{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## STEP 1: Look at the Big  Picture \n",
    "(I have understood the problem and know the goal of the model)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## STEP 2: GET THE DATA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import tarfile\n",
    "from six.moves import urllib\n",
    "\n",
    "DOWNLOAD_ROOT = \"https://raw.githubusercontent.com/ageron/handson-ml/master/\"\n",
    "HOUSING_PATH =  \"datasets/housing\"\n",
    "HOUSING_URL = DOWNLOAD_ROOT + HOUSING_PATH + \"/housing.tgz\"\n",
    "\n",
    "def fetch_housing_data(housing_url = HOUSING_URL, housing_path=HOUSING_PATH):\n",
    "    if not os.path.isdir(housing_path): #make the path if it doesn't exist\n",
    "        os.makedirs(housing_path)\n",
    "    tgz_path = os.path.join(housing_path, \"housing.tgz\")\n",
    "    urllib.request.urlretrieve(housing_url, tgz_path)\n",
    "    housing_tgz = tarfile.open(tgz_path)\n",
    "    housing_tgz.extractall(path=housing_path)\n",
    "    housing_tgz.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "      <th>median_house_value</th>\n",
       "      <th>ocean_proximity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-122.23</td>\n",
       "      <td>37.88</td>\n",
       "      <td>41.0</td>\n",
       "      <td>880.0</td>\n",
       "      <td>129.0</td>\n",
       "      <td>322.0</td>\n",
       "      <td>126.0</td>\n",
       "      <td>8.3252</td>\n",
       "      <td>452600.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-122.22</td>\n",
       "      <td>37.86</td>\n",
       "      <td>21.0</td>\n",
       "      <td>7099.0</td>\n",
       "      <td>1106.0</td>\n",
       "      <td>2401.0</td>\n",
       "      <td>1138.0</td>\n",
       "      <td>8.3014</td>\n",
       "      <td>358500.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-122.24</td>\n",
       "      <td>37.85</td>\n",
       "      <td>52.0</td>\n",
       "      <td>1467.0</td>\n",
       "      <td>190.0</td>\n",
       "      <td>496.0</td>\n",
       "      <td>177.0</td>\n",
       "      <td>7.2574</td>\n",
       "      <td>352100.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-122.25</td>\n",
       "      <td>37.85</td>\n",
       "      <td>52.0</td>\n",
       "      <td>1274.0</td>\n",
       "      <td>235.0</td>\n",
       "      <td>558.0</td>\n",
       "      <td>219.0</td>\n",
       "      <td>5.6431</td>\n",
       "      <td>341300.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-122.25</td>\n",
       "      <td>37.85</td>\n",
       "      <td>52.0</td>\n",
       "      <td>1627.0</td>\n",
       "      <td>280.0</td>\n",
       "      <td>565.0</td>\n",
       "      <td>259.0</td>\n",
       "      <td>3.8462</td>\n",
       "      <td>342200.0</td>\n",
       "      <td>NEAR BAY</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   longitude  latitude  housing_median_age  total_rooms  total_bedrooms  \\\n",
       "0    -122.23     37.88                41.0        880.0           129.0   \n",
       "1    -122.22     37.86                21.0       7099.0          1106.0   \n",
       "2    -122.24     37.85                52.0       1467.0           190.0   \n",
       "3    -122.25     37.85                52.0       1274.0           235.0   \n",
       "4    -122.25     37.85                52.0       1627.0           280.0   \n",
       "\n",
       "   population  households  median_income  median_house_value ocean_proximity  \n",
       "0       322.0       126.0         8.3252            452600.0        NEAR BAY  \n",
       "1      2401.0      1138.0         8.3014            358500.0        NEAR BAY  \n",
       "2       496.0       177.0         7.2574            352100.0        NEAR BAY  \n",
       "3       558.0       219.0         5.6431            341300.0        NEAR BAY  \n",
       "4       565.0       259.0         3.8462            342200.0        NEAR BAY  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "#Loading the data\n",
    "def load_housing_data(housing_path=HOUSING_PATH):\n",
    "    csv_path = os.path.join(housing_path,\"housing.csv\")\n",
    "    return pd.read_csv(csv_path)\n",
    "\n",
    "\n",
    "#Fetch and Load the data\n",
    "fetch_housing_data()\n",
    "housing = load_housing_data()\n",
    "housing.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<1H OCEAN     9136\n",
       "INLAND        6551\n",
       "NEAR OCEAN    2658\n",
       "NEAR BAY      2290\n",
       "ISLAND           5\n",
       "Name: ocean_proximity, dtype: int64"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#inspecting the  \"ocean_proximity\" attribute\n",
    "housing[\"ocean_proximity\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "      <th>median_house_value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>20640.000000</td>\n",
       "      <td>20640.000000</td>\n",
       "      <td>20640.000000</td>\n",
       "      <td>20640.000000</td>\n",
       "      <td>20433.000000</td>\n",
       "      <td>20640.000000</td>\n",
       "      <td>20640.000000</td>\n",
       "      <td>20640.000000</td>\n",
       "      <td>20640.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>-119.569704</td>\n",
       "      <td>35.631861</td>\n",
       "      <td>28.639486</td>\n",
       "      <td>2635.763081</td>\n",
       "      <td>537.870553</td>\n",
       "      <td>1425.476744</td>\n",
       "      <td>499.539680</td>\n",
       "      <td>3.870671</td>\n",
       "      <td>206855.816909</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.003532</td>\n",
       "      <td>2.135952</td>\n",
       "      <td>12.585558</td>\n",
       "      <td>2181.615252</td>\n",
       "      <td>421.385070</td>\n",
       "      <td>1132.462122</td>\n",
       "      <td>382.329753</td>\n",
       "      <td>1.899822</td>\n",
       "      <td>115395.615874</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-124.350000</td>\n",
       "      <td>32.540000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.499900</td>\n",
       "      <td>14999.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-121.800000</td>\n",
       "      <td>33.930000</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>1447.750000</td>\n",
       "      <td>296.000000</td>\n",
       "      <td>787.000000</td>\n",
       "      <td>280.000000</td>\n",
       "      <td>2.563400</td>\n",
       "      <td>119600.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>-118.490000</td>\n",
       "      <td>34.260000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>2127.000000</td>\n",
       "      <td>435.000000</td>\n",
       "      <td>1166.000000</td>\n",
       "      <td>409.000000</td>\n",
       "      <td>3.534800</td>\n",
       "      <td>179700.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>-118.010000</td>\n",
       "      <td>37.710000</td>\n",
       "      <td>37.000000</td>\n",
       "      <td>3148.000000</td>\n",
       "      <td>647.000000</td>\n",
       "      <td>1725.000000</td>\n",
       "      <td>605.000000</td>\n",
       "      <td>4.743250</td>\n",
       "      <td>264725.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>-114.310000</td>\n",
       "      <td>41.950000</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>39320.000000</td>\n",
       "      <td>6445.000000</td>\n",
       "      <td>35682.000000</td>\n",
       "      <td>6082.000000</td>\n",
       "      <td>15.000100</td>\n",
       "      <td>500001.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          longitude      latitude  housing_median_age   total_rooms  \\\n",
       "count  20640.000000  20640.000000        20640.000000  20640.000000   \n",
       "mean    -119.569704     35.631861           28.639486   2635.763081   \n",
       "std        2.003532      2.135952           12.585558   2181.615252   \n",
       "min     -124.350000     32.540000            1.000000      2.000000   \n",
       "25%     -121.800000     33.930000           18.000000   1447.750000   \n",
       "50%     -118.490000     34.260000           29.000000   2127.000000   \n",
       "75%     -118.010000     37.710000           37.000000   3148.000000   \n",
       "max     -114.310000     41.950000           52.000000  39320.000000   \n",
       "\n",
       "       total_bedrooms    population    households  median_income  \\\n",
       "count    20433.000000  20640.000000  20640.000000   20640.000000   \n",
       "mean       537.870553   1425.476744    499.539680       3.870671   \n",
       "std        421.385070   1132.462122    382.329753       1.899822   \n",
       "min          1.000000      3.000000      1.000000       0.499900   \n",
       "25%        296.000000    787.000000    280.000000       2.563400   \n",
       "50%        435.000000   1166.000000    409.000000       3.534800   \n",
       "75%        647.000000   1725.000000    605.000000       4.743250   \n",
       "max       6445.000000  35682.000000   6082.000000      15.000100   \n",
       "\n",
       "       median_house_value  \n",
       "count        20640.000000  \n",
       "mean        206855.816909  \n",
       "std         115395.615874  \n",
       "min          14999.000000  \n",
       "25%         119600.000000  \n",
       "50%         179700.000000  \n",
       "75%         264725.000000  \n",
       "max         500001.000000  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Describing the numerical fields \n",
    "housing.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving figure attribute_histogram_plots\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABZMAAAQwCAYAAACQdcAEAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAC6uklEQVR4nOzdf5yeV13n/9ebFkoFCq3QMU2qqRLUtpGyjd0q+2O0arMUSd3dsmELTbUaZauiG5XUdRf8EY0/QKxK3QjYVIEaBbaVUqUUR9f99octIqEtlSyNJTS28ksadGtTPt8/7hO4O52Za2Zyz8w997yej8f9mOs+1znXfc7JPTlzfa5znStVhSRJkiRJkiRJM3nSUldAkiRJkiRJkjT8DCZLkiRJkiRJkjoZTJYkSZIkSZIkdTKYLEmSJEmSJEnqZDBZkiRJkiRJktTJYLIkSZIkSZIkqZPBZGmJJdmf5NsW+DMOJfnqAR6vkjx3UMeTJGkpzXcsTvKvk9w7wHqMJzkwqONJklaexTi/nPR5Az3XXGz957ZJfivJf1/qOknD7tilroCkhVdVTz+yneRq4EBV/dTS1UiSpOUnSQHrqmofQFX9b+Br+/bvB763qt63NDWUJGlx9Z9rLndV9QNLXQdpOXBmsiRJkiRJkiSpk8FkaUgkOS7JG5I80F5vSHJc2zee5ECSbUkeSnIwyXf3lf3yJH+U5HNJ/jLJzyX5i779leS5SbYCFwM/0W5H+qP+/X35r07yc33vf7x95gNJvmeKev9KkvuTPNhuDTp+4XpKkqSFkeScJLck+Wwb934jyVPavj9v2f66jaH/qX9ZiiS/C3wl8Edt/09MtWxF/+3HSY5vY+5nktwNfOOkvKckeUeSv09yX5IfXug+kCSNhLOSfCjJPyT5/SRPBUjyfUn2Jfl0kuuTnNLS17Zzwi/evZ5kIsn3tu3nJvmzdrxPJvn9vnz9y0RcneQ3k9yQ5OEktyX5mr6835Hk3nacN7Zjfu9MDUlyaZL/k+RX2/j8sSTf3NI/3s6Pt/Tln/H8tOPc9ovnwUlOTPLuNgZ/pm2vmdQ/P9vq9nCS9yZ5dtc/TJI/SPJ3rQ/+PMkZffu6zuu/LslN7d/v3iQv7fo8aSEYTJaGx38DzgXOAp4PnAP0L0XxFcAzgdXAZcBvJjmx7ftN4PMtz5b2eoKq2gW8Ffilqnp6VX1nV6WSbAR+DPh2YB0wef2tXwSe1+r93Fa//9F1XEmShtBjwI8Czwa+CTgP+C8AVfVvWp7ntzH09/sLVtUrgPuB72z7f2kWn/ca4Gva63z6xu8kTwL+CPhremPrecCPJDl//s2TJK0QLwU2AqcB3wBcmuRbgV9o+1YBfwtcO8vj/SzwXuBEYA3w6zPkfRnw0y3vPmAHQAu0/iFwBfDlwL3AN8/y8/8l8KFW7m2t3t9I7/zz5cBvJDmy3Ma056ezOLft9yTgd4Cvonex+J+A35iU5z8D3w2cDDylHbvLje2zTwY+QO/8/Ihpz+uTPA24qbX/ZHr9/Mb+YLS0WAwmS8PjYuBnquqhqvp7egPwK/r2P9r2P1pV7wEOAV+b5BjgPwCvqap/rKq7gd0DrNdLgd+pqg9X1eeB1x7ZkSTA9wE/WlWfrqqHgZ8HNg/w8yVJWhRVdWdV3VpVh6tqP/A/gX+7gB/5UmBHG0M/DlzZt+8bgedU1c9U1T9X1ceA38YxVpLU7cqqeqCqPk3vwuRZ9M4331JVH6iqR+gFdb8pydpZHO9RekHVU6rq/1XVX8yQ951VdXtVHaYXKD2rpb8IuKuq3tn2XQn83Szbc19V/U5VPQb8PnAqvXPjR6rqvcA/A8+dxfnptOe2k1XVp6rqHe0c+2F6QfHJfxP8TlX9TVX9E7Cnr63Tqqq3VNXD7d/gtcDzkzxzFuf1Lwb2t344XFUfAN4B/Meuz5QGzQfwScPjFHpXh4/425Z2xKfaoHvEPwJPB55D73f54337+rcHUa87J9XriOcAXwbc2Ru3AQhwzAA/X5KkRZHkecDrgQ30xrdjefwYOGin8Pgxu3+M/SrglCSf7Us7BvjfC1gfSdJo6A/S/iO98ebL6c2EBaCqDiX5FL2Zu5/oON5P0JudfHuSzwCvq6q3zPKzj8wYftyYV1U1eSmoGTzYt/1PrfzktCPnxjOdn850bvs4Sb4M+FV6M7yP3BH8jCTHtKA2TN/W6Y55DL2g9EWtrl9ou54NHM/M5/VfBfzLSX8XHAv87kyfKS0EZyZLw+MBegPEEV/Z0rr8PXCY3u1GR5w6Q/6aIu0f6Q26R3xF3/bBScf7yr7tT9IbuM+oqme11zNH6Ym+kqQV5SrgI8C6qjoB+El6J6GzNXmM/Tx942s7iXxO3/6ZxtiP05uJ9ay+1zOq6kVzqI8kSUc87nyzLZvw5fQCyZ9vyVOeE1bV31XV91XVKcD301te4bnMzUH6zlnbLOI102efl67z05nG3cm2AV8L/Mv2N8GR5a7m8nfBZP8Z2ERveY1nAmv7jtl1Xv9x4M8m/V3w9Kp65VHUR5oXg8nS8Hg78FNJntPWk/ofwO91FWpXRd8JvDbJlyX5OuCSGYo8CHz1pLQPAv85yTFtHan+23f20Ftj6/R2dfY1fZ/9BXq33P5qkpMBkqx2PUdJ0jL1DOBzwKE2nk4+QZtqDJ1p/98AT01yQZIn03sWwnF9+/cAV7SH/KwBfqhv3+3A55K8Or0H9R2T5Mwkj3tInyRJs/Q24LuTnJXeg95/Hritqva3ZRY/Aby8jTffQ289fwCSXNT38LnP0Lt4+hhzcwOwPsmF6T3o73IeP4npqM3i/HTac9spPINeYPqzSU7qyDtbzwAeAT5FL3D/83117zqvfzfwvCSvSPLk9vrGJF8/gHpJc2IwWRoePwfcQe/BAnvp3YL0c7Ms+4P0rmz+Hb3bXN5Ob5CaypuB09uTcP9XS3sV8J3AZ+mtpXUknaq6EXgD8H56D1B4/6Tjvbql35rkc8D76F3BlSRpufkxerOGHqZ3Mvr7k/a/FtjdxtCpnqD+C/QuDH82yY9V1T/Qe4Dfm/jSzK/+W3p/mt4ttvfRe7DRF29VbSeV30lv/cX76M22ehO98V6SpDmpqpuB/05vnd2D9ILF/evwfx/w4/QCnWcA/1/fvm8EbktyCLgeeFVV3TfHz/8kveUdfql9xun0zn+nO2+dr2nPT2dxbtvvDfSWnvgkcCvwxwOo2zX0xv1PAHe34/ab9ry+rdv8HfT+zR5oeX6Rx1+klhZFqqa6413ScpbkF4GvqKotnZklSZIkSVpESZ5E7wLrxVX1p0tdn2Hkeb2GlTOTpRGQ5OuSfEN6zgEuA9611PWSJEmSJAkgyflJntWW2TjyXILJs3NXLM/rtVwYTJZGwzPora/0eXrrQL0OuG5JayRJkiRJ0pd8E/B/6S0d8Z3AhVX1T0l+K8mhKV6/tbTVnbskF0/TlrtmUdzzei0LLnMhSZIkSZIkSerkzGRJkiRJkiRJUqdjl7oCXZ797GfX2rVrl7QOn//853na0562pHVYbuyzubPP5s4+mzv7bO66+uzOO+/8ZFU9ZxGrtGCGYcxdzvz9Ggz7cTDsx8GwHwdnEH05SmMuzG3c9bs4N/bX3Nhfc2N/zY39NTfD1F/TjbtDH0xeu3Ytd9xxx5LWYWJigvHx8SWtw3Jjn82dfTZ39tnc2Wdz19VnSf528WqzsIZhzF3O/P0aDPtxMOzHwbAfB2cQfTlKYy7Mbdz1uzg39tfc2F9zY3/Njf01N8PUX9ONuy5zIUmSJEmSJEnqZDBZkiRJkiRJktTJYLIkSZIkSZIkqZPBZEmSJEmSJElSJ4PJkiRJkiRJkqROBpMlSZIkSZIkSZ0MJkuSJEmSJEmSOhlMliRJkiRJkiR1MpgsSZIkSZIkSepkMFmSJEmSJEmS1MlgsiRJkiRJkiSpU2cwOcmpSf40yT1J7kryqpb+2iSfSPLB9npRX5krkuxLcm+S8/vSz06yt+27MkkWplmSJEmSJEmSpEE6dhZ5DgPbquoDSZ4B3JnkprbvV6vqV/ozJzkd2AycAZwCvC/J86rqMeAqYCtwK/AeYCNw42CaIkmSJEmSJElaKJ3B5Ko6CBxs2w8nuQdYPUORTcC1VfUIcF+SfcA5SfYDJ1TVLQBJrgEuxGCyJGDt9htm3L9/5wWLVBNJOjoz/X/m/2WSJEmS5mNY4iazmZn8RUnWAi8AbgNeCPxgkkuAO+jNXv4MvUDzrX3FDrS0R9v25PSpPmcrvRnMjI2NMTExMZdqDtyhQ4eWvA7LjX02dyu9z7atPzzj/qn6ZqX32XzYZ3Nnn2my6f6I27b+MJd2/IEnSZIkScvZrIPJSZ4OvAP4kar6XJKrgJ8Fqv18HfA9wFTrINcM6U9MrNoF7ALYsGFDjY+Pz7aaC2JiYoKlrsNyY5/N3Urvs64AzP6Lx5+QttL7bD7ss7mzzyRJkiRJ6ul8AB9AkifTCyS/tareCVBVD1bVY1X1BeC3gXNa9gPAqX3F1wAPtPQ1U6RLkiRJkiRJkoZcZzA5SYA3A/dU1ev70lf1Zfsu4MNt+3pgc5LjkpwGrANub2svP5zk3HbMS4DrBtQOSZIkSZIkSdICms3M5BcCrwC+NckH2+tFwC8l2ZvkQ8C3AD8KUFV3AXuAu4E/Bi6vqsfasV4JvAnYB/xffPieJElflOSpSW5P8tdJ7kry0y39pCQ3Jflo+3liX5krkuxLcm+S8/vSz27j9L4kV7YLuZIkSZIkzVvnmslV9RdMvd7xe2YoswPYMUX6HcCZc6mgJEkryCPAt1bVobbE1F8kuRH498DNVbUzyXZgO/DqJKcDm4EzgFOA9yV5XruIexW9h9neSm/M3ogXcSVJkiRJR2FWayZLkqSFVz2H2tsnt1cBm4DdLX03cGHb3gRcW1WPVNV99O78OactRXVCVd1SVQVc01dGkiRJkqR5MZgsSdIQSXJMkg8CDwE3VdVtwFh79gDt58kt+2rg433FD7S01W17crokSZIkSfPWucyFJElaPG2JirOSPAt4V5KZloeaahmqmiH9iQdIttJbDoOxsTEmJibmVN+VaNv6w1Omjx0//b4j7N9uhw4dsp8GwH4cDPtxcOxLSZJGg8FkSZKGUFV9NskEvbWOH0yyqqoOtiUsHmrZDgCn9hVbAzzQ0tdMkT7V5+wCdgFs2LChxsfHB9mMkXTp9humTN+2/jCv2zvzn1b7Lx5fgBqNlomJCfweHj37cTDsx8GxLyVJGg0ucyFJ0pBI8pw2I5kkxwPfBnwEuB7Y0rJtAa5r29cDm5Mcl+Q0YB1we1sK4+Ek5yYJcElfGUmSJEmS5sWZyZIkDY9VwO4kx9C74Lunqt6d5BZgT5LLgPuBiwCq6q4ke4C7gcPA5W2ZDIBXAlcDxwM3tpckSZIkSfNmMFmSpCFRVR8CXjBF+qeA86YpswPYMUX6HcBM6y1LkiRJkjQnLnMhSZIkSZIkSepkMFmSJEmSJEmS1MlgsiRJkiRJkiSpk8FkSZIkSZIkSVIng8mSJEmSJEmSpE4GkyVJkiRJkiRJnQwmS5IkSZIkSZI6GUyWJEmSJEmSJHUymCxJkiRJkiRJ6mQwWZIkSZIkSZLUyWCyJEmSJGnFSfKWJA8l+XBf2i8n+UiSDyV5V5Jn9e27Ism+JPcmOb8v/ewke9u+K5NkkZsiSdKiMZgsSZIkSVqJrgY2Tkq7CTizqr4B+BvgCoAkpwObgTNamTcmOaaVuQrYCqxrr8nHlCRpZBhMliRJkiStOFX158CnJ6W9t6oOt7e3Amva9ibg2qp6pKruA/YB5yRZBZxQVbdUVQHXABcuSgMkSVoCBpMlSZIkSXqi7wFubNurgY/37TvQ0la37cnpkiSNpGOXugKSJEmSJA2TJP8NOAy89UjSFNlqhvTpjruV3pIYjI2NMTExMav6HDp0aNZ5ZX/Nlf01N/bX3NhfczNTf21bf3jK9CMWq58NJkuSJEmS1CTZArwYOK8tXQG9Gcen9mVbAzzQ0tdMkT6lqtoF7ALYsGFDjY+Pz6pOExMTzDav7K+5sr/mxv6aG/trbmbqr0u33zBj2f0XT11u0FzmQpIkSZIkIMlG4NXAS6rqH/t2XQ9sTnJcktPoPWjv9qo6CDyc5NwkAS4Brlv0ikuStEicmSxJkiRJWnGSvB0YB56d5ADwGuAK4Djgpl5smFur6geq6q4ke4C76S1/cXlVPdYO9UrgauB4emss34gkSSPKYLIkSZIkacWpqpdNkfzmGfLvAHZMkX4HcOYAqyZJ0tBymQtJkiRJkiRJUieDyZIkSZIkSZKkTgaTJUmSJEmSJEmdDCZLkiRJkiRJkjoZTJYkSZIkSZIkdTKYLEmSJEmSJEnqZDBZkiRJkiRJktTJYLIkSZIkSZIkqZPBZEmSJEmSJElSJ4PJkiRJkiRJkqROBpMlSZIkSZIkSZ0MJkuSJEmSJEmSOhlMliRJkiRJkiR1MpgsSZIkSZIkSepkMFmSJEmSJEmS1MlgsiRJkiRJkiSpk8FkSZIkSZIkSVIng8mSJEmSJEmSpE4GkyVJkiRJkiRJnQwmS5IkSZIkSZI6GUyWJEmSJEmSJHUymCxJ0pBIcmqSP01yT5K7kryqpb82ySeSfLC9XtRX5ook+5Lcm+T8vvSzk+xt+65MkqVokyRJkiRpdBy71BWQJElfdBjYVlUfSPIM4M4kN7V9v1pVv9KfOcnpwGbgDOAU4H1JnldVjwFXAVuBW4H3ABuBGxepHZIkSZKkEeTMZEmShkRVHayqD7Tth4F7gNUzFNkEXFtVj1TVfcA+4Jwkq4ATquqWqirgGuDCha29JEmSJGnUdc5MTnIqvZPQrwC+AOyqql9LchLw+8BaYD/w0qr6TCtzBXAZ8Bjww1X1Jy39bOBq4Hh6s6Re1U5yJUlSnyRrgRcAtwEvBH4wySXAHfRmL3+GXqD51r5iB1rao217cvpUn7OV3gxmxsbGmJiYGGg7RtG29YenTB87fvp9R9i/3Q4dOmQ/DYD9OBj24+DYl5IkjYbZLHMx3S23lwI3V9XOJNuB7cCrveVWkqSjk+TpwDuAH6mqzyW5CvhZoNrP1wHfA0y1DnLNkP7ExKpdwC6ADRs21Pj4+FHXf9Rduv2GKdO3rT/M6/bO/KfV/ovHF6BGo2ViYgK/h0fPfhwM+3Fw7EtJkkZD5zIXM9xyuwnY3bLt5ku3z3rLrSRJ85TkyfQCyW+tqncCVNWDVfVYVX0B+G3gnJb9AHBqX/E1wAMtfc0U6ZIkSZIkzducHsA36Zbbsao6CL2Ac5KTW7aRu+XWW7Lmzj6bu5XeZ/O5NXyl99l82Gdzt5h9liTAm4F7qur1femrjoy5wHcBH27b1wNvS/J6encDrQNur6rHkjyc5Fx6Y/YlwK8vSiMkSZIkSSNr1sHkKW65nTbrFGnL+pZbb8maO/ts7lZ6n0132/gRU90avtL7bD7ss7lb5D57IfAKYG+SD7a0nwReluQseuPmfuD7AarqriR7gLvpLUt1eVtWCuCVfOk5BTfislKSJEmSpKM0q2DyVLfcAg8emSnVlrB4qKV7y60kSfNQVX/B1Bdf3zNDmR3AjinS7wDOHFztJEmSJEkrXeeaydPdckvv1totbXsLcF1f+uYkxyU5jS/dcnsQeDjJue2Yl/SVkSRJkiRJkiQNsdnMTJ7ultudwJ4klwH3AxeBt9xKkiRJkiRJ0ijqDCbPcMstwHnTlPGWW0mSJEmSJEkaIZ3LXEiSJEmSJEmSZDBZkiRJkiRJktTJYLIkSZIkSZIkqZPBZEmSJEnSipPkLUkeSvLhvrSTktyU5KPt54l9+65Isi/JvUnO70s/O8netu/KJNM9c0iSpGXPYLIkSZIkaSW6Gtg4KW07cHNVrQNubu9JcjqwGTijlXljkmNamauArcC69pp8TEmSRobBZEmSJEnSilNVfw58elLyJmB3294NXNiXfm1VPVJV9wH7gHOSrAJOqKpbqqqAa/rKSJI0cgwmS5IkSZLUM1ZVBwHaz5Nb+mrg4335DrS01W17crokSSPp2KWugCRJkiRJQ26qdZBrhvSpD5JspbckBmNjY0xMTMzqww8dOjTrvLK/5sr+mhv7a27sr7mZqb+2rT88Y9nF6meDyZIkSZIk9TyYZFVVHWxLWDzU0g8Ap/blWwM80NLXTJE+paraBewC2LBhQ42Pj8+qUhMTE8w2r+yvubK/5sb+mhv7a25m6q9Lt98wY9n9F09dbtBc5kKSJEmSpJ7rgS1tewtwXV/65iTHJTmN3oP2bm9LYTyc5NwkAS7pKyNJ0shxZrIkSZIkacVJ8nZgHHh2kgPAa4CdwJ4klwH3AxcBVNVdSfYAdwOHgcur6rF2qFcCVwPHAze2lyRJI8lgsiRJkiRpxamql02z67xp8u8AdkyRfgdw5gCrJknS0HKZC0mSJEmSJElSJ4PJkiRJkiRJkqROBpMlSZIkSZIkSZ0MJkuSJEmSJEmSOhlMliRJkiRJkiR1MpgsSZIkSZIkSepkMFmSJEmSJEmS1MlgsiRJkiRJkiSpk8FkSZIkSZIkSVIng8mSJEmSJEmSpE4GkyVJkiRJkiRJnQwmS5IkSZIkSZI6HbvUFZC0cqzdfsNSV0GSJEmSJEnz5MxkSZIkSZIkSVIng8mSJEmSJEmSpE4GkyVJkiRJkiRJnVwzWdKyMNV6y9vWH+bSlr5/5wWLXSVJkiRJkqQVxZnJkiRJkiRJkqROBpMlSZIkSZIkSZ0MJkuSJEmSJEmSOhlMliRJkiRJkiR1MpgsSZIkSZIkSepkMFmSJEmSJEmS1MlgsiRJkiRJkiSpk8FkSZKGRJJTk/xpknuS3JXkVS39pCQ3Jflo+3liX5krkuxLcm+S8/vSz06yt+27MkmWok2SJEmSpNFhMFmSpOFxGNhWVV8PnAtcnuR0YDtwc1WtA25u72n7NgNnABuBNyY5ph3rKmArsK69Ni5mQyRJkiRJo+fYpa6AJEnqqaqDwMG2/XCSe4DVwCZgvGXbDUwAr27p11bVI8B9SfYB5yTZD5xQVbcAJLkGuBC4cbHaIkmSJGlprd1+w4z79++8YJFqolFiMFmSpCGUZC3wAuA2YKwFmqmqg0lObtlWA7f2FTvQ0h5t25PTp/qcrfRmMDM2NsbExMTgGjGitq0/PGX62PHT7zvC/u126NAh+2kA7MfBsB8Hx76UJGk0GEyWJGnIJHk68A7gR6rqczMsdzzVjpoh/YmJVbuAXQAbNmyo8fHxOdd3pbl0mhke29Yf5nV7Z/7Tav/F4wtQo9EyMTGB38OjZz8Ohv04OPalJEmjwTWTJUkaIkmeTC+Q/NaqemdLfjDJqrZ/FfBQSz8AnNpXfA3wQEtfM0W6JEmSJEnzZjBZkqQhkd4U5DcD91TV6/t2XQ9sadtbgOv60jcnOS7JafQetHd7WxLj4STntmNe0ldGkiRJkqR5cZkLSZKGxwuBVwB7k3ywpf0ksBPYk+Qy4H7gIoCquivJHuBu4DBweVU91sq9ErgaOJ7eg/d8+J4kSZIk6agYTJYkaUhU1V8w9XrHAOdNU2YHsGOK9DuAMwdXO0mSJEnSSucyF5IkSZIkSZKkTs5MliRJkiRJkpahtdtvWOoqaIVxZrIkSZIkSZIkqZPBZEmSJEmSJElSJ4PJkiRJkiT1SfKjSe5K8uEkb0/y1CQnJbkpyUfbzxP78l+RZF+Se5Ocv5R1lyRpIRlMliRJkiSpSbIa+GFgQ1WdCRwDbAa2AzdX1Trg5vaeJKe3/WcAG4E3JjlmKeouSdJC6wwmJ3lLkoeSfLgv7bVJPpHkg+31or59U16RTXJ2kr1t35VJMvjmSJIkSZJ01I4Fjk9yLPBlwAPAJmB3278buLBtbwKurapHquo+YB9wzuJWV5KkxXHsLPJcDfwGcM2k9F+tql/pT5h0RfYU4H1JnldVjwFXAVuBW4H30Ltie+NR1V6SJEmSpAGqqk8k+RXgfuCfgPdW1XuTjFXVwZbnYJKTW5HV9M5zjzjQ0p4gyVZ658WMjY0xMTExqzodOnRo1nllf82V/TU3w9Zf29YfnnfZxWjHsPXXsJupv7r+rRernzuDyVX150nWzvJ4X7wiC9yXZB9wTpL9wAlVdQtAkmvoXcU1mCxJkiRJGhptLeRNwGnAZ4E/SPLymYpMkVZTZayqXcAugA0bNtT4+Pis6jQxMcFs88r+miv7a26Grb8u3X7DvMvuv3h8cBWZxrD117Cbqb+6/q0X498TZjczeTo/mOQS4A5gW1V9humvyD7atienT2m+V2sXildR5s4+m7uV0GdHc8V0KmPHf+mYo953g7ISvmeDZp9JkrTifBtwX1X9PUCSdwLfDDyYZFWblbwKeKjlPwCc2ld+Db1lMSRJGjnzDSZfBfwsvautPwu8Dvgepr8iO+srtTD/q7ULxasoc2efzd1K6LOjuWI6lW3rD/O6vb3/xhbrCtxytxK+Z4Nmn0mStOLcD5yb5MvoLXNxHr1JVJ8HtgA728/rWv7rgbcleT295R7XAbcvdqUlSVoM8womV9WDR7aT/Dbw7vZ2uiuyB9r25HRJkiRJkoZGVd2W5A+BDwCHgb+iN9np6cCeJJfRCzhf1PLflWQPcHfLf3l7bpAkSSNnXsHkI7f2tLffBXy4bU95RbaqHkvycJJzgduAS4BfP7qqS5IkSZI0eFX1GuA1k5IfoTdLear8O4AdC10vSZKWWmcwOcnbgXHg2UkO0BtQx5OcRW+piv3A90PnFdlXAlcDx9N78J4P35MkSZIkSSvK2q6HaO28YJFqIklz1xlMrqqXTZH85hnyT3lFtqruAM6cU+0kSZIkSZIkSUPhSUtdAUmSJEmSJEnS8JvXmsmSJEmjquvWU0mSJElaqZyZLEmSJEmSJEnqZDBZkiRJkiRJktTJYLIkSZIkSZIkqZPBZEmSJEmSJElSJ4PJkiRJkiRJkqROBpMlSZIkSZIkSZ0MJkuSJEmSJEmSOhlMliRJkiRJkiR1MpgsSZIkSZIkSepkMFmSJEmSJEmS1MlgsiRJkiRJkiSpk8FkSZIkSZIkSVIng8mSJEmSJEmSpE7HLnUFJEmSJEmS1LN2+w0z7t+/84JFqokkPZHBZB2V6Qa5besPc+n2GxzkJEmSJEmSpBFhMFmSJEmSJEkaQl0z1aXFZjBZM/I/LUmSJEmSJEngA/gkSZIkSZIkSbPgzGRJkiRpCM10h5jPpZAkSdJScGayJEmSJEmSJKmTwWRJkiRJkiRJUieDyZIkSZIkSZKkTgaTJUkaIknekuShJB/uS3ttkk8k+WB7vahv3xVJ9iW5N8n5felnJ9nb9l2ZJIvdFkmSJEnSaDGYLEnScLka2DhF+q9W1Vnt9R6AJKcDm4EzWpk3Jjmm5b8K2Aqsa6+pjilJkiRJ0qwZTJYkaYhU1Z8Dn55l9k3AtVX1SFXdB+wDzkmyCjihqm6pqgKuAS5ckApLkiRJklaMY5e6ApIkaVZ+MMklwB3Atqr6DLAauLUvz4GW9mjbnpz+BEm20pvBzNjYGBMTE4Ov+TKzbf3heZUbO767rP3b7dChQ/ZTM9P3qauP7MfBsB8Hx76UJGk0GEyWJGn4XQX8LFDt5+uA7wGmWge5Zkh/YmLVLmAXwIYNG2p8fHwA1V3eLt1+w7zKbVt/mNftnflPq/0Xj8/r2CvJxMQEfg97Zvoudn2X7MfBsB8Hx76UJGk0uMyFJElDrqoerKrHquoLwG8D57RdB4BT+7KuAR5o6WumSJckSZIkad4MJkuSNOTaGshHfBfw4bZ9PbA5yXFJTqP3oL3bq+og8HCSc5MEuAS4blErLUmSJEkaOS5zIUnSEEnydmAceHaSA8BrgPEkZ9FbqmI/8P0AVXVXkj3A3cBh4PKqeqwd6pXA1cDxwI3tJUmSJEnSvBlMliRpiFTVy6ZIfvMM+XcAO6ZIvwM4c4BVkyRJkpbM2o7nWuzfecEi1URa2QwmS5IkSZIkSUugK0guDRvXTJYkSZIkqU+SZyX5wyQfSXJPkm9KclKSm5J8tP08sS//FUn2Jbk3yflLWXdJkhaSwWRJkiRJkh7v14A/rqqvA54P3ANsB26uqnXAze09SU4HNgNnABuBNyY5ZklqLUnSAjOYLEmSJElSk+QE4N/QnllQVf9cVZ8FNgG7W7bdwIVtexNwbVU9UlX3AfuAcxazzpIkLRbXTJYkSZIk6Uu+Gvh74HeSPB+4E3gVMFZVBwGq6mCSk1v+1cCtfeUPtLQnSLIV2AowNjbGxMTErCp06NChWefV8PfXtvWHj6r8oNs27P11RFe/LVYbBt1fR/t9OBqL0WfL5fs1LGbqr2H5HTCYLEmSJEnSlxwL/Avgh6rqtiS/RlvSYhqZIq2mylhVu4BdABs2bKjx8fFZVWhiYoLZ5tXw99elR/nAtf0Xjw+mIs2w99cRXf026H6ZzqD762i/D0djMfpsuXy/hsVM/TUsvwMucyFJkiRJ0pccAA5U1W3t/R/SCy4/mGQVQPv5UF/+U/vKrwEeWKS6SpK0qAwmS5IkSZLUVNXfAR9P8rUt6TzgbuB6YEtL2wJc17avBzYnOS7JacA64PZFrLIkSYvGZS4kSZIkSXq8HwLemuQpwMeA76Y3GWtPksuA+4GLAKrqriR76AWcDwOXV9VjS1NtSZIWlsFkSZIkSZL6VNUHgQ1T7Dpvmvw7gB0LWSdJkoaBwWRJ6rC2a5H7nRcsUk0kSZIkSZKWjsFkSZIkSZKkZWJUJ7t0tUvScDCYLEmSJEmSJC0Ag+QaNU9a6gpIkiRJkiRJkoafwWRJkiRJkiRJUieDyZIkSZIkSZKkTq6ZLGkkjOpDKCRJkiRJkoZFZzA5yVuAFwMPVdWZLe0k4PeBtcB+4KVV9Zm27wrgMuAx4Ier6k9a+tnA1cDxwHuAV1VVDbY5mg8Xg5ckSZIkSZLUZTYzk68GfgO4pi9tO3BzVe1Msr29f3WS04HNwBnAKcD7kjyvqh4DrgK2ArfSCyZvBG4cVEMkSZIkSZIkzY53+Go+OtdMrqo/Bz49KXkTsLtt7wYu7Eu/tqoeqar7gH3AOUlWASdU1S1tNvI1fWUkSZIkSZIkSUNuvmsmj1XVQYCqOpjk5Ja+mt7M4yMOtLRH2/bk9Ckl2UpvFjNjY2NMTEzMs5qDcejQoSWvw0Latv7wwI85dnzvuKPcb4M26t8zGPx37cj3bDaOpm+7PmM5/buthO/ZoNlnkiRJkiT1DPoBfJkirWZIn1JV7QJ2AWzYsKHGx8cHUrn5mpiYYKnrsJAuXYA1k7etP8zr9h7L/ovHB37sUTXq3zMY/HftyPdsNo7mu9hV7+X0PV8J37NBs88kSZIkSerpXOZiGg+2pStoPx9q6QeAU/vyrQEeaOlrpkiXJEmSJEmSJC0D852ZfD2wBdjZfl7Xl/62JK+n9wC+dcDtVfVYkoeTnAvcBlwC/PpR1VzLgou5S5IkSZIkSaOhM5ic5O3AOPDsJAeA19ALIu9JchlwP3ARQFXdlWQPcDdwGLi8qh5rh3olcDVwPHBje0mSJEmSJI2UrolVkrRcdQaTq+pl0+w6b5r8O4AdU6TfAZw5p9pJkiRJkiRJkobCoB/AJ0mSJEmSpCUy06xol5qUdLTm+wA+SZIkSZIkSdIKYjBZkiRJkiRJktTJZS4kSZIkSZK0rLm8h7Q4DCZL0lHqelKzf7hIkiRJkqRRYDBZkiRJkiRJC6prEo6k5cFgsiRJkiRJkjSN/kD4tvWHuXRSYNy7UbWSGEyWJEmSJEnSyHJpQmlwDCZLkiRJkiStAFMFVftn2hpUldTFYLIkSZIkSZJWLNdzlmbPYLIkSZIkSZI0TwajtZIYTJYkSZKWma6T1qs3Pm2RaiJJkqSVxGCyJEmSJEmSjvpBdc7QlUbfk5a6ApIk6UuSvCXJQ0k+3Jd2UpKbkny0/Tyxb98VSfYluTfJ+X3pZyfZ2/ZdmSSL3RZJkiRJ0mgxmCxJ0nC5Gtg4KW07cHNVrQNubu9JcjqwGTijlXljkmNamauArcC69pp8TEmSJEmS5sRgsiRJQ6Sq/hz49KTkTcDutr0buLAv/dqqeqSq7gP2AeckWQWcUFW3VFUB1/SVkSRJkiRpXgwmS5I0/Maq6iBA+3lyS18NfLwv34GWtrptT06XJEmSJGnefACfJEnL11TrINcM6U88QLKV3nIYjI2NMTExMbDKLVfb1h+eV7mx47vL2r/dDh06NDL9tPcT/zDj/vWrnznj/vl+F2G0+nEp2Y+DY19KkjQaDCZLkjT8HkyyqqoOtiUsHmrpB4BT+/KtAR5o6WumSH+CqtoF7ALYsGFDjY+PD7jqy8+l83wK+bb1h3nd3pn/tNp/8fi8jr2STExMMCrfw67vUtf3Yb7fRYCrNz5tZPpxKY3S93Gp2ZeSJI0Gl7mQJGn4XQ9sadtbgOv60jcnOS7JafQetHd7Wwrj4STnJglwSV8ZSZIkSZLmxZnJkiQNkSRvB8aBZyc5ALwG2AnsSXIZcD9wEUBV3ZVkD3A3cBi4vKoea4d6JXA1cDxwY3tJkqRZSnIMcAfwiap6cZKTgN8H1gL7gZdW1Wda3iuAy4DHgB+uqj9Zkkpr0aw9irtHJGk5M5gsSdIQqaqXTbPrvGny7wB2TJF+B3DmAKsmSdJK8yrgHuCE9n47cHNV7Uyyvb1/dZLTgc3AGcApwPuSPK/vAq8kSSPDZS4kSZIkSeqTZA1wAfCmvuRNwO62vRu4sC/92qp6pKruA/YB5yxSVSVJWlTOTJYkSZIk6fHeAPwE8Iy+tLH2XALaQ3FPbumrgVv78h1oaU+QZCuwFWBsbIyJiYlZVebQoUOzzqvF6a9t6w8v6PEX09jxs2/Pr7915sdwbFs/iBoNt7n013I3iN8j//+am5n6q+t7t1j9bDBZkiRJkqQmyYuBh6rqziTjsykyRVpNlbGqdgG7ADZs2FDj47M5fC9AMNu8Wpz+unSE1kzetv4wr9treGi2VlR/7f38jLv377yg8xD+/zU3M/VX1/87+y+eutygrZBvvyRJkiRJs/JC4CVJXgQ8FTghye8BDyZZ1WYlrwIeavkPAKf2lV8DPLCoNZYkaZG4ZrIkSZIkSU1VXVFVa6pqLb0H672/ql4OXA9sadm2AEfu978e2JzkuCSnAeuA2xe52pIkLQpnJkuSJEmS1G0nsCfJZcD9wEUAVXVXkj3A3cBh4PKqemzpqilJ0sIxmCxJkiRJ0hSqagKYaNufAs6bJt8OYMeiVUySpCXiMheSJEmSJEmSpE7OTJYkSZIkSZI0J2u33zDtvv07L1iwYw/i+Jo/ZyZLkiRJkiRJkjoZTJYkSZIkSZIkdXKZC0mSJEmSJEnLRtcyGDNxiYyjYzBZkiQtOtdAkyRJkqTlx2UuJEmSJEmSJEmdDCZLkiRJkiRJkjq5zIWGlrdAS5IkSZIkScPDYLKWLYPNkiRJkiRJ0uJxmQtJkiRJkiRJUieDyZIkSZIkSZKkTgaTJUmSJEmSJEmdDCZLkiRJkiRJkjoZTJYkSZIkSZIkdTp2qSsgSZIkSZIkaXSs3X4DANvWH+bStt1v/84LFrtKGhCDyZIkSZIkSX3WThH8kiQZTJYkSZIkSZK0iLxgs3wZTJYkSZIkSZK0InQFsl2CY2YGkyWtCDMNFg4UkiRJkiRJ3Z601BWQJEmSJEmSJA2/owomJ9mfZG+SDya5o6WdlOSmJB9tP0/sy39Fkn1J7k1y/tFWXpIkSZIkSZK0OAYxM/lbquqsqtrQ3m8Hbq6qdcDN7T1JTgc2A2cAG4E3JjlmAJ8vSZIkSZIkSVpgC7Fm8iZgvG3vBiaAV7f0a6vqEeC+JPuAc4BbFqAOkiRJ0lDzKeaSJElabo52ZnIB701yZ5KtLW2sqg4CtJ8nt/TVwMf7yh5oaZIkSZIkSZKkIXe0M5NfWFUPJDkZuCnJR2bImynSasqMvcD0VoCxsTEmJiaOsppH59ChQ0teh4W0bf3hgR9z7PjZHXemfj3aei23f7NR/57B4L9rs/2edenq91H6Lq6E79mg2WcapK6ZqPt3XrBINZEkafmbblzdtv4wl26/wXFVkhbAUQWTq+qB9vOhJO+it2zFg0lWVdXBJKuAh1r2A8CpfcXXAA9Mc9xdwC6ADRs21Pj4+NFU86hNTEyw1HVYSJcuwC2W29Yf5nV7u79e+y8en3bf0dZrpmMPo1H/nsHgv2uz/Z516fqujNJ3cSV8zwbNPpMkSZIkqWfeUZgkTwOeVFUPt+3vAH4GuB7YAuxsP69rRa4H3pbk9cApwDrg9qOouyRJkiRJ0pS8I0iSBu9opvSNAe9KcuQ4b6uqP07yl8CeJJcB9wMXAVTVXUn2AHcDh4HLq+qxo6q9JEmSpEVlcEaSJGnlmncwuao+Bjx/ivRPAedNU2YHsGO+nylJkiRJkhaPF5AkrTQz/b/n/3lH/wA+SZIkSZKkZacrUC5JeqInLXUFJEmSJEmSJEnDz2CyJEmSJEmSJKmTwWRJkpaJJPuT7E3ywSR3tLSTktyU5KPt54l9+a9Isi/JvUnOX7qaS5IkSZJGgcFkSZKWl2+pqrOqakN7vx24uarWATe39yQ5HdgMnAFsBN6Y5JilqLAkSZIkaTQYTJYkaXnbBOxu27uBC/vSr62qR6rqPmAfcM7iV0+SJEmSNCqOXeoKSJKkWSvgvUkK+J9VtQsYq6qDAFV1MMnJLe9q4Na+sgda2uMk2QpsBRgbG2NiYmIBq/8l29YfnnH/YtVjKl11m87Y8fMve8RStntYHDp0aGT64Wi/D0fjaPtx7yf+Ydp929bPXHZU/v1gtL6PS2059WWSU4FrgK8AvgDsqqpfS3IS8PvAWmA/8NKq+kwrcwVwGfAY8MNV9SdLUHVJkhacwWRJK97a7TcsdRWk2XphVT3QAsY3JfnIDHkzRVo9IaEXkN4FsGHDhhofHx9IRbtc2vF7t//ixanHVLrqNp1t6w/zur1H96fVUrZ7WExMTLBY38OFNt/v0iBcvfFpR9WPR1P3Ufoej9L3cakts748DGyrqg8keQZwZ5KbgEvpLS21M8l2ektLvXrS0lKnAO9L8ryqemyJ6i9J0oIxmCxJ0jJRVQ+0nw8leRe9ZSseTLKqzUpeBTzUsh8ATu0rvgZ4YFErLEnSMtTu+Dly18/DSe6hd3fPJmC8ZdsNTACvpm9pKeC+JEeWlrplcWu+PDmxQ5KWF4PJ0pDp+mNq/84LFqkmkoZJkqcBT2ontU8DvgP4GeB6YAuws/28rhW5HnhbktfTmyW1Drh90SsuSdIylmQt8ALgNo5yaal2vHktL7WUy4Qs9NJUC7Ek0CCWnlpJ7K+5sb/mZtT669ffet2M+9evfuZRHX+m/++HZalAg8mSvshAtjTUxoB3JYHe+P22qvrjJH8J7ElyGXA/cBFAVd2VZA9wN73bdS/3dlvpiUZ17Nv7iX+YcamK5douaTEleTrwDuBHqupzbQyeMusUaU9YWgrmv7zUUi4TstBLUy3EkkCDWHpqJbG/5sb+mpuV1l9H+3/iTP/fD8tSgSvnX1MrzqieHEpamarqY8Dzp0j/FHDeNGV2ADsWuGqSJI2cJE+mF0h+a1W9syWP7NJSLjUhSZotg8mSJGlF8YRZkjST9KYgvxm4p6pe37fLpaUkSSuewWRJkiRJkr7khcArgL1JPtjSfpJeENmlpSRJK5rBZC0pZ4ctPvtckiRJml5V/QVTr4MMLi0lSZrBSlhy1WCyJEmSJElaEE5mkaTRYjBZWmZWwlUuSaPBk0dJkiRJGi0Gk1cAT+YlSZIkSZIkHS2DyZIkSZIkaV6cvCRJK8uTlroCkiRJkiRJkqThZzBZkiRJkiRJktTJZS4kDYy3uEmSJEmSJI0ug8mSJEnSCuMFYEmSJM2HwWRJkiRJkiRJWmJ7P/EPXDrkF/0NJkvz0DWbZ//OCxapJpIkSZIkSVoOuuJJ29YvUkWOgg/gkyRJkiRJkiR1MpgsSZIkSZIkSepkMFmSJEmSJEmS1Mk1k7VizbROjWseS5IkSZIkSY9nMFmaQteC6JIkSZIkSdJKYzBZkiQtO10X/ZbrHSaj2q6ldLQXiL2TSZIkSfoS10yWJEmSJEmSJHVyZrIkSRo6LjckSZIkScPHmcmSJEmSJEmSpE4GkyVJkiRJkiRJnVzmQtKsedu5JElaaD70UJIkaXgZTJaWwFQnSdvWH+ZSg7WSJEmSJEkaUgaTJUmS1MnZok/kHTvzY79JkiQtXwaTR4B/kKuf3wdJ0nws1/FjudZbkiRJWo4MJkvSAnM2nyRJkiRJGgUGkyVJkiQNjLPFpcXn750kabEYTJYWgH/MSZIWgnc6SJIkSVpKBpMlSZJGQNeFzOUcbPYirSRJkjQcDCZLkiRJWhGW8qLLKF/wkSRJK4fBZEmSNHKcySqNJgOykiRJS8tgsiQNMU+aJUmSJEnSsDCY3Ge6oM229Ye5dPsNBm0krSgGsiVJWjw+YFOSJC0HKyqY7C2vkkaNAV9JszXMfwf11+3IRXxJkiRJw2dFBZOHmQEhSZK0XPl3jCRJkrQyGEyWpCU0zDMFJUlabhxXJUmSFpbB5GXCP4wlSZKkpbOUf487+1+SJA2LRQ8mJ9kI/BpwDPCmqtq52HWYr6N5KIbBYEmjZikfFORDimZnOY+5Gi3+HaTlYvJ31TW8NReOu5KklWBRg8lJjgF+E/h24ADwl0mur6q7F7MeC8GTJEnDyJPilWuUx1xJ0uI52vOclXKR13FXkrRSLPbM5HOAfVX1MYAk1wKbAAdYSZIGyzFXklYIJ7YMBcddSdKKkKpavA9L/iOwsaq+t71/BfAvq+oHJ+XbCmxtb78WuHfRKjm1ZwOfXOI6LDf22dzZZ3Nnn82dfTZ3XX32VVX1nMWqzGwt4zF3OfP3azDsx8GwHwfDfhycQfTlUI65sCjjrt/FubG/5sb+mhv7a27sr7kZpv6actxd7JnJmSLtCdHsqtoF7Fr46sxOkjuqasNS12M5sc/mzj6bO/ts7uyzuVvGfbYsx9zlbBl/V4aK/TgY9uNg2I+DswL6ckHH3RXQfwNlf82N/TU39tfc2F9zsxz660mL/HkHgFP73q8BHljkOkiStBI45kqStHgcdyVJK8JiB5P/EliX5LQkTwE2A9cvch0kSVoJHHMlSVo8jruSpBVhUZe5qKrDSX4Q+BPgGOAtVXXXYtZhnrz9d+7ss7mzz+bOPps7+2zulmWfLeMxdzlblt+VIWQ/Dob9OBj24+CMdF8uwrg70v23AOyvubG/5sb+mhv7a26Gvr8W9QF8kiRJkiRJkqTlabGXuZAkSZIkSZIkLUMGkyVJkiRJkiRJnQwm90lyUZK7knwhyYa+9G9PcmeSve3nt05R9vokH17cGi+9ufZZki9LckOSj7RyO5eu9ktjPt+zJGe39H1JrkySpan90pihz748yZ8mOZTkNyaVeVnrsw8l+eMkz178mi+defbZU5LsSvI37Xf0Pyx+zZfWfPqtL8+KHAdWoiRPTXJ7kr9u35efnrT/x5LUSvt/Z65m6sckP5Tk3pb+S0tZz+Vgur5MclaSW5N8MMkdSc5Z6rouB0mOSfJXSd7d3p+U5KYkH20/T1zqOi4HU/TjL7e/Lz6U5F1JnrXEVVw2kmxs/yfuS7J9qeszbJK8JclD/X+H+Xs7vSSntr9r72ljxqtaun02hRnGWPtrGo6js5dkf4tbfDDJHS1t6PvLYPLjfRj498CfT0r/JPCdVbUe2AL8bv/OJP8eOLQoNRw+8+mzX6mqrwNeALwwyb9blJoOj/n02VXAVmBde21chHoOk+n67P8B/x34sf7EJMcCvwZ8S1V9A/Ah4AcXoZ7DZE591vw34KGqeh5wOvBnC1rD4TSfflvp48BK9AjwrVX1fOAsYGOSc6F3ggZ8O3D/0lVv2ZiyH5N8C7AJ+IaqOgP4lSWs43Ix3Xfyl4CfrqqzgP/R3qvbq4B7+t5vB26uqnXAze29uk3ux5uAM9vfZn8DXLEktVpmkhwD/Cbw7+j9ffayJKcvba2GztU88fzI39vpHQa2VdXXA+cCl7fvlH02tenGWPtreo6jc/MtVXVWVR2ZzDT0/WUwuU9V3VNV906R/ldV9UB7exfw1CTHASR5OvBfgZ9bvJoOj7n2WVX9Y1X9acvzz8AHgDWLV+OlN9c+S7IKOKGqbqneEzOvAS5cvBovvRn67PNV9Rf0An390l5PSxLgBOCByeVH2Tz6DOB7gF9o+b5QVZ9c4GoOnfn020ofB1ai6jly8eDJ7XXkica/CvxE33tNY4Z+fCWws6oeafkeWqIqLhsz9GXRGwMBnskKGwvnI8ka4ALgTX3Jm4DdbXs3K+zvsPmYqh+r6r1Vdbi9vZUVdg5wFM4B9lXVx9r507X0vpNqqurPgU9PSvb3dhpVdbCqPtC2H6YX9FuNfTalGcZY+2sKjqMDMfT9ZTB57v4D8FdHTnCAnwVeB/zj0lVp6E3uMwDarW3fSe9Kix6vv89WAwf69h1oaZpGVT1KLxixl96J8+nAm5e0UkOu71bTn03ygSR/kGRsKeu0jDgOrEDt9r0PAg8BN1XVbUleAnyiqv56aWu3fEzVj8DzgH+d5LYkf5bkG5e0ksvENH35I8AvJ/k4vRnezgTt9gZ6F4S+0Jc2VlUHoReEAU5egnotN2/gif3Y73uAGxetNsvbauDjfe89F5gdf29nIclaencM34Z9Nq1pxlj7a2pvwHF0Lgp4b3pLnW5taUPfXysumJzkfUk+PMWr8+pukjOAXwS+v70/C3huVb1rYWu9tAbZZ33pxwJvB66sqo8tTM2XzoD7bKr1kUduxtvR9NkUx3oyvWDyC4BT6C1zMXIn0IPsM+BYejOE/k9V/QvgFkb01vIBf9fOYgWMA3qiqnqsLR2wBjgnyTfQWyrmfyxpxZaZKfrxTHr/H51I79bbHwf2tLtMNINp+vKVwI9W1anAj+KF1RkleTG95Z7uXOq6LGdd/Zjkv9G7zf6ti1qx5WtFnAto8bW7694B/EhVfW6p6zPMphljNYnj6Ly8sJ1//zt6S878m6Wu0Gwcu9QVWGxV9W3zKdem6r8LuKSq/m9L/ibg7CT76fXlyUkmqmp8EHUdFgPusyN2AR+tqjccZfWG0oD77ACPvw1wDSN4m+p8+2waZ7Vj/l+AJHsYwnWGjtaA++xT9GbWHgmK/gFw2QCPPzQG3G8rYhzQ9Krqs0km6N2Odhrw1y3uuQb4QJJzqurvlrCKy0JfP26kN+69sy3tdHuSLwDPBv5+Cau4bEzqyy301i2E3v/rb5qunAB4IfCSJC8CngqckOT3gAeTrKqqg235MZdemdmU/VhVL0+yBXgxcF77HVe3A8Cpfe9H8lxgAfh7O4M2+eYdwFur6p0t2T7rMGmMtb+eyHF0jo4sdVpVDyV5F72ljYa+v1bczOT5aLd/3wBcUVX/50h6VV1VVadU1VrgXwF/YwChZ7o+a/t+jt66fT+y+DUbXjN8zw4CD6f3UKIAlwDXLU0tl41PAKcneU57/+08/gEAmqSd0P0RMN6SzgPuXrIKLROOAytTkuccWRomyfHAt9Fbmujkqlrbvg8HgH9hIHl60/TjR4D/BXxrS38e8BR6D6nVNGboyweAf9uyfSvw0SWp4DJRVVdU1Zr2O7wZeH9VvRy4nl5gnvbTv8NmMF0/JtkIvBp4SVW5NNTs/SWwLslpSZ5Cr0+vX+I6LQf+3k6jnVO+Gbinql7ft8s+m8IMY6z9NYnj6NwkeVqSZxzZBr6D3gPhh76/4gXhL0nyXcCvA88BPgt8sKrOT/JT9G6R7/8D/Dv6HwiT3lpD766qFXW7w1z7jN4J4cfp/ed7ZA3l36iqFTNTZj7fsyQb6D2l+Hh668v90EqazTFdn7V9++k9XOgpbd93VNXdSX6A3mysR4G/BS6tqk8teuWXyDz77KuA3wWeRW8G4HdX1f2LXfelNJ9+6yu7lhU4DqxEbUmL3cAx9C7M76mqn5mUZz+woVbggyxna7p+bMGSt9C7y+SfgR+rqvcvWUWXgRn68l8Bv0bvzon/B/wXbz2dnSTj9L57L07y5cAe4CuB+4GLqmryw740hUn9uA84jt7dUAC3VtUPLFXdlpM2y+8N9H7H31JVO5a2RsMlydvpTYh4NvAg8Bp6Fyb9vZ1CGxv+N73nyxxZ1/Yn6a2bbJ9NMsMY69gwA8fRbkm+mi/dGXws8Laq2rEc+stgsiRJkiRJkiSpk8tcSJIkSZIkSZI6GUyWJEmSJEmSJHUymCxJkiRJkiRJ6mQwWZIkSZIkSZLUyWCyJEmSJEmSJKmTwWRJkiRJkiRJUieDyZIkSZIkSZKkTgaTJUmSJEmSJEmdDCZLkiRJkiRJkjoZTJYkSZIkSZIkdTKYLEmSJEmSJEnqZDBZkiRJkiRJktTJYLIkSZIkSZKWpSQTSb53nmW/MsmhJMcMul7SqDKYLC1zSa5O8nNLXQ9Jkpar2YylScaTHBjgZ1aS5w7qeJIkqVuS/Um+7cj7qrq/qp5eVY8tZb2k5cRgsrQIJg9Yg8orSdJK4VgqSZIkLT2DydIKlOTYpa6DJEmamrfaSpKWq3ZB94okdyf5TJLfSfLUtu/7kuxL8ukk1yc5pa9cJfnhJB9L8skkv5zkSW3fa5P8Xl/etS3/E85rk3xNkvcn+VQ7zluTPKvt+13gK4E/aktb/MTkYyU5pdXt062u39d37Ncm2ZPkmiQPJ7kryYYF6kppaBlMlhbYNAPWS9rA89m2vtPXT5e3pf9Bkr9L8g9J/jzJGXOsw3iSA0leneTvgN9JclySNyR5oL3ekOS4vjJdA/1/SfLRNoj+bBu0b0nyuTbAPqXlfXaSd7e2fjrJ/z7yR4EkSbMxDGNpX11+sp2c7k9ycV/6cUl+Jcn9SR5M8ltJju/b/+NJDrYx93smHfPqJFcleU+SzwPfkuTrW7s+29r5kr78z2wnsn+f5G+T/FTfCfelSf5Pkl9tZT+W5Jtb+seTPJRkS9+xXpTeCf/DST6R5Mfm0y+SJPW5GDgf+BrgecBPJflW4BeAlwKrgL8Frp1U7ruADcC/ADYB38PcpX3OKcDXA6cCrwWoqlcA9wPf2Za2+KUpyr8dONDK/0fg55Oc17f/Ja3ezwKuB35jHnWUljUDOtICmzxgAf+L3gD1I8BzgPfQO+F9ygyD243AOuBk4APAW+dRla8ATgK+CtgK/DfgXOAs4PnAOcBPAcxyoN8InN2O8RPALnp/NJwKnAm8rOXbRm8wfg4wBvwkUPOovyRphRqysfTZwGpgC7Aryde2fb9I74T5LOC5Lc//AEiyEfgx4NtbHaZaguM/AzuAZwC3AX8EvLfV94eAt/Z91q8DzwS+Gvi3wCXAd/cd618CHwK+HHgbvTH8G1u9Xg78RpKnt7xvBr6/qp5Bb/x+/5x7RZKkx/uNqvp4VX2a3tj2Mnrnim+pqg9U1SPAFcA3JVnbV+4Xq+rTVXU/8Aa+dE45a1W1r6puqqpHqurvgdfTGys7JTkV+FfAq6vq/1XVB4E3Aa/oy/YXVfWetsby79I7l5ZWFIPJ0uL7T8ANbYB7FPgV4Hjgm6crUFVvqaqH26D7WuD5SZ45x8/9AvCaNqj+E73B/Geq6qE2yP40XxokZzvQf66q7gI+DLy3qj5WVf9A74T9BS3fo/QC0l9VVY9W1f+uKoPJkqSjsVRjKcB/b2PpnwE3AC9NEuD7gB9tJ8EPAz8PbG5lXgr8TlV9uKo+3z5/suuq6v9U1RfoBaSfDuysqn+uqvcD7wZelt4SGP8JuKK1Zz/wOh5/ontfVf1OO9H9fXoXen+m1fu9wD/TCyxDb5w+PckJVfWZqvrAPPpEkqR+H+/b/lt6s3xPadsAVNUh4FP0Lr7OVG5Okpyc5Np2t83ngN+jdyF4Nk4Bjozj/fXor+Pf9W3/I/DUuIykVhiDydLimzyIfoHeoLl6qsxJjkmyM8n/bYPh/rZrtgPiEX9fVf9vunrw+MF6NgP9g33b/zTF+yMznn4Z2Ae8t91qu32O9ZYkabKlGks/04LBRxwZO58DfBlwZ1ta4rPAH7f0I/WdfII8Wf/+U4CPt3b1l1nd6vwUnjiGzzRGU1XTjdP/AXgR8LdJ/izJN01RN0mS5uLUvu2vBB5or686kpjkafTuoPlERzmAz9MbZ4/4ihk++xfo3Qn7DVV1Ar07ctK3f6aJTQ8AJyV5xqR6fGKa/NKKZDBZWhz9A9bkQTT0Bs1PTJEXere9bqJ3S+wzgbVHih5FHZ5QDx4/WM9moJ/dh/ZmTW2rqq8GvhP4r5PWnJIkaTaGYSw9sY2JRxwZOz9JL0B7RlU9q72e2ZbkADjIE0+QJ5vcvlMnPWPgyMnsJ+nNJp48hs/rRLeq/rKqNtFbTuN/AXvmcxxJkvpcnmRNkpPoLXP4+/SWXfruJGel96yenwdua3fYHPHjSU5sy028qpUD+CDwb5J8Zbur6IoZPvsZwCHgs0lWAz8+af+D9JaJeoKq+jjw/wG/kOSpSb4BuIz5LY0ljSyDydLi6B+w9gAXJDkvyZPprSn8CL1Ba3Je6A2Gj9CbGfxl9AbdQXg7vQchPCfJs+mt63jkCbmzGehnJcmLkzy3neh/DnisvSRJmothGUt/OslTkvxr4MXAH7QZxL8N/GqSkwGSrE5yfl99L01yepIvA17T8Rm30ZuF9RNJnpxknN4F2Wvb0hV7gB1JnpHkq4D/ypfG8Flr7bg4yTPbciFHxmlJko7G2+it+/+x9vq5qroZ+O/AO+hdZP0avrQc1BHXAXfSCx7fQG9df6rqJnqB5Q+1/e+e4bN/mt4D/P6hHeOdk/b/Ar3z4M9O89DZl9G76PwA8C56S0Xe1NVgaSUxmCwtji8OWPROBl9O7+E5n2zvv7Oq/nly3ja4XUPv9tVPAHcDtw6oTj8H3EFvQN5L72FEPwcwy4F+ttYB76N3dfgW4I1VNXE0FZckrUjDMJb+HfAZeieYbwV+oKo+0va9mt6yTre2pTTeB3wtQFXdSO9BQu9veWZ8yF1rx0uAf9fa90bgkr7P+iF6weaPAX9B76T9LfNs0yuA/a3OP0CvXyVJOhp/WVWntzt1tlTVPwJU1W9V1ddU1UlV9eKqOjCp3Huq6qur6svb3a1fvMBZVZe34z23qn67qlJVh9u+8ap6U9u+q6rOrt5DeM+qqtdV1Zq+41xXVV/ZjvUrVbV/0rEOtLqd1Or6W31lX1tVL+97/7iy0koRn4MlSZIkSZKko5VkP/C9VfW+OZYrYF1V7VuQikkaGGcmS5IkSZIkSZI6GUyWRkSSn0xyaIrXjUtdN0mSlgPHUkmSjk5VrZ3rrORWLs5KlpYHl7mQJEmSJEmSJHU6dqkr0OXZz352rV27diDH+vznP8/Tnva0gRxrqdiG4TEK7bANw2EU2gCj0Y75tOHOO+/8ZFU9Z5D1aGvNPQw8Bhyuqg1JTqL3FOu1wH7gpVX1mZb/CuCylv+Hq+pPWvrZwNXA8cB7gFfVDFeRBznmwmh8J+bC9o422zvabO/ysBBj7lLyXLfbqLYLRrdttmt5GdV2wei2bTHbNd24O/TB5LVr13LHHXcM5FgTExOMj48P5FhLxTYMj1Foh20YDqPQBhiNdsynDUn+dmFqw7dU1Sf73m8Hbq6qnUm2t/evTnI6sBk4AzgFeF+S57WnX18FbAVupRdM3ghMe7v+IMdcGI3vxFzY3tFme0eb7V0eFnDMXRKe63Yb1XbB6LbNdi0vo9ouGN22LWa7pht3XTNZkqTlYROwu23vBi7sS7+2qh6pqvuAfcA5SVYBJ1TVLW028jV9ZSRJkiRJmrOhn5ksSdIKVMB7kxTwP6tqFzBWVQcBqupgkpNb3tX0Zh4fcaClPdq2J6c/TpKt9GYvMzY2xsTExMAacejQoYEeb9jZ3tFme0eb7ZUkSZodg8mSJA2fF1bVAy1gfFOSj8yQN1Ok1Qzpj0/oBap3AWzYsKEGecvUqN5aNh3bO9ps72izvZIkSbPjMheSJA2Zqnqg/XwIeBdwDvBgW7qC9vOhlv0AcGpf8TXAAy19zRTpkiRJkiTNi8FkSZKGSJKnJXnGkW3gO4APA9cDW1q2LcB1bft6YHOS45KcBqwDbm9LYjyc5NwkAS7pKyNJkiRJ0py5zIUkScNlDHhXL/7LscDbquqPk/wlsCfJZcD9wEUAVXVXkj3A3cBh4PKqeqwd65XA1cDxwI3tJUmSJEnSvBhMliRpiFTVx4DnT5H+KeC8acrsAHZMkX4HcOag6yhJkiRJWplc5kKSJEmSJEmS1MlgsiRJkiRJkiSpk8FkSZIkSZIkSVIng8mSJEmSJEmSpE4GkyVJkiRJkiRJnQwmS5IkSZIkSZI6HbvUFVhO1m6/Ydp9+3desIg1kSRptM005oLjriRJc+G4KkkaFGcmS5IkSZIkSZI6GUyWJEmSJEmSJHUymCxJkiRJkiRJ6mQwWZIkSZIkSZLUyWCyJEmSJEmSJKmTwWRJkiRJkiRJUieDyZIkSZIkSZKkTgaTJUmSJEmSJEmdDCZLkiRJkiRJkjoZTJYkSZIkSZIkdTKYLEmSJEmSJEnqZDBZkiRJkiRJktTJYLIkSZIkSZIkqZPBZEmSJEmSJElSJ4PJkiRJkiRJkqROBpMlSZIkSZIkSZ0MJkuSJEmSJEmSOhlMliRJkiRJkiR1MpgsSZIkSZIkSepkMFmSJEmSJEmS1MlgsiRJkiRJkiSpk8FkSZIkSZIkSVIng8mSJEmSJEmSpE4GkyVJkiRJkiRJnQwmS5IkSZIkSZI6GUyWJEmSJEmSJHUymCxJkiRJkiRJ6mQwWZIkSZIkSZLUyWCyJEmSJGlFSvKjSe5K8uEkb0/y1CQnJbkpyUfbzxP78l+RZF+Se5Oc35d+dpK9bd+VSbI0LZIkaWEZTJYkSZIkrThJVgM/DGyoqjOBY4DNwHbg5qpaB9zc3pPk9Lb/DGAj8MYkx7TDXQVsBda118ZFbIokSYtmVsHkJPvbVdYPJrmjpXm1VpIkSZK0nB0LHJ/kWODLgAeATcDutn83cGHb3gRcW1WPVNV9wD7gnCSrgBOq6paqKuCavjKSJI2UY+eQ91uq6pN9749crd2ZZHt7/+pJV2tPAd6X5HlV9Rhfulp7K/AeeldrbxxAOyRJkiRJmrWq+kSSXwHuB/4JeG9VvTfJWFUdbHkOJjm5FVlN71z2iAMt7dG2PTn9CZJspXdOzNjYGBMTEwNpy6FDh2Y81rb1h2csP6h6DFpXu5azUW2b7VpeRrVdMLptG4Z2zSWYPNkmYLxt7wYmgFfTd7UWuC/Jkau1+2lXawGSHLlaazBZkiRJkrSo2t21m4DTgM8Cf5Dk5TMVmSKtZkh/YmLVLmAXwIYNG2p8fHwONZ7exMQEMx3r0u03zFh+/8WDqcegdbVrORvVttmu5WVU2wWj27ZhaNdsg8kFvDdJAf+zDYAjd7W2y0xXcxfrqsAwXIE4WqPQBhiNdtiG4TAKbYDRaMcotEGSJM3atwH3VdXfAyR5J/DNwINJVrXz3FXAQy3/AeDUvvJr6C2LcaBtT06XJGnkzDaY/MKqeqAFjG9K8pEZ8i7bq7VdZrqau1hXcofhCsTRGoU2wGi0wzYMh1FoA4xGO0ahDZIkadbuB85N8mX0lrk4D7gD+DywBdjZfl7X8l8PvC3J6+kt6bgOuL2qHkvycJJzgduAS4BfX9SWSJK0SGYVTK6qB9rPh5K8CzgHr9ZKkiRJkpapqrotyR8CHwAOA39Fb1LT04E9SS6jF3C+qOW/K8ke4O6W//L2bCCAVwJXA8fTW8rR5RwlSSOpM5ic5GnAk6rq4bb9HcDP0Lsq69VaSZIkSdKyVFWvAV4zKfkRerOUp8q/A9gxRfodwJkDr6AkSUNmNjOTx4B3JTmS/21V9cdJ/hKv1kqSJEmSJEnSitAZTK6qjwHPnyL9U3i1VpIkSZIkSZJWhCctdQUkSZIkSZIkScPPYLIkSZIkSZIkqZPBZEmSJEmSJElSp9k8gE+SJGmorN1+w4z79++8YJFqIkmSJEkrhzOTJUmSJEmSJEmdDCZLkiRJkiRJkjoZTJYkSZIkSZIkdTKYLEnSkElyTJK/SvLu9v6kJDcl+Wj7eWJf3iuS7Etyb5Lz+9LPTrK37bsySZaiLZIkSZKk0WEwWZKk4fMq4J6+99uBm6tqHXBze0+S04HNwBnARuCNSY5pZa4CtgLr2mvj4lRdkiRJkjSqDCZLkjREkqwBLgDe1Je8CdjdtncDF/alX1tVj1TVfcA+4Jwkq4ATquqWqirgmr4ykiRJkiTNy7FLXQFJkvQ4bwB+AnhGX9pYVR0EqKqDSU5u6auBW/vyHWhpj7btyelPkGQrvRnMjI2NMTExcfQtaA4dOjTv421bf/ioPnuQ7Zito2nvcmR7R5vtHW0rrb2SJGlwDCZLkjQkkrwYeKiq7kwyPpsiU6TVDOlPTKzaBewC2LBhQ42Pz+ZjZ2diYoL5Hu/S7Tcc1Wfvv3h+n3s0jqa9y5HtHW22d7SttPZKkqTBMZjcZ+1RnrhKknSUXgi8JMmLgKcCJyT5PeDBJKvarORVwEMt/wHg1L7ya4AHWvqaKdIlSZIkSZo310yWJGlIVNUVVbWmqtbSe7De+6vq5cD1wJaWbQtwXdu+Htic5Lgkp9F70N7tbUmMh5OcmyTAJX1lJEmSJEmaF2cmS5I0/HYCe5JcBtwPXARQVXcl2QPcDRwGLq+qx1qZVwJXA8cDN7aXJEmSJEnzZjBZkqQhVFUTwETb/hRw3jT5dgA7pki/Azhz4WooSZIkSVppXOZCkiRJkiRJktTJYLIkSZIkSZIkqZPBZEmSJEmSJElSJ4PJkiRJkiRJkqROBpMlSZIkSZIkSZ0MJkuSJEmSJEmSOhlMliRJkiRJkiR1MpgsSZIkSZIkSepkMFmSJEmSJEmS1MlgsiRJkiRJkiSpk8FkSZIkSZIkSVIng8mSJEmSJEmSpE4GkyVJkiRJkiRJnQwmS5IkSZIkSZI6GUyWJEmSJEmSJHUymCxJkiRJkiRJ6mQwWZIkSZIkSZLUyWCyJEmSJEmSJKmTwWRJkiRJkiRJUieDyZIkSZIkSZKkTgaTJUmSJEmSJEmdDCZLkiRJkiRJkjoZTJYkSZIkSZIkdTKYLEmSJEmSJEnqZDBZkiRJkiRJktTJYLIkSZIkSZIkqZPBZEmSJEmSJElSJ4PJkiRJkiRJkqROsw4mJzkmyV8leXd7f1KSm5J8tP08sS/vFUn2Jbk3yfl96Wcn2dv2XZkkg22OJEmSJEmSJGkhzGVm8quAe/rebwdurqp1wM3tPUlOBzYDZwAbgTcmOaaVuQrYCqxrr41HVXtJkiRJkiRJ0qKYVTA5yRrgAuBNfcmbgN1tezdwYV/6tVX1SFXdB+wDzkmyCjihqm6pqgKu6SsjSZIkSZIkSRpix84y3xuAnwCe0Zc2VlUHAarqYJKTW/pq4Na+fAda2qNte3L6EyTZSm8GM2NjY0xMTMyymjM7dOjQjMfatv7wvI89qDp26WrDcjAKbYDRaIdtGA6j0AYYjXaMQhskSZIkSVooncHkJC8GHqqqO5OMz+KYU62DXDOkPzGxahewC2DDhg01Pj6bj+02MTHBTMe6dPsN8z72/ounP+4gdbVhORiFNsBotMM2DIdRaAOMRjtGoQ2SJEmSJC2U2cxMfiHwkiQvAp4KnJDk94AHk6xqs5JXAQ+1/AeAU/vKrwEeaOlrpkiXJEmSJEmSJA25zjWTq+qKqlpTVWvpPVjv/VX1cuB6YEvLtgW4rm1fD2xOclyS0+g9aO/2tiTGw0nOTRLgkr4ykiRJkiRJkqQhNts1k6eyE9iT5DLgfuAigKq6K8ke4G7gMHB5VT3WyrwSuBo4HrixvSRJkiRJkiRJQ25OweSqmgAm2vangPOmybcD2DFF+h3AmXOtpCRJkiRJkiRpaXUucyFJkiRJkiRJksFkSZIkSZIkSVIng8mSJEmSpBUpybOS/GGSjyS5J8k3JTkpyU1JPtp+ntiX/4ok+5Lcm+T8vvSzk+xt+65sD52XJGnkGEyWJEmSJK1Uvwb8cVV9HfB84B5gO3BzVa0Dbm7vSXI6sBk4A9gIvDHJMe04VwFbgXXttXExGyFJ0mIxmCxJkiRJWnGSnAD8G+DNAFX1z1X1WWATsLtl2w1c2LY3AddW1SNVdR+wDzgnySrghKq6paoKuKavjCRJI+XYpa6AJEmSJElL4KuBvwd+J8nzgTuBVwFjVXUQoKoOJjm55V8N3NpX/kBLe7RtT05/giRb6c1gZmxsjImJiYE05NChQzMea9v6wzOWH1Q9Bq2rXcvZqLbNdi0vo9ouGN22DUO7DCZLkiRJklaiY4F/AfxQVd2W5NdoS1pMY6p1kGuG9CcmVu0CdgFs2LChxsfH51Th6UxMTDDTsS7dfsOM5fdfPJh6DFpXu5azUW2b7VpeRrVdMLptG4Z2ucyFJEmSJGklOgAcqKrb2vs/pBdcfrAtXUH7+VBf/lP7yq8BHmjpa6ZIlyRp5BhMliRJkiStOFX1d8DHk3xtSzoPuBu4HtjS0rYA17Xt64HNSY5Lchq9B+3d3pbEeDjJuUkCXNJXRpKkkeIyF5IkSZKkleqHgLcmeQrwMeC76U262pPkMuB+4CKAqroryR56AefDwOVV9Vg7ziuBq4HjgRvbS5KkkWMwWZIkSZK0IlXVB4ENU+w6b5r8O4AdU6TfAZw50MpJkjSEDCYPyNquBxrsvGCRaiJJkiRJkiRJg+eayZIkSZIkSZKkTgaTJUmSJEmSJEmdDCZLkiRJkiRJkjoZTJYkSZIkSZIkdTKYLEnSkEjy1CS3J/nrJHcl+emWflKSm5J8tP08sa/MFUn2Jbk3yfl96Wcn2dv2XZkkS9EmSZIkSdLoMJgsSdLweAT41qp6PnAWsDHJucB24OaqWgfc3N6T5HRgM3AGsBF4Y5Jj2rGuArYC69pr4yK2Q5IkSZI0ggwmS5I0JKrnUHv75PYqYBOwu6XvBi5s25uAa6vqkaq6D9gHnJNkFXBCVd1SVQVc01dGkiRJkqR5OXapKyBJkr6kzSy+E3gu8JtVdVuSsao6CFBVB5Oc3LKvBm7tK36gpT3atienT/V5W+nNYGZsbIyJiYmBteXQoUPzPt629YeP6rMH2Y7ZOpr2Lke2d7TZ3tG20torSZIGx2CyJElDpKoeA85K8izgXUnOnCH7VOsg1wzpU33eLmAXwIYNG2p8fHxO9Z3JxMQE8z3epdtvOKrP3n/x/D73aBxNe5cj2zvabO9oW2ntlSRJg+MyF5IkDaGq+iwwQW+t4wfb0hW0nw+1bAeAU/uKrQEeaOlrpkiXJEmSJGneDCZLkjQkkjynzUgmyfHAtwEfAa4HtrRsW4Dr2vb1wOYkxyU5jd6D9m5vS2I8nOTcJAEu6SsjSZIkSdK8uMyFJEnDYxWwu62b/CRgT1W9O8ktwJ4klwH3AxcBVNVdSfYAdwOHgcvbMhkArwSuBo4HbmwvSZIkSZLmzWCyJElDoqo+BLxgivRPAedNU2YHsGOK9DuAmdZbliRJkiRpTlzmQpIkSZIkSZLUyWCyJEmSJEmSJKmTwWRJkiRJkiRJUieDyZIkSZIkSZKkTgaTJUmSJEmSJEmdDCZLkiRJkiRJkjoZTJYkSZIkSZIkdTKYLEmSJEmSJEnqdOxSV0CSJK08a7ffsNRVkCRJkiTNkTOTJUmSJEmSJEmdDCZLkiRJkiRJkjoZTJYkSZIkSZIkdTKYLEmSJEmSJEnqZDBZkiRJkiRJktTp2KWugCRJ0qCt3X7DtPv277xgEWsiSZIkSaPDmcmSJEmSJEmSpE4GkyVJkiRJkiRJnQwmS5IkSZIkSZI6dQaTkzw1ye1J/jrJXUl+uqWflOSmJB9tP0/sK3NFkn1J7k1yfl/62Un2tn1XJsnCNEuSJEmSJEmSNEizmZn8CPCtVfV84CxgY5Jzge3AzVW1Dri5vSfJ6cBm4AxgI/DGJMe0Y10FbAXWtdfGwTVFkiRJkiRJkrRQOoPJ1XOovX1yexWwCdjd0ncDF7btTcC1VfVIVd0H7APOSbIKOKGqbqmqAq7pKyNJkiRJkiRJGmLHziZTm1l8J/Bc4Der6rYkY1V1EKCqDiY5uWVfDdzaV/xAS3u0bU9On+rzttKbwczY2BgTExOzbtBMDh06NOOxtq0/PJDPmcpitWE5GIU2wGi0wzYMh1FoA4xGO0ahDZIkSZIkLZRZBZOr6jHgrCTPAt6V5MwZsk+1DnLNkD7V5+0CdgFs2LChxsfHZ1PNThMTE8x0rEu33zCQz5nK/oun/9y56GrDcjAKbYDRaIdtGA6j0AYYjXaMQhskSZIkSVoos1kz+Yuq6rPABL21jh9sS1fQfj7Ush0ATu0rtgZ4oKWvmSJdkiRJkiRJkjTkOoPJSZ7TZiST5Hjg24CPANcDW1q2LcB1bft6YHOS45KcRu9Be7e3JTEeTnJukgCX9JWRJEmSJEmSJA2x2SxzsQrY3dZNfhKwp6reneQWYE+Sy4D7gYsAququJHuAu4HDwOVtmQyAVwJXA8cDN7aXJEmSJEmSJGnIdQaTq+pDwAumSP8UcN40ZXYAO6ZIvwOYab1lSZIkSZIkSdIQmtOayZIkSZIkSZKklclgsiRJkiRJkiSpk8FkSZIkSZIkSVIng8mSJEmSJEmSpE4GkyVJkiRJkiRJnQwmS5IkSZIkSZI6GUyWJEmSJEmSJHUymCxJkiRJkiRJ6nTsUldAkiRJkiQtnbXbb5h23/6dFyxiTSRJw86ZyZIkSZIkSZKkTgaTJUmSJEmSJEmdDCZLkiRJkiRJkjoZTJYkSZIkSZIkdfIBfJIkSZIkLWN7P/EPXDrDQ/QkSRoUZyZLkiRJkiRJkjoZTJYkSZIkSZIkdTKYLEmSJElasZIck+Svkry7vT8pyU1JPtp+ntiX94ok+5Lcm+T8vvSzk+xt+65MkqVoiyRJC81gsiRJkiRpJXsVcE/f++3AzVW1Dri5vSfJ6cBm4AxgI/DGJMe0MlcBW4F17bVxcaouSdLiMpgsSZIkSVqRkqwBLgDe1Je8CdjdtncDF/alX1tVj1TVfcA+4Jwkq4ATquqWqirgmr4ykiSNFIPJkiRJkqSV6g3ATwBf6Esbq6qDAO3nyS19NfDxvnwHWtrqtj05XZKkkXPsUldAkiRJkqTFluTFwENVdWeS8dkUmSKtZkif6jO30lsOg7GxMSYmJmZV1y5jx8O29YcHcqzJBlXH+Th06NCSfv5CGtW22a7lZVTbBaPbtmFol8FkSZIkSdJK9ELgJUleBDwVOCHJ7wEPJllVVQfbEhYPtfwHgFP7yq8BHmjpa6ZIf4Kq2gXsAtiwYUONj48PpCG//tbreN3ehTm933/x+IIcdzYmJiYYVB8Nm1Ftm+1aXka1XTC6bRuGdrnMhSRJkiRpxamqK6pqTVWtpfdgvfdX1cuB64EtLdsW4Lq2fT2wOclxSU6j96C929tSGA8nOTdJgEv6ykiSNFKcmSxJkiRJ0pfsBPYkuQy4H7gIoKruSrIHuBs4DFxeVY+1Mq8ErgaOB25sL0mSRo7BZEmSJEnSilZVE8BE2/4UcN40+XYAO6ZIvwM4c+FqKEnScHCZC0mSJEmSJElSJ4PJkiQNiSSnJvnTJPckuSvJq1r6SUluSvLR9vPEvjJXJNmX5N4k5/eln51kb9t3ZVvDUZIkSZKkeTOYLEnS8DgMbKuqrwfOBS5PcjqwHbi5qtYBN7f3tH2bgTOAjcAbkxzTjnUVsJXew4HWtf2SJEmSJM2bwWRJkoZEVR2sqg+07YeBe4DVwCZgd8u2G7iwbW8Crq2qR6rqPmAfcE6SVcAJVXVLVRVwTV8ZSZIkSZLmxQfwSZI0hJKsBV4A3AaMVdVB6AWck5zcsq0Gbu0rdqClPdq2J6dP9Tlb6c1gZmxsjImJiYG14dChQ9Meb9v6wwP7nLkaZBv7zdTeUWR7R5vtHW0rrb2SJGlwDCZLkjRkkjwdeAfwI1X1uRmWO55qR82Q/sTEql3ALoANGzbU+Pj4nOs7nYmJCaY73qXbbxjY58zV/ovHF+S4M7V3FNne0WZ7R9tKa68kSRocl7mQJGmIJHkyvUDyW6vqnS35wbZ0Be3nQy39AHBqX/E1wAMtfc0U6ZIkSZIkzZvBZEmShkR6U5DfDNxTVa/v23U9sKVtbwGu60vfnOS4JKfRe9De7W1JjIeTnNuOeUlfGUmSJEmS5sVlLiRJGh4vBF4B7E3ywZb2k8BOYE+Sy4D7gYsAququJHuAu4HDwOVV9Vgr90rgauB44Mb2kiRJkiRp3gwmS5I0JKrqL5h6vWOA86YpswPYMUX6HcCZg6udJEmSJGmlc5kLSZIkSZIkSVIng8mSJEmSJEmSpE4GkyVJkiRJkiRJnQwmS5IkSZIkSZI6GUyWJEmSJEmSJHUymCxJkiRJkiRJ6mQwWZIkSZIkSZLUyWCyJEmSJEmSJKmTwWRJkiRJkiRJUqfOYHKSU5P8aZJ7ktyV5FUt/aQkNyX5aPt5Yl+ZK5LsS3JvkvP70s9OsrftuzJJFqZZkiRJkiRJkqRBOnYWeQ4D26rqA0meAdyZ5CbgUuDmqtqZZDuwHXh1ktOBzcAZwCnA+5I8r6oeA64CtgK3Au8BNgI3DrpRw2jt9htm3L9/5wWLVBNJkiRJkiRJmrvOmclVdbCqPtC2HwbuAVYDm4DdLdtu4MK2vQm4tqoeqar7gH3AOUlWASdU1S1VVcA1fWUkSZIkSZIkSUNsNjOTvyjJWuAFwG3AWFUdhF7AOcnJLdtqejOPjzjQ0h5t25PTp/qcrfRmMDM2NsbExMRcqjmtQ4cOzXisbesPD+Rz5mO2bexqw3IwCm2A0WiHbRgOo9AGGI12jEIb1M27hSRJkiRpfmYdTE7ydOAdwI9U1edmWO54qh01Q/oTE6t2AbsANmzYUOPj47Ot5owmJiaY6ViXdpxcLqT9F4/PKl9XG5aDUWgDjEY7bMNwGIU2wGi0YxTaoP+/vX8Pl/Ss60Tv748EQjhEEpFlSMfp4ARn56AcemKU2b49RiQahuS99uCECZJonLzDRMGZdkNHrxl078m72xlhFBV8M4AJCoQM4CSbgBCjS7bXACGcDEnI0JAWmrRp5KBpxol0uN8/6ulQ6a61avU61Kp66vO5rrrWU/dzqPv31Kq66/nVXfcNAADARhk7zEWSVNWjM0gkv6W19q6u+P5u6Ip0f/d35XuTnDq0+5Yk93XlW0aUAwAAAAAw5cYmk2vQBfmNSe5urb1maNVNSS7tli9NcuNQ+cVVdVxVnZbk9CS3dUNiPFBV53bHfMnQPgAAAAAATLGVDHPxnCQ/meSOqvpEV/aLSXYluaGqLk/y+SQvTJLW2p1VdUOSu5IcTHJla+2hbr+XJrk2yfFJ3tvdAAAAAACYcmOTya21P8vo8Y6T5Lwl9rk6ydUjym9PctbRVBAAAAAAgM23ojGTAQAAAACYb5LJAAAAAACMJZkMAAAAAMBYkskAAAAAAIwlmQwAAAAAwFiSyQAAAAAAjCWZDAAAAADAWJLJAAAAAACMJZkMAAAAAMBYkskAAAAAAIwlmQwAAAAAwFiSyQAAAAAAjCWZDAAAAADAWJLJAAAAAACMJZkMAAAAAMBYkskAAAAAAIwlmQwAAAAAwFiSyQAAAAAAjCWZDAAAAADAWJLJAAAAAACMJZkMAAAAAMBYx252BSbpji/+dS7befNmVwMAAAAAYObomQwAAAAAwFiSyQAAAAAAjCWZDAAAAADAWJLJAAAAAACMJZkMAAAAAMBYkskAAAAAAIwlmQwAAAAAwFiSyQAAAAAAjCWZDAAAwNypqlOr6k+q6u6qurOqXt6Vn1RVt1TVZ7q/Jw7tc1VV7a6qe6rqeUPlz66qO7p1r62q2oyYAGCjSSYDAAAwjw4m2dFa+1+SnJvkyqo6I8nOJLe21k5Pcmt3P926i5OcmeT8JK+rqmO6Y70+yRVJTu9u508yEACYFMlkAAAA5k5rbV9r7WPd8gNJ7k5ySpILk1zXbXZdkou65QuTXN9ae7C1dm+S3UnOqaqTk5zQWvtga60lefPQPgDQK8dudgUAAABgM1XV1iTPTPLhJAuttX3JIOFcVU/pNjslyYeGdtvblX2jWz68fNTjXJFBD+YsLCxkcXFxXeq/cHyy4+yD63Ksw61XHVfjwIEDm/r4G6mvsYlrtvQ1rqS/sU1DXJLJADBFqupNSZ6fZH9r7ayu7KQkb0+yNcmeJD/RWvtqt+6qJJcneSjJy1pr7+vKn53k2iTHJ3lPkpd3vaUYY+vOm5ddv2fXBROqCQCTUFVPSPLOJD/fWvubZYY7HrWiLVN+ZGFr1yS5Jkm2bdvWtm/fftT1HeU333JjXn3Hxlze77lk+4YcdyUWFxezXudo2vQ1NnHNlr7GlfQ3tmmIyzAXADBdrs2R4ywauxEANkBVPTqDRPJbWmvv6orv74auSPd3f1e+N8mpQ7tvSXJfV75lRDkA9I5kMgBMkdbaB5J85bBiYzcCwDqrQRfkNya5u7X2mqFVNyW5tFu+NMmNQ+UXV9VxVXVaBl/W3tYNifFAVZ3bHfMlQ/sAQK8Y5gIApt+Gjd0IAHPsOUl+MskdVfWJruwXk+xKckNVXZ7k80lemCSttTur6oYkdyU5mOTK1tpD3X4vzbeGl3pvdwOA3pFMBoDZteaxGzdqIqBk+ckhNmqSoElYKqZpmAxjksTbb+Ltt3mLdymttT/L6DYzSc5bYp+rk1w9ovz2JGetX+0AYDpJJgPA9Lu/qk7ueiWv69iNGzURULL85BCXjZnkbpotNRHRNEyGMUni7Tfx9tu8xQsArB9jJgPA9DN2IwAAAJtOz+QpsXVMD609uy6YUE0A2ExV9bYk25M8uar2JnlVjN0IAADAFJBMBoAp0lp70RKrZm7sxju++NczPZwFAAAAj2SYCwAAAAAAxhqbTK6qN1XV/qr61FDZSVV1S1V9pvt74tC6q6pqd1XdU1XPGyp/dlXd0a17bTeGIwAAAAAAM2AlPZOvTXL+YWU7k9zaWjs9ya3d/VTVGUkuTnJmt8/rquqYbp/XJ7kig8mBTh9xTAAAAAAAptTYZHJr7QNJvnJY8YVJruuWr0ty0VD59a21B1tr9ybZneScqjo5yQmttQ+21lqSNw/tAwAAAADAlFvtBHwLrbV9SdJa21dVT+nKT0nyoaHt9nZl3+iWDy8fqaquyKAXcxYWFrK4uLjKah5W6eOTHWcfXJdjTdqhc3DgwIF1Ox+bpQ8xJP2IQwzToQ8xJP2Iow8xAACsp61jJtPds+uCCdUEgGmw2mTyUkaNg9yWKR+ptXZNkmuSZNu2bW379u3rUrnffMuNefUd6x3yZOy5ZHuSQVJ5vc7HZulDDEk/4hDDdOhDDEk/4uhDDAAAALBRVjJm8ij3d0NXpPu7vyvfm+TUoe22JLmvK98yohwAAAAAgBmw2mTyTUku7ZYvTXLjUPnFVXVcVZ2WwUR7t3VDYjxQVedWVSV5ydA+AAAAAABMubFjPlTV25JsT/Lkqtqb5FVJdiW5oaouT/L5JC9MktbanVV1Q5K7khxMcmVr7aHuUC9Ncm2S45O8t7sBAAAAADADxiaTW2svWmLVeUtsf3WSq0eU357krKOqHQAAAAAAU2G1w1wAAAAAADBHJJMBAAAAABhLMhkAAAAAgLEkkwEAAAAAGEsyGQAAAACAsSSTAQAAAAAYSzIZAAAAAICxJJMBAAAAABjr2M2uAADALNm68+aR5TvOPpjLdt6cPbsumHCNAAAAJkPPZAAAAAAAxpJMBgAAAABgLMlkAAAAAADGkkwGAAAAAGAsyWQAAAAAAMaSTAYAAAAAYKxjN7sCrMzWnTcnSXacfTCXdcuH7Nl1wWZUCQAAAACYI3omAwAAAAAwlmQyAAAAAABjSSYDAAAAADCWZDIAAAAAAGOZgA8AYB1tPWyi3GEmzQUAAGaZnskAAAAAAIwlmQwAAAAAwFiGuQAAAABWZbnhnRJDPAH0jZ7JAAAAAACMpWdyD/gmGAAAAADYaJLJAAAT4gtgAABglhnmAgAAAACAsSSTAQAAAAAYSzIZAAAAAICxJJMBAAAAABhLMhkAAAAAgLGO3ewKsPHMHA8As0GbDQAATDM9kwEAAAAAGEsyGQAAAACAsQxzAQAAAGwIQzgB9ItkMgDAjFjugtzFOAAAsNEMcwEAAAAAwFh6JqOXEwD0gJ8RAwAAG03PZAAAAAAAxtIzmWXp5QQAAMBGWe6a89rzHz/BmgCwEpLJAABzwBfEAADAWkkmsybjLkxH2XH2wVzW7efCFQCmg2QzAAAwzsSTyVV1fpLfSHJMkje01nZNug4AMA+0uaynw5PNvhwGeCTt7vq744t//XBbM4r2B2DyJppMrqpjkvx2kucm2ZvkI1V1U2vtrknWg+mxXC8oHwwAVk+byySt5pdKw8a1+T4vANNOuwvAvJh0z+RzkuxurX0uSarq+iQXJtHAcoS1XpiO4+IT6DltLjNjLW3+RieyAVZIu7sJNvKaUfsAMNqkk8mnJPnC0P29Sb7/8I2q6ookV3R3D1TVPev0+E9O8lfrdKxN8TIxrJv61TUfYiriWCMxTIc+xJD0I47VxPD3NqIi62Cz29ykH/8TKzYt7duk9CXeo/g80It4j4J4+21W453WNjfZ/HZ3Vp/TZW1mW7MO14vj9PI5i7hmTV/jSvob2yTjGtnuTjqZXCPK2hEFrV2T5Jp1f/Cq21tr29b7uJMkhunRhzjEMB36EEPSjzj6EMOQTW1zk96dz7HE22/i7Tfxsg5c626AvsaV9Dc2cc2WvsaV9De2aYjrURN+vL1JTh26vyXJfROuAwDMA20uAEyOdheAuTDpZPJHkpxeVadV1WOSXJzkpgnXAQDmgTYXACZHuwvAXJjoMBettYNV9bNJ3pfkmCRvaq3dOcEqbMjPeCdMDNOjD3GIYTr0IYakH3H0IYYkU9HmJj06nysk3n4Tb7+JlzWZgna3r89pX+NK+hubuGZLX+NK+hvbpsdVrR0xjBMAAAAAADzCpIe5AAAAAABgBkkmAwAAAAAw1lwkk6vq/Kq6p6p2V9XOza7P4apqT1XdUVWfqKrbu7KTquqWqvpM9/fEoe2v6mK5p6qeN1T+7O44u6vqtVVVG1zvN1XV/qr61FDZutW7qo6rqrd35R+uqq0TiuGXq+qL3fPxiar68SmP4dSq+pOquruq7qyql3flM/NcLBPDzDwXVfXYqrqtqj7ZxfArXfksPQ9LxTAzz8PQ4x9TVR+vqnd392fmeeiDmvJ2d6Vqg9u5abPMe3EvY57E+/Y02sj3x2lTM/oZe7Wq6klV9Y6q+nT3Ov6BPsfLwCy2uX16bVYPromPIq6ZuyYYEdfMXzsfZVwz/ZxVD66xVxHbbDxnrbVe3zKY/OCzSZ6W5DFJPpnkjM2u12F13JPkyYeV/YckO7vlnUl+tVs+o4vhuCSndbEd0627LckPJKkk703yYxtc7x9K8qwkn9qIeif5V0l+p1u+OMnbJxTDLyf5hRHbTmsMJyd5Vrf8xCT/vavrzDwXy8QwM89F93hP6JYfneTDSc6dsedhqRhm5nkYqtu/SfLWJO/u7s/M8zDrt8xAu3sUsWxoOzdtt0ygPZmm2zLveb2MdyjuDXt/nLZbZvQz9hrivS7Jz3TLj0nypD7H6za7bW6fXpvpwTXxUcT1y5mxa4IRdZ35a+ejjGumn7P04Bp7FbHNxHM2Dz2Tz0myu7X2udba3yW5PsmFm1ynlbgwgw+E6f5eNFR+fWvtwdbavUl2Jzmnqk5OckJr7YNt8J/y5qF9NkRr7QNJvrKB9R4+1juSnHfoG5YNjmEp0xrDvtbax7rlB5LcneSUzNBzsUwMS5nGGFpr7UB399HdrWW2noelYljK1MWQJFW1JckFSd5wWF1n4nnogVltd48wgXZuqkyoPZkaE3rfnioTeH+cBb2Mt6pOyCD588Ykaa39XWvta+lpvDysN21uZvR/tQ/XxKP04Tp5lD5cOx9lXEuZlbhm/hp7KbN+7T0PyeRTknxh6P7eLP+i2gwtyfur6qNVdUVXttBa25cM3hiSPKUrXyqeU7rlw8snbT3r/fA+rbWDSf46ybdvWM0f6Wer6s9r8POeEw+vz2F1nZoYup8tPDODb7Vm8rk4LIZkhp6LGvx0+BNJ9ie5pbU2c8/DEjEkM/Q8JPn1JK9I8s2hspl6HmbcLLS7azGr7fNR2cD2ZKpM4H172vx6Nvb9cdr06TP2OE9L8qUkv1uDYUzeUFWPT3/jZWBW29y+vzb7/Llzlq4JltWHa+dRZvl6epQ+XGMvZZavvechmTwq675ctn8zPKe19qwkP5bkyqr6oWW2XSqeaY9zNfXerJhen+S7kzwjyb4krx5Tn6mIoaqekOSdSX6+tfY3y226RJ02PY4RMczUc9Fae6i19owkWzL4lvCsZTafpRhm5nmoqucn2d9a++hKd1miPpv+ephh83p+ZrV9PsIGtydTZQLv21NjQu+P02YePmMfcmwGP0l/fWvtmUm+nsFPf5cy6/EyMKvP1zy9NofN+ufOmbkmGKcP186jzPr19Ch9uMZeyixfe89DMnlvklOH7m9Jct8m1WWk1tp93d/9Sf4gg58r3d91V0/3d3+3+VLx7O2WDy+ftPWs98P7VNWxSb4tK/+pzaq11u7vXtTfTPKfM3g+HlGfw+q66TFU1aMzaDTe0lp7V1c8U8/FqBhm8bno6v21JItJzs+MPQ+jYpix5+E5SV5QVXsy+KnnD1fV72dGn4cZNfXt7hrNavu8IhNoT6bSBr5vT5NJvD9OlZ59xh5nb5K9Q72a3pFBcrmv8TIwk23uHLw2e/m5c8auCZbUh2vnUfp0PT1KH66xlzKL197zkEz+SJLTq+q0qnpMBoNO37TJdXpYVT2+qp54aDnJjyb5VAZ1vLTb7NIkN3bLNyW5uAazMp6W5PQkt3Vd+x+oqnO7MVBeMrTPJK1nvYeP9U+T/HE3BsyGOvSm1Pl/Z/B8TG0M3WO+McndrbXXDK2amediqRhm6bmoqu+oqid1y8cn+ZEkn85sPQ8jY5il56G1dlVrbUtrbWsG7/d/3Fp7cWboeeiBqW5318Gsts9jTag9mRoTet+eGhN6f5waPfyMvazW2l8m+UJVfU9XdF6Su9LTeHnYzLW5c/La7OXnzlm6Jlgmhpm/dh6lD9fTo/ThGnspM3/t3SYwS+Fm35L8eAazWX42yS9tdn0Oq9vTMpiR8ZNJ7jxUvwzGMbk1yWe6vycN7fNLXSz3ZGjG2iTbun+0zyb5rSS1wXV/Wwbd7r+RwTcel69nvZM8Nsl/yWBg8duSPG1CMfxekjuS/HkGL76TpzyGf5TBTxX+PMknutuPz9JzsUwMM/NcJPneJB/v6vqpJP9uvV/LmxjDzDwPh8WzPcm7Z+156MMtU9zuHmUcG9rOTdstE2hPpum2zHteL+M9LPYNeX+cpltm+DP2GmJ+RpLbu//p/5rkxD7H6/bw8zVTbW7fXpvpwTXxUcQ1k9cEh8U189fORxnXTD9n6cE19ipim4nn7NADAAAAAADAkuZhmAsAAAAAANZIMhkAAAAAgLEkkwEAAAAAGEsyGQAAAACAsSSTAQAAAAAYSzIZAAAAAICxJJMBAAAAABhLMhkAAAAAgLEkkwEAAAAAGEsyGQAAAACAsSSTAQAAAAAYSzIZAAAAAICxJJNhQqpqT1X9yGbXI0mq6tqq+ver3Hexqn5miXVbq6pV1bFrqyEAAAAA00YyGQCAuVRVv1xVv98tf1dVHaiqYzbw8X6nqv7tRh0fADbSJrSbq+4ENSuW66wF00rvQQAA5l5r7fNJnrDBj/EvN/L4ADApk2g3gemkZzJM1jOq6s+r6q+r6u1V9dgkqap/UVW7q+orVXVTVT21Kz9i2Ijhby6r6u9X1Z92x/urqnr70Hb/oKpu6Y55T1X9xGF1ObGqbq6qB6rqw1X13UP7/mBVfaQ77keq6gdHBVNVx1TVr3WP/bkkFxy2/rKq+lz3GPdW1SVrPYEAAAAAbA7JZJisn0hyfpLTknxvksuq6oeT/F/dupOT/EWS61d4vP8zyfuTnJhkS5LfTJKqenySW5K8NclTkrwoyeuq6syhfV+U5Fe6fXcnubrb96QkNyd5bZJvT/KaJDdX1bePePx/keT5SZ6ZZFuSf3poRVeH1yb5sdbaE5P8YJJPrDAuAHhYN+/A/959Ifv1qnpjVS1U1Xu7Lyz/qKpO7LY9t6r+W1V9rao+WVXbh45zWvcl7ANVdUuSJw+te8QXuFX1U1V1d7ft56rq/zO07faq2ltVO6pqf1Xtq6qfWkEcD/9cd9wxqur4qnp1Vf1F9+Xun1XV8d26F1TVnV2Mi1X1v6zmXI07XwDMpr60m50TaxWdoOqwOYvqkUN0PLaqfr+qvtzF/ZGqWujWfVt3vvZV1Rer6t/XMkN5VNVx3THOGir7jqr626p6SlWdWFXvrqovVdVXu+UtSxzr4ToucY6Pqm6wUSSTYbJe21q7r7X2lST/d5JnJLkkyZtaax9rrT2Y5KokP1BVW1dwvG8k+XtJntpa+5+ttT/ryp+fZE9r7Xdbawdbax9L8s4MJXuTvKu1dltr7WCSt3R1SQa9iz/TWvu9bt+3Jfl0kn8y4vF/Ismvt9a+0MX0fx22/ptJzqqq41tr+1prd64gJgAY5X9L8twkT8+gTXpvkl/M4ML2UUleVlWnZPCF6L9PclKSX0jyzqr6ju4Yb03y0W6f/zPJpcs83v4M2tMTkvxUkv9UVc8aWv+dSb4tySlJLk/y28NJ2hVa7hi/luTZGXwZe1KSVyT5ZlU9Pcnbkvx8ku9I8p4k/3dVPWbouGPPVZKs4HwBMLv60m6uRyeow13a1eXUbt9/meRvu3XXJTmY5O9n0GnqR5MsOaZxdw3/rq6eh/xEkj9tre3P4Fz/bgbX7d/VPc5vraCOoxxV3WCjSCbDZP3l0PL/yGCMqadm0Bs5SdJaO5Dkyxk0suO8Ikklua3rofTTXfnfS/L93TekX6uqr2WQtP7OMXXJ4fXp/MUS9Xlqki8ctt2hOL6e5J9l0DDv675N/gcriAkARvnN1tr9rbUvJvl/kny4tfbx7iLuDzK4qHpxkve01t7TWvtma+2WJLcn+fGq+q4k/zDJv22tPdha+0AGX+yO1Fq7ubX22Tbwpxn8Euh/HdrkG0n+j9baN1pr70lyIMn3HGVMI49RVY9K8tNJXt5a+2Jr7aHW2n/rYv1nSW5urd3SWvtGBknn4zNIOh/Nucpy5+so4wBg+vSl3VyPTlCH+0YGSeS/37WxH22t/U3XO/nHkvx8a+3rXTL4PyW5eMzx3ppHJpP/eVeW1tqXW2vvbK39j9baAxkkw/9fK6jjI6yhbrDuTMAHm+++DJK/SR4eHuLbk3wxyde74scl+Ztu+eGEcGvtLzMYaiJV9Y+S/FFVfSCDBO+fttaeu9b6dL4ryR+O2HZfBt/mDm/3sNba+5K8rwY/y/33Sf5zHvmBAgBW6v6h5b8dcf8JGbRfL6yq4QvJRyf5kwy+AP1q92XnIX+RR7ZjD6uqH0vyqgx6dD0qg7b4jqFNvtxd2B4y/MXsSi11jCcneWySz47Y5/Avob9ZVV/II7/0Xcm5SpY/XwDMtr60m+vRCepwv5dBHNdX1ZOS/H6SX8rgfDw6g85Qh7Z9VB7ZgWqUP05yfFV9f1ffZ2SQsE9VPS6DpO/5GfSuTpInVtUxrbWHVlDXQ1ZbN1h3eibD5ntrkp+qqmdU1XFJ/r8ZfGu8p7X2pQySyi+uwWR3P51keIyoFw6Nt/TVJC3JQ0neneTpVfWTVfXo7vYPa2hMxWW8p9v3n1fVsVX1z5Kc0R3zcDdk8POoLd1PlHYO1W2hBmM6Pj7Jgxl883w0jSUAHK0vJPm91tqThm6Pb63tyuAL0BO7dumQ7xp1kK49fmcGvX4XWmtPyqB9rFHbb4C/SvI/M9TmDzn8S+jK4IL4i6t4nOXOFwD9N8vt5lKdoA61h1/PIKF9yHCnrG+01n6ltXZGBr/seX6Sl2RwPh5M8uSh83FCa2147qEjtNa+mcG18Ysy6JX87q4XcpLsyKAH9ve31k5I8kNd+ahzs2SdV1s32AiSybDJWmu3Jvm3GTS++zK4cBz+qcq/SPK/ZzD0xZlJ/tvQun+Y5MNVdSDJTRn8HPberuH60e4492Xw7eivJjluBfX5cgaN6Y7uMV+R5Pmttb8asfl/TvK+JJ9M8rEMxoo65FHdMe5L8pUMfsrzr8Y9PgCswe8n+SdV9bzuS9jH1mDSny2ttb/I4Ke7v1JVj+l+0bPUT2Efk0Gb+aUkB7veVj86kQjy8EXpm5K8pqqe2sXyA93F+g1JLqiq86rq0Rm0tQ/mkZ8PVmrJ87VuwQAwzWa53RzXCeoTSS7uOlYdPln8P66qs7vJ6/4mg2EvHmqt7ctgeI5XV9UJVfWoqvruqlrJsBRvzWAoqku65UOemEFP8K914zy/apljfCLJD1XVd1XVt2Uwn1KSZI11g3VlmAuYkNba1sPu//LQ8u8k+Z0l9ntvktOWWPeKDJK9o9bdk8E4UqPWXXbY/cUkW4bu/1kGk/6M2nf70PLBJP+6ux3y293ffVnFWFAAsFqttS9U1YVJ/kMGk9Q9lOS2JC/tNvnnGUxe85UkH0zy5iRPGnGcB6rqZRkkbo/LYIzImza6/of5hQwmtv1IBj/p/WSS57XW7qmqFyf5zQx+yvuJJP+ktfZ3R/sAKzhfAPTYLLebrbUvV9Xzk/xGktdnMDnfcCeof5tBTF9N8qcZJHhP6tZ9ZwbX31sy+AXt2zNIrCeDHsq7ktyVQSL4cxl0zBpXnw9X1dczGH7jvUOrfr177L/KoKPVq5NctMQxbqmqtyf58277X03ygqFNVlU3WG/VWtvsOgAAAAAAMOUMcwEAAAAAwFiSyQAAsE6q6s6qOjDidslm1w0Aps2stJtV9TtL1HPkcJXQZ4a5AAAAAABgrKmfgO/JT35y27p167oc6+tf/3oe//jHr8uxZtE8xz/PsSfin+f45zn2ZDLxf/SjH/2r1tp3bOiDTMhSbW5f/4/6GJeYZkcf4+pjTEk/45rVmPrU5ibLX+vO6nM0LZy/tXH+1sb5Wz3nbm3W+/wt1e5OfTJ569atuf3229flWIuLi9m+ffu6HGsWzXP88xx7Iv55jn+eY08mE39V/cWGPsAELdXm9vX/qI9xiWl29DGuPsaU9DOuWY2pT21usvy17qw+R9PC+Vsb529tnL/Vc+7WZr3P31LtrjGTAQAAAAAYa0XJ5KraU1V3VNUnqur2ruykqrqlqj7T/T1xaPurqmp3Vd1TVc8bKn92d5zdVfXaqqr1DwkAAAAAgPV2ND2T/3Fr7RmttW3d/Z1Jbm2tnZ7k1u5+quqMJBcnOTPJ+UleV1XHdPu8PskVSU7vbuevPQQAAAAAADbaWoa5uDDJdd3ydUkuGiq/vrX2YGvt3iS7k5xTVScnOaG19sHWWkvy5qF9AAAAAACYYiudgK8leX9VtST/v9baNUkWWmv7kqS1tq+qntJte0qSDw3tu7cr+0a3fHj5Earqigx6MGdhYSGLi4srrObyDhw4sG7HmkXzHP88x56If57jn+fYE/EDAADAelppMvk5rbX7uoTxLVX16WW2HTUOclum/MjCQbL6miTZtm1bW6+ZCOd9Vsh5jn+eY0/EP8/xz3PsifgBAABgPa1omIvW2n3d3/1J/iDJOUnu74auSPd3f7f53iSnDu2+Jcl9XfmWEeUAAAAAAEy5scnkqnp8VT3x0HKSH03yqSQ3Jbm02+zSJDd2yzclubiqjquq0zKYaO+2bkiMB6rq3KqqJC8Z2gcAAAAAgCm2kmEuFpL8wSD/m2OTvLW19odV9ZEkN1TV5Uk+n+SFSdJau7OqbkhyV5KDSa5srT3UHeulSa5NcnyS93Y3AAAAAACm3Nhkcmvtc0m+b0T5l5Oct8Q+Vye5ekT57UnOOvpqAgAAAACwmVY0ZjIAAAAAAPNNMhkAAAAAgLEkkwEAAAAAGGslE/DR2brz5iXX7dl1wQRrAkBfVdWbkjw/yf7W2llD5T+X5GczmNz25tbaK7ryq5JcnuShJC9rrb2vK392vjXp7XuSvLy11iYYCiMs91ki8XkCAAAYbdy1xLXnP34i9dAzGQCmy7VJzh8uqKp/nOTCJN/bWjszya915WckuTjJmd0+r6uqY7rdXp/kiiSnd7dHHBMAAACOlmQyAEyR1toHknzlsOKXJtnVWnuw22Z/V35hkutbaw+21u5NsjvJOVV1cpITWmsf7HojvznJRRMJAAAAgN4yzAUATL+nJ/lfq+rqJP8zyS+01j6S5JQkHxrabm9X9o1u+fDyI1TVFRn0YM7CwkIWFxeP2ObAgQMjy2fdZsS14+yDy65fa336+Fz1Maakn3H1Maakn3H1MSYAYDIkkwFg+h2b5MQk5yb5h0luqKqnJakR27Zlyo8sbO2aJNckybZt29r27duP2GZxcTGjymfdZsR12bgxky/Zvqbj9/G56mNMST/j6mNMST/j6mNMAMBkSCYDwPTbm+Rd3ZAVt1XVN5M8uSs/dWi7LUnu68q3jChnHZhEDwAAmFfGTAaA6fdfk/xwklTV05M8JslfJbkpycVVdVxVnZbBRHu3tdb2JXmgqs6tqkrykiQ3bkrNAWBKVdWbqmp/VX1qqOykqrqlqj7T/T1xaN1VVbW7qu6pqucNlT+7qu7o1r22a3sBoJckkwFgilTV25J8MMn3VNXeqro8yZuSPK272L0+yaVt4M4kNyS5K8kfJrmytfZQd6iXJnlDBpPyfTbJeyccCgBMu2uTnH9Y2c4kt7bWTk9ya3c/VXVGkouTnNnt87qqOqbb5/UZzD9wenc7/JgA0BuGuQCAKdJae9ESq168xPZXJ7l6RPntSc5ax6oBQK+01j5QVVsPK74wyfZu+boki0le2ZVf31p7MMm9VbU7yTlVtSfJCa21DyZJVb05yUXxJS4APSWZDAAAAAML3XBRaa3tq6qndOWnJPnQ0HZ7u7JvdMuHl49UVVdk0Is5CwsLWVxcHLndgQMHllzHeM7f2jh/a+P8rZ5zt7wdZx9cdv2kzp9kMgAAACxv1DjIbZnykVpr1yS5Jkm2bdvWtm/fPnK7xcXFLLWO8Zy/tXH+1sb5Wz3nbnmXjZkI/NrzHz+R82fMZAAAABi4v6pOTpLu7/6ufG+SU4e225Lkvq58y4hyAOglyWQAAAAYuCnJpd3ypUluHCq/uKqOq6rTMpho77ZuSIwHqurcqqokLxnaBwB6xzAXAAAAzJ2qelsGk+09uar2JnlVkl1Jbqiqy5N8PskLk6S1dmdV3ZDkriQHk1zZWnuoO9RLk1yb5PgMJt4z+R4AvSWZDAAAwNxprb1oiVXnLbH91UmuHlF+e5Kz1rFqADC1DHMBAAAAAMBYkskAAAAAAIxlmAsAgHW0defNm10FAACADaFnMgAAAAAAY+mZDAAwJcb1at6z64IJ1QQAAOBIkskAAAAAU8KXy8A0M8wFAAAAAABjSSYDAAAAADCWZDIAAAAAAGNJJgMAAAAAMJZkMgAAAAAAY0kmAwAAAAAwlmQyAAAAAABjSSYDAAAAADCWZDIAAAAAAGNJJgMAAAAAMJZkMgAAAAAAYx272RUAAAAAmBdbd9682VUAWDU9kwEAAAAAGEsyGQAAAACAsSSTAQAAAAAYSzIZAAAAAICxJJMBYIpU1Zuqan9VfWrEul+oqlZVTx4qu6qqdlfVPVX1vKHyZ1fVHd2611ZVTSoGAAAA+kkyGQCmy7VJzj+8sKpOTfLcJJ8fKjsjycVJzuz2eV1VHdOtfn2SK5Kc3t2OOCYAAAAcDclkAJgirbUPJPnKiFX/KckrkrShsguTXN9ae7C1dm+S3UnOqaqTk5zQWvtga60leXOSiza25gAAAPTdsZtdAQBgeVX1giRfbK198rDRKk5J8qGh+3u7sm90y4eXjzr2FRn0YM7CwkIWFxeP2ObAgQMjy2fdauPacfbB9a/MCi1X3zu++NdZOD75zbfcOHL92ad82wbVamP5/5sdfYwp6WdcfYwJAJgMyeR1snXnzcuu37PrggnVBIA+qarHJfmlJD86avWIsrZM+ZGFrV2T5Jok2bZtW9u+ffsR2ywuLmZU+axbbVyXjWnzN9KeS7Yvue6ynTdnx9kH8+o7Rn+8W27faeb/b3b0Maakn3H1MSYAYDIkkwFgun13ktOSHOqVvCXJx6rqnAx6HJ86tO2WJPd15VtGlAMAAMCqSSYDwBRrrd2R5CmH7lfVniTbWmt/VVU3JXlrVb0myVMzmGjvttbaQ1X1QFWdm+TDSV6S5DcnX3sAgI2z1C+Ed5x9MJftvNkvhAE2gAn4AGCKVNXbknwwyfdU1d6qunypbVtrdya5IcldSf4wyZWttYe61S9N8oYMJuX7bJL3bmjFAQAA6D09kwFgirTWXjRm/dbD7l+d5OoR292e5Kx1rRwAAABzTc9kAAAAAADG0jMZAGDIUuMvAgAAzDs9kwEAAAAAGGvFyeSqOqaqPl5V7+7un1RVt1TVZ7q/Jw5te1VV7a6qe6rqeUPlz66qO7p1r62qWt9wAAAAAADYCEfTM/nlSe4eur8zya2ttdOT3NrdT1WdkeTiJGcmOT/J66rqmG6f1ye5Isnp3e38NdUeAAAAAICJWNGYyVW1JckFGcwW/2+64guTbO+Wr0uymOSVXfn1rbUHk9xbVbuTnFNVe5Kc0Fr7YHfMNye5KMl71yEOAIAVOTQm8o6zD+Yy4yMDAACs2Ep7Jv96klck+eZQ2UJrbV+SdH+f0pWfkuQLQ9vt7cpO6ZYPLwcAAAAAYMqN7ZlcVc9Psr+19tGq2r6CY44aB7ktUz7qMa/IYDiMLCwsZHFxcQUPO96BAwfWdKwdZx9c9b7rFcNarDX+WTbPsSfin+f45zn2RPwAAACwnlYyzMVzkrygqn48yWOTnFBVv5/k/qo6ubW2r6pOTrK/235vklOH9t+S5L6ufMuI8iO01q5Jck2SbNu2rW3fvn3lES1jcXExaznWWn4Ku+eS1T/uellr/LNsnmNPxD/P8c9z7In4AQAAYD2NHeaitXZVa21La21rBhPr/XFr7cVJbkpyabfZpUlu7JZvSnJxVR1XVadlMNHebd1QGA9U1blVVUleMrQPAAAAAABTbEUT8C1hV5IbquryJJ9P8sIkaa3dWVU3JLkrycEkV7bWHur2eWmSa5Mcn8HEeybfAwAAAGbK1jG/XN6z64IJ1QRgso4qmdxaW0yy2C1/Ocl5S2x3dZKrR5TfnuSso60kAAAAAACba+wwFwAAAAAAIJkMAAAAAMBYkskAAAAAAIwlmQwAAAAAwFhHNQEfAAAAAMvbuvPmqXzsPbsumGBNgD6STAYAAIAhVfWvk/xMkpbkjiQ/leRxSd6eZGuSPUl+orX21W77q5JcnuShJC9rrb1v8rXmaEm6Ahw9w1wAAABAp6pOSfKyJNtaa2clOSbJxUl2Jrm1tXZ6klu7+6mqM7r1ZyY5P8nrquqYzag7AGw0yWQAAAB4pGOTHF9Vx2bQI/m+JBcmua5bf12Si7rlC5Nc31p7sLV2b5LdSc6ZbHUBYDIMcwEAMCM2c/xFgHnRWvtiVf1aks8n+dsk72+tvb+qFlpr+7pt9lXVU7pdTknyoaFD7O3KWCPtHsD0kUwGAACATlWdmEFv49OSfC3Jf6mqFy+3y4iytsSxr0hyRZIsLCxkcXFx5AEPHDiw5Lp5suPsg6vab+H4wb7jzuFyx1/LvhttI+NK/P+tlfO3es7d8sa970zq/EkmAwAAwLf8SJJ7W2tfSpKqeleSH0xyf1Wd3PVKPjnJ/m77vUlOHdp/SwbDYhyhtXZNkmuSZNu2bW379u0jK7C4uJil1s2Ty1bZM3nH2Qfz6juOzZ5Ltq/6+GvZd6NtZFyJ/7+1cv5Wz7lb3rj3nWvPf/xEzp8xkwEAAOBbPp/k3Kp6XFVVkvOS3J3kpiSXdttcmuTGbvmmJBdX1XFVdVqS05PcNuE6A8BE6JkMAAAAndbah6vqHUk+luRgko9n0Jv4CUluqKrLM0g4v7Db/s6quiHJXd32V7bWHtqUygPABpNMBgAAgCGttVcledVhxQ9m0Et51PZXJ7l6o+sFiYkJgc1lmAsAAAAAAMaSTAYAAAAAYCzJZACYIlX1pqraX1WfGir7j1X16ar686r6g6p60tC6q6pqd1XdU1XPGyp/dlXd0a17bTeBEAAAAKyaZDIATJdrk5x/WNktSc5qrX1vkv+e5Kokqaozklyc5Mxun9dV1THdPq9PckUGM8qfPuKYAAAAcFQkkwFgirTWPpDkK4eVvb+1drC7+6EkW7rlC5Nc31p7sLV2b5LdSc6pqpOTnNBa+2BrrSV5c5KLJhIAAAAAvXXsZlcAADgqP53k7d3yKRkklw/Z25V9o1s+vPwIVXVFBj2Ys7CwkMXFxSO2OXDgwMjyWbXj7EFefuH4by33xXIxzepz2Lf/v0P6GFcfY0r6GVcfYwIAJkMyGQBmRFX9UpKDSd5yqGjEZm2Z8iMLW7smyTVJsm3btrZ9+/YjtllcXMyo8ll12c6bkwySrq++o18fhZaLac8l2ydbmXXSt/+/Q/oYVx9jSvoZVx9jAgAmo19XUADQU1V1aZLnJzmvG7oiGfQ4PnVosy1J7uvKt4woBwAAgFUzZjIATLmqOj/JK5O8oLX2P4ZW3ZTk4qo6rqpOy2Civdtaa/uSPFBV51ZVJXlJkhsnXnEAAAB6Rc9kAJgiVfW2JNuTPLmq9iZ5VZKrkhyX5JZBbjgfaq39y9banVV1Q5K7Mhj+4srW2kPdoV6a5Nokxyd5b3cDAACAVZNMBoAp0lp70YjiNy6z/dVJrh5RfnuSs9axagAAAMw5w1wAAAAAADCWZDIAAAAAAGNJJgMAAAAAMJZkMgAAAAAAY0kmAwAAAAAwlmQyAAAAAABjSSYDAAAAADCWZDIAAAAAAGMdu9kVAAAAAJgmW3fevNlVAJhKeiYDAAAAADCWZDIAAAAAAGMZ5gIAAADoHUNVAKw/yWQAAACAOTAuwb5n1wUTqgkwqwxzAQAAAADAWJLJAAAAAACMJZkMAAAAAMBYkskAAAAAAIwlmQwAAAAAwFiSyQAAAAAAjHXsZlcAAAAA6KetO29edv2eXRdMqCYArAc9kwEAAAAAGEsyGQAAAACAsSSTAQAAAAAYSzIZAAAAAICxTMAHAAAAwJotN+GiyRahH/RMBgAAAABgLMlkAAAAAADGkkwGgClSVW+qqv1V9amhspOq6paq+kz398ShdVdV1e6quqeqnjdU/uyquqNb99qqqknHAgAAQL+MHTO5qh6b5ANJjuu2f0dr7VVVdVKStyfZmmRPkp9orX212+eqJJcneSjJy1pr7+vKn53k2iTHJ3lPkpe31tr6hgQAM+3aJL+V5M1DZTuT3Npa21VVO7v7r6yqM5JcnOTMJE9N8kdV9fTW2kNJXp/kiiQfyqDNPT/JeycWxSZbbrw+AAAAVmclPZMfTPLDrbXvS/KMJOdX1bn51oXt6Ulu7e7nsAvb85O8rqqO6Y516ML29O52/vqFAgCzr7X2gSRfOaz4wiTXdcvXJbloqPz61tqDrbV7k+xOck5VnZzkhNbaB7svbd88tA8AAACsythkchs40N19dHdrcWELAJOy0FrblyTd36d05ack+cLQdnu7slO65cPLAQAAYNXGDnORJF3P4o8m+ftJfru19uGqesSFbVUNX9h+aGj3Qxew38gKL2yr6ooMejBnYWEhi4uLKw5oOQcOHFjTsXacfXDV+65XDGux1vhn2TzHnoh/nuOf59iTuYh/1DjIbZnyIw+wgjZ3Fs/jStrshePX1rZPo+VimrXn8JBZ/P9biT7G1ceYkn7G1ceYAIDJWFEyuRt78RlV9aQkf1BVZy2z+ZovbFtr1yS5Jkm2bdvWtm/fvpJqjrW4uJi1HOuyNYy/uOeS1T/uellr/LNsnmNPxD/P8c9z7Emv4r+/qk7uvrw9Ocn+rnxvklOHttuS5L6ufMuI8iOspM2dxfO4kjZ7x9kH8+o7VvRRaGYsG9MdX1923z27LtiAGq3dLP7/rUQf4+pjTEk/4+pjTADAZKxkzOSHtda+lmQxg7GO7+8uaLPeF7YAwCPclOTSbvnSJDcOlV9cVcdV1WkZzEdwW/fLoQeq6tyqqiQvGdoHAAAAVmVsMrmqvqPrkZyqOj7JjyT5dFzYAsC6q6q3Jflgku+pqr1VdXmSXUmeW1WfSfLc7n5aa3cmuSHJXUn+MMmV3a+JkuSlSd6QwdwFn03y3okGAgAAQO+s5LedJye5rhs3+VFJbmitvbuqPpjkhu4i9/NJXpgMLmyr6tCF7cEceWF7bZLjM7iodWELAENaay9aYtV5S2x/dZKrR5TfnmS5YakAAADgqIxNJrfW/jzJM0eUfzkubAEAAADmwtY1zCUF9MNRjZkMAAAAAMB8kkwGAACAIVX1pKp6R1V9uqrurqofqKqTquqWqvpM9/fEoe2vqqrdVXVPVT1vM+sOABtJMhkAAAAe6TeS/GFr7R8k+b4kdyfZmeTW1trpSW7t7qeqzkhycZIzk5yf5HXdnEMA0DuSyQAAANCpqhOS/FCSNyZJa+3vWmtfS3Jhkuu6za5LclG3fGGS61trD7bW7k2yO8k5k6wzAEzK2An4AAAAYI48LcmXkvxuVX1fko8meXmShdbaviRpre2rqqd025+S5END++/tyo5QVVckuSJJFhYWsri4OLICBw4cWHLdrNlx9sFl1y8X57h9l7Jw/Or3nXeLi4vL/v+t5bz25X96nD69fifNuVveuNffpM6fZDIAAAB8y7FJnpXk51prH66q30g3pMUSakRZG7Vha+2aJNckybZt29r27dtHHnBxcTFLrZs1l+28edn1ey7Zvup9l7Lj7IN59R3SHaux55Lty/7/rfY5OXTsedCn1++kOXfLG/f6u/b8x0/k/Hl3BQAgW8dd7O+6YEI1Adh0e5Psba19uLv/jgySyfdX1cldr+STk+wf2v7Uof23JLlvYrUFgAmSTB4y7iIKAACAfmut/WVVfaGqvqe1dk+S85Lc1d0uTbKr+3tjt8tNSd5aVa9J8tQkpye5bfI1B4CNJ5kMAAAAj/RzSd5SVY9J8rkkP5XBBPY3VNXlST6f5IVJ0lq7s6puyCDZfDDJla21hzan2gCwsSSTAQAAYEhr7RNJto1Ydd4S21+d5OqNrBMATAPJ5AkxDiEAAAAAMMskkwEAAADYUDrZQT88arMrAAAAAADA9JNMBgAAAABgLMlkAAAAAADGkkwGAAAAAGAsyWQAAAAAAMaSTAYAAAAAYCzJZAAAAAAAxpJMBgAAAABgrGM3uwIAAADAfNq68+bNrgIAR0HPZAAAAAAAxtIzGQAAAFgVPYsB5oueyQAAAAAAjCWZDAAAAADAWJLJAAAAAACMJZkMAAAAAMBYkskAMCOq6l9X1Z1V9amqeltVPbaqTqqqW6rqM93fE4e2v6qqdlfVPVX1vM2sOwAAALPv2M2uAAAwXlWdkuRlSc5orf1tVd2Q5OIkZyS5tbW2q6p2JtmZ5JVVdUa3/swkT03yR1X19NbaQ5sUAjNu686bl1y3Z9cFE6wJAACwWfRMBoDZcWyS46vq2CSPS3JfkguTXNetvy7JRd3yhUmub6092Fq7N8nuJOdMtroAAAD0iZ7JADADWmtfrKpfS/L5JH+b5P2ttfdX1UJrbV+3zb6qekq3yylJPjR0iL1d2SNU1RVJrkiShYWFLC4uHvHYBw4cGFk+zXacfXDsNgvHr2y7WbJZMW3k/8cs/v+tRB/j6mNMST/j6mNMAMBkSCYDwAzoxkK+MMlpSb6W5L9U1YuX22VEWTuioLVrklyTJNu2bWvbt28/YqfFxcWMKp9mly0zJMMhO84+mFff0a+PQpsV055Ltm/YsWfx/28l+hhXH2NK+hlXH2MCZt9yQ2olhtWCadGvKygA6K8fSXJva+1LSVJV70ryg0nur6qTu17JJyfZ322/N8mpQ/tvyWBYDAAAGGnrzpuz4+yDK/piHphPxkwGgNnw+STnVtXjqqqSnJfk7iQ3Jbm02+bSJDd2yzclubiqjquq05KcnuS2CdcZAACAHtEzGQBmQGvtw1X1jiQfS3IwycczGJ7iCUluqKrLM0g4v7Db/s6quiHJXd32V7bWHtqUygMAANALkskAMCNaa69K8qrDih/MoJfyqO2vTnL1RtcLAACA+WCYCwAAAAAAxpJMBgAAAABgLMlkAAAAAADGkkwGAAAAAGAsyWQAAAAAAMaSTAYAAAAAYCzJZAAAAAAAxpJMBgAAAABgrGM3uwIAAAAAsFG27rx52fV7dl0woZrA7NMzGQAAAACAsfRMBgBgTfT2AQCA+aBnMgAAAAAAY0kmAwAAAAAwlmQyAAAAAABjSSYDAAAAADCWCfgAAAAAmGom/IXpMDaZXFWnJnlzku9M8s0k17TWfqOqTkry9iRbk+xJ8hOtta92+1yV5PIkDyV5WWvtfV35s5Ncm+T4JO9J8vLWWlvfkAAAAACYJ+OSzcD6WEnP5INJdrTWPlZVT0zy0aq6JcllSW5tre2qqp1JdiZ5ZVWdkeTiJGcmeWqSP6qqp7fWHkry+iRXJPlQBsnk85O8d72DAgD6zcUCAADA5I0dM7m1tq+19rFu+YEkdyc5JcmFSa7rNrsuyUXd8oVJrm+tPdhauzfJ7iTnVNXJSU5orX2w64385qF9AAAAAACYYkc1ZnJVbU3yzCQfTrLQWtuXDBLOVfWUbrNTMuh5fMjeruwb3fLh5aMe54oMejBnYWEhi4uLR1PNJR04cGDZY+04++C6PM5qrFeMyxkXf5/Nc+yJ+Oc5/nmOPRE/AAAArKcVJ5Or6glJ3pnk51trf1NVS246oqwtU35kYWvXJLkmSbZt29a2b9++0moua3FxMcsd67JN/Mnsnku2b/hjjIu/z+Y59kT88xz/PMeeiB8AAADW09hhLpKkqh6dQSL5La21d3XF93dDV6T7u78r35vk1KHdtyS5ryvfMqIcAAAAAIApNzaZXIMuyG9Mcndr7TVDq25Kcmm3fGmSG4fKL66q46rqtCSnJ7mtGxLjgao6tzvmS4b2AQAAAABgiq2kZ/Jzkvxkkh+uqk90tx9PsivJc6vqM0me291Pa+3OJDckuSvJHya5srX2UHeslyZ5QwaT8n02yXvXMxgAAABYD1V1TFV9vKre3d0/qapuqarPdH9PHNr2qqraXVX3VNXzNq/WALCxxo6Z3Fr7s4we7zhJzltin6uTXD2i/PYkZx1NBQEAAGATvDzJ3UlO6O7vTHJra21XVe3s7r+yqs5IcnGSM5M8NckfVdXThzpVAUBvrGjMZAAAAJgXVbUlyQUZ/LL2kAuTXNctX5fkoqHy61trD7bW7s3gl7jnTKiqADBRY3smAwAAwJz59SSvSPLEobKFbi6gtNb2VdVTuvJTknxoaLu9XdkRquqKJFckycLCQhYXF0c++IEDB5ZcN212nH1ws6twhIXjp7Nes2Iez996vt5m6fU7bZy75Y17XU7q/EkmAwAAQKeqnp9kf2vto1W1fSW7jChrozZsrV2T5Jok2bZtW9u+ffThFxcXs9S6aXPZzps3uwpH2HH2wbz6DumO1ZrH87fnku3rdqxZev1OG+dueePeb689//ETOX+GuQCAGVFVT6qqd1TVp6vq7qr6AZMBAcC6e06SF1TVniTXZzAZ/e8nub+qTk6S7u/+bvu9SU4d2n9LkvsmV10AmBzJZACYHb+R5A9ba/8gyfdlMCnQocmATk9ya3c/h00GdH6S11XVMZtSawCYIa21q1prW1prWzNoS/+4tfbiJDclubTb7NIkN3bLNyW5uKqOq6rTkpye5LYJVxsAJkIyGQBmQFWdkOSHkrwxSVprf9da+1pMBgQAk7IryXOr6jNJntvdT2vtziQ3JLkryR8mubK19tCm1RIANtB8DYIDALPraUm+lOR3q+r7knw0ycuzxsmAVjIR0DROhLEek8L0cXKZaY1pLf8/0/j/tx76GFcfY0r6GVcfY9oorbXFJIvd8peTnLfEdlcnuXpiFQOATSKZDACz4dgkz0ryc621D1fVb6Qb0mIJK5oMaCUTAU3jRBjrMdlPHyeXmdaY1jKpzTT+/62HPsbVx5iSfsbVx5gAgMkwzAUAzIa9Sfa21j7c3X9HBsllkwEBAAAwEZLJADADWmt/meQLVfU9XdF5GYzNaDIgAAAAJmL6fgcJACzl55K8paoek+RzSX4qgy+Gb6iqy5N8PskLk8FkQFV1aDKggzEZEJto65hhSfbsumBCNQEAANZCMnlKuMgCYJzW2ieSbBuxymRAAAAAbDjDXAAAAAAAMJZkMgAAAAAAY0kmAwAAAAAwlmQyAAAAAABjSSYDAAAAADCWZDIAAAAAAGNJJgMAAAAAMJZkMgAAAAAAY0kmAwAAAAAw1rGbXQEAAAAA2Cxbd9687Po9uy6YUE1g+umZDAAAAADAWJLJAAAAAACMZZgLAAAAAFiCYTDgW/RMBgAAAABgLMlkAAAAAADGkkwGAAAAAGAsYyYDADDVlhun0BiFAAAwOXomAwAAAAAwlmQyAAAAAABjSSYDAAAAADCWZDIAAAAAAGNJJgMAAAAAMNaxm10BAAAAAGB9bd1587Lr9+y6YEI1oU/0TAYAAAAAYCzJZAAAAAAAxpJMBgAAAABgLMlkAAAAAADGkkwGAAAAAGAsyWQAmCFVdUxVfbyq3t3dP6mqbqmqz3R/Txza9qqq2l1V91TV8zav1gAAAPSBZDIAzJaXJ7l76P7OJLe21k5Pcmt3P1V1RpKLk5yZ5Pwkr6uqYyZcVwAAAHpEMhkAZkRVbUlyQZI3DBVfmOS6bvm6JBcNlV/fWnuwtXZvkt1JzplQVQEAAOihYze7AgDAiv16klckeeJQ2UJrbV+StNb2VdVTuvJTknxoaLu9XdkjVNUVSa5IkoWFhSwuLh7xoAcOHBhZvpl2nH1wzcdYOH59jjNNZjWm33zLjUuuG8S09L7T9r+5UtP4ulqrPsaU9DOuPsbExtm68+bNrgIAU0QyGQBmQFU9P8n+1tpHq2r7SnYZUdaOKGjtmiTXJMm2bdva9u1HHnpxcTGjyjfTZetwYbvj7IN59R39+ig0jzHtuWT75CqzjqbxdbVWfYwp6WdcfYwJYF75wodJ69fVBgD013OSvKCqfjzJY5OcUFW/n+T+qjq565V8cpL93fZ7k5w6tP+WJPdNtMYAAAD0imQyAMyA1tpVSa5Kkq5n8i+01l5cVf8xyaVJdnV/D40XcFOSt1bVa5I8NcnpSW6bcLUBAIA10POYaSOZDACzbVeSG6rq8iSfT/LCJGmt3VlVNyS5K8nBJFe21h7avGoCAAAw6ySTAWDGtNYWkyx2y19Oct4S212d5OqJVQwAAIBee9RmVwAAAAAAgOmnZzIAAAAAbAJjIjNr9EwGAAAAAGCssT2Tq+pNSZ6fZH9r7ayu7KQkb0+yNcmeJD/RWvtqt+6qJJcneSjJy1pr7+vKn53k2iTHJ3lPkpe31tr6htNfy31TtWfXBROsCQAAAAAroecxfbOSYS6uTfJbSd48VLYzya2ttV1VtbO7/8qqOiPJxUnOTPLUJH9UVU/vZo9/fZIrknwog2Ty+Uneu16BAAD94UM3AADA9BmbTG6tfaCqth5WfGGS7d3ydRnMKP/Krvz61tqDSe6tqt1JzqmqPUlOaK19MEmq6s1JLopkMgAAAAAzbLgjxI6zD+YyHSPosdVOwLfQWtuXJK21fVX1lK78lAx6Hh+ytyv7Rrd8ePlE3fHFv/aCBgAAAABYhdUmk5dSI8raMuWjD1J1RQZDYmRhYSGLi4vrUrmF4wffEPXNSs/PgQMH1u1czpp5jj0R/zzHP8+xJ+IHAI5eVZ2awTCP35nkm0muaa39xmrmDgKAvlltMvn+qjq565V8cpL9XfneJKcObbclyX1d+ZYR5SO11q5Jck2SbNu2rW3fvn2V1Xyk33zLjXn1HeudP998ey7ZvqLtFhcXs17nctbMc+yJ+Oc5/nmOPRE/ALAqB5PsaK19rKqemOSjVXVLksty9HMHAUCvPGqV+92U5NJu+dIkNw6VX1xVx1XVaUlOT3JbNyTGA1V1blVVkpcM7QMAAABTobW2r7X2sW75gSR3ZzBM44UZzBmU7u9F3fLDcwe11u5NsjvJOROtNABMyNhuulX1tgwm23tyVe1N8qoku5LcUFWXJ/l8khcmSWvtzqq6IcldGXybe+XQt7EvTXJtkuMzmHjP5HsAAABMrW4y+mcm+XCOfu6gUcdb0ZCO0zRU1ywOFdnXIS4nxflbm1k6f9PyPnPINL33TaNx/1eTOn9jk8mttRctseq8Jba/OsnVI8pvT3LWUdUOAACWsXXM5Mp7dl0woZoAfVNVT0jyziQ/31r7m8GPbEdvOqJs5BxBKx3ScZqG6prFSex3nH2wl0NcTorztzazdP5WOmzqpEzTe980Gvd+fO35j5/I+ZuN/24AAACYkKp6dAaJ5Le01t7VFR/t3EEAU82X8qzGasdMBgAAgN7p5vl5Y5K7W2uvGVp1VHMHTaq+ADBJeiYDAADAtzwnyU8muaOqPtGV/WJWN3cQAPSKZDIAAAB0Wmt/ltHjICdHOXcQAPSNZDIAAL1lLEAAAFg/xkwGAAAAAGAsyWQAAAAAAMaSTAYAAAAAYCzJZAAAAAAAxpJMBgAAAABgrGM3uwIAAAAAwHTZuvPmZdfv2XXBhGrCNNEzGQAAAACAsSSTAQAAAAAYyzAXAAAAAMBRWW4YDENg9JeeyQAAAAAAjCWZDAAAAADAWJLJADADqurUqvqTqrq7qu6sqpd35SdV1S1V9Znu74lD+1xVVbur6p6qet7m1R4AAIA+MGYyAMyGg0l2tNY+VlVPTPLRqrolyWVJbm2t7aqqnUl2JnllVZ2R5OIkZyZ5apI/qqqnt9Ye2qT6AwAAJBk93vKOsw/msq7cmMvTSzK5B5Yb8DzxAgTog9baviT7uuUHquruJKckuTDJ9m6z65IsJnllV359a+3BJPdW1e4k5yT54GRrDgAAQF9IJgPAjKmqrUmemeTDSRa6RHNaa/uq6indZqck+dDQbnu7ssOPdUWSK5JkYWEhi4uLRzzegQMHRpZvpB1nH9zwx1g4fjKPM0liOnqT/t8+ZDNeVxutjzEl/YyrjzEBAJMhmQwAM6SqnpDknUl+vrX2N1W15KYjytoRBa1dk+SaJNm2bVvbvn37ETstLi5mVPlGumzMr27Ww46zD+bVd/Tro5CYjt6eS7Zv2LGXsxmvq43Wx5iSfsbVx5gAgMkwAR8AzIiqenQGieS3tNbe1RXfX1Und+tPTrK/K9+b5NSh3bckuW9SdQUAAKB/+tV1BQB6qgZdkN+Y5O7W2muGVt2U5NIku7q/Nw6Vv7WqXpPBBHynJ7ltcjUGAADm1bj5vZhdkskAMBuek+Qnk9xRVZ/oyn4xgyTyDVV1eZLPJ3lhkrTW7qyqG5LcleRgkitbaw9NvNYAAAD0hmQyAMyA1tqfZfQ4yEly3hL7XJ3k6g2rFAAAAHNFMhkAgLk17ieYe3ZdMKGaAAAwDZb7fOizoWQyALAJjKEGAAAsxRf+00syGQAAVsFFDgDA5tB7ePNIJs+BQy+wHWcfzGUjXmxeZAAAAADAOJLJAACwBEOyAADAt0gmAwAAAABzQWeBtZFMBgCADbDchcqOsw9m++SqAgAA6+JRm10BAAAAAACmn57JAAAAAEAvGMZiY+mZDAAAAADAWJLJAAAAAACMZZgLAADYBON+grln1wUTqgkAAKyMZDLLXsi4iAEAAAAAnQESyWQAAACYayarAmCljJkMAAAAAMBYeiYDAMAUMhQZAADTRjIZAAAAAGCD9WHMZclkltWHf3IAAAAA2GhrHYN+FsawN2YyAAAAAABjSSYDAAAAADCWYS4AAGDGTPNQZNNcNwAA1kYyGQAAgFXx5QEAzBfJZNZkrQOD+3AJ0F+zMHkEAAAAK2fMZAAAAAAAxtIzGQAAemaahx5Yy68W/KoNAGBzSSYDAAArtlQyeMfZB3OZ4W0AAHpNMplNtVzPFD1PAAA2xjT3XF7OrNYbAKAvJJOZWi4WAAA2hwk0oV+8pgFYLxNPJlfV+Ul+I8kxSd7QWts16TrQD5LNAMvT5gI8ks+PbKTNbHcliwGYlIkmk6vqmCS/neS5SfYm+UhV3dRau2uS9WA+DH+gOtox/MZdSLgQAaadNheYRxuZUFvrsdf6+XItx2bjaXcBmBeT7pl8TpLdrbXPJUlVXZ/kwiQaWKbKWi8WNrNnwEZeqKz1sYGJ0uYCHKVRn5PWa2LBaUp0myxxQ2h3AZgL1Vqb3INV/dMk57fWfqa7/5NJvr+19rOHbXdFkiu6u9+T5J51qsKTk/zVOh1rFs1z/PMceyL+eY5/nmNPJhP/32utfccGP8ZRW+c2t6//R32MS0yzo49x9TGmpJ9xzWpMU9nmJhtyrTurz9G0cP7WxvlbG+dv9Zy7tVnv8zey3Z10z+QaUXZENru1dk2Sa9b9watub61tW+/jzop5jn+eY0/EP8/xz3PsydzHv25tbl/PYx/jEtPs6GNcfYwp6WdcfYxpCqzrta7naG2cv7Vx/tbG+Vs9525tJnX+HrXRD3CYvUlOHbq/Jcl9E64DAMwDbS4ATI52F4C5MOlk8keSnF5Vp1XVY5JcnOSmCdcBAOaBNhcAJke7C8BcmOgwF621g1X1s0nel+SYJG9qrd05wSqs+9AZM2ae45/n2BPxz3P88xx7Msfxr3Ob29fz2Me4xDQ7+hhXH2NK+hlXH2PaVBtwres5Whvnb22cv7Vx/lbPuVubiZy/iU7ABwAAAADAbJr0MBcAAAAAAMwgyWQAAAAAAMaai2RyVZ1fVfdU1e6q2rnZ9VkvVfWmqtpfVZ8aKjupqm6pqs90f08cWndVdw7uqarnDZU/u6ru6Na9tqpq0rEcrao6tar+pKrurqo7q+rlXfm8xP/Yqrqtqj7Zxf8rXflcxJ8kVXVMVX28qt7d3Z+n2Pd09f5EVd3elc1T/E+qqndU1ae794AfmKf4J61vbehS7UcfHP6+2AejXu+bXae1qqp/3f3vfaqq3lZVj93sOq1GHeXn0FmwREz/sfv/+/Oq+oOqetImVnFVRsU1tO4XqqpV1ZM3o26M1re292gc7XvL0X7Oq6rjqurtXfmHq2rr0D6Xdo/xmaq6dEIhr5ulPuM4fytTE7jG7vP5O+Twz6PO38rVBl/nr+v5a631+pbB5AefTfK0JI9J8skkZ2x2vdYpth9K8qwknxoq+w9JdnbLO5P8ard8Rhf7cUlO687JMd2625L8QJJK8t4kP7bZsa0g9pOTPKtbfmKS/97FOC/xV5IndMuPTvLhJOfOS/xdvf9NkrcmeXd3f55i35PkyYeVzVP81yX5mW75MUmeNE/xT/hc964NXar92Ox6rVNsj3hf7MNt1Ot9s+u0xnhOSXJvkuO7+zckuWyz67XKWFb8OXRWbkvE9KNJju2Wf3XWYloqrq781Awmi/uLwz9XuG3q89W7tvco49/Qa9wk/yrJ73TLFyd5e7d8UpLPdX9P7JZP3OzzcZTnbsOvkXt+/jb8GrvP52/oPG7YdXrfz182+Dp/Pc/fPPRMPifJ7tba51prf5fk+iQXbnKd1kVr7QNJvnJY8YUZXHil+3vRUPn1rbUHW2v3Jtmd5JyqOjnJCa21D7bBf9Gbh/aZWq21fa21j3XLDyS5O4MLtHmJv7XWDnR3H93dWuYk/qrakuSCJG8YKp6L2JcxF/FX1QkZXGS8MUlaa3/XWvta5iT+TdC7NnSZ9mOmLfG+ONOWeb3PumOTHF9VxyZ5XJL7Nrk+q3KUn0NnwqiYWmvvb60d7O5+KMmWiVdsjZZ4rpLkPyV5RQafIZkevWt7j8YErnGHj/WOJOd1vfael+SW1tpXWmtfTXJLkvPXO76NNKFr5D6fv0lcY/f2/CUTuU7v9flbwlSev3lIJp+S5AtD9/emBxeNy1hore1LBo1Jkqd05Uudh1O65cPLZ0bXNf+ZGXxzODfxdz8f+USS/Rm88Ocp/l/P4OLnm0Nl8xJ7MvhQ8/6q+mhVXdGVzUv8T0vypSS/2/186g1V9fjMT/yT1us29LD2Y9b9eo58X5x1S73eZ1Zr7YtJfi3J55PsS/LXrbX3b26t1tVS78V98dMZ9PCZeVX1giRfbK19crPrwhF63fau0np+znt4n+6Lor9O8u3LHGsmbeA1cq/P3wSusXt9/rLx1+l9P38bfZ2/budvHpLJo8bAnMdv35c6DzN9fqrqCUnemeTnW2t/s9ymI8pmOv7W2kOttWdk0EPmnKo6a5nNexN/VT0/yf7W2kdXusuIspmMfchzWmvPSvJjSa6sqh9aZtu+xX9sBj99fH1r7ZlJvp7Bz32W0rf4J6235+ko2o+pt4r3xVlxtK/3qdeNc3dhBj9HfGqSx1fVize3VqxEVf1SkoNJ3rLZdVmrqnpckl9K8u82uy6M1Nu2dwOs5nNe7z8bbvA1cq/P3wSusXt7/iZ0nd7b89fZ6Ov8dTt/85BM3pvBeGCHbMmM/pxwhe7vurWn+7u/K1/qPOzNI3+uNzPnp6oenUEj+ZbW2ru64rmJ/5DuJ7+LGfwMYR7if06SF1TVngx+9vfDVfX7mY/YkySttfu6v/uT/EEGP4ecl/j3Jtnb9RJIBj/PeVbmJ/5J62UbukT7McuWel+cdUu93mfZjyS5t7X2pdbaN5K8K8kPbnKd1tNS78UzrZuI5vlJLul+MjrrvjuDLzQ+2b1vbEnysar6zk2tFYf0su1do/X8nPfwPt1wQ9+WwbAavTjvE7hG7vX5O2QDr7H7fP4mcZ3e5/M3iev8dTt/85BM/kiS06vqtKp6TAaDTN+0yXXaSDclubRbvjTJjUPlF3ezN56W5PQkt3Xd5B+oqnO7sVJeMrTP1Orq+sYkd7fWXjO0al7i/47qZhOvquMzuDj9dOYg/tbaVa21La21rRm8nv+4tfbizEHsSVJVj6+qJx5azmBioE9lTuJvrf1lki9U1fd0RecluStzEv8m6F0bukz7MbOWeV+cacu83mfZ55OcW1WP6/4Xz8tgTMu+WOq9eGZV1flJXpnkBa21/7HZ9VkPrbU7WmtPaa1t7d439mYwaddfbnLVGOhd27sO1vNz3vCx/mkGbWbLYDLKH62qE7tfkfxoVzYzJnSN3OfzN4lr7N6evwldp/f2/E3oOn/9zl+bghkLN/qW5MczmMn0s0l+abPrs45xvS2D8fa+kcGHwMszGO/k1iSf6f6eNLT9L3Xn4J50szl25du6f9LPJvmtJLXZsa0g9n+UQbf7P0/yie7243MU//cm+XgX/6eS/LuufC7iH6r79nxrlti5iD2DMUQ/2d3uPPSeNi/xd/V+RpLbu////5rBjLNzE/8mnO9etaFLtR+bXa91jO/h98U+3Ea93je7TusQ069kcHH6qSS/l+S4za7TKuM4qs+hs3BbIqbdGYwjeOj94nc2u57rEddh6/fksNnj3Tb9OetV23uUsW/oNW6Sxyb5L91r+7YkTxva56e78t1Jfmqzz8Uqzt2GXyP3/Pxt+DV2n8/fYedyezbgOr3P5y8TuM5fz/N36IAAAAAAALCkeRjmAgAAAACANZJMBgAAAABgLMlkAAAAAADGkkwGAAAAAGAsyWQAAAAAAMaSTAYAAAAAYCzJZAAAAAAAxvr/A55Pa4PYddMBAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1440x1080 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Where to save the figures\n",
    "PROJECT_ROOT_DIR = \".\"\n",
    "CHAPTER_ID = \"end_to_end_project_practice\"\n",
    "IMAGES_PATH = os.path.join(PROJECT_ROOT_DIR, \"images\", CHAPTER_ID)\n",
    "os.makedirs(IMAGES_PATH, exist_ok=True)\n",
    "\n",
    "def save_fig(fig_id, tight_layout=True, fig_extension=\"png\", resolution=300):\n",
    "    path = os.path.join(IMAGES_PATH, fig_id + \".\" + fig_extension)\n",
    "    print(\"Saving figure\", fig_id)\n",
    "    if tight_layout:\n",
    "        plt.tight_layout()\n",
    "    plt.savefig(path, format=fig_extension, dpi=resolution)\n",
    "\n",
    "\n",
    "#To gain quick insight on the data, we'll use some histograms\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "housing.hist(bins=50, figsize=(20,15))\n",
    "save_fig(\"attribute_histogram_plots\", fig_extension=\"jpg\")\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create a Test Set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "#We'll fix the tail heaviness, unit problem, hard capping and scaling problems later.\n",
    "#We need to create a test set first before looking at the data too much\n",
    "#We also want to keep the test set constant per run to avoid the mistakenly exposing the test set to the training model. Well use a hash function\n",
    "\n",
    "import numpy as np\n",
    "np.random.seed(42)\n",
    "import hashlib\n",
    "\n",
    "def test_set_check(identifier, test_ratio, hash=hashlib.md5):\n",
    "    #keep only the last byte of the hash, and put the instance in the test set if this value <= 51 (~20% of 256 possible values) or our chosen test ratio.\n",
    "    return bytearray(hash(np.int64(identifier)).digest())[-1] < 256 * test_ratio    \n",
    "    \n",
    "def split_train_set_by_id(data, test_ratio, id_column, hash=hashlib.md5):\n",
    "    ids = data[id_column]\n",
    "    in_test_set = ids.apply(lambda id_: test_set_check(id_, test_ratio, hash))\n",
    "    return data.loc[-in_test_set], data.loc[in_test_set]\n",
    "\n",
    "#Generate a suitable ID using longitude and latitude\n",
    "housing_with_id = housing.copy() #make a copy of the housing dataframe\n",
    "housing_with_id[\"id\"] = housing[\"longitude\"] * 1000 + housing[\"latitude\"]\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "      <th>median_house_value</th>\n",
       "      <th>id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4373.000000</td>\n",
       "      <td>4373.000000</td>\n",
       "      <td>4373.000000</td>\n",
       "      <td>4373.000000</td>\n",
       "      <td>4325.000000</td>\n",
       "      <td>4373.000000</td>\n",
       "      <td>4373.000000</td>\n",
       "      <td>4373.000000</td>\n",
       "      <td>4373.000000</td>\n",
       "      <td>4373.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>-119.711685</td>\n",
       "      <td>35.783169</td>\n",
       "      <td>27.697462</td>\n",
       "      <td>2674.638235</td>\n",
       "      <td>538.448324</td>\n",
       "      <td>1424.038418</td>\n",
       "      <td>500.897553</td>\n",
       "      <td>3.964195</td>\n",
       "      <td>210401.988338</td>\n",
       "      <td>-119675.902172</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>2.041053</td>\n",
       "      <td>2.209536</td>\n",
       "      <td>12.138070</td>\n",
       "      <td>2176.718349</td>\n",
       "      <td>411.781289</td>\n",
       "      <td>1068.917156</td>\n",
       "      <td>374.259657</td>\n",
       "      <td>1.906693</td>\n",
       "      <td>114201.150923</td>\n",
       "      <td>2039.015719</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-124.300000</td>\n",
       "      <td>32.560000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.499900</td>\n",
       "      <td>22500.000000</td>\n",
       "      <td>-124258.200000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-121.920000</td>\n",
       "      <td>33.940000</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>1487.000000</td>\n",
       "      <td>299.000000</td>\n",
       "      <td>792.000000</td>\n",
       "      <td>282.000000</td>\n",
       "      <td>2.606500</td>\n",
       "      <td>122900.000000</td>\n",
       "      <td>-121881.980000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>-118.970000</td>\n",
       "      <td>34.400000</td>\n",
       "      <td>28.000000</td>\n",
       "      <td>2191.000000</td>\n",
       "      <td>441.000000</td>\n",
       "      <td>1178.000000</td>\n",
       "      <td>417.000000</td>\n",
       "      <td>3.640600</td>\n",
       "      <td>187300.000000</td>\n",
       "      <td>-118932.360000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>-117.950000</td>\n",
       "      <td>37.790000</td>\n",
       "      <td>36.000000</td>\n",
       "      <td>3221.000000</td>\n",
       "      <td>647.000000</td>\n",
       "      <td>1754.000000</td>\n",
       "      <td>607.000000</td>\n",
       "      <td>4.928600</td>\n",
       "      <td>267600.000000</td>\n",
       "      <td>-117916.330000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>-114.630000</td>\n",
       "      <td>41.860000</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>37937.000000</td>\n",
       "      <td>5471.000000</td>\n",
       "      <td>16122.000000</td>\n",
       "      <td>5189.000000</td>\n",
       "      <td>15.000100</td>\n",
       "      <td>500001.000000</td>\n",
       "      <td>-114597.240000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         longitude     latitude  housing_median_age   total_rooms  \\\n",
       "count  4373.000000  4373.000000         4373.000000   4373.000000   \n",
       "mean   -119.711685    35.783169           27.697462   2674.638235   \n",
       "std       2.041053     2.209536           12.138070   2176.718349   \n",
       "min    -124.300000    32.560000            1.000000      2.000000   \n",
       "25%    -121.920000    33.940000           18.000000   1487.000000   \n",
       "50%    -118.970000    34.400000           28.000000   2191.000000   \n",
       "75%    -117.950000    37.790000           36.000000   3221.000000   \n",
       "max    -114.630000    41.860000           52.000000  37937.000000   \n",
       "\n",
       "       total_bedrooms    population   households  median_income  \\\n",
       "count     4325.000000   4373.000000  4373.000000    4373.000000   \n",
       "mean       538.448324   1424.038418   500.897553       3.964195   \n",
       "std        411.781289   1068.917156   374.259657       1.906693   \n",
       "min          1.000000      6.000000     1.000000       0.499900   \n",
       "25%        299.000000    792.000000   282.000000       2.606500   \n",
       "50%        441.000000   1178.000000   417.000000       3.640600   \n",
       "75%        647.000000   1754.000000   607.000000       4.928600   \n",
       "max       5471.000000  16122.000000  5189.000000      15.000100   \n",
       "\n",
       "       median_house_value             id  \n",
       "count         4373.000000    4373.000000  \n",
       "mean        210401.988338 -119675.902172  \n",
       "std         114201.150923    2039.015719  \n",
       "min          22500.000000 -124258.200000  \n",
       "25%         122900.000000 -121881.980000  \n",
       "50%         187300.000000 -118932.360000  \n",
       "75%         267600.000000 -117916.330000  \n",
       "max         500001.000000 -114597.240000  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#split the data into train set and test set using a random split\n",
    "train_set_rand, test_set_rand = split_train_set_by_id(housing_with_id, 0.2, \"id\")\n",
    "test_set_rand.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "      <th>median_house_value</th>\n",
       "      <th>id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>16267.000000</td>\n",
       "      <td>16267.000000</td>\n",
       "      <td>16267.000000</td>\n",
       "      <td>16267.000000</td>\n",
       "      <td>16108.000000</td>\n",
       "      <td>16267.000000</td>\n",
       "      <td>16267.000000</td>\n",
       "      <td>16267.000000</td>\n",
       "      <td>16267.000000</td>\n",
       "      <td>16267.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>-119.531536</td>\n",
       "      <td>35.591186</td>\n",
       "      <td>28.892728</td>\n",
       "      <td>2625.312412</td>\n",
       "      <td>537.715421</td>\n",
       "      <td>1425.863404</td>\n",
       "      <td>499.174648</td>\n",
       "      <td>3.845529</td>\n",
       "      <td>205902.512203</td>\n",
       "      <td>-119495.945053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1.991663</td>\n",
       "      <td>2.113958</td>\n",
       "      <td>12.691610</td>\n",
       "      <td>2182.878515</td>\n",
       "      <td>423.939077</td>\n",
       "      <td>1148.976727</td>\n",
       "      <td>384.480832</td>\n",
       "      <td>1.897243</td>\n",
       "      <td>115699.564560</td>\n",
       "      <td>1989.706730</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>-124.350000</td>\n",
       "      <td>32.540000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>0.499900</td>\n",
       "      <td>14999.000000</td>\n",
       "      <td>-124309.460000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>-121.710000</td>\n",
       "      <td>33.930000</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>1438.500000</td>\n",
       "      <td>295.000000</td>\n",
       "      <td>786.000000</td>\n",
       "      <td>279.000000</td>\n",
       "      <td>2.553900</td>\n",
       "      <td>118800.000000</td>\n",
       "      <td>-121673.170000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>-118.450000</td>\n",
       "      <td>34.240000</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>2112.000000</td>\n",
       "      <td>432.000000</td>\n",
       "      <td>1164.000000</td>\n",
       "      <td>408.000000</td>\n",
       "      <td>3.512500</td>\n",
       "      <td>177400.000000</td>\n",
       "      <td>-118415.700000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>-118.030000</td>\n",
       "      <td>37.690000</td>\n",
       "      <td>37.000000</td>\n",
       "      <td>3124.500000</td>\n",
       "      <td>647.000000</td>\n",
       "      <td>1718.000000</td>\n",
       "      <td>603.500000</td>\n",
       "      <td>4.703250</td>\n",
       "      <td>264000.000000</td>\n",
       "      <td>-117995.920000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>-114.310000</td>\n",
       "      <td>41.950000</td>\n",
       "      <td>52.000000</td>\n",
       "      <td>39320.000000</td>\n",
       "      <td>6445.000000</td>\n",
       "      <td>35682.000000</td>\n",
       "      <td>6082.000000</td>\n",
       "      <td>15.000100</td>\n",
       "      <td>500001.000000</td>\n",
       "      <td>-114275.810000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          longitude      latitude  housing_median_age   total_rooms  \\\n",
       "count  16267.000000  16267.000000        16267.000000  16267.000000   \n",
       "mean    -119.531536     35.591186           28.892728   2625.312412   \n",
       "std        1.991663      2.113958           12.691610   2182.878515   \n",
       "min     -124.350000     32.540000            1.000000      6.000000   \n",
       "25%     -121.710000     33.930000           18.000000   1438.500000   \n",
       "50%     -118.450000     34.240000           29.000000   2112.000000   \n",
       "75%     -118.030000     37.690000           37.000000   3124.500000   \n",
       "max     -114.310000     41.950000           52.000000  39320.000000   \n",
       "\n",
       "       total_bedrooms    population    households  median_income  \\\n",
       "count    16108.000000  16267.000000  16267.000000   16267.000000   \n",
       "mean       537.715421   1425.863404    499.174648       3.845529   \n",
       "std        423.939077   1148.976727    384.480832       1.897243   \n",
       "min          2.000000      3.000000      2.000000       0.499900   \n",
       "25%        295.000000    786.000000    279.000000       2.553900   \n",
       "50%        432.000000   1164.000000    408.000000       3.512500   \n",
       "75%        647.000000   1718.000000    603.500000       4.703250   \n",
       "max       6445.000000  35682.000000   6082.000000      15.000100   \n",
       "\n",
       "       median_house_value             id  \n",
       "count        16267.000000   16267.000000  \n",
       "mean        205902.512203 -119495.945053  \n",
       "std         115699.564560    1989.706730  \n",
       "min          14999.000000 -124309.460000  \n",
       "25%         118800.000000 -121673.170000  \n",
       "50%         177400.000000 -118415.700000  \n",
       "75%         264000.000000 -117995.920000  \n",
       "max         500001.000000 -114275.810000  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_set_rand.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "16267 train + 4373 test\n"
     ]
    }
   ],
   "source": [
    "print(len(train_set_rand), 'train +', len(test_set_rand), 'test')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Stratified Splitting of the Dataset for Data with clear stratas (Layers)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving figure median_income\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABZgAAAQwCAYAAABolTvzAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAA3hUlEQVR4nO3dcayld33n98+3noQ4OIApYep6rNpbWewauyHxyPUWsRqvSXEWhPmjSI7YYLdUrhBJk8qrMu5KrfqHW0st2wax0FohayNoRhYbhBXL2bW8O4pWghCbkEwMcXHDrDH22tksEIYg0qHf/nGfrK7Hd+be+XLnnjn26yVd3fP8zvM853fgN+Pjt5/73OruAAAAAADA2fp3Vj0BAAAAAADWk8AMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMDIvlVPYDuve93r+vLLL9+Vc333u9/NK1/5yl05F+wV65Z1ZN2yjqxb1o01yzqybllH1i3ryLrlXHjsscf+dXf/5Knj531gvvzyy/Poo4/uyrmOHj2aQ4cO7cq5YK9Yt6wj65Z1ZN2ybqxZ1pF1yzqybllH1i3nQlX9y63G3SIDAAAAAIARgRkAAAAAgBGBGQAAAACAEYEZAAAAAIARgRkAAAAAgBGBGQAAAACAEYEZAAAAAIARgRkAAAAAgBGBGQAAAACAEYEZAAAAAIARgRkAAAAAgBGBGQAAAACAEYEZAAAAAIARgRkAAAAAgBGBGQAAAACAEYEZAAAAAIARgRkAAAAAgBGBGQAAAACAEYEZAAAAAIARgRkAAAAAgBGBGQAAAACAEYEZAAAAAIARgRkAAAAAgBGBGQAAAACAEYEZAAAAAIARgRkAAAAAgBGBGQAAAACAEYEZAAAAAIARgRkAAAAAgBGBGQAAAACAEYEZAAAAAIARgRkAAAAAgBGBGQAAAACAEYEZAAAAAIARgRkAAAAAgBGBGQAAAACAEYEZAAAAAIARgRkAAAAAgBGBGQAAAACAEYEZAAAAAIARgRkAAAAAgBGBGQAAAACAEYEZAAAAAIARgRkAAAAAgBGBGQAAAACAEYEZAAAAAIARgRkAAAAAgBGBGQAAAACAkX2rngDwQpcffvAF23dcczK3nTLGS8/xu9++6ikAAAAAnDVXMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADCybWCuqjdU1Zc2ff15Vf1KVb22qh6uqq8u3y/edMydVfVkVT1RVW/bNH5tVR1bnvtwVdW5emMAAAAAAJxb2wbm7n6iu9/U3W9Kcm2Sv0jymSSHkzzS3VcmeWTZTlVdleSWJG9MclOSj1bVBcvpPpbk9iRXLl837eq7AQAAAABgz5ztLTJuTPL/dPe/THJzkvuW8fuSvGt5fHOSI939/e7+WpInk1xXVZckeVV3f667O8knNh0DAAAAAMCaqY3Wu8Odq349yRe7+yNV9a3ufs2m577Z3RdX1UeSfL67P7mMfzzJQ0mOJ7m7u9+6jL8lyQe7+x1bvM7t2bjSOfv377/2yJEj0/f3AidOnMhFF120K+eCc+XYN779gu39FybPfW9Fk2HPXHPpq1c9hV3l71vWkXXLurFmWUfWLevIumUdWbecCzfccMNj3X3w1PF9Oz1BVf1okncmuXO7XbcY6zOMv3iw+54k9yTJwYMH+9ChQzud5hkdPXo0u3UuOFduO/zgC7bvuOZkPnRsx39UWVPH33No1VPYVf6+ZR1Zt6wba5Z1ZN2yjqxb1pF1y146m1tk/Fw2rl5+btl+brntRZbvzy/jTye5bNNxB5I8s4wf2GIcAAAAAIA1dDaB+eeT/Mam7QeS3Lo8vjXJZzeN31JVr6iqK7Lxy/y+0N3PJvlOVV1fVZXkvZuOAQAAAABgzezo5+6r6seT/GyS/2rT8N1J7q+q9yV5Ksm7k6S7H6+q+5N8OcnJJB/o7h8sx7w/yb1JLszGfZkf2oX3AAAAAADACuwoMHf3XyT5d08Z+7MkN55m/7uS3LXF+KNJrj77aQIAAAAAcL45m1tkAAAAAADAvyUwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADCyb9UTACC5/PCDq57CrrrjmpO57SX2ns6F43e/fdVTAAAAgB+KK5gBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAY2VFgrqrXVNWnq+qPq+orVfU3q+q1VfVwVX11+X7xpv3vrKonq+qJqnrbpvFrq+rY8tyHq6rOxZsCAAAAAODc2+kVzL+a5Le7+68n+akkX0lyOMkj3X1lkkeW7VTVVUluSfLGJDcl+WhVXbCc52NJbk9y5fJ10y69DwAAAAAA9ti2gbmqXpXkbyX5eJJ0919297eS3JzkvmW3+5K8a3l8c5Ij3f397v5akieTXFdVlyR5VXd/rrs7ySc2HQMAAAAAwJrZyRXMfy3Jnyb5R1X1+1X1a1X1yiT7u/vZJFm+v37Z/9IkX990/NPL2KXL41PHAQAAAABYQ/t2uM/PJPml7v7dqvrVLLfDOI2t7qvcZxh/8Qmqbs/GrTSyf//+HD16dAfT3N6JEyd27VxwrtxxzckXbO+/8MVjcL6zbnfGP5POLz4nsG6sWdaRdcs6sm5ZR9Yte2kngfnpJE939+8u25/ORmB+rqou6e5nl9tfPL9p/8s2HX8gyTPL+IEtxl+ku+9Jck+SHDx4sA8dOrSzd7ONo0ePZrfOBefKbYcffMH2HdeczIeO7eSPKpw/rNudOf6eQ6ueApv4nMC6sWZZR9Yt68i6ZR1Zt+ylbW+R0d3/KsnXq+oNy9CNSb6c5IEkty5jtyb57PL4gSS3VNUrquqKbPwyvy8st9H4TlVdX1WV5L2bjgEAAAAAYM3s9PKyX0ryqar60SR/kuQ/z0acvr+q3pfkqSTvTpLufryq7s9GhD6Z5APd/YPlPO9Pcm+SC5M8tHwBAAAAALCGdhSYu/tLSQ5u8dSNp9n/riR3bTH+aJKrz2J+AAAAAACcp7a9RQYAAAAAAGxFYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYGRHgbmqjlfVsar6UlU9uoy9tqoerqqvLt8v3rT/nVX1ZFU9UVVv2zR+7XKeJ6vqw1VVu/+WAAAAAADYC2dzBfMN3f2m7j64bB9O8kh3X5nkkWU7VXVVkluSvDHJTUk+WlUXLMd8LMntSa5cvm764d8CAAAAAACr8MPcIuPmJPctj+9L8q5N40e6+/vd/bUkTya5rqouSfKq7v5cd3eST2w6BgAAAACANVMbrXebnaq+luSbSTrJ/9nd91TVt7r7NZv2+WZ3X1xVH0ny+e7+5DL+8SQPJTme5O7ufusy/pYkH+zud2zxerdn40rn7N+//9ojR478cO9yceLEiVx00UW7ci44V45949sv2N5/YfLc91Y0GRiybnfmmktfveopsInPCawba5Z1ZN2yjqxb1pF1y7lwww03PLbp7hb/1r4dHv/m7n6mql6f5OGq+uMz7LvVfZX7DOMvHuy+J8k9SXLw4ME+dOjQDqd5ZkePHs1unQvOldsOP/iC7TuuOZkPHdvpH1U4P1i3O3P8PYdWPQU28TmBdWPNso6sW9aRdcs6sm7ZSzu6RUZ3P7N8fz7JZ5Jcl+S55bYXWb4/v+z+dJLLNh1+IMkzy/iBLcYBAAAAAFhD2wbmqnplVf3EXz1O8p8m+aMkDyS5ddnt1iSfXR4/kOSWqnpFVV2RjV/m94XufjbJd6rq+qqqJO/ddAwAAAAAAGtmJz+/vD/JZzaacPYl+b+6+7er6veS3F9V70vyVJJ3J0l3P15V9yf5cpKTST7Q3T9YzvX+JPcmuTAb92V+aBffCwAAAAAAe2jbwNzdf5Lkp7YY/7MkN57mmLuS3LXF+KNJrj77aQIAAAAAcL7Z0T2YAQAAAADgVAIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIzsOzFV1QVX9flX91rL92qp6uKq+uny/eNO+d1bVk1X1RFW9bdP4tVV1bHnuw1VVu/t2AAAAAADYK2dzBfMvJ/nKpu3DSR7p7iuTPLJsp6quSnJLkjcmuSnJR6vqguWYjyW5PcmVy9dNP9TsAQAAAABYmR0F5qo6kOTtSX5t0/DNSe5bHt+X5F2bxo909/e7+2tJnkxyXVVdkuRV3f257u4kn9h0DAAAAAAAa6Y2Wu82O1V9Osn/nOQnkvy97n5HVX2ru1+zaZ9vdvfFVfWRJJ/v7k8u4x9P8lCS40nu7u63LuNvSfLB7n7HFq93ezaudM7+/fuvPXLkyA/3LhcnTpzIRRddtCvngnPl2De+/YLt/Rcmz31vRZOBIet2Z6659NWrngKb+JzAurFmWUfWLevIumUdWbecCzfccMNj3X3w1PF92x1YVe9I8nx3P1ZVh3bwWlvdV7nPMP7iwe57ktyTJAcPHuxDh3bysts7evRodutccK7cdvjBF2zfcc3JfOjYtn9U4bxi3e7M8fccWvUU2MTnBNaNNcs6sm5ZR9Yt68i6ZS/t5N/+35zknVX1d5L8WJJXVdUnkzxXVZd097PL7S+eX/Z/Osllm44/kOSZZfzAFuMAAAAAAKyhbe/B3N13dveB7r48G7+87591999N8kCSW5fdbk3y2eXxA0luqapXVNUV2fhlfl/o7meTfKeqrq+qSvLeTccAAAAAALBmfpifX747yf1V9b4kTyV5d5J09+NVdX+SLyc5meQD3f2D5Zj3J7k3yYXZuC/zQz/E6wMAAAAAsEJnFZi7+2iSo8vjP0ty42n2uyvJXVuMP5rk6rOdJAAAAAAA559tb5EBAAAAAABbEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAY2TYwV9WPVdUXquoPqurxqvofl/HXVtXDVfXV5fvFm465s6qerKonquptm8avrapjy3Mfrqo6N28LAAAAAIBzbSdXMH8/yd/u7p9K8qYkN1XV9UkOJ3mku69M8siynaq6KsktSd6Y5KYkH62qC5ZzfSzJ7UmuXL5u2r23AgAAAADAXto2MPeGE8vmjyxfneTmJPct4/cledfy+OYkR7r7+939tSRPJrmuqi5J8qru/lx3d5JPbDoGAAAAAIA1s6N7MFfVBVX1pSTPJ3m4u383yf7ufjZJlu+vX3a/NMnXNx3+9DJ26fL41HEAAAAAANbQvp3s1N0/SPKmqnpNks9U1dVn2H2r+yr3GcZffIKq27NxK43s378/R48e3ck0t3XixIldOxecK3dcc/IF2/svfPEYnO+s253xz6Tzi88JrBtrlnVk3bKOrFvWkXXLXtpRYP4r3f2tqjqajXsnP1dVl3T3s8vtL55fdns6yWWbDjuQ5Jll/MAW41u9zj1J7kmSgwcP9qFDh85mmqd19OjR7Na54Fy57fCDL9i+45qT+dCxs/qjCitn3e7Qse+uegZscsc1P8iH/sW5///k+N1vP+evwcuDz7asI+uWdWTdso6sW/bStrfIqKqfXK5cTlVdmOStSf44yQNJbl12uzXJZ5fHDyS5papeUVVXZOOX+X1huY3Gd6rq+qqqJO/ddAwAAAAAAGtmJ5eXXZLkvqq6IBtB+v7u/q2q+lyS+6vqfUmeSvLuJOnux6vq/iRfTnIyyQeWW2wkyfuT3JvkwiQPLV8AAAAAAKyhbQNzd/9hkp/eYvzPktx4mmPuSnLXFuOPJjnT/ZsBAAAAAFgT294iAwAAAAAAtiIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwIjADAAAAADAiMAMAAAAAMCIwAwAAAAAwsm/VE+D0Lj/84KqnAAAAAABwWq5gBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBk28BcVZdV1T+vqq9U1eNV9cvL+Gur6uGq+ury/eJNx9xZVU9W1RNV9bZN49dW1bHluQ9XVZ2btwUAAAAAwLm2kyuYTya5o7v/RpLrk3ygqq5KcjjJI919ZZJHlu0sz92S5I1Jbkry0aq6YDnXx5LcnuTK5eumXXwvAAAAAADsoW0Dc3c/291fXB5/J8lXklya5OYk9y273ZfkXcvjm5Mc6e7vd/fXkjyZ5LqquiTJq7r7c93dST6x6RgAAAAAANZMbbTeHe5cdXmS30lydZKnuvs1m577ZndfXFUfSfL57v7kMv7xJA8lOZ7k7u5+6zL+liQf7O53bPE6t2fjSufs37//2iNHjoze3KlOnDiRiy66aFfOtReOfePbq54C54H9FybPfW/Vs4CzY92yjvZq3V5z6avP/YvwsrBun20hsW5ZT9Yt68i65Vy44YYbHuvug6eO79vpCarqoiT/OMmvdPefn+H2yVs90WcYf/Fg9z1J7kmSgwcP9qFDh3Y6zTM6evRodutce+G2ww+uegqcB+645mQ+dGzHf1ThvGDdso72at0ef8+hc/4avDys22dbSKxb1pN1yzqybtlLO7kHc6rqR7IRlz/V3b+5DD+33PYiy/fnl/Gnk1y26fADSZ5Zxg9sMQ4AAAAAwBraNjDXxqXKH0/yle7+B5ueeiDJrcvjW5N8dtP4LVX1iqq6Ihu/zO8L3f1sku9U1fXLOd+76RgAAAAAANbMTn4O9M1JfiHJsar60jL23yW5O8n9VfW+JE8leXeSdPfjVXV/ki8nOZnkA939g+W49ye5N8mF2bgv80O78zYAAAAAANhr2wbm7v4X2fr+yUly42mOuSvJXVuMP5qNXxAIAAAAAMCa29E9mAEAAAAA4FQCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAI/tWPQEAgJeTyw8/uOopsALH7377qqcAAADnhCuYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAY2TYwV9WvV9XzVfVHm8ZeW1UPV9VXl+8Xb3ruzqp6sqqeqKq3bRq/tqqOLc99uKpq998OAAAAAAB7ZSdXMN+b5KZTxg4neaS7r0zyyLKdqroqyS1J3rgc89GqumA55mNJbk9y5fJ16jkBAAAAAFgj2wbm7v6dJP/mlOGbk9y3PL4vybs2jR/p7u9399eSPJnkuqq6JMmruvtz3d1JPrHpGAAAAAAA1lBt9N5tdqq6PMlvdffVy/a3uvs1m57/ZndfXFUfSfL57v7kMv7xJA8lOZ7k7u5+6zL+liQf7O53nOb1bs/G1c7Zv3//tUeOHBm/wc1OnDiRiy66aFfOtReOfePbq54C54H9FybPfW/Vs4CzY92yjqxbzqVrLn31rp9z3T7bQmLdsp6sW9aRdcu5cMMNNzzW3QdPHd+3y6+z1X2V+wzjW+rue5LckyQHDx7sQ4cO7crkjh49mt0611647fCDq54C54E7rjmZDx3b7T+qcG5Zt6wj65Zz6fh7Du36Odftsy0k1i3rybplHVm37KWd3IN5K88tt73I8v35ZfzpJJdt2u9AkmeW8QNbjAMAAAAAsKamgfmBJLcuj29N8tlN47dU1Suq6ops/DK/L3T3s0m+U1XXV1Ulee+mYwAAAAAAWEPb/hxoVf1GkkNJXldVTyf5H5LcneT+qnpfkqeSvDtJuvvxqro/yZeTnEzyge7+wXKq9ye5N8mF2bgv80O7+k4AAAAAANhT2wbm7v750zx142n2vyvJXVuMP5rk6rOaHQAAAAAA563pLTIAAAAAAHiZE5gBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABgRmAEAAAAAGBGYAQAAAAAYEZgBAAAAABjZt+oJAADAS93lhx/c9XPecc3J3HYOzsvuOX7321c9BQCAc84VzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACM7Fv1BAAAAF6KLj/84KqncN6545qTue0l/r/L8bvfvuopAMCecgUzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAjAjMAAAAAACMCMwAAAAAAIwIzAAAAAAAj+1Y9AQAAAHipuPzwg6ueArvsjmtO5rZt/n89fvfb92g2AOcfVzADAAAAADAiMAMAAAAAMCIwAwAAAAAw4h7MAAAAAHCWzud7ru/k3uHMuOf6i7mCGQAAAACAEYEZAAAAAIARt8gAAAAA+CGcz7dKADjXXMEMAAAAAMCIwAwAAAAAwMieB+aquqmqnqiqJ6vq8F6/PgAAAAAAu2NPA3NVXZDkHyb5uSRXJfn5qrpqL+cAAAAAAMDu2OsrmK9L8mR3/0l3/2WSI0lu3uM5AAAAAACwC6q79+7Fqv6zJDd193+5bP9Ckv+4u3/xlP1uT3L7svmGJE/s0hRel+Rf79K5YK9Yt6wj65Z1ZN2ybqxZ1pF1yzqybllH1i3nwn/Q3T956uC+PZ5EbTH2osLd3fckuWfXX7zq0e4+uNvnhXPJumUdWbesI+uWdWPNso6sW9aRdcs6sm7ZS3t9i4ynk1y2aftAkmf2eA4AAAAAAOyCvQ7Mv5fkyqq6oqp+NMktSR7Y4zkAAAAAALAL9vQWGd19sqp+Mck/SXJBkl/v7sf3cAq7ftsN2APWLevIumUdWbesG2uWdWTdso6sW9aRdcue2dNf8gcAAAAAwEvHXt8iAwAAAACAlwiBGQAAAACAkZdNYK6qm6rqiap6sqoOr3o+sJ2quqyq/nlVfaWqHq+qX171nGAnquqCqvr9qvqtVc8FdqKqXlNVn66qP17+zv2bq54TbKeq/pvl88EfVdVvVNWPrXpOcKqq+vWqer6q/mjT2Gur6uGq+ury/eJVzhFOdZp1+78snxP+sKo+U1WvWeEU4QW2WrObnvt7VdVV9bpVzI2Xj5dFYK6qC5L8wyQ/l+SqJD9fVVetdlawrZNJ7ujuv5Hk+iQfsG5ZE7+c5CurngSchV9N8tvd/deT/FSsX85zVXVpkv86ycHuvjobvzz7ltXOCrZ0b5KbThk7nOSR7r4yySPLNpxP7s2L1+3DSa7u7v8oyf+d5M69nhScwb158ZpNVV2W5GeTPLXXE+Ll52URmJNcl+TJ7v6T7v7LJEeS3LziOcEZdfez3f3F5fF3shE8Ll3trODMqupAkrcn+bVVzwV2oqpeleRvJfl4knT3X3b3t1Y6KdiZfUkurKp9SX48yTMrng+8SHf/TpJ/c8rwzUnuWx7fl+Rdezkn2M5W67a7/2l3n1w2P5/kwJ5PDE7jNH/XJsn/luS/TdJ7OyNejl4ugfnSJF/ftP10hDrWSFVdnuSnk/zuiqcC2/nfs/Eh5v9b8Txgp/5akj9N8o+WW7v8WlW9ctWTgjPp7m8k+V+zcUXSs0m+3d3/dLWzgh3b393PJhsXVCR5/YrnA2frv0jy0KonAWdSVe9M8o3u/oNVz4WXh5dLYK4txvwXHNZCVV2U5B8n+ZXu/vNVzwdOp6rekeT57n5s1XOBs7Avyc8k+Vh3/3SS78aPa3OeW+5Ze3OSK5L8+0leWVV/d7WzAnjpq6q/n41bGX5q1XOB06mqH0/y95P896ueCy8fL5fA/HSSyzZtH4gfI2QNVNWPZCMuf6q7f3PV84FtvDnJO6vqeDZuRfS3q+qTq50SbOvpJE9391/9hMinsxGc4Xz21iRf6+4/7e7/N8lvJvlPVjwn2KnnquqSJFm+P7/i+cCOVNWtSd6R5D3d7YI1zmf/YTb+I/QfLP9udiDJF6vq31vprHhJe7kE5t9LcmVVXVFVP5qNX4LywIrnBGdUVZWNe4J+pbv/warnA9vp7ju7+0B3X56Nv2f/WXe7oo7zWnf/qyRfr6o3LEM3JvnyCqcEO/FUkuur6seXzws3xi+nZH08kOTW5fGtST67wrnAjlTVTUk+mOSd3f0Xq54PnEl3H+vu13f35cu/mz2d5GeWz71wTrwsAvNyM/5fTPJPsvHh+/7ufny1s4JtvTnJL2TjKtAvLV9/Z9WTAngJ+qUkn6qqP0zypiT/02qnA2e2XHH/6SRfTHIsG5/p71nppGALVfUbST6X5A1V9XRVvS/J3Ul+tqq+muRnl204b5xm3X4kyU8keXj597L/Y6WThE1Os2ZhT5Wf7AAAAAAAYOJlcQUzAAAAAAC7T2AGAAAAAGBEYAYAAAAAYERgBgAAAABgRGAGAAAAAGBEYAYAAAAAYERgBgAAAABg5P8HzajRdpkdvyYAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 1440x1080 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#I'll look for these stratas first by visualising the median income\n",
    "housing[\"median_income\"].hist(bins = 10, figsize= (20,15))\n",
    "save_fig(\"median_income\")\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3.0    7236\n",
       "2.0    6581\n",
       "4.0    3639\n",
       "5.0    2362\n",
       "1.0     822\n",
       "Name: income_cat, dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Then I will split the data based on what i can see (divide the income to 5 categories)\n",
    "housing[\"income_cat\"] = np.ceil(housing[\"median_income\"] / 1.5) # divide by 1.5 to limit the number of categories\n",
    "housing[\"income_cat\"].where(housing[\"income_cat\"]<5, 5.0, inplace=True) #Values above 5 are labelled as 5\n",
    "\n",
    "housing[\"income_cat\"].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving figure income categories\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "housing[\"income_cat\"].hist()\n",
    "save_fig(\"income categories\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Stratified Sampling\n",
    "from sklearn.model_selection import StratifiedShuffleSplit\n",
    "\n",
    "split = StratifiedShuffleSplit(n_splits=1, test_size=0.2, random_state=42)\n",
    "for train_index, test_index in split.split(housing, housing[\"income_cat\"]):\n",
    "    strat_train_set = housing.loc[train_index]\n",
    "    strat_test_set = housing.loc[test_index]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Drop off the income_cat attribute form the train and test sets\n",
    "for set in (strat_train_set, strat_test_set):\n",
    "    set.drop([\"income_cat\"], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## STEP 3: DISCOVER AND VISUALIZE THE DATA\n",
    "\n",
    "(our Dataset is small so there isn't a need for an exploration set)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving figure First_Viz_Plot\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#visualising graphical data by plotting Longitude against Latitude\n",
    "\n",
    "#we set s(radius of each circle), c(color of each circle), and cmap (color map) -> Jet ranges from blue to red\n",
    "#This makes the plot show more information \n",
    "#The argument `sharex=False` fixes a display bug (the x-axis values and legend were not displayed)\n",
    "housing.plot(kind=\"scatter\", x=\"longitude\", y=\"latitude\", alpha = 0.4,\n",
    "             s=housing[\"population\"]/100, label=\"population\",\n",
    "             c=\"median_house_value\", cmap = plt.get_cmap(\"jet\"), colorbar = True,\n",
    "             sharex=False)\n",
    "save_fig(\"First_Viz_Plot\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that housing prices is related to the location and population density. Clustering Algorithms could provide more features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading california.png\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "('.\\\\images\\\\end_to_end_project_practice\\\\california.png',\n",
       " <http.client.HTTPMessage at 0x22b77a89e08>)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#A more sophisticated plot\n",
    "#Download an image of california from using already created directories from above\n",
    "filename = \"california.png\"\n",
    "print(\"Downloading\", filename)\n",
    "url = DOWNLOAD_ROOT + \"images/end_to_end_project/\" + filename\n",
    "urllib.request.urlretrieve(url, os.path.join(IMAGES_PATH, filename))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\envs\\myMLenv\\lib\\site-packages\\ipykernel_launcher.py:24: UserWarning: FixedFormatter should only be used together with FixedLocator\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving figure california_housing_prices_advanced_plt\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x504 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#We create a more complex plot and use the image lib as well\n",
    "import matplotlib.image as mpimg\n",
    "#Read the image\n",
    "california_img=mpimg.imread(PROJECT_ROOT_DIR + \"/images/\" + CHAPTER_ID +\"/california.png\")\n",
    "\n",
    "#Our initial scatter plot but remove color bar because we would creat one\n",
    "ax = housing.plot(kind = \"scatter\", x=\"longitude\", y=\"latitude\", figsize=(10,7),\n",
    "                  s=housing[\"population\"]/100, label=\"population\",\n",
    "                  c=\"median_house_value\", cmap = plt.get_cmap(\"jet\"), \n",
    "                  colorbar = False, alpha=0.4\n",
    "                  )\n",
    "#Also plot the image\n",
    "plt.imshow(california_img, extent=[-124.55, -113.80, 32.45, 42.05], alpha=0.5,\n",
    "           cmap=plt.get_cmap(\"jet\"))\n",
    "plt.ylabel(\"Latitude\", fontsize=14)\n",
    "plt.xlabel(\"Longitude\", fontsize=14)\n",
    "\n",
    "\n",
    "\n",
    "#For the Colour bar\n",
    "prices = housing[\"median_house_value\"]\n",
    "tick_values = np.linspace(prices.min(), prices.max(), 11)\n",
    "cbar = plt.colorbar()\n",
    "cbar.ax.set_yticklabels([\"$%dk\"%(round(v/1000)) for v in tick_values], fontsize=14)\n",
    "cbar.set_label(\"Median House Value\", fontsize=16)\n",
    "\n",
    "plt.legend(fontsize=16)\n",
    "save_fig(\"california_housing_prices_advanced_plt\")\n",
    "plt.show()\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We look for correlation to get useful features (Linear correlation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "median_house_value    1.000000\n",
       "median_income         0.688075\n",
       "income_cat            0.643892\n",
       "total_rooms           0.134153\n",
       "housing_median_age    0.105623\n",
       "households            0.065843\n",
       "total_bedrooms        0.049686\n",
       "population           -0.024650\n",
       "longitude            -0.045967\n",
       "latitude             -0.144160\n",
       "Name: median_house_value, dtype: float64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corr_matrix = housing.corr()\n",
    "\n",
    "corr_matrix[\"median_house_value\"].sort_values(ascending=False) #To get correlation with our label/target"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "median income is the most promising attribute to predict median house value, we'll get it's scatter plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving figure income_vs_house_value_scatterplot\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "housing.plot(kind=\"scatter\", x=\"median_income\", y=\"median_house_value\", alpha=0.1)\n",
    "plt.axis([0,16,0,550000])\n",
    "save_fig(\"income_vs_house_value_scatterplot\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Above, we notice that the correlation is strong but there are also some price caps at specific prices. We can also fix the tail heaviness of the distribution by computing logarithms.\n",
    "\n",
    "We'll create other attributes that we think matters, this is an iterative process. I might come back here"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "housing[\"rooms_per_household\"] = housing[\"total_rooms\"]/housing[\"households\"]\n",
    "housing[\"bedrooms_per_room\"] = housing[\"total_bedrooms\"]/housing[\"total_rooms\"]\n",
    "housing[\"population_per_household\"] = housing[\"population\"]/housing[\"households\"]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "median_house_value          1.000000\n",
       "median_income               0.688075\n",
       "income_cat                  0.643892\n",
       "rooms_per_household         0.151948\n",
       "total_rooms                 0.134153\n",
       "housing_median_age          0.105623\n",
       "households                  0.065843\n",
       "total_bedrooms              0.049686\n",
       "population_per_household   -0.023737\n",
       "population                 -0.024650\n",
       "longitude                  -0.045967\n",
       "latitude                   -0.144160\n",
       "bedrooms_per_room          -0.255880\n",
       "Name: median_house_value, dtype: float64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing.corr()[\"median_house_value\"].sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So bedrooms per room is a much better attribute (better correlation) than total bedrooms or total rooms"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## STEP 4: Prep the Data for ML Algorithms"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Rather than doing the transformations separately, we would create a pipeline for everything later, for now we shall describe all the transformations taking place."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-118.51  ,   34.26  ,   29.    , 2119.5   ,  433.    , 1164.    ,\n",
       "        408.    ,    3.5409])"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Lets revert the training set by copying it  from the strat_train_set\n",
    "#and separate the predictors and labels\n",
    "\n",
    "housing =  strat_train_set.drop(\"median_house_value\", axis = 1) # drop the lables for the training set\n",
    "housing_labels = strat_train_set[\"median_house_value\"].copy()\n",
    "\n",
    "#There are some missing  features which we will fill with the median\n",
    "try:\n",
    "    from sklearn.impute import SimpleImputer #Scikit-Learn 0.20+\n",
    "except ImportError:\n",
    "    from sklearn.preprocessing import Imputer as SimpleImputer\n",
    "    \n",
    "imputer = SimpleImputer(strategy=\"median\")\n",
    "\n",
    "\n",
    "#we remove the text attribute because median is only calculated on numerical attributes\n",
    "housing_num = housing.select_dtypes(include=[np.number]) # or the code below\n",
    "#housing_num = housing.drop(\"ocean_proximity\", axis =1)\n",
    "\n",
    "\n",
    "\n",
    "# we fit the Imputer Instance to our data\n",
    "imputer.fit(housing_num)\n",
    "\n",
    "#view the values\n",
    "imputer.statistics_\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-118.51  ,   34.26  ,   29.    , 2119.5   ,  433.    , 1164.    ,\n",
       "        408.    ,    3.5409])"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_num.median().values #To verify that it worked"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>17606</th>\n",
       "      <td>-121.89</td>\n",
       "      <td>37.29</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1568.0</td>\n",
       "      <td>351.0</td>\n",
       "      <td>710.0</td>\n",
       "      <td>339.0</td>\n",
       "      <td>2.7042</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18632</th>\n",
       "      <td>-121.93</td>\n",
       "      <td>37.05</td>\n",
       "      <td>14.0</td>\n",
       "      <td>679.0</td>\n",
       "      <td>108.0</td>\n",
       "      <td>306.0</td>\n",
       "      <td>113.0</td>\n",
       "      <td>6.4214</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14650</th>\n",
       "      <td>-117.20</td>\n",
       "      <td>32.77</td>\n",
       "      <td>31.0</td>\n",
       "      <td>1952.0</td>\n",
       "      <td>471.0</td>\n",
       "      <td>936.0</td>\n",
       "      <td>462.0</td>\n",
       "      <td>2.8621</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3230</th>\n",
       "      <td>-119.61</td>\n",
       "      <td>36.31</td>\n",
       "      <td>25.0</td>\n",
       "      <td>1847.0</td>\n",
       "      <td>371.0</td>\n",
       "      <td>1460.0</td>\n",
       "      <td>353.0</td>\n",
       "      <td>1.8839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3555</th>\n",
       "      <td>-118.59</td>\n",
       "      <td>34.23</td>\n",
       "      <td>17.0</td>\n",
       "      <td>6592.0</td>\n",
       "      <td>1525.0</td>\n",
       "      <td>4459.0</td>\n",
       "      <td>1463.0</td>\n",
       "      <td>3.0347</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       longitude  latitude  housing_median_age  total_rooms  total_bedrooms  \\\n",
       "17606    -121.89     37.29                38.0       1568.0           351.0   \n",
       "18632    -121.93     37.05                14.0        679.0           108.0   \n",
       "14650    -117.20     32.77                31.0       1952.0           471.0   \n",
       "3230     -119.61     36.31                25.0       1847.0           371.0   \n",
       "3555     -118.59     34.23                17.0       6592.0          1525.0   \n",
       "\n",
       "       population  households  median_income  \n",
       "17606       710.0       339.0         2.7042  \n",
       "18632       306.0       113.0         6.4214  \n",
       "14650       936.0       462.0         2.8621  \n",
       "3230       1460.0       353.0         1.8839  \n",
       "3555       4459.0      1463.0         3.0347  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#We Transform the training set\n",
    "X = imputer.transform(housing_num)\n",
    "housing_tr = pd.DataFrame(X, columns=housing_num.columns, index=housing.index)\n",
    "housing_tr.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4629</th>\n",
       "      <td>-118.30</td>\n",
       "      <td>34.07</td>\n",
       "      <td>18.0</td>\n",
       "      <td>3759.0</td>\n",
       "      <td>433.0</td>\n",
       "      <td>3296.0</td>\n",
       "      <td>1462.0</td>\n",
       "      <td>2.2708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6068</th>\n",
       "      <td>-117.86</td>\n",
       "      <td>34.01</td>\n",
       "      <td>16.0</td>\n",
       "      <td>4632.0</td>\n",
       "      <td>433.0</td>\n",
       "      <td>3038.0</td>\n",
       "      <td>727.0</td>\n",
       "      <td>5.1762</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17923</th>\n",
       "      <td>-121.97</td>\n",
       "      <td>37.35</td>\n",
       "      <td>30.0</td>\n",
       "      <td>1955.0</td>\n",
       "      <td>433.0</td>\n",
       "      <td>999.0</td>\n",
       "      <td>386.0</td>\n",
       "      <td>4.6328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13656</th>\n",
       "      <td>-117.30</td>\n",
       "      <td>34.05</td>\n",
       "      <td>6.0</td>\n",
       "      <td>2155.0</td>\n",
       "      <td>433.0</td>\n",
       "      <td>1039.0</td>\n",
       "      <td>391.0</td>\n",
       "      <td>1.6675</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19252</th>\n",
       "      <td>-122.79</td>\n",
       "      <td>38.48</td>\n",
       "      <td>7.0</td>\n",
       "      <td>6837.0</td>\n",
       "      <td>433.0</td>\n",
       "      <td>3468.0</td>\n",
       "      <td>1405.0</td>\n",
       "      <td>3.1662</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       longitude  latitude  housing_median_age  total_rooms  total_bedrooms  \\\n",
       "4629     -118.30     34.07                18.0       3759.0           433.0   \n",
       "6068     -117.86     34.01                16.0       4632.0           433.0   \n",
       "17923    -121.97     37.35                30.0       1955.0           433.0   \n",
       "13656    -117.30     34.05                 6.0       2155.0           433.0   \n",
       "19252    -122.79     38.48                 7.0       6837.0           433.0   \n",
       "\n",
       "       population  households  median_income  \n",
       "4629       3296.0      1462.0         2.2708  \n",
       "6068       3038.0       727.0         5.1762  \n",
       "17923       999.0       386.0         4.6328  \n",
       "13656      1039.0       391.0         1.6675  \n",
       "19252      3468.0      1405.0         3.1662  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#to view some of the incomplete rows\n",
    "sample_incomplete_rows = housing[housing.isnull().any(axis=1)].head()\n",
    "housing_tr.loc[sample_incomplete_rows.index.values] # quite verbose"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>17606</th>\n",
       "      <td>-121.89</td>\n",
       "      <td>37.29</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1568.0</td>\n",
       "      <td>351.0</td>\n",
       "      <td>710.0</td>\n",
       "      <td>339.0</td>\n",
       "      <td>2.7042</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18632</th>\n",
       "      <td>-121.93</td>\n",
       "      <td>37.05</td>\n",
       "      <td>14.0</td>\n",
       "      <td>679.0</td>\n",
       "      <td>108.0</td>\n",
       "      <td>306.0</td>\n",
       "      <td>113.0</td>\n",
       "      <td>6.4214</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14650</th>\n",
       "      <td>-117.20</td>\n",
       "      <td>32.77</td>\n",
       "      <td>31.0</td>\n",
       "      <td>1952.0</td>\n",
       "      <td>471.0</td>\n",
       "      <td>936.0</td>\n",
       "      <td>462.0</td>\n",
       "      <td>2.8621</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3230</th>\n",
       "      <td>-119.61</td>\n",
       "      <td>36.31</td>\n",
       "      <td>25.0</td>\n",
       "      <td>1847.0</td>\n",
       "      <td>371.0</td>\n",
       "      <td>1460.0</td>\n",
       "      <td>353.0</td>\n",
       "      <td>1.8839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3555</th>\n",
       "      <td>-118.59</td>\n",
       "      <td>34.23</td>\n",
       "      <td>17.0</td>\n",
       "      <td>6592.0</td>\n",
       "      <td>1525.0</td>\n",
       "      <td>4459.0</td>\n",
       "      <td>1463.0</td>\n",
       "      <td>3.0347</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       longitude  latitude  housing_median_age  total_rooms  total_bedrooms  \\\n",
       "17606    -121.89     37.29                38.0       1568.0           351.0   \n",
       "18632    -121.93     37.05                14.0        679.0           108.0   \n",
       "14650    -117.20     32.77                31.0       1952.0           471.0   \n",
       "3230     -119.61     36.31                25.0       1847.0           371.0   \n",
       "3555     -118.59     34.23                17.0       6592.0          1525.0   \n",
       "\n",
       "       population  households  median_income  \n",
       "17606       710.0       339.0         2.7042  \n",
       "18632       306.0       113.0         6.4214  \n",
       "14650       936.0       462.0         2.8621  \n",
       "3230       1460.0       353.0         1.8839  \n",
       "3555       4459.0      1463.0         3.0347  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_tr = pd.DataFrame(X, columns=housing_num.columns, index=housing_num.index)\n",
    "housing_tr.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Handling Text and Categorical Attributes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "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>ocean_proximity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>17606</th>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18632</th>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14650</th>\n",
       "      <td>NEAR OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3230</th>\n",
       "      <td>INLAND</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3555</th>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19480</th>\n",
       "      <td>INLAND</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8879</th>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13685</th>\n",
       "      <td>INLAND</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4937</th>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4861</th>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      ocean_proximity\n",
       "17606       <1H OCEAN\n",
       "18632       <1H OCEAN\n",
       "14650      NEAR OCEAN\n",
       "3230           INLAND\n",
       "3555        <1H OCEAN\n",
       "19480          INLAND\n",
       "8879        <1H OCEAN\n",
       "13685          INLAND\n",
       "4937        <1H OCEAN\n",
       "4861        <1H OCEAN"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#We would use Transformers \n",
    "\n",
    "#Let us create a categorical input feature\n",
    "housing_cat = housing[['ocean_proximity']] #note the double square brackets to return a DF\n",
    "housing_cat.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<16512x5 sparse matrix of type '<class 'numpy.float64'>'\n",
       "\twith 16512 stored elements in Compressed Sparse Row format>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#We'll use a OneHotEncoder for the categories of the ocean proximity.\n",
    "\n",
    "try:\n",
    "    from sklearn.preprocessing import OrdinalEncoder #just to raise an ImportError if Scikit-Learn < 0.20\n",
    "    from sklearn.preprocessing import OneHotEncoder\n",
    "except ImportError:\n",
    "    from future_encoders import OneHotEncoder #Scikit-Learn < 0.20\n",
    "    \n",
    "    \n",
    "cat_encoder = OneHotEncoder()\n",
    "housing_cat_1hot = cat_encoder.fit_transform(housing_cat)\n",
    "housing_cat_1hot"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Creation of Custom Transformers\n",
    "\n",
    "(Scikit-Learn relies on duck typing, we'll create a transformer class and implement fit(), transform() and fit_transform() and adding TransformerMixin and BaseEstimator as base classes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "' from sklearn.preprocessing import FunctionTransformer\\n\\ndef add_extra_features (X, add_bedrooms_per_room=True):\\n        rooms_per_household = X[:, rooms_ix] / X[:, household_ix]\\n        population_per_household = X[:, population_ix] / X[:, household_ix]\\n        if self.add_bedrooms_per_room:\\n            bedrooms_per_room = X[:, bedrooms_ix] / X[:, rooms_ix]\\n            return np.c_[X, rooms_per_household, population_per_household, bedrooms_per_room]\\n        \\n        else:\\n            return np.c_[X, rooms_per_household, population_per_household]  \\n        \\n\\nattr_adder =FunctionTransformer(add_extra_features, validate=False,\\n                                kw_args={\"add_bedrooms_per_room\": False}) #Validate is set to False because of non-float values in the data\\nhousing_extra_attribs = attr_adder.fit_transform(housing.values)  '"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.base import BaseEstimator, TransformerMixin\n",
    "\n",
    "#We get the colum location of the following attributes using a list comprehension\n",
    "rooms_ix, bedrooms_ix, population_ix, household_ix = [\n",
    "    list(housing.columns).index(col)\n",
    "    for col in (\"total_rooms\", \"total_bedrooms\", \"population\", \"households\")]\n",
    "\n",
    "#Previous hard coded implementation\n",
    "#rooms_ix, bedrooms_ix, population_ix, household_ix = 3, 4, 5, 6  #The column location of these attributes in the DataFrame\n",
    "\n",
    "\n",
    "class CombinedAttributesAdder(BaseEstimator, TransformerMixin):\n",
    "    def __init__(self, add_bedrooms_per_room = True):  #no args or *kargs since we are using BaseEstimator as a base class\n",
    "        self.add_bedrooms_per_room = add_bedrooms_per_room\n",
    "        \n",
    "    def fit(self, X, y=None):\n",
    "        return self #nothing else to do\n",
    "    def transform(self, X, y=None):\n",
    "        rooms_per_household = X[:, rooms_ix] / X[:, household_ix]\n",
    "        population_per_household = X[:, population_ix] / X[:, household_ix]\n",
    "        if self.add_bedrooms_per_room:\n",
    "            bedrooms_per_room = X[:, bedrooms_ix] / X[:, rooms_ix]\n",
    "            return np.c_[X, rooms_per_household, population_per_household, bedrooms_per_room]\n",
    "        \n",
    "        else:\n",
    "            return np.c_[X, rooms_per_household, population_per_household]\n",
    "        \n",
    "        \n",
    "attr_adder = CombinedAttributesAdder(add_bedrooms_per_room=False)\n",
    "housing_extra_attribs = attr_adder.transform(housing.values)\n",
    "\n",
    "#Code for using a FunctionTransformer rather than creating your own class\n",
    "\n",
    "\"\"\" from sklearn.preprocessing import FunctionTransformer\n",
    "\n",
    "def add_extra_features (X, add_bedrooms_per_room=True):\n",
    "        rooms_per_household = X[:, rooms_ix] / X[:, household_ix]\n",
    "        population_per_household = X[:, population_ix] / X[:, household_ix]\n",
    "        if self.add_bedrooms_per_room:\n",
    "            bedrooms_per_room = X[:, bedrooms_ix] / X[:, rooms_ix]\n",
    "            return np.c_[X, rooms_per_household, population_per_household, bedrooms_per_room]\n",
    "        \n",
    "        else:\n",
    "            return np.c_[X, rooms_per_household, population_per_household]  \n",
    "        \n",
    "\n",
    "attr_adder =FunctionTransformer(add_extra_features, validate=False,\n",
    "                                kw_args={\"add_bedrooms_per_room\": False}) #Validate is set to False because of non-float values in the data\n",
    "housing_extra_attribs = attr_adder.fit_transform(housing.values)  \"\"\" \n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "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>longitude</th>\n",
       "      <th>latitude</th>\n",
       "      <th>housing_median_age</th>\n",
       "      <th>total_rooms</th>\n",
       "      <th>total_bedrooms</th>\n",
       "      <th>population</th>\n",
       "      <th>households</th>\n",
       "      <th>median_income</th>\n",
       "      <th>ocean_proximity</th>\n",
       "      <th>rooms_per_household</th>\n",
       "      <th>population_per_household</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>17606</th>\n",
       "      <td>-121.89</td>\n",
       "      <td>37.29</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1568.0</td>\n",
       "      <td>351.0</td>\n",
       "      <td>710.0</td>\n",
       "      <td>339.0</td>\n",
       "      <td>2.7042</td>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "      <td>4.625369</td>\n",
       "      <td>2.094395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18632</th>\n",
       "      <td>-121.93</td>\n",
       "      <td>37.05</td>\n",
       "      <td>14.0</td>\n",
       "      <td>679.0</td>\n",
       "      <td>108.0</td>\n",
       "      <td>306.0</td>\n",
       "      <td>113.0</td>\n",
       "      <td>6.4214</td>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "      <td>6.00885</td>\n",
       "      <td>2.707965</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14650</th>\n",
       "      <td>-117.2</td>\n",
       "      <td>32.77</td>\n",
       "      <td>31.0</td>\n",
       "      <td>1952.0</td>\n",
       "      <td>471.0</td>\n",
       "      <td>936.0</td>\n",
       "      <td>462.0</td>\n",
       "      <td>2.8621</td>\n",
       "      <td>NEAR OCEAN</td>\n",
       "      <td>4.225108</td>\n",
       "      <td>2.025974</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3230</th>\n",
       "      <td>-119.61</td>\n",
       "      <td>36.31</td>\n",
       "      <td>25.0</td>\n",
       "      <td>1847.0</td>\n",
       "      <td>371.0</td>\n",
       "      <td>1460.0</td>\n",
       "      <td>353.0</td>\n",
       "      <td>1.8839</td>\n",
       "      <td>INLAND</td>\n",
       "      <td>5.232295</td>\n",
       "      <td>4.135977</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3555</th>\n",
       "      <td>-118.59</td>\n",
       "      <td>34.23</td>\n",
       "      <td>17.0</td>\n",
       "      <td>6592.0</td>\n",
       "      <td>1525.0</td>\n",
       "      <td>4459.0</td>\n",
       "      <td>1463.0</td>\n",
       "      <td>3.0347</td>\n",
       "      <td>&lt;1H OCEAN</td>\n",
       "      <td>4.50581</td>\n",
       "      <td>3.047847</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      longitude latitude housing_median_age total_rooms total_bedrooms  \\\n",
       "17606   -121.89    37.29               38.0      1568.0          351.0   \n",
       "18632   -121.93    37.05               14.0       679.0          108.0   \n",
       "14650    -117.2    32.77               31.0      1952.0          471.0   \n",
       "3230    -119.61    36.31               25.0      1847.0          371.0   \n",
       "3555    -118.59    34.23               17.0      6592.0         1525.0   \n",
       "\n",
       "      population households median_income ocean_proximity rooms_per_household  \\\n",
       "17606      710.0      339.0        2.7042       <1H OCEAN            4.625369   \n",
       "18632      306.0      113.0        6.4214       <1H OCEAN             6.00885   \n",
       "14650      936.0      462.0        2.8621      NEAR OCEAN            4.225108   \n",
       "3230      1460.0      353.0        1.8839          INLAND            5.232295   \n",
       "3555      4459.0     1463.0        3.0347       <1H OCEAN             4.50581   \n",
       "\n",
       "      population_per_household  \n",
       "17606                 2.094395  \n",
       "18632                 2.707965  \n",
       "14650                 2.025974  \n",
       "3230                  4.135977  \n",
       "3555                  3.047847  "
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#we convert it to a dataframe and add the new Column names\n",
    "housing_extra_attribs = pd.DataFrame(housing_extra_attribs,\n",
    "                                     columns= list(housing.columns)+[\"rooms_per_household\", \"population_per_household\"],\n",
    "                                     index= housing.index)\n",
    "\n",
    "housing_extra_attribs.head()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Feature Scaling (Transformation)\n",
    "(to make the scale of the data similar so that algorithms can perform well)\n",
    "\n",
    "We'll use Standardization"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Transformation Pipeline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "#You can use a custom DataFrameSelector transformer and a FeatureUnion to combine different \n",
    "#pipelines for different types of data or attributes but\n",
    "\n",
    "#We'll use a ColumnTransformer to combine the pipelines so no need for a DataFrameSelector\n",
    "try:\n",
    "    from sklearn.compose import ColumnTransformer\n",
    "except ImportError:\n",
    "    from future_encoders import ColumnTransformer #Scikit-Learn <0.20\n",
    "\n",
    "#import Pipelines and the standard scaler   \n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "    \n",
    "    \n",
    "num_attribs = list(housing_num)\n",
    "cat_attribs = [\"ocean_proximity\"]\n",
    "\n",
    "num_pipeline = Pipeline([\n",
    "    (\"imputer\", SimpleImputer(strategy='median')),\n",
    "    (\"attribs_adder\", CombinedAttributesAdder()),\n",
    "    (\"std_scaler\", StandardScaler())\n",
    "])\n",
    "\n",
    "\n",
    "full_pipeline = ColumnTransformer([\n",
    "    (\"num\", num_pipeline, num_attribs),\n",
    "    (\"cat\", OneHotEncoder(),cat_attribs)\n",
    "])\n",
    "\n",
    "housing_prepared = full_pipeline.fit_transform(housing)\n",
    "    \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16512, 16)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Lets peep the shape\n",
    "housing_prepared.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## STEP 5: SELECT AND TRAIN A MODEL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None, normalize=False)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#1. Linear Regression\n",
    "\n",
    "from sklearn.linear_model import LinearRegression\n",
    "\n",
    "lin_reg = LinearRegression()\n",
    "lin_reg.fit(housing_prepared, housing_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predictions: [210644.60459286 317768.80697211 210956.43331178  59218.98886849\n",
      " 189747.55849879]\n"
     ]
    }
   ],
   "source": [
    "#we would have to run the full preprocessing pipeline on a few training instances\n",
    "some_data = housing.iloc[:5]\n",
    "some_labels = housing_labels.iloc[:5]\n",
    "some_data_prepared = full_pipeline.transform(some_data)\n",
    "\n",
    "\n",
    "print(\"Predictions:\", lin_reg.predict(some_data_prepared))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Labels: [286600.0, 340600.0, 196900.0, 46300.0, 254500.0]\n"
     ]
    }
   ],
   "source": [
    "#We would compare this result to the actual value (The labels)\n",
    "print(\"Labels:\", list(some_labels))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "68628.19819848922"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Let us use a better method of evaluating the model\n",
    "#We'll use the mean_squared_error function\n",
    "\n",
    "from sklearn.metrics import mean_squared_error\n",
    "\n",
    "housing_predictions = lin_reg.predict(housing_prepared)\n",
    "lin_mse = mean_squared_error(housing_labels, housing_predictions)\n",
    "lin_rmse = np.sqrt(lin_mse)\n",
    "lin_rmse\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "49439.89599001897"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#we'll also calculate the mean absolute error so we import the necessary files\n",
    "\n",
    "from sklearn.metrics import mean_absolute_error\n",
    "\n",
    "lin_mae = mean_absolute_error(housing_labels, housing_predictions)\n",
    "lin_mae"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Our Linear Model is clearing Underfitting (the training data) so we will use a more powerful model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeRegressor(criterion='mse', max_depth=None, max_features=None,\n",
       "                      max_leaf_nodes=None, min_impurity_decrease=0.0,\n",
       "                      min_impurity_split=None, min_samples_leaf=1,\n",
       "                      min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "                      presort=False, random_state=None, splitter='best')"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#2. Decision Tree  Regressor\n",
    "\n",
    "from sklearn.tree import DecisionTreeRegressor\n",
    "\n",
    "tree_reg = DecisionTreeRegressor()\n",
    "tree_reg.fit(housing_prepared, housing_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Lets evaluate this model on the training set\n",
    "housing_predictions = tree_reg.predict(housing_prepared)\n",
    "tree_mse = mean_squared_error(housing_labels, housing_predictions)\n",
    "tree_rmse = np.sqrt(tree_mse)\n",
    "tree_rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This error is unrealistic. It is probably overfitting the training data. We'll use cross validation to check for this."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We'll use some Cross validation Techniques and Try some other models\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Scores: [68238.05196774 66817.04578495 70753.16954743 69884.28682243\n",
      " 70568.69371449 74169.72026563 71656.23162198 70466.17619771\n",
      " 76555.53603628 69481.84893535]\n",
      "Mean: 70859.07608940084\n",
      "Standard deviation 2652.359918121003\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "#This splits the training data and test set into smaller training and validation set\n",
    "#We are performing a K-fold cross validation, training and evaluating the model K times and\n",
    "#and picking a different fold for evaluation every time and training on the other K-1 times\n",
    "scores = cross_val_score(tree_reg, housing_prepared, housing_labels,\n",
    "                         scoring=\"neg_mean_squared_error\", cv=10)\n",
    "\n",
    "tree_rmse_scores = np.sqrt(-scores) #-ve because the scoring was done using a utility function not an error func\n",
    "\n",
    "#we'll display the scores using this \n",
    "def display_scores(scores):\n",
    "    print(\"Scores:\", scores)\n",
    "    print(\"Mean:\", scores.mean())\n",
    "    print(\"Standard deviation\", scores.std())\n",
    "    \n",
    "display_scores(tree_rmse_scores)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Scores: [66782.73843989 66960.118071   70347.95244419 74739.57052552\n",
      " 68031.13388938 71193.84183426 64969.63056405 68281.61137997\n",
      " 71552.91566558 67665.10082067]\n",
      "Mean: 69052.46136345083\n",
      "Standard deviation 2731.674001798347\n"
     ]
    }
   ],
   "source": [
    "#Doing the cross validation for the linear regression model\n",
    "\n",
    "lin_scores = cross_val_score(lin_reg, housing_prepared, housing_labels,\n",
    "                             scoring=\"neg_mean_squared_error\", cv=10)\n",
    "lin_rmse_scores = np.sqrt(-lin_scores)\n",
    "display_scores(lin_rmse_scores)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So the Decision Tree overfits and performs worse than the linear regression model\n",
    "\n",
    "We'll try a Random Forest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,\n",
       "                      max_features='auto', max_leaf_nodes=None,\n",
       "                      min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                      min_samples_leaf=1, min_samples_split=2,\n",
       "                      min_weight_fraction_leaf=0.0, n_estimators=10,\n",
       "                      n_jobs=None, oob_score=False, random_state=42, verbose=0,\n",
       "                      warm_start=False)"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestRegressor\n",
    "\n",
    "#We are using a random forest of 10 decision trees\n",
    "forest_reg = RandomForestRegressor(n_estimators=10, random_state=42)\n",
    "forest_reg.fit(housing_prepared, housing_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "21933.31414779769"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#predicting with the model and calculating the rmse score\n",
    "housing_predictions = forest_reg.predict(housing_prepared)\n",
    "forest_mse = mean_squared_error(housing_labels, housing_predictions)\n",
    "forest_rmse = np.sqrt(forest_mse)\n",
    "forest_rmse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Scores: [51646.44545909 48940.60114882 53050.86323649 54408.98730149\n",
      " 50922.14870785 56482.50703987 51864.52025526 49760.85037653\n",
      " 55434.21627933 53326.10093303]\n",
      "Mean: 52583.72407377466\n",
      "Standard deviation 2298.353351147122\n"
     ]
    }
   ],
   "source": [
    "#Running a cross validation on the model\n",
    "\n",
    "forest_scores = cross_val_score(forest_reg, housing_prepared, housing_labels,\n",
    "                                scoring =\"neg_mean_squared_error\", cv=10)\n",
    "forest_rmse_scores = np.sqrt(-forest_scores)\n",
    "display_scores(forest_rmse_scores)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is more promising but it is still being overfitted (The score on training set is lower than on the validation sets)\n",
    "\n",
    "We would regularise it later (constrain the model by tweaking the hyper parameters)\n",
    "\n",
    "Lets try a Support Vector Machine (SVM) Regressor (SVR)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "111094.6308539982"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from  sklearn.svm import SVR\n",
    "\n",
    "svm_reg = SVR(kernel=\"linear\")\n",
    "svm_reg.fit(housing_prepared, housing_labels)\n",
    "housing_predictions = svm_reg.predict(housing_prepared)\n",
    "svm_mse = mean_squared_error(housing_labels, housing_predictions)\n",
    "svm_rmse = np.sqrt(svm_mse)\n",
    "svm_rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "An SVR is considerably worse, we'll stick with a Random Forest"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## STEP 6: Finetune your model\n",
    "\n",
    "We make a list of the promising models and fine tune them"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score='raise-deprecating',\n",
       "             estimator=RandomForestRegressor(bootstrap=True, criterion='mse',\n",
       "                                             max_depth=None,\n",
       "                                             max_features='auto',\n",
       "                                             max_leaf_nodes=None,\n",
       "                                             min_impurity_decrease=0.0,\n",
       "                                             min_impurity_split=None,\n",
       "                                             min_samples_leaf=1,\n",
       "                                             min_samples_split=2,\n",
       "                                             min_weight_fraction_leaf=0.0,\n",
       "                                             n_estimators='warn', n_jobs=None,\n",
       "                                             oob_score=False, random_state=42,\n",
       "                                             verbose=0, warm_start=False),\n",
       "             iid='warn', n_jobs=None,\n",
       "             param_grid=[{'max_features': [2, 4, 6, 8],\n",
       "                          'n_estimators': [3, 10, 30]},\n",
       "                         {'bootstrap': [False], 'max_features': [2, 3, 4],\n",
       "                          'n_estimators': [3, 10]}],\n",
       "             pre_dispatch='2*n_jobs', refit=True, return_train_score=True,\n",
       "             scoring='neg_mean_squared_error', verbose=0)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Using Grid Search rather than fiddling with the hyperparameters\n",
    "#(values and combination) yourself \n",
    "\n",
    "#We would state the hyperparameters we want to test and what values to try out\n",
    "#we would use it for the random forest\n",
    "\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "param_grid = [\n",
    "    #try 12 (3x4) combinations of hyperparameters\n",
    "    {\"n_estimators\": [3, 10, 30], \"max_features\": [2, 4, 6, 8]},\n",
    "    #then try 6 (2x3) combinations with bootstrap set as False\n",
    "    {\"bootstrap\" : [False], \"n_estimators\" : [3, 10], \"max_features\" : [2, 3, 4]}\n",
    "] \n",
    "\n",
    "forest_reg = RandomForestRegressor(random_state=42)\n",
    "#train across 5 folds,  that's a total of (12+6)x5 = 90 rounds of training\n",
    "\n",
    "grid_search = GridSearchCV(forest_reg, param_grid, cv=5,\n",
    "                           scoring=\"neg_mean_squared_error\", return_train_score=True)\n",
    "grid_search.fit(housing_prepared, housing_labels)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'max_features': 8, 'n_estimators': 30}"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#We can print out the best parameters thus\n",
    "grid_search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "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>mean_fit_time</th>\n",
       "      <th>std_fit_time</th>\n",
       "      <th>mean_score_time</th>\n",
       "      <th>std_score_time</th>\n",
       "      <th>param_max_features</th>\n",
       "      <th>param_n_estimators</th>\n",
       "      <th>param_bootstrap</th>\n",
       "      <th>params</th>\n",
       "      <th>split0_test_score</th>\n",
       "      <th>split1_test_score</th>\n",
       "      <th>...</th>\n",
       "      <th>mean_test_score</th>\n",
       "      <th>std_test_score</th>\n",
       "      <th>rank_test_score</th>\n",
       "      <th>split0_train_score</th>\n",
       "      <th>split1_train_score</th>\n",
       "      <th>split2_train_score</th>\n",
       "      <th>split3_train_score</th>\n",
       "      <th>split4_train_score</th>\n",
       "      <th>mean_train_score</th>\n",
       "      <th>std_train_score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.061631</td>\n",
       "      <td>0.003490</td>\n",
       "      <td>0.002593</td>\n",
       "      <td>4.885974e-04</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 2, 'n_estimators': 3}</td>\n",
       "      <td>-3.837622e+09</td>\n",
       "      <td>-4.147108e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-4.053749e+09</td>\n",
       "      <td>1.519609e+08</td>\n",
       "      <td>18</td>\n",
       "      <td>-1.064113e+09</td>\n",
       "      <td>-1.105142e+09</td>\n",
       "      <td>-1.116550e+09</td>\n",
       "      <td>-1.112342e+09</td>\n",
       "      <td>-1.129650e+09</td>\n",
       "      <td>-1.105559e+09</td>\n",
       "      <td>2.220402e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.187891</td>\n",
       "      <td>0.001368</td>\n",
       "      <td>0.007580</td>\n",
       "      <td>4.890647e-04</td>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 2, 'n_estimators': 10}</td>\n",
       "      <td>-3.047771e+09</td>\n",
       "      <td>-3.254861e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.094381e+09</td>\n",
       "      <td>1.327046e+08</td>\n",
       "      <td>11</td>\n",
       "      <td>-5.927175e+08</td>\n",
       "      <td>-5.870952e+08</td>\n",
       "      <td>-5.776964e+08</td>\n",
       "      <td>-5.716332e+08</td>\n",
       "      <td>-5.802501e+08</td>\n",
       "      <td>-5.818785e+08</td>\n",
       "      <td>7.345821e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.639690</td>\n",
       "      <td>0.061539</td>\n",
       "      <td>0.025724</td>\n",
       "      <td>2.703511e-03</td>\n",
       "      <td>2</td>\n",
       "      <td>30</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 2, 'n_estimators': 30}</td>\n",
       "      <td>-2.689185e+09</td>\n",
       "      <td>-3.021086e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.849913e+09</td>\n",
       "      <td>1.626879e+08</td>\n",
       "      <td>9</td>\n",
       "      <td>-4.381089e+08</td>\n",
       "      <td>-4.391272e+08</td>\n",
       "      <td>-4.371702e+08</td>\n",
       "      <td>-4.376955e+08</td>\n",
       "      <td>-4.452654e+08</td>\n",
       "      <td>-4.394734e+08</td>\n",
       "      <td>2.966320e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.104520</td>\n",
       "      <td>0.002130</td>\n",
       "      <td>0.003192</td>\n",
       "      <td>3.989697e-04</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 4, 'n_estimators': 3}</td>\n",
       "      <td>-3.730181e+09</td>\n",
       "      <td>-3.786886e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.716852e+09</td>\n",
       "      <td>1.631421e+08</td>\n",
       "      <td>16</td>\n",
       "      <td>-9.865163e+08</td>\n",
       "      <td>-1.012565e+09</td>\n",
       "      <td>-9.169425e+08</td>\n",
       "      <td>-1.037400e+09</td>\n",
       "      <td>-9.707739e+08</td>\n",
       "      <td>-9.848396e+08</td>\n",
       "      <td>4.084607e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.355848</td>\n",
       "      <td>0.025740</td>\n",
       "      <td>0.009175</td>\n",
       "      <td>3.990174e-04</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 4, 'n_estimators': 10}</td>\n",
       "      <td>-2.666283e+09</td>\n",
       "      <td>-2.784511e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.781611e+09</td>\n",
       "      <td>1.268562e+08</td>\n",
       "      <td>8</td>\n",
       "      <td>-5.097115e+08</td>\n",
       "      <td>-5.162820e+08</td>\n",
       "      <td>-4.962893e+08</td>\n",
       "      <td>-5.436192e+08</td>\n",
       "      <td>-5.160297e+08</td>\n",
       "      <td>-5.163863e+08</td>\n",
       "      <td>1.542862e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1.023062</td>\n",
       "      <td>0.034984</td>\n",
       "      <td>0.025139</td>\n",
       "      <td>2.027250e-03</td>\n",
       "      <td>4</td>\n",
       "      <td>30</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 4, 'n_estimators': 30}</td>\n",
       "      <td>-2.387153e+09</td>\n",
       "      <td>-2.588448e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.537877e+09</td>\n",
       "      <td>1.214603e+08</td>\n",
       "      <td>3</td>\n",
       "      <td>-3.838835e+08</td>\n",
       "      <td>-3.880268e+08</td>\n",
       "      <td>-3.790867e+08</td>\n",
       "      <td>-4.040957e+08</td>\n",
       "      <td>-3.845520e+08</td>\n",
       "      <td>-3.879289e+08</td>\n",
       "      <td>8.571233e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.139820</td>\n",
       "      <td>0.003107</td>\n",
       "      <td>0.002992</td>\n",
       "      <td>1.507891e-07</td>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 6, 'n_estimators': 3}</td>\n",
       "      <td>-3.119657e+09</td>\n",
       "      <td>-3.586319e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.441447e+09</td>\n",
       "      <td>1.893141e+08</td>\n",
       "      <td>14</td>\n",
       "      <td>-9.245343e+08</td>\n",
       "      <td>-8.886939e+08</td>\n",
       "      <td>-9.353135e+08</td>\n",
       "      <td>-9.009801e+08</td>\n",
       "      <td>-8.624664e+08</td>\n",
       "      <td>-9.023976e+08</td>\n",
       "      <td>2.591445e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.477523</td>\n",
       "      <td>0.016671</td>\n",
       "      <td>0.009575</td>\n",
       "      <td>1.016925e-03</td>\n",
       "      <td>6</td>\n",
       "      <td>10</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 6, 'n_estimators': 10}</td>\n",
       "      <td>-2.549663e+09</td>\n",
       "      <td>-2.782039e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.704640e+09</td>\n",
       "      <td>1.471542e+08</td>\n",
       "      <td>6</td>\n",
       "      <td>-4.980344e+08</td>\n",
       "      <td>-5.045869e+08</td>\n",
       "      <td>-4.994664e+08</td>\n",
       "      <td>-4.990325e+08</td>\n",
       "      <td>-5.055542e+08</td>\n",
       "      <td>-5.013349e+08</td>\n",
       "      <td>3.100456e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1.422996</td>\n",
       "      <td>0.069055</td>\n",
       "      <td>0.024135</td>\n",
       "      <td>3.988505e-04</td>\n",
       "      <td>6</td>\n",
       "      <td>30</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 6, 'n_estimators': 30}</td>\n",
       "      <td>-2.370010e+09</td>\n",
       "      <td>-2.583638e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.514668e+09</td>\n",
       "      <td>1.285063e+08</td>\n",
       "      <td>2</td>\n",
       "      <td>-3.838538e+08</td>\n",
       "      <td>-3.804711e+08</td>\n",
       "      <td>-3.805218e+08</td>\n",
       "      <td>-3.856095e+08</td>\n",
       "      <td>-3.901917e+08</td>\n",
       "      <td>-3.841296e+08</td>\n",
       "      <td>3.617057e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.173337</td>\n",
       "      <td>0.004058</td>\n",
       "      <td>0.002992</td>\n",
       "      <td>2.611745e-07</td>\n",
       "      <td>8</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 8, 'n_estimators': 3}</td>\n",
       "      <td>-3.353504e+09</td>\n",
       "      <td>-3.348552e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.348851e+09</td>\n",
       "      <td>1.241864e+08</td>\n",
       "      <td>13</td>\n",
       "      <td>-9.228123e+08</td>\n",
       "      <td>-8.553031e+08</td>\n",
       "      <td>-8.603321e+08</td>\n",
       "      <td>-8.881964e+08</td>\n",
       "      <td>-9.151287e+08</td>\n",
       "      <td>-8.883545e+08</td>\n",
       "      <td>2.750227e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.540555</td>\n",
       "      <td>0.024479</td>\n",
       "      <td>0.008179</td>\n",
       "      <td>3.990412e-04</td>\n",
       "      <td>8</td>\n",
       "      <td>10</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 8, 'n_estimators': 10}</td>\n",
       "      <td>-2.571970e+09</td>\n",
       "      <td>-2.718994e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.674037e+09</td>\n",
       "      <td>1.392720e+08</td>\n",
       "      <td>5</td>\n",
       "      <td>-4.932416e+08</td>\n",
       "      <td>-4.815238e+08</td>\n",
       "      <td>-4.730979e+08</td>\n",
       "      <td>-5.155367e+08</td>\n",
       "      <td>-4.985555e+08</td>\n",
       "      <td>-4.923911e+08</td>\n",
       "      <td>1.459294e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1.550066</td>\n",
       "      <td>0.007026</td>\n",
       "      <td>0.021931</td>\n",
       "      <td>2.162476e-05</td>\n",
       "      <td>8</td>\n",
       "      <td>30</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'max_features': 8, 'n_estimators': 30}</td>\n",
       "      <td>-2.357390e+09</td>\n",
       "      <td>-2.546640e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.468326e+09</td>\n",
       "      <td>1.091647e+08</td>\n",
       "      <td>1</td>\n",
       "      <td>-3.841658e+08</td>\n",
       "      <td>-3.744500e+08</td>\n",
       "      <td>-3.773239e+08</td>\n",
       "      <td>-3.882250e+08</td>\n",
       "      <td>-3.810005e+08</td>\n",
       "      <td>-3.810330e+08</td>\n",
       "      <td>4.871017e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.087365</td>\n",
       "      <td>0.001016</td>\n",
       "      <td>0.002992</td>\n",
       "      <td>7.776979e-07</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>False</td>\n",
       "      <td>{'bootstrap': False, 'max_features': 2, 'n_est...</td>\n",
       "      <td>-3.785816e+09</td>\n",
       "      <td>-4.166012e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.955792e+09</td>\n",
       "      <td>1.900966e+08</td>\n",
       "      <td>17</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.291826</td>\n",
       "      <td>0.004319</td>\n",
       "      <td>0.008776</td>\n",
       "      <td>3.987590e-04</td>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>False</td>\n",
       "      <td>{'bootstrap': False, 'max_features': 2, 'n_est...</td>\n",
       "      <td>-2.810721e+09</td>\n",
       "      <td>-3.107789e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.987513e+09</td>\n",
       "      <td>1.539231e+08</td>\n",
       "      <td>10</td>\n",
       "      <td>-6.056477e-02</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-2.967449e+00</td>\n",
       "      <td>-6.056027e-01</td>\n",
       "      <td>1.181156e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.113681</td>\n",
       "      <td>0.002276</td>\n",
       "      <td>0.003005</td>\n",
       "      <td>1.552074e-05</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>False</td>\n",
       "      <td>{'bootstrap': False, 'max_features': 3, 'n_est...</td>\n",
       "      <td>-3.618324e+09</td>\n",
       "      <td>-3.441527e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.536728e+09</td>\n",
       "      <td>7.795196e+07</td>\n",
       "      <td>15</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-6.072840e+01</td>\n",
       "      <td>-1.214568e+01</td>\n",
       "      <td>2.429136e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.384569</td>\n",
       "      <td>0.013293</td>\n",
       "      <td>0.008789</td>\n",
       "      <td>4.052561e-04</td>\n",
       "      <td>3</td>\n",
       "      <td>10</td>\n",
       "      <td>False</td>\n",
       "      <td>{'bootstrap': False, 'max_features': 3, 'n_est...</td>\n",
       "      <td>-2.757999e+09</td>\n",
       "      <td>-2.851737e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.779927e+09</td>\n",
       "      <td>6.286611e+07</td>\n",
       "      <td>7</td>\n",
       "      <td>-2.089484e+01</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-5.465556e+00</td>\n",
       "      <td>-5.272080e+00</td>\n",
       "      <td>8.093117e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>0.140025</td>\n",
       "      <td>0.002566</td>\n",
       "      <td>0.002992</td>\n",
       "      <td>1.239777e-06</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>False</td>\n",
       "      <td>{'bootstrap': False, 'max_features': 4, 'n_est...</td>\n",
       "      <td>-3.134040e+09</td>\n",
       "      <td>-3.559375e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-3.305171e+09</td>\n",
       "      <td>1.879203e+08</td>\n",
       "      <td>12</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>0.464757</td>\n",
       "      <td>0.002264</td>\n",
       "      <td>0.008378</td>\n",
       "      <td>4.887931e-04</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>False</td>\n",
       "      <td>{'bootstrap': False, 'max_features': 4, 'n_est...</td>\n",
       "      <td>-2.525578e+09</td>\n",
       "      <td>-2.710011e+09</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.601971e+09</td>\n",
       "      <td>1.088031e+08</td>\n",
       "      <td>4</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-1.514119e-02</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-0.000000e+00</td>\n",
       "      <td>-3.028238e-03</td>\n",
       "      <td>6.056477e-03</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>18 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    mean_fit_time  std_fit_time  mean_score_time  std_score_time  \\\n",
       "0        0.061631      0.003490         0.002593    4.885974e-04   \n",
       "1        0.187891      0.001368         0.007580    4.890647e-04   \n",
       "2        0.639690      0.061539         0.025724    2.703511e-03   \n",
       "3        0.104520      0.002130         0.003192    3.989697e-04   \n",
       "4        0.355848      0.025740         0.009175    3.990174e-04   \n",
       "5        1.023062      0.034984         0.025139    2.027250e-03   \n",
       "6        0.139820      0.003107         0.002992    1.507891e-07   \n",
       "7        0.477523      0.016671         0.009575    1.016925e-03   \n",
       "8        1.422996      0.069055         0.024135    3.988505e-04   \n",
       "9        0.173337      0.004058         0.002992    2.611745e-07   \n",
       "10       0.540555      0.024479         0.008179    3.990412e-04   \n",
       "11       1.550066      0.007026         0.021931    2.162476e-05   \n",
       "12       0.087365      0.001016         0.002992    7.776979e-07   \n",
       "13       0.291826      0.004319         0.008776    3.987590e-04   \n",
       "14       0.113681      0.002276         0.003005    1.552074e-05   \n",
       "15       0.384569      0.013293         0.008789    4.052561e-04   \n",
       "16       0.140025      0.002566         0.002992    1.239777e-06   \n",
       "17       0.464757      0.002264         0.008378    4.887931e-04   \n",
       "\n",
       "   param_max_features param_n_estimators param_bootstrap  \\\n",
       "0                   2                  3             NaN   \n",
       "1                   2                 10             NaN   \n",
       "2                   2                 30             NaN   \n",
       "3                   4                  3             NaN   \n",
       "4                   4                 10             NaN   \n",
       "5                   4                 30             NaN   \n",
       "6                   6                  3             NaN   \n",
       "7                   6                 10             NaN   \n",
       "8                   6                 30             NaN   \n",
       "9                   8                  3             NaN   \n",
       "10                  8                 10             NaN   \n",
       "11                  8                 30             NaN   \n",
       "12                  2                  3           False   \n",
       "13                  2                 10           False   \n",
       "14                  3                  3           False   \n",
       "15                  3                 10           False   \n",
       "16                  4                  3           False   \n",
       "17                  4                 10           False   \n",
       "\n",
       "                                               params  split0_test_score  \\\n",
       "0              {'max_features': 2, 'n_estimators': 3}      -3.837622e+09   \n",
       "1             {'max_features': 2, 'n_estimators': 10}      -3.047771e+09   \n",
       "2             {'max_features': 2, 'n_estimators': 30}      -2.689185e+09   \n",
       "3              {'max_features': 4, 'n_estimators': 3}      -3.730181e+09   \n",
       "4             {'max_features': 4, 'n_estimators': 10}      -2.666283e+09   \n",
       "5             {'max_features': 4, 'n_estimators': 30}      -2.387153e+09   \n",
       "6              {'max_features': 6, 'n_estimators': 3}      -3.119657e+09   \n",
       "7             {'max_features': 6, 'n_estimators': 10}      -2.549663e+09   \n",
       "8             {'max_features': 6, 'n_estimators': 30}      -2.370010e+09   \n",
       "9              {'max_features': 8, 'n_estimators': 3}      -3.353504e+09   \n",
       "10            {'max_features': 8, 'n_estimators': 10}      -2.571970e+09   \n",
       "11            {'max_features': 8, 'n_estimators': 30}      -2.357390e+09   \n",
       "12  {'bootstrap': False, 'max_features': 2, 'n_est...      -3.785816e+09   \n",
       "13  {'bootstrap': False, 'max_features': 2, 'n_est...      -2.810721e+09   \n",
       "14  {'bootstrap': False, 'max_features': 3, 'n_est...      -3.618324e+09   \n",
       "15  {'bootstrap': False, 'max_features': 3, 'n_est...      -2.757999e+09   \n",
       "16  {'bootstrap': False, 'max_features': 4, 'n_est...      -3.134040e+09   \n",
       "17  {'bootstrap': False, 'max_features': 4, 'n_est...      -2.525578e+09   \n",
       "\n",
       "    split1_test_score  ...  mean_test_score  std_test_score  rank_test_score  \\\n",
       "0       -4.147108e+09  ...    -4.053749e+09    1.519609e+08               18   \n",
       "1       -3.254861e+09  ...    -3.094381e+09    1.327046e+08               11   \n",
       "2       -3.021086e+09  ...    -2.849913e+09    1.626879e+08                9   \n",
       "3       -3.786886e+09  ...    -3.716852e+09    1.631421e+08               16   \n",
       "4       -2.784511e+09  ...    -2.781611e+09    1.268562e+08                8   \n",
       "5       -2.588448e+09  ...    -2.537877e+09    1.214603e+08                3   \n",
       "6       -3.586319e+09  ...    -3.441447e+09    1.893141e+08               14   \n",
       "7       -2.782039e+09  ...    -2.704640e+09    1.471542e+08                6   \n",
       "8       -2.583638e+09  ...    -2.514668e+09    1.285063e+08                2   \n",
       "9       -3.348552e+09  ...    -3.348851e+09    1.241864e+08               13   \n",
       "10      -2.718994e+09  ...    -2.674037e+09    1.392720e+08                5   \n",
       "11      -2.546640e+09  ...    -2.468326e+09    1.091647e+08                1   \n",
       "12      -4.166012e+09  ...    -3.955792e+09    1.900966e+08               17   \n",
       "13      -3.107789e+09  ...    -2.987513e+09    1.539231e+08               10   \n",
       "14      -3.441527e+09  ...    -3.536728e+09    7.795196e+07               15   \n",
       "15      -2.851737e+09  ...    -2.779927e+09    6.286611e+07                7   \n",
       "16      -3.559375e+09  ...    -3.305171e+09    1.879203e+08               12   \n",
       "17      -2.710011e+09  ...    -2.601971e+09    1.088031e+08                4   \n",
       "\n",
       "    split0_train_score  split1_train_score  split2_train_score  \\\n",
       "0        -1.064113e+09       -1.105142e+09       -1.116550e+09   \n",
       "1        -5.927175e+08       -5.870952e+08       -5.776964e+08   \n",
       "2        -4.381089e+08       -4.391272e+08       -4.371702e+08   \n",
       "3        -9.865163e+08       -1.012565e+09       -9.169425e+08   \n",
       "4        -5.097115e+08       -5.162820e+08       -4.962893e+08   \n",
       "5        -3.838835e+08       -3.880268e+08       -3.790867e+08   \n",
       "6        -9.245343e+08       -8.886939e+08       -9.353135e+08   \n",
       "7        -4.980344e+08       -5.045869e+08       -4.994664e+08   \n",
       "8        -3.838538e+08       -3.804711e+08       -3.805218e+08   \n",
       "9        -9.228123e+08       -8.553031e+08       -8.603321e+08   \n",
       "10       -4.932416e+08       -4.815238e+08       -4.730979e+08   \n",
       "11       -3.841658e+08       -3.744500e+08       -3.773239e+08   \n",
       "12       -0.000000e+00       -0.000000e+00       -0.000000e+00   \n",
       "13       -6.056477e-02       -0.000000e+00       -0.000000e+00   \n",
       "14       -0.000000e+00       -0.000000e+00       -0.000000e+00   \n",
       "15       -2.089484e+01       -0.000000e+00       -0.000000e+00   \n",
       "16       -0.000000e+00       -0.000000e+00       -0.000000e+00   \n",
       "17       -0.000000e+00       -1.514119e-02       -0.000000e+00   \n",
       "\n",
       "    split3_train_score  split4_train_score  mean_train_score  std_train_score  \n",
       "0        -1.112342e+09       -1.129650e+09     -1.105559e+09     2.220402e+07  \n",
       "1        -5.716332e+08       -5.802501e+08     -5.818785e+08     7.345821e+06  \n",
       "2        -4.376955e+08       -4.452654e+08     -4.394734e+08     2.966320e+06  \n",
       "3        -1.037400e+09       -9.707739e+08     -9.848396e+08     4.084607e+07  \n",
       "4        -5.436192e+08       -5.160297e+08     -5.163863e+08     1.542862e+07  \n",
       "5        -4.040957e+08       -3.845520e+08     -3.879289e+08     8.571233e+06  \n",
       "6        -9.009801e+08       -8.624664e+08     -9.023976e+08     2.591445e+07  \n",
       "7        -4.990325e+08       -5.055542e+08     -5.013349e+08     3.100456e+06  \n",
       "8        -3.856095e+08       -3.901917e+08     -3.841296e+08     3.617057e+06  \n",
       "9        -8.881964e+08       -9.151287e+08     -8.883545e+08     2.750227e+07  \n",
       "10       -5.155367e+08       -4.985555e+08     -4.923911e+08     1.459294e+07  \n",
       "11       -3.882250e+08       -3.810005e+08     -3.810330e+08     4.871017e+06  \n",
       "12       -0.000000e+00       -0.000000e+00      0.000000e+00     0.000000e+00  \n",
       "13       -0.000000e+00       -2.967449e+00     -6.056027e-01     1.181156e+00  \n",
       "14       -0.000000e+00       -6.072840e+01     -1.214568e+01     2.429136e+01  \n",
       "15       -0.000000e+00       -5.465556e+00     -5.272080e+00     8.093117e+00  \n",
       "16       -0.000000e+00       -0.000000e+00      0.000000e+00     0.000000e+00  \n",
       "17       -0.000000e+00       -0.000000e+00     -3.028238e-03     6.056477e-03  \n",
       "\n",
       "[18 rows x 23 columns]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#To see the scores of all the hyperparameter combination\n",
    "pd.DataFrame(grid_search.cv_results_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "63669.05791727153 {'max_features': 2, 'n_estimators': 3}\n",
      "55627.16171305252 {'max_features': 2, 'n_estimators': 10}\n",
      "53384.57867637289 {'max_features': 2, 'n_estimators': 30}\n",
      "60965.99185930139 {'max_features': 4, 'n_estimators': 3}\n",
      "52740.98248528835 {'max_features': 4, 'n_estimators': 10}\n",
      "50377.344409590376 {'max_features': 4, 'n_estimators': 30}\n",
      "58663.84733372485 {'max_features': 6, 'n_estimators': 3}\n",
      "52006.15355973719 {'max_features': 6, 'n_estimators': 10}\n",
      "50146.465964159885 {'max_features': 6, 'n_estimators': 30}\n",
      "57869.25504027614 {'max_features': 8, 'n_estimators': 3}\n",
      "51711.09443660957 {'max_features': 8, 'n_estimators': 10}\n",
      "49682.25345942335 {'max_features': 8, 'n_estimators': 30}\n",
      "62895.088889905004 {'bootstrap': False, 'max_features': 2, 'n_estimators': 3}\n",
      "54658.14484390074 {'bootstrap': False, 'max_features': 2, 'n_estimators': 10}\n",
      "59470.399594730654 {'bootstrap': False, 'max_features': 3, 'n_estimators': 3}\n",
      "52725.01091081235 {'bootstrap': False, 'max_features': 3, 'n_estimators': 10}\n",
      "57490.612956065226 {'bootstrap': False, 'max_features': 4, 'n_estimators': 3}\n",
      "51009.51445842374 {'bootstrap': False, 'max_features': 4, 'n_estimators': 10}\n"
     ]
    }
   ],
   "source": [
    "#A better view of the means scores of all the combinations\n",
    "\n",
    "cvres = grid_search.cv_results_\n",
    "for mean_score, params in zip(cvres[\"mean_test_score\"], cvres[\"params\"]):\n",
    "    print(np.sqrt(-mean_score), params)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=None,\n",
       "                      max_features=8, max_leaf_nodes=None,\n",
       "                      min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                      min_samples_leaf=1, min_samples_split=2,\n",
       "                      min_weight_fraction_leaf=0.0, n_estimators=30,\n",
       "                      n_jobs=None, oob_score=False, random_state=42, verbose=0,\n",
       "                      warm_start=False)"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#We print out the best estimators \n",
    "grid_search.best_estimator_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is the best, we can specify the refit=True  (This is the default) in the GridSearchCV constructor \n",
    "to make it retrain the whole dataset using this combination"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv=5, error_score='raise-deprecating',\n",
       "                   estimator=RandomForestRegressor(bootstrap=True,\n",
       "                                                   criterion='mse',\n",
       "                                                   max_depth=None,\n",
       "                                                   max_features='auto',\n",
       "                                                   max_leaf_nodes=None,\n",
       "                                                   min_impurity_decrease=0.0,\n",
       "                                                   min_impurity_split=None,\n",
       "                                                   min_samples_leaf=1,\n",
       "                                                   min_samples_split=2,\n",
       "                                                   min_weight_fraction_leaf=0.0,\n",
       "                                                   n_estimators='warn',\n",
       "                                                   n_jobs=None, oob_score=False,\n",
       "                                                   random_sta...\n",
       "                                                   warm_start=False),\n",
       "                   iid='warn', n_iter=10, n_jobs=None,\n",
       "                   param_distributions={'max_features': <scipy.stats._distn_infrastructure.rv_frozen object at 0x0000022B77515AC8>,\n",
       "                                        'n_estimators': <scipy.stats._distn_infrastructure.rv_frozen object at 0x0000022B77515108>},\n",
       "                   pre_dispatch='2*n_jobs', random_state=42, refit=True,\n",
       "                   return_train_score=False, scoring='neg_mean_squared_error',\n",
       "                   verbose=0)"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Trying a Randomized Search\n",
    "\n",
    "#This gives you more control on the number of combination especially when \n",
    "# the search space is large\n",
    "#The benefits is that you can control your computing budget and\n",
    "# an run more specialized searches\n",
    "\n",
    "from sklearn.model_selection import RandomizedSearchCV\n",
    "from scipy.stats import randint\n",
    "\n",
    "#we would have 10 iterations but test 1-200 estimators and 1-8 features at each \n",
    "#iteration between 5 fold trainings\n",
    "param_distribs = {\n",
    "    \"n_estimators\" : randint(low=1, high=200),\n",
    "    \"max_features\" : randint(low=1, high=8)\n",
    "}\n",
    "\n",
    "forest_reg = RandomForestRegressor(random_state=42) \n",
    "rnd_search = RandomizedSearchCV(forest_reg, param_distributions= param_distribs,\n",
    "                                n_iter = 10, cv = 5, scoring = \"neg_mean_squared_error\",\n",
    "                                random_state=42)\n",
    "\n",
    "rnd_search.fit(housing_prepared, housing_labels)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "49150.657232934034 {'max_features': 7, 'n_estimators': 180}\n",
      "51389.85295710133 {'max_features': 5, 'n_estimators': 15}\n",
      "50796.12045980556 {'max_features': 3, 'n_estimators': 72}\n",
      "50835.09932039744 {'max_features': 5, 'n_estimators': 21}\n",
      "49280.90117886215 {'max_features': 7, 'n_estimators': 122}\n",
      "50774.86679035961 {'max_features': 3, 'n_estimators': 75}\n",
      "50682.75001237282 {'max_features': 3, 'n_estimators': 88}\n",
      "49608.94061293652 {'max_features': 5, 'n_estimators': 100}\n",
      "50473.57642831875 {'max_features': 3, 'n_estimators': 150}\n",
      "64429.763804893395 {'max_features': 5, 'n_estimators': 2}\n"
     ]
    }
   ],
   "source": [
    "#printing the results\n",
    "cvres = rnd_search.cv_results_\n",
    "for mean_score, params in zip(cvres[\"mean_test_score\"], cvres[\"params\"]):\n",
    "    print(np.sqrt(-mean_score), params)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So the Random Search Hypertuning was better with a lower RMSE value of 49150.66 (Random Search) vs 49682.25 (Grid Search)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ensemble Methods\n",
    "\n",
    "We would combine the models together (ensemble them)to get a  better model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([7.33442355e-02, 6.29090705e-02, 4.11437985e-02, 1.46726854e-02,\n",
       "       1.41064835e-02, 1.48742809e-02, 1.42575993e-02, 3.66158981e-01,\n",
       "       5.64191792e-02, 1.08792957e-01, 5.33510773e-02, 1.03114883e-02,\n",
       "       1.64780994e-01, 6.02803867e-05, 1.96041560e-03, 2.85647464e-03])"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#First we need to see the feature importances from the grid search\n",
    "feature_importances = grid_search.best_estimator_.feature_importances_\n",
    "feature_importances"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(0.36615898061813423, 'median_income'),\n",
       " (0.16478099356159054, 'INLAND'),\n",
       " (0.10879295677551575, 'pop_per_hhold'),\n",
       " (0.07334423551601243, 'longitude'),\n",
       " (0.06290907048262032, 'latitude'),\n",
       " (0.056419179181954014, 'rooms_per_hhold'),\n",
       " (0.053351077347675815, 'bedrooms_per_room'),\n",
       " (0.04114379847872964, 'housing_median_age'),\n",
       " (0.014874280890402769, 'population'),\n",
       " (0.014672685420543239, 'total_rooms'),\n",
       " (0.014257599323407808, 'households'),\n",
       " (0.014106483453584104, 'total_bedrooms'),\n",
       " (0.010311488326303788, '<1H OCEAN'),\n",
       " (0.0028564746373201584, 'NEAR OCEAN'),\n",
       " (0.0019604155994780706, 'NEAR BAY'),\n",
       " (6.0280386727366e-05, 'ISLAND')]"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Let's display the importance between scores beside the corresponding attribute\n",
    "extra_attribs = [\"rooms_per_hhold\", \"pop_per_hhold\", \"bedrooms_per_room\"]\n",
    "cat_one_hot_attribs = list(cat_encoder.categories_[0])\n",
    "attributes = num_attribs + extra_attribs + cat_one_hot_attribs\n",
    "sorted(zip(feature_importances, attributes), reverse = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally we would evaluate the model on the test set "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "#We would get the predictors and the label from the test set and run our full\n",
    "#transformation pipeline on the test data \n",
    "#We call transform() not fit_transform() since we are not training on this set\n",
    "\n",
    "\n",
    "\n",
    "#we can use the best estimator from the grid search\n",
    "#final_model = grid_search.best_estimator_\n",
    "\n",
    "#or the best estimator from the randomcv search (It scored better i.e lower rmse)\n",
    "final_model = rnd_search.best_estimator_\n",
    "\n",
    "X_test = strat_test_set.drop(\"median_house_value\", axis=1)\n",
    "y_test = strat_test_set[\"median_house_value\"].copy()\n",
    "\n",
    "X_test_prepared = full_pipeline.transform(X_test)\n",
    "final_predictions = final_model.predict(X_test_prepared)\n",
    "\n",
    "#Let's calculate the RMSE\n",
    "final_mse = mean_squared_error(y_test, final_predictions)\n",
    "final_rmse = np.sqrt(final_mse)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "46910.92117024934"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Display the rmse value\n",
    "final_rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's also compute a confidence level for the test RMSE\n",
    "We are computing a 95% confidence level\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([44945.41068188, 48797.32686039])"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from scipy import stats\n",
    "\n",
    "confidence = 0.95\n",
    "squared_errors = (final_predictions - y_test) ** 2\n",
    "mean_of_squared_errors = squared_errors.mean()\n",
    "m = len(squared_errors)\n",
    "\n",
    "#We use T-scores to compute the 95% confidence level\n",
    "np.sqrt(stats.t.interval(confidence, m - 1, \n",
    "                         loc = mean_of_squared_errors,\n",
    "                         scale = stats.sem(squared_errors)))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(44945.4106818769, 48797.32686039384)"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "######\n",
    "# Alternate codes\n",
    "\n",
    "#Computing the interval manually\n",
    "\n",
    "#ppf is the percent point function\n",
    "tscore = stats.t.ppf((1+confidence) / 2, df= m - 1)\n",
    "tmargin = tscore * squared_errors.std(ddof=1) / np.sqrt(m)\n",
    "np.sqrt(mean_of_squared_errors-tmargin), np.sqrt(mean_of_squared_errors + tmargin)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(44945.999723355504, 48796.78430953023)"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#computing the interval manually but with zscores\n",
    "\n",
    "zscore = stats.norm.ppf((1+confidence) / 2)\n",
    "zmargin = zscore * squared_errors.std(ddof=1) / np.sqrt(m)\n",
    "np.sqrt(mean_of_squared_errors-zmargin), np.sqrt(mean_of_squared_errors + zmargin)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## STEP 7: Present our Solution"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## STEP 8: Launch, Monitor, Maintain your system."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## SOLUTION TO THE EXERCISES "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. \n",
    "Question: Try a Support Vector Machine regressor (`sklearn.svm.SVR`), with various hyperparameters such as `kernel=\"linear\"` (with various values for the `C` hyperparameter) or `kernel=\"rbf\"` (with various values for the `C` and `gamma` hyperparameters). Don't worry about what these hyperparameters mean for now. How does the best `SVR` predictor perform?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 50 candidates, totalling 250 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=4)]: Using backend LokyBackend with 4 concurrent workers.\n",
      "[Parallel(n_jobs=4)]: Done  33 tasks      | elapsed:  1.5min\n",
      "[Parallel(n_jobs=4)]: Done 154 tasks      | elapsed:  7.6min\n",
      "[Parallel(n_jobs=4)]: Done 250 out of 250 | elapsed: 11.6min finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score='raise-deprecating',\n",
       "             estimator=SVR(C=1.0, cache_size=200, coef0=0.0, degree=3,\n",
       "                           epsilon=0.1, gamma='auto_deprecated', kernel='rbf',\n",
       "                           max_iter=-1, shrinking=True, tol=0.001,\n",
       "                           verbose=False),\n",
       "             iid='warn', n_jobs=4,\n",
       "             param_grid=[{'C': [10.0, 30.0, 100.0, 300.0, 1000.0, 3000.0,\n",
       "                                10000.0, 30000.0],\n",
       "                          'kernel': ['linear']},\n",
       "                         {'C': [1.0, 3.0, 10.0, 30.0, 100.0, 300.0, 1000.0],\n",
       "                          'gamma': [0.01, 0.03, 0.1, 0.3, 1.0, 3.0],\n",
       "                          'kernel': ['rbf']}],\n",
       "             pre_dispatch='2*n_jobs', refit=True, return_train_score=False,\n",
       "             scoring='neg_mean_squared_error', verbose=2)"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "##We previously imported GridSearch and SVR\n",
    "\n",
    "#each possible combination within a dict is tested\n",
    "param_grid = [\n",
    "    {\"kernel\" : [\"linear\"], \"C\": [10., 30., 100., 300., 1000., 3000., 10000., 30000.0]},\n",
    "    {\"kernel\" : [\"rbf\"], \"C\": [1.0, 3.0, 10., 30., 100., 300., 1000.0],\n",
    "     \"gamma\" : [0.01, 0.03, 0.1, 0.3, 1.0, 3.0]},\n",
    "]\n",
    "\n",
    "svm_reg = SVR()\n",
    "grid_search = GridSearchCV(svm_reg, param_grid, cv=5, scoring=\"neg_mean_squared_error\",\n",
    "                           verbose=2, n_jobs=4)\n",
    "grid_search.fit(housing_prepared, housing_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "70363.90312768027"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Printing out the score of the best model\n",
    "negative_mse = grid_search.best_score_\n",
    "rmse = np.sqrt(-negative_mse)\n",
    "rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is still much worse than the RandomForestRegressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'C': 30000.0, 'kernel': 'linear'}"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#To get the best hyperparameters found\n",
    "grid_search.best_params_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Linear kernel seems better than the RBF kernel. I'll use a higher \"C\" value since the best parameter is the highest C value tested. (I'll also launch the grid search again removing the smallest values and putting higher C values.)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. \n",
    "Question: Try replacing `GridSearchCV` with `RandomizedSearchCV`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 50 candidates, totalling 250 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=4)]: Using backend LokyBackend with 4 concurrent workers.\n",
      "C:\\ProgramData\\Anaconda3\\envs\\myMLenv\\lib\\site-packages\\joblib\\externals\\loky\\process_executor.py:706: UserWarning: A worker stopped while some jobs were given to the executor. This can be caused by a too short worker timeout or by a memory leak.\n",
      "  \"timeout or by a memory leak.\", UserWarning\n",
      "[Parallel(n_jobs=4)]: Done  33 tasks      | elapsed:  1.6min\n",
      "[Parallel(n_jobs=4)]: Done 154 tasks      | elapsed: 11.6min\n",
      "[Parallel(n_jobs=4)]: Done 250 out of 250 | elapsed: 19.7min finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "RandomizedSearchCV(cv=5, error_score='raise-deprecating',\n",
       "                   estimator=SVR(C=1.0, cache_size=200, coef0=0.0, degree=3,\n",
       "                                 epsilon=0.1, gamma='auto_deprecated',\n",
       "                                 kernel='rbf', max_iter=-1, shrinking=True,\n",
       "                                 tol=0.001, verbose=False),\n",
       "                   iid='warn', n_iter=50, n_jobs=4,\n",
       "                   param_distributions={'C': <scipy.stats._distn_infrastructure.rv_frozen object at 0x0000022B77522D88>,\n",
       "                                        'gamma': <scipy.stats._distn_infrastructure.rv_frozen object at 0x0000022B77522188>,\n",
       "                                        'kernel': ['linear', 'rbf']},\n",
       "                   pre_dispatch='2*n_jobs', random_state=42, refit=True,\n",
       "                   return_train_score=False, scoring='neg_mean_squared_error',\n",
       "                   verbose=2)"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#We have already imported RandomizedSearchCV\n",
    "from scipy.stats import expon, reciprocal\n",
    "\n",
    "# see https://docs.scipy.org/doc/scipy/reference/stats.html\n",
    "# for `expon()` and `reciprocal()` documentation and more probability distribution functions.\n",
    "\n",
    "#Note: gamma is ignored when the kernel is \"linear\"\n",
    "param_distribs = {\n",
    "    \"kernel\" : [\"linear\", \"rbf\"],\n",
    "    \"C\" : reciprocal(20, 200000),\n",
    "    \"gamma\" : expon(scale=1.0),\n",
    "}\n",
    "\n",
    "svm_reg = SVR()\n",
    "rnd_search_svm = RandomizedSearchCV(svm_reg, param_distributions=param_distribs,\n",
    "                                n_iter = 50, cv=5, scoring=\"neg_mean_squared_error\",\n",
    "                                verbose=2, n_jobs=4, random_state=42)\n",
    "rnd_search_svm.fit(housing_prepared, housing_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "54767.99053704409"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#The best model achieved the following score (5 fold cross validation)\n",
    "\n",
    "negative_mse = rnd_search_svm.best_score_\n",
    "rmse = np.sqrt(-negative_mse)\n",
    "rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is closer to the performance of the RandomForestRegressor. Progress!!!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'C': 157055.10989448498, 'gamma': 0.26497040005002437, 'kernel': 'rbf'}"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#To get the best hyperparameters\n",
    "rnd_search_svm.best_params_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This time the search found a good set of hyperparameters for the RBF kernel. Randomized search tends to find better hyperparameters than grid search in the same amount of time."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Extra material"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's look at the exponential distribution we used, with `scale=1.0`. Note that some samples are much larger or smaller than 1.0, but when you look at the log of the distribution, you can see that most values are actually concentrated roughly in the range of exp(-2) to exp(+2), which is about 0.1 to 7.4.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "expon_distrib = expon(scale=1.0)\n",
    "samples = expon_distrib.rvs(10000, random_state=42)\n",
    "plt.figure(figsize=(10,4))\n",
    "plt.subplot(121)\n",
    "plt.title(\"Exponential distribution (scale=1.0)\")\n",
    "plt.hist(samples, bins =50)\n",
    "plt.subplot(122)\n",
    "plt.title(\"Log of this distribution\")\n",
    "plt.hist(np.log(samples), bins=50)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The distribution we used for `C` looks quite different: the scale of the samples is picked from a uniform distribution within a given range, which is why the right graph, which represents the log of the samples, looks roughly constant. This distribution is useful when you don't have a clue of what the target scale is:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 720x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "reciprocal_distrib = reciprocal(20, 200000)\n",
    "samples = reciprocal_distrib.rvs(10000, random_state=42)\n",
    "plt.figure(figsize=(10, 4))\n",
    "plt.subplot(121)\n",
    "plt.title(\"Reciprocal distribution (scale=1.0)\")\n",
    "plt.hist(samples, bins=50)\n",
    "plt.subplot(122)\n",
    "plt.title(\"Log of this distribution\")\n",
    "plt.hist(np.log(samples), bins=50)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The reciprocal distribution is useful when you have no idea what the scale of the hyperparameter should be (indeed, as you can see on the figure on the right, all scales are equally likely, within the given range), whereas the exponential distribution is best when you know (more or less) what the scale of the hyperparameter should be."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3\n",
    "\n",
    "Question: Try adding a transformer in the preparation pipeline to select only the most important attributes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "#We are going to create a custom transformer\n",
    "from sklearn.base import BaseEstimator, TransformerMixin\n",
    "\n",
    "def indices_of_top_k(arr, k):\n",
    "    return np.sort(np.argpartition(np.array(arr), -k)[-k:]) #select the top k indices \n",
    "\n",
    "class TopFeatureSelector(BaseEstimator, TransformerMixin):\n",
    "    def __init__(self, feature_importances, k):\n",
    "        self.feature_importances = feature_importances   \n",
    "        self.k = k\n",
    "    def fit(self, X, y=None):\n",
    "        self.feature_indices = indices_of_top_k(self.feature_importances, self.k)\n",
    "        return self\n",
    "    def transform(self, X):\n",
    "        return X[:, self.feature_indices]\n",
    "        "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note: this feature selector assumes that you have already computed the feature importances somehow (for example using a `RandomForestRegressor`). You may be tempted to compute them directly in the `TopFeatureSelector`'s `fit()` method, however this would likely slow down grid/randomized search since the feature importances would have to be computed for every hyperparameter combination (unless you implement some sort of cache)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's define the number of top features we want to keep:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "k = 5\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 0,  1,  7,  9, 12], dtype=int64)"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Let's look for the indices of the top k features\n",
    "top_k_feature_indices = indices_of_top_k(feature_importances, k)\n",
    "top_k_feature_indices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['longitude', 'latitude', 'median_income', 'pop_per_hhold',\n",
       "       'INLAND'], dtype='<U18')"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.array(attributes)[top_k_feature_indices]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[(0.36615898061813423, 'median_income'),\n",
       " (0.16478099356159054, 'INLAND'),\n",
       " (0.10879295677551575, 'pop_per_hhold'),\n",
       " (0.07334423551601243, 'longitude'),\n",
       " (0.06290907048262032, 'latitude')]"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Double checking that these are the top k features\n",
    "sorted(zip(feature_importances, attributes), reverse=True)[:k]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "FULL PIPELINE WITH FEATURE SELECTION"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "preparation_and_feature_selection_pipeline = Pipeline([\n",
    "        (\"preparation\", full_pipeline),\n",
    "        (\"feature_selection\", TopFeatureSelector(feature_importances,k))\n",
    "    ])\n",
    "\n",
    "housing_prepared_top_k_features = preparation_and_feature_selection_pipeline.fit_transform(housing)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-1.15604281,  0.77194962, -0.61493744, -0.08649871,  0.        ],\n",
       "       [-1.17602483,  0.6596948 ,  1.33645936, -0.03353391,  0.        ],\n",
       "       [ 1.18684903, -1.34218285, -0.5320456 , -0.09240499,  0.        ]])"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Lets look at the features of the first 3 instances\n",
    "\n",
    "housing_prepared_top_k_features[0:3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[-1.15604281,  0.77194962, -0.61493744, -0.08649871,  0.        ],\n",
       "       [-1.17602483,  0.6596948 ,  1.33645936, -0.03353391,  0.        ],\n",
       "       [ 1.18684903, -1.34218285, -0.5320456 , -0.09240499,  0.        ]])"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Let's compare it with the top k features\n",
    "\n",
    "housing_prepared[0:3, top_k_feature_indices]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We are good to go then"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4.\n",
    "\n",
    "Question: Try creating a single pipeline that does the full data preparation plus the final prediction."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Full pipeline (Prepare select and predict pipeline)\n",
    "\n",
    "prepare_select_and_predict_pipeline = Pipeline([\n",
    "    (\"preparation\", full_pipeline),\n",
    "    (\"feature_selection\", TopFeatureSelector(feature_importances, k)),\n",
    "    (\"svm_reg\", rnd_search_svm.best_estimator_)  # or SVR(**rnd_search.best_params_)\n",
    "])\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I'll use the best Random Forest Regressor later after I fx the naming of variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pipeline(memory=None,\n",
       "         steps=[('preparation',\n",
       "                 ColumnTransformer(n_jobs=None, remainder='drop',\n",
       "                                   sparse_threshold=0.3,\n",
       "                                   transformer_weights=None,\n",
       "                                   transformers=[('num',\n",
       "                                                  Pipeline(memory=None,\n",
       "                                                           steps=[('imputer',\n",
       "                                                                   SimpleImputer(add_indicator=False,\n",
       "                                                                                 copy=True,\n",
       "                                                                                 fill_value=None,\n",
       "                                                                                 missing_values=nan,\n",
       "                                                                                 strategy='median',\n",
       "                                                                                 verbose=0)),\n",
       "                                                                  ('attribs_adder',\n",
       "                                                                   CombinedAttributesAdder(add_...\n",
       "       1.41064835e-02, 1.48742809e-02, 1.42575993e-02, 3.66158981e-01,\n",
       "       5.64191792e-02, 1.08792957e-01, 5.33510773e-02, 1.03114883e-02,\n",
       "       1.64780994e-01, 6.02803867e-05, 1.96041560e-03, 2.85647464e-03]),\n",
       "                                    k=5)),\n",
       "                ('svm_reg',\n",
       "                 SVR(C=157055.10989448498, cache_size=200, coef0=0.0, degree=3,\n",
       "                     epsilon=0.1, gamma=0.26497040005002437, kernel='rbf',\n",
       "                     max_iter=-1, shrinking=True, tol=0.001, verbose=False))],\n",
       "         verbose=False)"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Let's Train the model\n",
    "prepare_select_and_predict_pipeline.fit(housing, housing_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Predictions:\t [203214.28978849 371846.88152572 173295.65441612  47328.3970888 ]\n",
      "Labels:\t\t [286600.0, 340600.0, 196900.0, 46300.0]\n"
     ]
    }
   ],
   "source": [
    "##Trying the pipeline on a few instances\n",
    "\n",
    "some_data = housing.iloc[:4]\n",
    "some_labels = housing_labels.iloc[:4]\n",
    "\n",
    "print(\"Predictions:\\t\", prepare_select_and_predict_pipeline.predict(some_data))\n",
    "print(\"Labels:\\t\\t\", list(some_labels))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Predictions are not fantastic, I'll try the RandomForestRegressor"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5.\n",
    "\n",
    "Question: Automatically explore some preparation options using `GridSearchCV`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fitting 5 folds for each of 48 candidates, totalling 240 fits\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[Parallel(n_jobs=4)]: Using backend LokyBackend with 4 concurrent workers.\n",
      "C:\\ProgramData\\Anaconda3\\envs\\myMLenv\\lib\\site-packages\\joblib\\externals\\loky\\process_executor.py:706: UserWarning: A worker stopped while some jobs were given to the executor. This can be caused by a too short worker timeout or by a memory leak.\n",
      "  \"timeout or by a memory leak.\", UserWarning\n",
      "[Parallel(n_jobs=4)]: Done  33 tasks      | elapsed:  1.0min\n",
      "[Parallel(n_jobs=4)]: Done 154 tasks      | elapsed:  8.3min\n",
      "[Parallel(n_jobs=4)]: Done 240 out of 240 | elapsed: 17.6min finished\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=5, error_score='raise-deprecating',\n",
       "             estimator=Pipeline(memory=None,\n",
       "                                steps=[('preparation',\n",
       "                                        ColumnTransformer(n_jobs=None,\n",
       "                                                          remainder='drop',\n",
       "                                                          sparse_threshold=0.3,\n",
       "                                                          transformer_weights=None,\n",
       "                                                          transformers=[('num',\n",
       "                                                                         Pipeline(memory=None,\n",
       "                                                                                  steps=[('imputer',\n",
       "                                                                                          SimpleImputer(add_indicator=False,\n",
       "                                                                                                        copy=True,\n",
       "                                                                                                        fill_value=None,\n",
       "                                                                                                        missing_values=nan,\n",
       "                                                                                                        strategy='media...\n",
       "                                            kernel='rbf', max_iter=-1,\n",
       "                                            shrinking=True, tol=0.001,\n",
       "                                            verbose=False))],\n",
       "                                verbose=False),\n",
       "             iid='warn', n_jobs=4,\n",
       "             param_grid=[{'feature_selection__k': [1, 2, 3, 4, 5, 6, 7, 8, 9,\n",
       "                                                   10, 11, 12, 13, 14, 15, 16],\n",
       "                          'preparation__num__imputer__strategy': ['mean',\n",
       "                                                                  'median',\n",
       "                                                                  'most_frequent']}],\n",
       "             pre_dispatch='2*n_jobs', refit=True, return_train_score=False,\n",
       "             scoring='neg_mean_squared_error', verbose=2)"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "param_grid = [{\n",
    "    \"preparation__num__imputer__strategy\" : [\"mean\", \"median\", \"most_frequent\"],\n",
    "    \"feature_selection__k\" : list(range(1, len(feature_importances)+ 1))\n",
    "}]\n",
    "\n",
    "grid_search_prep = GridSearchCV(prepare_select_and_predict_pipeline, param_grid, cv=5,\n",
    "                                scoring=\"neg_mean_squared_error\", verbose=2, n_jobs=4)\n",
    "\n",
    "grid_search_prep.fit(housing, housing_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'feature_selection__k': 15,\n",
       " 'preparation__num__imputer__strategy': 'most_frequent'}"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grid_search_prep.best_params_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The best imputer strategy is `most_frequent` and apparently almost all features are useful (15 out of 16). The last one (`ISLAND`) seems to just add some noise."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## EXTRA MATERIAL AND RESOLVED PROBLEMS"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Final solution\n",
    "We would implement all the information that we have learnt and do the final tweaks to the model.\n",
    "\n",
    "They are\n",
    "Best Features = 15/16\n",
    "Best Imputer Strategy = \"Most Frequent\" instead of \"Median\"\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "##I changed the Imputer strategy and recomputed the pipeline\n",
    "\n",
    "num_pipeline = Pipeline([\n",
    "    (\"imputer\", SimpleImputer(strategy='most_frequent')),\n",
    "    (\"attribs_adder\", CombinedAttributesAdder()),\n",
    "    (\"std_scaler\", StandardScaler())\n",
    "])\n",
    "\n",
    "\n",
    "full_pipeline = ColumnTransformer([\n",
    "    (\"num\", num_pipeline, num_attribs),\n",
    "    (\"cat\", OneHotEncoder(),cat_attribs)\n",
    "])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "###Final Model\n",
    "\n",
    "#Full pipeline (Prepare select and predict pipeline)\n",
    "\n",
    "final_prepare_select_and_predict_pipeline = Pipeline([\n",
    "    (\"preparation\", full_pipeline),\n",
    "    (\"feature_selection\", TopFeatureSelector(feature_importances, k=15)),\n",
    "    (\"forest_reg\", rnd_search.best_estimator_)  # or SVR(**rnd_search.best_params_)\n",
    "])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pipeline(memory=None,\n",
       "         steps=[('preparation',\n",
       "                 ColumnTransformer(n_jobs=None, remainder='drop',\n",
       "                                   sparse_threshold=0.3,\n",
       "                                   transformer_weights=None,\n",
       "                                   transformers=[('num',\n",
       "                                                  Pipeline(memory=None,\n",
       "                                                           steps=[('imputer',\n",
       "                                                                   SimpleImputer(add_indicator=False,\n",
       "                                                                                 copy=True,\n",
       "                                                                                 fill_value=None,\n",
       "                                                                                 missing_values=nan,\n",
       "                                                                                 strategy='most_frequent',\n",
       "                                                                                 verbose=0)),\n",
       "                                                                  ('attribs_adder',\n",
       "                                                                   CombinedAttributesAdd...\n",
       "       1.64780994e-01, 6.02803867e-05, 1.96041560e-03, 2.85647464e-03]),\n",
       "                                    k=15)),\n",
       "                ('forest_reg',\n",
       "                 RandomForestRegressor(bootstrap=True, criterion='mse',\n",
       "                                       max_depth=None, max_features=7,\n",
       "                                       max_leaf_nodes=None,\n",
       "                                       min_impurity_decrease=0.0,\n",
       "                                       min_impurity_split=None,\n",
       "                                       min_samples_leaf=1, min_samples_split=2,\n",
       "                                       min_weight_fraction_leaf=0.0,\n",
       "                                       n_estimators=180, n_jobs=None,\n",
       "                                       oob_score=False, random_state=42,\n",
       "                                       verbose=0, warm_start=False))],\n",
       "         verbose=False)"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Fit the model\n",
    "final_prepare_select_and_predict_pipeline.fit(housing, housing_labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Lets evaluate this model on the training set\n",
    "housing_predictions = final_prepare_select_and_predict_pipeline.predict(housing) #Remember we run the predict on the DF since the pipeline handles transforms\n",
    "final_tree_mse = mean_squared_error(housing_labels, housing_predictions)\n",
    "final_tree_rmse = np.sqrt(tree_mse)\n",
    "final_tree_rmse"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Scores: [48418.75612834 46584.32714032 48553.24516237 50371.84796564\n",
      " 47885.68524655 52230.55615726 47245.98047895 46775.16748442\n",
      " 51443.51061293 48280.12167478]\n",
      "Mean: 48778.919805157406\n",
      "Standard deviation 1842.8167456314804\n"
     ]
    }
   ],
   "source": [
    "## Score the model using Cross validation\n",
    "\n",
    "final_scores = cross_val_score(final_prepare_select_and_predict_pipeline, housing, housing_labels,\n",
    "                         scoring=\"neg_mean_squared_error\", cv=10)\n",
    "\n",
    "final_rmse_scores = np.sqrt(-final_scores) \n",
    "\n",
    "display_scores(final_rmse_scores)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "### I will create a preparation and feature selection pipeline for the test data\n",
    "\n",
    "final_prepare_and_select_pipeline = Pipeline([\n",
    "    (\"preparation\", full_pipeline),\n",
    "    (\"feature_selection\", TopFeatureSelector(feature_importances, k=15)),\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Finally Test it on the Test set and calculate a rmse possibly a 95% confidence\n",
    "\n",
    "X_test = strat_test_set.drop(\"median_house_value\", axis=1)\n",
    "y_test = strat_test_set[\"median_house_value\"].copy()\n",
    "\n",
    "\n",
    "final_predictions = final_prepare_select_and_predict_pipeline.predict(X_test) #This is a full preparation, selection and predict pipeline, we just plug the test set there\n",
    "\n",
    "#Let's calculate the RMSE\n",
    "final_mse = mean_squared_error(y_test, final_predictions)\n",
    "final_rmse = np.sqrt(final_mse)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "46742.84638753447"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Display the rmse values\n",
    "final_rmse"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Yup! we achieved a lower rmse value than all previous models. Further tweaking is still possible 👏🏾"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Model Persistence using joblib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "my_complete_model = final_prepare_select_and_predict_pipeline\n",
    "\n",
    "\n",
    "best_svm_model_pipeline = Pipeline([\n",
    "    (\"preparation\", full_pipeline),\n",
    "    (\"feature_selection\", TopFeatureSelector(feature_importances, k)),\n",
    "    (\"svm_reg\", grid_search_prep.best_estimator_)  # or SVR(**rnd_search_prep.best_params_)\n",
    "])\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['best_svm_model.pkl']"
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#from sklearn.externals import joblib # deprecated, use import joblib instead\n",
    "import joblib\n",
    "\n",
    "joblib.dump(my_complete_model, \"complete_forest_reg_model.pkl\") # DIFF\n",
    "joblib.dump(best_svm_model_pipeline, \"best_svm_model.pkl\")\n",
    "#...\n",
    "\n",
    "#To load the model\n",
    "#my_model_loaded = joblib.load(\"my_model.pkl\") # If you want to load the model"
   ]
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "d18c3d99e326808c0ea9776033fedd1fe4d7964bdcf2f58ebb2d89bb2832f5d0"
  },
  "kernelspec": {
   "display_name": "Python 3.7.9 64-bit ('myMLenv': conda)",
   "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": "3.7.9"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}