8PFq3bk1BQQFTpkzhnnvuYcqUKef1q+7nLiIiIlIdZpbqnBtS3j5fzvLRKEyPml6jBFp8a86cOaxYsYJTp04xfvx4Jk+efOGDREREROpQs0+oa1t4eLhGp2tReSs4ioiIiDQkzbqGWkRERESkppRQi4iIiIjUgBJqEREREZEaUEItIiIiIlIDSqgbuFWrVjFx4kQAli5dyjPPPFNh3+PHj/PSSy9V+xpz5swp9+a/t956i+3bt1f7fACbN2/mnXfeKdm+UOwiIiIijZUS6npSWFhY7WMmTZrE7NmzK9x/sQl1RWozob5Q7CIiIiKNVbNPqJOTITwc/Pw8j8nJNTtfRkYGkZGR3HXXXURHR5OQkEB+fj7gmVLv6aefJj4+ngULFrB8+XKGDx/O4MGDmTZtWskS4f/85z+JjIwkPj6eRYsWlZx7/vz5PPjggwAcOHCAKVOmEBMTQ0xMDOvWrWP27Nmkp6cTGxvLrFmzAHj22WcZOnQo0dHRPPnkkyXnSkpKom/fvlxzzTXs3LnzvNexbt06li5dyqxZs4iNjSU9PZ309HQmTJhAXFwco0aNIi0tDYAFCxYwcOBAYmJiGD16NGfOnOGJJ57gjTfeIDY2ljfeeKNM7HfffTczZ85kxIgR9O7dm5SUFACKioq4//77GTBgABMnTuSGG24o2SciIiLSUDXreaiTk2HGDPDmu2RmerYBptdgrZedO3cyb948Ro4cyT333MNLL73Eww8/DEBQUBBr1qzh8OHD3HLLLaxYsYJWrVrxm9/8ht/97nc88sgj/OAHP2DlypVcdtll3HrrreVeY+bMmVx11VUsXryYwsJC8vLyeOaZZ9i6dSubN28GYPny5ezatYv169fjnGPSpEmsXr2aVq1a8frrr7Np0yYKCgoYPHgwcXFxZc4/YsQIJk2axMSJE0lISABg3LhxvPzyy/Tp04dPPvmE+++/n5UrV/L000/z7rvv0qNHD44fP07Lli15+umn2bBhA3/4wx8Azy8Dpe3fv581a9aQlpbGpEmTSEhIYNGiRWRkZLBlyxYOHjxIv379uOeeey7+gxARERGpA806oU5M/CaZLpaf72mvSULds2dPRo4cCcB3v/tdXnzxxZKEujhB/vjjj9m+fXtJvzNnzjB8+HDS0tKIiIigT58+JcfPnTv3vGusXLmSv/zlLwD4+/sTEhLCsWPHyvRZvnw5y5cvZ9CgQYBnGe9du3aRm5vLlClTCA4OBjzlGBeSl5fHunXrmDZtWknb6dOnARg5ciR333033/72t7nllluq9B5NnjwZPz8/+vfvz4EDBwBYs2YN06ZNw8/Pj65duzJ27NgqnUtERESkPjXrhDorq3rtVWVmFW63atUKAOcc1157La+99lqZvps3bz7v+IvlnOPRRx/l3nvvLdP++9//vtrXKCoqol27diWj36W9/PLLfPLJJ/z9738nNja23D7nCgwMLBNn6UcRERGRxqRZ11CHhVWvvaqysrL46KOPAHjttdeIj48/r8+wYcNYu3Ytu3fvBiA/P5/PP/+cyMhIvvzyS9LT00uOL8+4ceP405/+BHhucDxx4gRt2rQhNze3pM91113HK6+8UlKbvXfvXg4ePMjo0aNZvHgxJ0+eJDc3l2XLlpV7jdLna9u2LRERESxYsADwJL+ffvopAOnp6Vx55ZU8/fTTdOzYkT179pwXS1XEx8ezcOFCioqKOHDgAKtWrarW8SIiIiL1oVkn1ElJ4K16KBEc7GmviX79+vHqq68SHR3N0aNHue+++87r06lTJ+bPn89tt91GdHQ0w4YNIy0tjaCgIObOncuNN95IfHw8vXr1KvcaL7zwAu+//z5RUVHExcWxbds2OnTowMiRIxk4cCCzZs1i/Pjx3H777QwfPpyoqCgSEhLIzc1l8ODB3HrrrcTGxjJ16lRGjRpV7jW+853v8OyzzzJo0CDS09NJTk5m3rx5xMTEMGDAAJYsWQLArFmziIqKYuDAgYwePZqYmBjGjh3L9u3bS25KrIqpU6cSGhrKwIEDuffee7nyyisJCQmp4rsuIiIiUj+sMf+ZfciQIW7Dhg1l2nbs2EG/fv2qfI7kZE/NdFaWZ2Q6Kalm9dMZGRlMnDiRrVu3XvxJmrG8vDxat27NkSNHuOKKK1i7di1du3a94HHV/dxFREREqsPMUp1zQ8rb16xrqMGTPNckgZbaNXHiRI4fP86ZM2d4/PHHq5RMi4iIiNSnZp9Q17bw8HCNTteA6qZFRESksWnWNdQiIiIiIjXVJBPqxlwXLtWnz1tERETqU5NLqIOCgjhy5IiSrGbCOceRI0cICgqq71BERESkmWpyNdShoaFkZ2dz6NCh+g5F6khQUBChoaH1HYaI1LXanqZJROQiNbmEOiAggIiIiPoOQ0REfCk5GWbMgPx8z3ZmpmcblFSLSJ1rciUfIiLSDCQmfpNMF8vP97SLiNQxJdQiItL4ZGVVr11ExIeUUIuISOMTFla9dhERH1JCLSIijU9SEgQHl20LDuh35KIAACAASURBVPa0i4jUMSXUIiLS+EyfDnPnQq9eYOZ5nDtXNySKSL1ocrN8iIhIMzF9uhJoEWkQfDZCbWY9zex9M9thZtvM7Mfe9hgz+8jMtpjZMjNr620PN7OTZrbZ+/Wyr2ITEREREaktvhyhLgD+wzm30czaAKlm9i/gf4GHnXMfmNk9wCzgce8x6c65WB/GJCIiIiJSq3w2Qu2c2++c2+h9ngvsAHoAfYHV3m7/Aqb6KgYREREREV+rk5sSzSwcGAR8AmwFJnl3TQN6luoaYWabzOwDMxtVwblmmNkGM9ug5cVFRETOl5wM4eHg5+d5TE6u74hEmjafJ9Rm1hpYCPzEOXcCuAd4wMxSgTbAGW/X/UCYc24Q8BDwt+L66tKcc3Odc0Occ0M6derk6/BFREQaleJV2TMzwblvVmVXUi3iOz5NqM0sAE8yneycWwTgnEtzzo13zsUBrwHp3vbTzrkj3uep3vbLfRmfiIhIU6NV2UXqni9n+TBgHrDDOfe7Uu2dvY9+wM+Bl73bnczM3/u8N9AH+MJX8YmIiDRFWpVdpO75coR6JHAHcHWpqfBuAG4zs8+BNGAf8Gdv/9HAZ2b2KZAC/NA5d9SH8YmIiDQ5WpVdpO75bNo859wawCrY/UI5/RfiKQ8RERGRi5SU5KmZLl32oVXZRXxLS4+LiIg0IVqVXaTuaelxERGRJkarsovULY1Qi4iIiIjUgBJqEREREZEaUEItIiIiIlIDSqhFRERERGpACbWIiIiISA0ooRYRERERqQEl1CIiIiIiNaCEWkRqLjkZwsPBz8/zmJxc3xGJiIjUGS3sIiI1k5xcdp3jzEzPNmhlCRERaRY0Qi0iNZOY+E0yXSw/39MuIiLSDCihFpGaycqqXruIiEgTo4RaRGomLKx67SIiIk2MEmoRqZmkJAgOLtsWHOxpFxERaQaUUItIzUyfDnPnQq9eYOZ5nDtXNyRKs1BY5Fj9+SFSM4/WdygiUo80y4eI1Nz06UqgpVlJP5THwtRsFm3cy1cnTnH9wK7E9Wpf32GJSD1RQi0iIlIFOSfP8vZn+0hJzWZT1nH8/YyrLu/EEzf1Z1y/zvUdnojUIyXUIiIiFSgscny46xApqdks336AMwVF9O3ShsduiGTyoB50bhNU3yGKSAOghFpEROQcuw7kkrIxm7c27eXAidO0Cw7g9ivCSIgLZUD3tphZfYcoIg2IEmoRERHgeP4Zln26j5SNe/l0j6ekY2zfTjw1KZSxkZ0JbOFf3yGKSAOlhFpERJqtgsIiVu86xMLUvfxr+wHOFBYR2bUNP7+xHzfH9qBTm8D6DlFEGgEl1CIi0iwkb0km8b1EsnKyCAu+kvjOD5O2J4TDeadp36ol04cVl3SE1HeoItLIKKEWEZFGqUyCHBJG0rgkpkeVP31j8pZk7l3yH/idHkqXgp/AyT58eKSA/qG5/GrKcMb07UzLFlqaQUQujhJqERFpdJK3JDNj2Qzyz+YDkJmTyYxlMwDKJNVnC4tYtfMQiSlf0uH0yxgBnLbdHA2Yy9f+q7CCSxk/IKM+XoKINCFKqEVEpNFJfC+xJJkuln82n8T3EpkeNZ0d+0+QkuqZpePI12coJJyvW7xNnv97nPXLKDkmKye3jiMXkaZICbWIiDQ6WTlZ57X5ubYcOxLDDS98yPb9JwjwN67p14Wpg0P5/rvDOXbiy/OOCQsJq4twRaSJU0ItIiKNTlhIGJk5meBacEnREFoXjOOSoqEYLWjhbzw1aQCTYrpzaauWACQV/qJMiQhAcEAwSeOS6usliEgTooRaREQanQcGPcOz760m8Ew8/oRQyDFOtnyH2deO4aFRN57Xv7iuuqo3MYqIVIc55+o7hos2ZMgQt2HDhvoOQ0RE6sCh3NMs2byXlNRs0r7Kxd/PUdhyEweLltL50iMkXfMLJcgi4jNmluqcG1LePo1Qi4hIg3W6oJD30w6SkprN+zsPUVjkiOnZjl/cPICbYrrTLngi8Hh9hykizZwSahERaVCcc2zde4KU1D0s+XQfx/PP0rlNIN8fFUHC4FD6dGlT3yF+IzkZEhMhKwvCwiApCaZrlFykufFZQm1mPYG/AF2BImCuc+4FM4sBXgZaAxnAdOfcCe8xjwLfAwqBmc65d30Vn4iINCwHT5ziLW9Jx+cH8mjZwo/rBnRl6uAexF/WkRb+DWzhleRkmDED8r03OmZmerZBSbVIM+OzGmoz6wZ0c85tNLM2QCowGXgVeNg594GZ3QNEOOceN7P+wGvAFUB3YAVwuXOusKJrqIZaRKRxO3W2kPd2HCQldQ+rdx2msMgxOKwdU+NCmRjdnZBLAuo7xIqFh3uS6HP16gUZGXUdjYj4WL3UUDvn9gP7vc9zzWwH0APoC6z2dvsX8C6eAribgdedc6eBL81sN57k+iNfxSgiInXPOcen2TksTM1m6af7yDl5lq5tg5gxujcJcaF8q1Pr+g6xarLOnwu70nYRabLqpIbazMKBQcAnwFZgErAEmAb09HbrAXxc6rBsb9u555oBzAAIC9OE/CIijcVXOadYvGkvCzdms/tgHoEt/JgwsCtTB4cy8rKO+PtZfYdYPWFh5Y9Q6/8mkWbH5wm1mbUGFgI/cc6d8JZ5vGhmTwBLgTPFXcs5/Lx6FOfcXGAueEo+fBO1iIjUhlNnC/nX9gOkpGbz4a5DFDkY0utSfn1LFDdGd6NtUAMu6biQpKSyNdQAwcGedhFpVnyaUJtZAJ5kOtk5twjAOZcGjPfuvxwonoE/m29GqwFCgX2+jE9ERGqfc45Ne46XlHTkniqge0gQ9435FglxPYno2Kq+Q6wdxTceapYPkWbPlzclGp4bEI86535Sqr2zc+6gmfkB84FVzrlXzGwA8De+uSnxPaCPbkoUEWkc9uecZNFGT0nHF4e+JijAjxsGdmNqXCjDe3fAr7GVdIiIlFJfC7uMBO4AtpjZZm/bY0AfM3vAu70I+DOAc26bmb0JbAcKgAcqS6ZFRKT+nTxTyPLtX5GSms2a3YdxDq6IaM8PR3+L66O60qYxl3SIiFSRlh4XEZFqcc6RmnmMlNRs3v5sP3mnC+jR7hKmDu7B1LhQenVoIiUdIiKlaOlxERGpsb3HT7J4YzYpqdlkHMnnkgB/bojqxtS4HgyLUEmHiDRfSqhFRKRC+WcKeHebp6RjXfoRnINhvdvzwNjLuD6qG60D9d+IiIh+EoqISBnOOf6dcYyU1D28s+Ur8k4X0LP9Jfx4XB+mDg6lZ/vg+g5RRKRBUUItIiIA7DmaXzJLR9bRfFq19OfG6G5MHRzK0PD2KukQEamAEmoRkWbs69MF/GPrV6Sk7uHjL44CMOJbHfjJNX2YMLArwS3134SIyIXoJ6WISDNTVORYn3GUlNRs3tmyn/wzhfTqEMxD117OLYN7EHqpSjpERKrjggm1mXUBfgV0d85db2b9geHOuXk+j05ERGpN1pF8Fm7MZuHGbLKPnaR1YAsmxXQnIS6UuF6X4lmPS0REqqsqI9Tz8Sy+kujd/hx4A1BCLSLSwOWdLuCdLftJSc1m/ZdHMYP4yzoy67q+jO/flUta+td3iCIijV5VEuqOzrk3zexRAOdcgZlpBUMRkQaqqMjx8RdHSEnN5h9bv+Lk2UJ6d2zFrOv6MmVQD7q3u6S+QxQRaVKqklB/bWYdAAdgZsOAHJ9GJSIi1ZZx+GsWbsxm0ca97D1+kjaBLZg8qAcJcaEMDmunkg4RER+pSkL9ELAU+JaZrQU6AQk+jUpERKok99RZ/v7ZfhZuzObfGcfwM4jv04mfXR/J+P5dCApQSYeIiK9VmlCbmT9wlferL2DATufc2TqITUREylFY5FiXfpiFqdn8c9tXnDpbxLc6teKRCX25ZVAoXUOC6jtEEZFmpdKE2jlXaGY3O+eeB7bVUUwiIo1e8pZkEt9LJCsni7CQMJLGJTE9anrZPsmQmAhZWRAWBklJMH16BScEvjiUV1LSsT/nFG2DWpAQF0pCXE9iQkNU0iEiUk+qUvKx1sz+gGdmj6+LG51zG30WlYhII5a8JZkZy2aQfzYfgMycTGYsmwFQklQnJ8OMGZDv6UJmpmcbyibVOSc9JR0pqXvYmHUcP4OrLu/Ez2/sz7h+nVXSISLSAJhzrvIOZu+X0+ycc1f7JqSqGzJkiNuwYUN9hyEiUkb478PJzMk8r71XSC8yfpLh6RPuSaLP69ML0r9wrNl9mJTUbN7d9hVnCoro07k1CXGhTBnUg85tVdIhIlLXzCzVOTekvH0XHKF2zo2t/ZBERJqurJysC7ZnldOlRYdccsL3MuKZbA6cOE3IJQHcOqQn04aEEtVDJR0iIg1VVVZKDAGeBEZ7mz4AnnbOaeo8EZFyhIWElTtCHRYS9s3zMM8ItV/gWYL77aN1VDaB3Y9DkTGweyfm3BTK1f06E9hCJR0iIg2dXxX6vALkAt/2fp3As3KiiIiUI2lcEsEBwWXaggOCSRqXBEBBYRF3/uwgXW7ZSOiDK+hw3VYsoJC8Nf2Y3f9q5t09lOujuimZFhFpJKpyU+K3nHNTS20/ZWabfRWQiEhjV3zj4bmzfAztPIlfv7ODRZv2cij3NO0jA8jfEcb+j0LpFtiWZ5Os0lk+RESkYapKQn3SzOKdc2sAzGwkcNK3YYk0ENWd10zEa3rUdKZHTed4/hmWfrqPNz7IJjF7NS38jLGRnZk6OJSrIzvTskVV/lAoIiINWVUS6vuAV7211ADHgLt9FpFIQ1HVec1EznG2sIjVnx9i4cZsVmw/yJnCIvp3a8sTE/szKbY7HVsH1neIIiJSiy44bV5JR7O2AM65Ez6NqBo0bZ74VGXzmmVk1HU00gikfXWClA3ZvLV5H4fzTtOhVUtuju3B1LgeDOgecuETiIhIg1WjafPM7FfAb51zx73blwL/4Zz7ee2GKdLAlDevWWXt0iwd/foMSzbvZeHGbLbuPUGAv3F1ZGcS4noypm8nAvxV0iEi0tRVpeTjeufcY8UbzrljZnYDoIRamrbiec3Ka5dm7WxhEe+nHWThxmxWph3kbKFjYI+2zLmpP5Nie9C+Vcv6DlFEROpQVRJqfzMLdM6dBjCzSwAVAErTl5RUtoYaIDjY0y7N0rZ9OSxM3cuSzXs58vUZOrYO5K7h4SQMCSWya9v6Dk9EROpJVRLqvwLvmdmfAQfcA7zq06hEGoLiGw81y0ezdjjvNEs27yMlNZsd+z0lHdf278LUwaGMvlwlHSIiUsWbEs1sAnANYMBy59y7vg6sKnRTooj4wpmCIlamHSQlNZtVOw9SUOSICQ1halwoN0V351KVdIiINDs1vSmxFZ4k+p9m1hfoa2YBzrmztR2oiEh9cc6xbd8JUlKzWbJ5L8fyz9K5TSDfi49galwol3dpU98hiohIA1WVko/VwCjv7B4rgA3ArYD+7i0ijd6h3NMs2byXlNRs0r7KpaW/H9cO6EJCXCijLutIC5V01EjyluTzVowsXkmyKWjqr09EqqYqCbU55/LN7HvAfznnfmtmm3wdmIiIr5wuKOS9HQdZmJrNqs8PUVjkiO3Zjl9OHshN0d0JCQ6o7xCbhOQtycxYNoP8s54bezNzMpmxzLM4UlNIOpv66xORqrtgDbU3eb4feB74nnNum5ltcc5F1UWAlVENtYhUlXOOz7JzWLgxmyWb95Fz8ixd2gYyZVAoCXGhXNa5dX2H2OSE/z6czJzzp57sFdKLjJ9k1H1Ataypvz4RKatGNdTAj4FHgcXeZLo38H5tBigi4isHT5xi8SZPSceug3kEtvDjugFdSYgLZeRlHfH3s/oOscnKyil/EaSK2hubpv76RKTqLphQO+dW46mjLt7+Aph5oePMrCfwF6ArUATMdc69YGaxwMtAEFAA3O+cW29m4cAOYKf3FB87535YrVcjIgKcOlvIih0HSEnNZvXnhyhyENfrUn59SxQ3RnejbVAdlHQkJzf7KRfDQsLKHcENC2kaiyM19dcnIlVXlRHqi1WAZ4nyjWbWBkg1s38BvwWecs79w7vi4m+BMd5j0p1zsT6MSUSaKOccm/ccJyU1m2Wf7uPEqQK6hQRx/5jLuGVwD3p3qsWSjgsly8nJZRcFysz0bEOzSqqTxiWVqTEGCA4IJmlc01gcqam/PhGpOp8l1M65/cB+7/NcM9sB9MCzOEzxkmIhwD5fxSAiTd9XOcUlHXtIP/Q1QQGeko5pcT0Z/q0OtV/SUZVkOTGx7Aqb4NlOTGxWCXXxjXlNdRaMpv76RKTqqrSwS40v4innWA0MxJNUv4tnkRg/YIRzLtPbZxvwOXAC+Llz7sNyzjUDmAEQFhYWl5l5/p/bRKRpO3W2kHe3fcXCjXtZs8tT0jE0/FIS4kK5IaobbXxZ0hEe7kmiz9WrF2RkeJ77+UF5P1vNoKjId7FJ5VSGIyI1UNlNiRVOsGpm7c3sCTP7vnkkmtnbZvasd07qql68NbAQ+Ilz7gRwH/BT51xP4KfAPG/X/UCYc24Q8BDwNzNre+75nHNznXNDnHNDOnXqVNUwRKQeJCd78k8/P89jcvLFn8s5R2rmUR5dtIWhv1zBj1/fTPrBPB4cexmrHh7Dgh+O4NahYb5NpsGTjF2oPayCGtqK2stRm++d8M1fFjIzPb/sFP9lQW+siNSCCkeozewdYAue8ox+3udvAtcCMc65my94crMA4G3gXefc77xtOUA755wzMwNynHPnJc5mtgp42DlX4bx4mjZPpOE6tzICIDgY5s6t3qDgvuMnS2bp+PLw11wS4M/1UZ5ZOoZFdMCvrmfpqMoIdQ1ffG29d1JKVT43EZFKVDZCXVlCvdk5F+tNerOdcz3O3XeBixrwKnDUOfeTUu07gPucc6vMbBzwW+dcnJl18vYt9E7N9yEQ5Zw7WtE1mk1CrT9TSiNUk/zl5BlPSUdKajZr0w/jHFwZ0Z6EuFCuj+pG60Bf3k99AVXNdmvw77a+c78mufqfynBEpIYuNqH+DLgKaINndDrGOZdhZh2AD51z/S9w0Xg8SfEWPNPmATyGpz76BTw3RJ7CM21eqplNBZ7GMztIIfCkc25ZZddoFgm1hqqkkapu/uKcY0PmMVI2ZPP3LfvJO11A6KWXMHVwKFMHhxLWIdj3QVeVj3/Jrc/c79zV/8Azc8Xcm+Y27qS6vn9LEZFG72IT6tuA33s378dT++yA/nimvZvrg1irpVkk1PpPQBqB8kY0E2+aXqVv3exj+SzauJeFG7PJPJJPcEt/rh/YjWlDQrkivH3dl3Q0APX5z77Jrv6nwQkRqaGLSqi9B/p7+xSYWQsgFtjrnRKv3jWLhFp/ppQGrqIRzbv83uXVX8SXm79MmVbAP7Z4Sjo++uIIAMN7d2BqXCjXD+xKq/os6WgA6jP383vKD8f5P3MMo+jJRv4zR+VzIlIDF51QN3TNIqFuqiPU+o+tyahsRDOpU0apj9nx/UePktM+m3e27OfrM4WEtQ8mIS6UKYN60LN9AyrpaADq659Ikx2hFhGpISXUjVlT/DNlU3xNzdiFRjT3HM1n4cZsFm7MZs/Rk7QObMENUV1JiOvJ0PBL8dy/LA1Fk62hFhGpISXUjV1TG81tqqPuzVR5I5rmLiGs5SSu7PBjPvnyKGYw8lsdmRrXg+sGdCW4ZfMu6WjomuQsHyIiNVTjhNpbS92FUkuVO+cqWN2g7jSbhLqpUV14g1LT39dKRjTPnCSoKIpWheNoVTgSI4iIjq1IiAtl8qAe9Gh3ie9ehIiIiI9VllBfcJjIzH4EPAkc4Jvp7xwQXWsRNiZNbbS4PoSFlT9CXY1V5KR2nFt9U7x4HFT923pkt8l8OyyI97fnQ1F7sJMMucx49NrhDA5rmiUdGsEVEZHSLjhCbWa7gSudc0fqJqSqq/MRatX+1g69jw3GxVbf5J46yztb9pOSms2/M45hBvGXdSQhLpTrBnQlKMDfVyHXO9UYi4g0TzUq+TCz94FrnXMFvgiuJuo8oVbtb+3RSH+DUJ3qm8Iix0fpR0hJ3cM/t33FqbNF9O7UqmSWjm4hzaOkQ7NgiIg0TxdV8mFmD3mffgGsMrO/A6eL9zvnflerUTYGWRWUjVfULhWbPl0JdANQleqbLw9/zcLUbBZtzGZfzinaBLXglsGhJMSFMqhnuyZZ0lGZrJzy/71X1C4iIk1fZTXUbbyPWd6vlt4voJw5spoD1f5KA3Yxg/5JSeVX3zz+9FleW+8p6UjNPIafwejLO/HoDf24tn+XJl3ScSFhIWHljlCHhejngIhIc1VhQu2cewrAzKY55xaU3mdm03wdWINUUfaRlFR/MYlw8TcXFu9LTISsPY6woYeJvjmb337+Fae3F3FZ59bMvj6SKYN60KVtkG9fRCORNC6p3BrqpHH6OSAi0lxVpYZ6o3Nu8IXa6kO9TJun2l9pgGpS3p9+KM9b0rGXr06cIuSSACbFdGdqXCgxoSHNrqSjKjTLh4hI83NRNyWa2fXADcC3gTdK7WoL9HfOXVHbgVaX5qEW8aju1N45+WdZ9tk+Fm7MZlPWcfz9jNF9OpIQ15Nx/To365IOERGR8lzsPNT7gA3AJCC1VHsu8NPaC09Eaqoq5f0FhUV8uPswC1OzWb79AGcKiujbpQ2JN/Tj5kHd6dxGJR0iIiIXo7Ia6k+BT83sb865s3UYk4hUU2Xl/bsO5JKyMZvFG/dyMPc07YIDuP2KMKYODmVgj7a1VtKhaigREWmuLrhSIhBuZr8G+gMlQ1jOud4+i0pEqqXMzYVZEHbZGW56cB9vHssm8fkc/P2MsX07kRAXytjIzgS2qN2SjtpYcVFERKSxqspNiWvwLD3+PHAT8P+8xz3p+/Aqpxpqaax8MZpbUFjE6l2HSEnNZsX2g5wpLCKyaxsS4kK5ObYHndoE1k7w5dCaRyIi0tRdbA11sUucc++ZmTnnMoE5ZvYhniRbRKqptkdzd36VS0rqHhZv2sfhvNO0b9WS6cOKSzpCai/wSmjNIxERac6qklCfMjM/YJeZPQjsBTr7NiyRpisxsWytM3i2ExOrnlAf+/oMSz/dR0pqNlv25tDCzxgb2ZlpcaGM6duZli38aj/wSmjNIxERac6qklD/BAgGZgK/AK4G7vJlUCJNybnlHeUlnnDh0dyzhUWs2nmIhanZvJd2gLOFjgHd2/LExP7cHNudDq19V9JxIVrzSEREmrMLJtTOuX97n+bhqZ8WkSoqr7zDrPw5oysazd2x/wQpqdks2byXw3ln6Ni6JXcOD2fq4FD6d2/ru+Cr4bybIjXLh4iINCMVJtRmtrSyA51zk2o/HJGmpbzyDufOT6rPHc09kneaJZs9C69s23eCAH9jXGQXEuJCuapvJwL867akoyqmT1cCLSIizVNlI9TDgT3Aa8AngNYfFqmmiso4nPPMgFE8mnvDDZD4eBEznj5Ip6HZ+IUepAhHVI8Qnpo0gEkx3bm0Vctaj09zR4uIiNRcZQl1V+Ba4DbgduDvwGvOuW11EZhIU1BRzXTxdHLOOZ6bd4JnU7IJnLiPTsFnKMwLJH9jBLNvDeWh77XxWWyaO1pERKR2XHAeagAzC8STWD8LPO2c+y9fB1YVmodaGrpzk1bwlHf87o+n8f/WXlJSs0n7KhdX4Ef+7s7kbenJqS87gvPz+RzODWXuaI2Si4hIY3DR81B7E+kb8STT4cCLwKLaDlCkqSpzs97eQsKuPEjk9dn89vNDFKY5Ynq24+jyAXy9oztFp8qWdPh6DueGMHe0RslFRKQpqPDOJjN7FVgHDAaecs4Ndc79wjm3t86iE6lnyVuS6fjdmVi7DMyK6Ng9j+Tkqh/vnCP6qhyGPbCVsAffg/iNpB3MYUT7CP7109EseWAk7Y+Gn5dMg+/ncK7o/HU5d3Rlc3KLiIg0FpVNFXAHcDnwY2CdmZ3wfuWa2Ym6CU+kbiUne0oh/PygY/c87vx+Hkfe/DXkhAN+HNnfmu9+12Hm6Xf//Z5HM2jRgpL2P80/xdzV6Uz4/Yfc9Ic1fHRgD19/0ZEDbw4l87+uZtGT/Vi/wlMfnZTkKQMprS7mcK6v65ZWX6PkpT/n8HCq9UuSiIjIuapUQ91QqYZaalN59c5QROW/d5biX0jwZQdpFbWHSyIOYX4wKKwdmxaGsmdtd4pOB5TpXrpWub7qiOu7frk+6rgrqmufO1dlJiIiUrHKaqiVUIt4VZTcVc7RslsOrQfuIbj/PvyDCijIDeLrrT0IORpKxpbW+PmVv5CLGRQV1ULgjVh9JLcN5WZMERFpXC76pkSR5qQ6ZQb+rU/RasBeWg3MpmXHPIrO+nHy867kbQ3lVGZHcEaOd+b2iqbOq8ta5YaqPlZYbAg3Y4r8//buPbzOukr0+HclvZEWQikgpSEJYJERykBbkZHBC8UHb4gHPCrmOF7O2HG8AeOMjmQuR8cwHsdxwBnnklFHj5NBZwoooI5AhxnlIHBIqa1QFJCmpC1QCm2h1zRd54+9A7tp0tvOzt5Nvp/n6ZP9/t53v3tl84as/ct610/S2GJCLRUNl/gOlH3EhH4Om/0k007vZUproaRjW+901v9wDpsfmknumLjH+aCQIA41Czuatcq1bLRXWPQDjiRppFVs/eKIOCEi7oiIFRHxQERcXhw/MyLujoilEXFfRJxd8pxPR8QjEfGLiLiwUrFJQxnqJr1JU/povPA6jnrLXTR95HaOeev9TDz6OTbd/VJWd76GJ7texfPLmvdIpksT5ra2QglDS0uhzKOlxXrdaqqFmzHHFe8AlTQOVKyGOiJmAjMzc0lEHA50A28DrgH+KjN/GBFvAj6Zma+NiJdTWOb8bOB44HbglMzsH+41rKHWSBu4SW/1M1uZde5qjn5FL0/v2MyUiXW86fSZHLmhia//+QxWrYoXlgz/wQ8KM5719dDfX0iYXZyktlX7ZsxxwztAJY0h48PTMAAAGz9JREFUVamhzsy1wNri4+ciYgUwC0jgiOJhjcCa4uOLgW9n5nbgsYh4hEJy/dNKxSiV2rqjn2mnPcF5f9zLnY88TSacdPxRfHLeybxxznEcPqUwC/0nv1PlQFW20S4zGbf21mjc/wCSxpBRqaGOiFbgLOAe4ArgRxHxRQolJ68qHjYLuLvkab3FscHnWggsBGi26HHc6VreRfvidlZtXEVzYzMdCzpom3Pwv5gzk+6eZ1nU3csty9by/PadNE0/jI+dP5tL586iZcbUEYxeGme8A1TSOFHxhDoipgHXA1dk5qaI+BxwZWZeHxHvAL4GXADEEE/fox4lMzuBTiiUfFQuctWaruVdLLx5IVv6CjNePRt7WHhzcZ3qZW0H9Cf81Ru2cuOSXhZ197Jy/RYaJtXzxtNncum8WZxz4gzq6oa6HCUdEO8AlTROVDShjoiJFJLprsy8oTj8XgqrLwL8G/DV4uNe4ISSpzfxYjmIxrGBeteensug8VxYcBWccR0AW/q2cPn/voetN7a98Jflnp5C2SbsnlRv2bGTf//5E1y/pJe7Hl1PJpxz0lF89PzZvPH045g62aY30oiyxY2kcaJiGUREBIXZ5xWZ+aWSXWuA1wD/CZwPPFwcvwn4l4j4EoWbEmcD91YqPh0adr+nqa6wBPjN/1jYWUyq19/wx7B19+cNlGm++93JvY89w/VLevn+srVs3tFP81ENXLHgFC6ZO4sTjhrU7kHSyKlGo3FJqoJKdvn4TeAnwHIKjXwBrgI2AddSSOa3AR/OzO7ic9qBDwA7KZSI/HBvr2GXj7Fv2NULG1fClSfCssvghi4GVwzVH7GFaXNWc8Zbe1n1zBamTqrnzWfM5NK5TZx94lEUPu9JkiTtn2p1+biToeuiAeYN85wOwL8FjnOlLc2G/by3sVCDGf/xebJ4mcXEnTS87Ammnf44U1qegYSm6TO48vWzufC042iYZEmHJEkaeWYYqhldXXD55bB+fTL8Z7GixlW0NLbQs7GJySesZ9qcXhpetpa6Sf30PdPAhh+fwuc+OIuPvt+SDkmSVFkm1KoJu9dK7z2ZnjRlJ0dc/HdsfPpcZn1oMROO2MGu7RPYvOJ4Ni9vYvvq6cyYEXz0/aMSuiRJGudMqFUThlr/YXdJ3eR+jnnFSmLunUw+7NXkzl1sq+/h2R+cz9YVLeTOeqDQRODaa0clbEmSJBNqVcfgBVp6Vj3GcK3IpzSvZ+rchzjm9OfZ2tdPX2zn2fpvsrn+DvoPexpmX0Z97xfYtaHJJgKSJGnUmVBr1A21QEs0Pk5ueHGxhwlHbmbqnF6mnbaaCY1byV19vO2sk/mrpe9ke92K3XPvM65j1xnfZtef7kKSJGm0mVBr1LUvbn8hmR6Q5/8hdT/6expO3sDUOb1MaXqW3AXbHm9k07Y7+PLHT+V9Z83hulVb6Nm45zmbG115TZIkVYcJtUbdqo2rXtzIOqbsOoNpp87isJPvoK5uAn3rp7Lxnpk8d8zf0vSq2/nSgg7a5rTR1QXPf+HnsLYBGle9sGJiw8QGOhbYbVGSJFWHCbVGXXNjM6uf3cm0/gVM7X8dE/IY+nmeuql3ceP7/5AzTziyuPDKV194zotdQKYVBoorJs447Giu/dQraZtj0bQkSaoOE2qNmo1b+7hl2Rqad14L2yeQ9LOt7n6enfhVmPxzOt/6Fc5qnj7kc4fsAtI3lWl3fpm2f6587JIkScMxoVZF9e9KfvLwOhZ193Lrg0+yY+cuZh97JG84ax3/3vtZ1j3/c5obm+lY8JW9zjKvWnVg45IkSaPFhFoV8chTz7GoezU33t/Lk5u2c2TDRC57xQlcOq+JObMaiyUdb9/v8zU3Q0/P0OOSJEnVZEKtEbNxSx83LVvD9d29LH18A/V1wWtPOYY/vaiJBb92LJMn1B/0uTs6SldSLGhoKIxLkiRVkwm1yrKzfxc/efhpFi3p5bZiSQcbDufZJb/G9E3Hc/4fT+FNc8p/nYGFWtrbC2UeLuAiVVBXlz9sknQATKh1UH755HMs6u7lxvtXs+657UxvmMi8xmZ+8JUmNvUcAQSbKMwqw8j8Lm5rG7+/081vNGpebKlT2O7pGdkfZEkagyIzqx3DQZs/f37ed9991Q5j3Hh28w5uXraGRd29LOvdyIS64HWnHsulc5s4/9RjOeWldUPWObe0wMqVox7umDE4v4FCuUtnp/mNKqC1degbFvxBljTORUR3Zs4fcp8Jtfamr38Xn7313/jW3Q+T208nmMjM6Tv54LlncPGZxzNj2uQXjq2rg6EupwjY5argB838RqPKH2RJGtLeEuq60Q5Gh4aHntjE5255kLP+7Pt867+m0b/9RJ6r/z5rJn+MZf3vZnLjXbsl0zB8x40R7cTR1VXIMOvqCl+7ukbw5LXJloEaVaPygyxJY4s11HrBM5t38L2lq1nU3csDazYxsT7om/Aznpp0M1vruiH6Aejrg/bF7Xv0ja54J45xWttpy0CNKlvqSNIBc4Z6nOvr38WtDzzB73zrPl559e185uYHAfhfF72ce666gFV1f8LW+ntfSKYHrNq45/RoW1uhrrelpfDX4ZaWEa7zHWq5xC1bCuNjWEdHIZ8pZX6jiqn4D7IkjT3WUI9BXcu7aF/czqqNq4qrEHbsMZv8wJqNXN+9mu8tXc36zTs4etpk/ttZx3PpvCZOPe6IF45rvaaVno17To+2NLaw8oqVlf5WdjeOazvt8iFJUnXtrYbako8xpmt5FwtvXsiWvsJMbs/GHhbeXCiLuPDEt/O9pYUuHSvWbmJSfR0XvLzQpeM1pxzDhPo9/2DRsaBjt/MBNExsoGNBFaZHx3Htw3huGShJUq2z5GOMaV/cvlvyS04gt/067YtWcc7Vi/mzWx5kYn3wZxefxr3tC/jbtnks+LWXDJlMA7TNaaPzok5aGlsIgpbGFjov6txjxntUWPswYsq9t3Mc3hsqSdKwLPkYY+o+U0dmMilPZurOBUztfw31NNLPM3z41fO5dF4Tp7zk8GqHefCsfShbuX2t7YstSRqP7EM9Tjz13DbO/vJH2PH8XCZlK0kfW+p/yvP1i3nJ9GdYeeWvqh2iakC5fa3tiy1JGo+soR7Dtu/sZ/GKp1jU3ct//XId7LqEqHuY9RO+wpb6H7MrNhdqni/orHaoqhHl9rW2L7YkSbszoT4EZSbLejeyqLuXm362ho1b+3jJEZNZ+OqTuHRuE/c8uYH2xQ+yeeMWWhpbhuzyofGr3Hs7x/G9oZIkDcmE+hDy1KZt3Hh/YeGVh596nskT6rjwtOO4dF4Tv/nSo6mvCwBeemybCbSGVe66Ha77IUnS7kyoa9y2vn5uX/Eki7p7+fEv17ErYW7zkfz5JXN48xkzOWLKxGqHqEPMwI2DB3tvZ7nPlyRprPGmxBqUmSx9fAOLunu5+Wdr2LRtJzMbp3DJ3FlcOreJk46ZVu0QJUmSxhVvSjxEPLFxGzfc38v13b08um4zUybW8YbTjuPt807gN06e8UJJhyRJkmqHCXWVbevr59YHCyUddz5cKOl4Ret0PnjeSbz5jJkcbkmHJElSTTOhPkAjsa5IZrJk1bMs6l7NLcvW8Ny2ncw68jA++rqXcsncJlqPnlqZ4CVJkjTiKpZQR8QJwP8BjgN2AZ2ZeW1EfAd4WfGwI4ENmXlmRLQCK4BfFPfdnZkfqlR8B2PwCnE9PYVt2L+kes2GrS906Xjs6c0cNrGeN845jrfPa+KcE2dQZ0mHJEnSIaeSM9Q7gU9k5pKIOBzojojbMvOdAwdExF8CG0ue82hmnlnBmMrS3r57qzAobLe3D59Qb93Rz48eeIJF3b3830efJhNeeeJR/O5rT+ZNc2YybXIF/hO4PLckSdKoqVhCnZlrgbXFx89FxApgFvAgQEQE8A7g/ErFMNL2d4W4zOT/rXyW67t7+f7ytTy/fSdN0w/j4+fP5tK5TTTPaKhckOVOo0uSJOmAjEoNdbGc4yzgnpLh84AnM/PhkrETI+J+YBPwR5n5k9GIb3/ta4W43me3cMOS1Vy/pJee9VtomFTPm+fM5NJ5TZzdetTolHQczDS6JEmSDlrFE+qImAZcD1yRmZtKdl0GXFeyvRZozsz1ETEP+G5EnDboOUTEQmAhQPMor3Xc0QEf+O2d7Nj24ts2edo23v77T/Puf+zlrkfXA/Cqk2fw8fNn84bTj2NqJUo69mZ/p9ElSZI0Iiqa7UXERArJdFdm3lAyPgG4BJg3MJaZ24HtxcfdEfEocAqw28otmdkJdEJhYZdKxr+HM7rIi26HW/+EyUdMY9rcX9JwyjoW9U6kZUYDv/f6U7hk7iyaplewpGNf9jWNLkmSpBFVyS4fAXwNWJGZXxq0+wLgoczsLTn+GOCZzOyPiJOA2cCvKhXfwWhf3E7faT1Me9lTzOj7MLvYwub6n3B44wP85+/fTuFbrrKOjt1rqAEaGgrjkiRJGnF1FTz3ucB7gPMjYmnx35uK+97F7uUeAK8GlkXEz4BFwIcy85kKxnfAVm0slE1sqb+Lpyd+kd4p7+GZSX/Nqq13VDSZ7lreRes1rdR9po7Wa1rpWt41/MFtbdDZCS0tEFH42tk5Juqnu7qgtRXqImmd0EtXtBUGuvbyfkiSJFVYZI5u1cRImj9/ft533337PnCEtF7TSs/GPcspWhpbWHnFyoq8ZtfyLhbevJAtfS/OODdMbKDzok7a5hz6SfL+Gty8BKCBzXTyQdoavjdmPjRIkqTaFBHdmTl/qH2VnKEeczoWdNAwcff66IaJDXQsqFw5Rfvi9t2SaYAtfVtoX9xesdesRUM2L2Eq7Vz9YhcTSZKkKjChPgBtc9rovKiTlsYWgqClsaXiM8UDZSb7Oz5WDdu8hOa9HyBJklRho9zT7dDXNqdtVEstmhubhywzaW4cX107hm1ewqoXD5AkSaoCZ6hrXDXKTGpRR0ehWUmpBjbTwVV2MZEkSVVlQl3jqlFmUot2a15C0lLfSycLaWu5yxsSJUlSVdnlQ5IkSdoHu3xIkiRJFWJCLUmSJJXBhFqSJEkqgwm1JEmSVAYTakmSJKkMJtSSJElSGUyoJUmSpDKYUEuSJEllMKGWJEmSymBCLUmSJJXBhFqSJEkqgwm1JEmSVAYTakmSJKkMJtSSJElSGUyoJUmSpDKYUEuSJEllMKGWJEmSymBCLUmSJJXBhFqSJEkqgwm1JEmSVAYTakmSJKkMJtSSJElSGUyoJUmSpDKYUEuSJEllMKGWJEmSymBCLUmSJJWhYgl1RJwQEXdExIqIeCAiLi+Ofycilhb/rYyIpSXP+XREPBIRv4iICysVmyRJkjRSJlTw3DuBT2Tmkog4HOiOiNsy850DB0TEXwIbi49fDrwLOA04Hrg9Ik7JzP4KxihJkiSVpWIz1Jm5NjOXFB8/B6wAZg3sj4gA3gFcVxy6GPh2Zm7PzMeAR4CzKxWfJEmSNBJGpYY6IlqBs4B7SobPA57MzIeL27OAx0v291KSgEuSJEm1qOIJdURMA64HrsjMTSW7LuPF2WmAGOLpOcT5FkbEfRFx37p160Y2WEmSJOkAVTShjoiJFJLprsy8oWR8AnAJ8J2Sw3uBE0q2m4A1g8+ZmZ2ZOT8z5x9zzDGVCVySJEnaT5Xs8hHA14AVmfmlQbsvAB7KzN6SsZuAd0XE5Ig4EZgN3Fup+CRJkqSRUMkuH+cC7wGWl7TGuyozf0Chm0dpuQeZ+UBE/CvwIIUOIR+xw4ckSZJqXcUS6sy8k6HrosnM9w0z3gF0VComSZIkaaS5UqIkSZJUBhNqSZIkqQwm1JIkSVIZTKglSZKkMphQS5IkSWUwoZYkSZLKYEItSZIklcGEWpIkSSqDCbUkSZJUBhNqSZIkqQwm1JIkSVIZTKi1b11d0NoKdXWFr11d1Y5IkiSpZkyodgCqcV1dsHAhbNlS2O7pKWwDtLVVLy5JkqQa4Qy19q69/cVkesCWLYVxSZIkmVBrH1atOrBxSZKkccaEWnvX3Hxg45IkSeOMCbX2rqMDGhp2H2toKIxLkiTJhFr70NYGnZ3Q0gIRha+dnd6QKEmSVGRCrX1ra4OVK2HXrsLXSifTtumTJEmHENvmqbbYpk+SJB1inKFWbbFNnyRJOsSYUKu22KZPkiQdYkyoVVts0ydJkg4xJtSqLbbpkyRJhxgTatUW2/RJkqRDjF0+VHva2kygJUnSIcMZakmSJKkMJtSSJElSGUyoJUmSpDKYUEuSJEllMKGWJEmSymBCLUmSJJWhYgl1RJwQEXdExIqIeCAiLi/Z97GI+EVx/AvFsdaI2BoRS4v//r5SsUmSJEkjpZJ9qHcCn8jMJRFxONAdEbcBLwEuBs7IzO0RcWzJcx7NzDMrGJMkSZI0oiqWUGfmWmBt8fFzEbECmAV8EPh8Zm4v7nuqUjFIkiRJlTYqNdQR0QqcBdwDnAKcFxH3RMR/RcQrSg49MSLuL46fNxqxSZIkSeWo+NLjETENuB64IjM3RcQEYDpwDvAK4F8j4iQKs9nNmbk+IuYB342I0zJz06DzLQQWFjefj4hfVPp7OIQcDTxd7SBU07xGtD+8TrQvXiPal7F4jbQMt6OiCXVETKSQTHdl5g3F4V7ghsxM4N6I2AUcnZnrgIEykO6IeJTCbPZ9pefMzE6gs5JxH6oi4r7MnF/tOFS7vEa0P7xOtC9eI9qX8XaNVLLLRwBfA1Zk5pdKdn0XOL94zCnAJODpiDgmIuqL4ycBs4FfVSo+SZIkaSRUcob6XOA9wPKIWFocuwr4OvD1iPg5sAN4b2ZmRLwa+GxE7AT6gQ9l5jMVjE+SJEkqWyW7fNwJxDC7/8cQx19PoTxEB89SGO2L14j2h9eJ9sVrRPsyrq6RKJQyS5IkSToYLj0uSZIklcGEegyJiPpiH+9bqh2Lak9EHBkRiyLioYhYERG/Ue2YVFsi4sqIeCAifh4R10XElGrHpOqLiK9HxFPFe58Gxo6KiNsi4uHi1+nVjFHVNcw18hfF3zfLIuLGiDiymjFWmgn12HI5sKLaQahmXQv8e2aeCvw6XisqERGzgI8D8zPzdKAeeFd1o1KN+AbwhkFjfwgszszZwOLitsavb7DnNXIbcHpmngH8Evj0aAc1mkyox4iIaALeDHy12rGo9kTEEcCrKbSyJDN3ZOaG6kalGjQBOKy4AFcDsKbK8agGZOaPgcFdty4Gvll8/E3gbaMalGrKUNdIZt6amTuLm3cDTaMe2CgyoR47rgE+CeyqdiCqSScB64B/KpYFfTUiplY7KNWOzFwNfBFYRWHl2o2ZeWt1o1INe0lmrgUofj22yvGotn0A+GG1g6gkE+oxICLeAjyVmd3VjkU1awIwF/i7zDwL2Ix/olWJYg3sxcCJwPHA1IjYo8WpJB2IiGgHdgJd1Y6lkkyox4ZzgbdGxErg28D5EfHP1Q1JNaYX6M3Me4rbiygk2NKAC4DHMnNdZvYBNwCvqnJMql1PRsRMgOLXp6ocj2pQRLwXeAvQlmO8T7MJ9RiQmZ/OzKbMbKVwE9F/ZKYzS3pBZj4BPB4RLysOLQAerGJIqj2rgHMioiEigsI14o2rGs5NwHuLj98LfK+KsagGRcQbgE8Bb83MLdWOp9IqufS4pNryMaArIiYBvwLeX+V4VEMy856IWAQsofDn2fsZZyudaWgRcR3wWuDoiOgF/hT4PPCvEfE/KXwY++/Vi1DVNsw18mlgMnBb4TM6d2fmh6oWZIW5UqIkSZJUBks+JEmSpDKYUEuSJEllMKGWJEmSymBCLUmSJJXBhFqSJEkqgwm1JEmSVAYTakkaQkRkRHyrZHtCRKyLiFtKxt4WEcsi4qGIWB4RbyvZ942IWB0Rk4vbRxdXMyUiWiNia0QsLfn3WxHxLxHxuyXneGXx/EOuGRARK4uv+7OIuDUijhs0PnDuL5fE9Fhx7GcRsWAf78FbIuL+4rEPRsTvRER7yXn7Sx5/vHj+tw86x/ODvuf7I2JFRNxbXEVt4Lj3Fd/f0vfk5cXnZUR8rOTYvyke/5XicQ8Oej93i0GSKs2FXSRpaJuB0yPisMzcCrweWD2wMyJ+Hfgi8PrMfCwiTqSwgMGvMnNZ8bB+4APA3w1x/kcz88zSgYj4EfDT4gIr64G/AT6cmTv3EufrMvPpiLgauAr4eOn4EMf/QWYuiojXUVi4ZfZQJ42IicX9Z2dmb/GDQWtm/gLoKB7zfOn3EBHf2EucA9/zWcVjTwJuiIi6zPyn4v7vZOZHB8XRSmFZ68sj4h8yc8fAvsz8SMkxtwx+PyVptDhDLUnD+yHw5uLjy4DrSvb9PnB1Zj4GUPz658AflBxzDXDlcDPMg2XmkxSS9C8AHwKWZead+xnrj4GX7uexAD8FZu1l/+EUJl3WF2PbXkymR0Rm/gr4PV78ALA364DFvLjUtSTVFBNqSRret4F3RcQU4AzgnpJ9pwHdg46/rzg+YBVwJ/CeIc598qDyhvOK438PvJxCYv7JA4j1LcDyku07Ss595RDHvwH47nAny8xngJuAnoi4LiLaImKkf2csAU4t2X7noPfksJJ9nwc+ERH1IxyDJJXNkg9JGkZmLiuWE1wG/GDQ7gByP8auppCYfn/Q+B4lH8XX3BUR/wDMz8z1+xHmHRHRDywD/qhkfLiSj7+IiC8AxwLn7O3EmfnbETEHuIDCjPzrgfft7Sn7OTYgBm0PVfIxEMtjEXEv8O69xSxJ1eAMtSTt3U0UyjCuGzT+ADB/0Nhc4MHSgcx8BFgKvOMAXnNX8d/+eF1mnpmZv5WZG/bj+D+gUBryR8A393VwZi7PzL+ikExfuo/D1wPTBzYi4ihgqKR+wFnAin3FUOJq4FP4u0tSjfF/SpK0d18HPpuZyweNfxH4dHEGe+DGuKuAvxziHB0UZnhrQmbuAq4F6iLiwqGOiYhpEfHakqEzgZ59nPo/KZRtTCpuvw+4Y5jzt1J4D/96P8MmMx+i8IHlLfv7HEkaDZZ8SNJeZGYvheRz8PjSiPgUcHOxI0Yf8MnMXDrEsQ9ExBIKM9gDTo6I0mO/nplfHsHQB0pBoHBz428Niikj4nMU6rR/NMTzA/hksfxkK4WuJ+/b2wtm5i0RMQ/oLr72oxRurhxwckTcD0wBngP+uqTDBxSS8d8s2f4wsGbQy3QA9+8tDkkabZG5t/I2SZIkSXtjyYckSZJUBks+JKnGRcQ9wORBw+8Zoq77YM9/I3DioOFPZeZQpSCSpEEs+ZAkSZLKYMmHJEmSVAYTakmSJKkMJtSSJElSGUyoJUmSpDKYUEuSJEll+P+ULNSzBuCGUgAAAABJRU5ErkJggg==\n",
+      "text/plain": [
+       "<Figure size 864x432 with 1 Axes>"
+      ]
+     },
+     "metadata": {
+      "needs_background": "light"
+     },
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# for regression: (pick a single column to visualize results)\n",
+    "\n",
+    "# Results from this graph _should not_ be used as a part of your results -- it is just here to help with intuition. \n",
+    "# Instead, look at the error values and individual intercepts.\n",
+    "\n",
+    "col_name = 'MONEY_PER_STUDENT'\n",
+    "col_index = X1_train.columns.get_loc(col_name)\n",
+    "\n",
+    "f = plt.figure(figsize=(12,6))\n",
+    "plt.scatter(X1_train[col_name], y1_train, color = \"red\")\n",
+    "plt.scatter(X1_train[col_name], model.predict(X1_train), color = \"green\")\n",
+    "plt.scatter(X1_test[col_name], model.predict(X1_test), color = \"blue\")\n",
+    "\n",
+    "new_x1 = np.linspace(X1_train[col_name].min(),X1_train[col_name].max(),100)\n",
+    "intercept = model.predict([X1_train.sort_values(col_name).iloc[0]]) - X1_train[col_name].min()*model.coef_[col_index]\n",
+    "plt.plot(new_x1, intercept+new_x1*model.coef_[col_index])\n",
+    "\n",
+    "plt.legend(['controlled model','true training','predicted training','predicted testing'])\n",
+    "plt.xlabel(col_name)\n",
+    "plt.ylabel('Math 8 score')\n",
+    "plt.title('Prediction with 25% trained')\n",
+    "\n",
+    "col_name = 'MONEY_PER_STUDENT'\n",
+    "col_index = X2_train.columns.get_loc(col_name)\n",
+    "\n",
+    "f = plt.figure(figsize=(12,6))\n",
+    "plt.scatter(X2_train[col_name], y2_train, color = \"red\")\n",
+    "plt.scatter(X2_train[col_name], model.predict(X2_train), color = \"green\")\n",
+    "plt.scatter(X2_test[col_name], model.predict(X2_test), color = \"blue\")\n",
+    "\n",
+    "new_x2 = np.linspace(X2_train[col_name].min(),X2_train[col_name].max(),100)\n",
+    "intercept = model.predict([X2_train.sort_values(col_name).iloc[0]]) - X2_train[col_name].min()*model.coef_[col_index]\n",
+    "plt.plot(new_x2, intercept+new_x2*model.coef_[col_index])\n",
+    "\n",
+    "plt.legend(['controlled model','true training','predicted training','predicted testing'])\n",
+    "plt.xlabel(col_name)\n",
+    "plt.ylabel('Math 8 score')\n",
+    "plt.title('Prediction with 35% trained')"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "<h2> Summary </h2>"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "**<WRITE A PARAGRAPH SUMMARIZING YOUR WORK AND FINDINGS\\>**"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 325,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "\"\\nLooking at the 8 scatter plots in the first section, there looks to be a slight positive correlation between money\\ninvested in each student and their test scores. The average investment per student is a bit over $6 with a standard\\ndeviation of $1.73. Excluding the high outliers, the bulk of the investments range from $3(minimum) to $7.50, and\\nwithin that subsect, as the investments increase, the math scores do as well. The training model analyzed the data\\nin the same way. In both graphs with 25% and 35% of the data allocated towards training, respectively, the training\\npredicted the math scores to rise with the investment the same way the scatter plot showed. It's a limited analysis,\\nbut the correlation shows that, if possible, increasing the investment per student relates to an increase in their\\nmath scores.\\n\""
+      ]
+     },
+     "execution_count": 325,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "\"\"\"\n",
+    "Looking at the 8 scatter plots in the first section, there looks to be a slight positive correlation between money\n",
+    "invested in each student and their test scores. The average investment per student is a bit over $6 with a standard\n",
+    "deviation of $1.73. Excluding the high outliers, the bulk of the investments range from $3(minimum) to $7.50, and\n",
+    "within that subsect, as the investments increase, the math scores do as well. The training model analyzed the data\n",
+    "in the same way. In both graphs with 25% and 35% of the data allocated towards training, respectively, the training\n",
+    "predicted the math scores to rise with the investment the same way the scatter plot showed. It's a limited analysis,\n",
+    "but the correlation shows that, if possible, increasing the investment per student relates to an increase in their\n",
+    "math scores.\n",
+    "\"\"\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}