04.05-Frequent-Category-Imputation.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Frequent category imputation | Mode imputation\n",
    "\n",
    "Imputation is the act of replacing missing data with statistical estimates of the missing values. The goal of any imputation technique is to produce a **complete dataset** that can be used to train machine learning models.\n",
    "\n",
    "Mode imputation consists of replacing all occurrences of missing values (NA) within a variable by the mode, which in other words refers to the **most frequent value** or **most frequent category**.\n",
    "\n",
    "\n",
    "### Which variables can I impute with most frequent / mode Imputation?\n",
    "\n",
    "Although the mode, or most frequent value can be calculated for both numerical and categorical variables, in practice, we only use this technique on categorical variables. The reason is that for numerical variables, the mean or the median tend to better represent the average value of the population.\n",
    "\n",
    "\n",
    "### Assumptions\n",
    "\n",
    "- Data is missing completely at random (MCAR)\n",
    "- The missing observations, most likely look like the majority of the observations in the variable (aka, the mode)\n",
    "\n",
    "The rationale is that the value that should have been seen for the missing observation is most likely the most frequent value or category of the variable.\n",
    "\n",
    "\n",
    "### Advantages\n",
    "\n",
    "- Easy to implement\n",
    "- Fast way of obtaining complete datasets\n",
    "- Can be integrated in production (during model deployment)\n",
    "\n",
    "\n",
    "### Limitations\n",
    "\n",
    "- Distortion the relation of the most frequent label with other variables within the dataset\n",
    "- May lead to an over-representation of the most frequent label if there is a big number of NA\n",
    "\n",
    "\n",
    "### When to use mode / most frequent category imputation?\n",
    "\n",
    "- Data is missing completely at random\n",
    "- No more than 5% of the variable contains missing data\n",
    "\n",
    "## In this demo:\n",
    "\n",
    "We will use the House Prices dataset to demonstrate how to do mode imputation in categorical variables.\n",
    "\n",
    "- To download the dataset please refer to the lecture **Datasets** in **Section 1** of the course."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# to split the datasets\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### House Price dataset"
   ]
  },
  {
   "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>BsmtQual</th>\n",
       "      <th>FireplaceQu</th>\n",
       "      <th>SalePrice</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Gd</td>\n",
       "      <td>NaN</td>\n",
       "      <td>208500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Gd</td>\n",
       "      <td>TA</td>\n",
       "      <td>181500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Gd</td>\n",
       "      <td>TA</td>\n",
       "      <td>223500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>TA</td>\n",
       "      <td>Gd</td>\n",
       "      <td>140000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Gd</td>\n",
       "      <td>TA</td>\n",
       "      <td>250000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  BsmtQual FireplaceQu  SalePrice\n",
       "0       Gd         NaN     208500\n",
       "1       Gd          TA     181500\n",
       "2       Gd          TA     223500\n",
       "3       TA          Gd     140000\n",
       "4       Gd          TA     250000"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# let's load the dataset with a few columns for the demonstration\n",
    "\n",
    "# these are categorical columns and the target SalePrice\n",
    "cols_to_use = ['BsmtQual', 'FireplaceQu', 'SalePrice']\n",
    "\n",
    "data = pd.read_csv('../houseprice.csv', usecols=cols_to_use)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "BsmtQual       0.025342\n",
       "FireplaceQu    0.472603\n",
       "SalePrice      0.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# let's inspect the percentage of missing values in each variable\n",
    "\n",
    "data.isnull().mean()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The variable BsmtQual contains only a few observations with NA (2.5), therefore, replacing those observations with the value of the most frequent category is a good approach.\n",
    "\n",
    "On the other hand, the variable FireplaceQu, contains missing data in almost half of the observations. Therefore, replacing those with the most frequent category is most likely going to distort the relationship of the most frequent category with the sale price. \n",
    "\n",
    "In addition, if so many observations in FireplaceQu are missing, it could be for a reason and not just at random, so we could be better off capturing that somehow with a different imputation technique. We will see how in the next notebook. For now, let's explore the effect of replacing NA by the most frequent categories when missing data is few or a lot."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Imputation important\n",
    "\n",
    "Imputation should be done over the training set, and then propagated to the test set. This means that the most frequent category has to be selected from the training set, and used to replace NA both in train and test sets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1022, 3), (438, 3))"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# let's separate into training and testing set\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(\n",
    "    data,\n",
    "    data['SalePrice'], # target\n",
    "    test_size=0.3, # percentage of observations in the test set\n",
    "    random_state=0) # seed for reproducibility\n",
    "\n",
    "X_train.shape, X_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### BsmtQual"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.023483365949119372"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# let's remind ourselves of the % of missing values\n",
    "X_train['BsmtQual'].isnull().mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Number of houses')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# let's inspect the number of observations per category in BsmtQual\n",
    "\n",
    "# we used this line of code in section 3 on variable characteristics\n",
    "\n",
    "# value_counts() counts the amount of houses that show each of the labels\n",
    "# in the variable indicated below\n",
    "\n",
    "X_train['BsmtQual'].value_counts().sort_values(ascending=False).plot.bar()\n",
    "plt.xlabel('BsmtQual')\n",
    "plt.ylabel('Number of houses')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    TA\n",
       "dtype: object"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let's find the most frequent category for BsmtQual\n",
    "\n",
    "X_train['BsmtQual'].mode()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'BsmtQual')"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# let's plot the distribution of the target for the houses that:\n",
    "\n",
    "# * show the most frequent category or\n",
    "# * those that show missing data\n",
    "\n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "# select and plot houses with the most frequent label\n",
    "X_train[X_train['BsmtQual']=='TA']['SalePrice'].plot(kind='kde', ax=ax)\n",
    "\n",
    "# select and plot houses with missing data in the variable\n",
    "X_train[X_train['BsmtQual'].isnull()]['SalePrice'].plot(kind='kde', ax=ax, color='red')\n",
    "\n",
    "# add the legend\n",
    "lines, labels = ax.get_legend_handles_labels()\n",
    "labels = ['Houses with TA', 'Houses with NA']\n",
    "ax.legend(lines, labels, loc='best')\n",
    "\n",
    "# add figure title\n",
    "plt.title('BsmtQual')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The plot above indicates that the houses with missing data cost overall less than those with the label TA. Therefore, replacing missing values by TA, could affect the overall distribution, if there were a lot of NA."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# I keep the original distribution of SalePrice for the most\n",
    "# frequent category for later\n",
    "\n",
    "# the use of tmp is common to name temporal variables\n",
    "\n",
    "tmp = X_train[X_train['BsmtQual']=='TA']['SalePrice']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Sole\\Anaconda3\\envs\\feml\\lib\\site-packages\\pandas\\core\\generic.py:6130: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  self._update_inplace(new_data)\n"
     ]
    }
   ],
   "source": [
    "# Let's fill na in both train and test\n",
    "\n",
    "# I use the fillna() method from pandas\n",
    "\n",
    "# with the argument inplace=True, I indicate to pandas\n",
    "# that I want the values replaced in the same dataset\n",
    "\n",
    "X_train['BsmtQual'].fillna('TA', inplace=True)\n",
    "X_test['BsmtQual'].fillna('TA', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# check that NA have been filled\n",
    "\n",
    "X_train['BsmtQual'].isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Number of houses')"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# evaluate effect on distribution of categories\n",
    "\n",
    "X_train['BsmtQual'].value_counts().sort_values(ascending=False).plot.bar()\n",
    "plt.xlabel('BsmtQual')\n",
    "plt.ylabel('Number of houses')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the above plot, we see that now there are a few more houses with the label TA, which are those that before contained NA, but the overall effect on the distribution is not dramatic. The plot looks almost identical to that one before the imputation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'BsmtQual')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# did the distribution of SalePrice for the most frequent category change?\n",
    "# let's have a look\n",
    "\n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "# original distribution of salePrice for houses with most frequent label\n",
    "# remember I captured this a few cells up in the notebook\n",
    "tmp.plot(kind='kde', ax=ax)\n",
    "\n",
    "# distribution of the variable after imputation\n",
    "X_train[X_train['BsmtQual'] == 'TA']['SalePrice'].plot(\n",
    "    kind='kde', ax=ax, color='red')\n",
    "\n",
    "# add the legend\n",
    "lines, labels = ax.get_legend_handles_labels()\n",
    "labels = ['Original variable', 'Imputed variable']\n",
    "ax.legend(lines, labels, loc='best')\n",
    "\n",
    "# add title\n",
    "plt.title('BsmtQual')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The distribution of house prices between the original and imputed variable is quite similar, because the number of missing data was small. We expected this behaviour.\n",
    "\n",
    "Let's explore the effect of mode imputation when the percentage of missing data is high.\n",
    "\n",
    "#### FirePlaceQu"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.46771037181996084"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# let's remind ourselves of the % of missing values\n",
    "X_train['FireplaceQu'].isnull().mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Number of houses')"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# let's inspect the number of observations per category\n",
    "\n",
    "X_train['FireplaceQu'].value_counts().sort_values(ascending=False).plot.bar()\n",
    "plt.xlabel('FireplaceQu')\n",
    "plt.ylabel('Number of houses')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    Gd\n",
       "dtype: object"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Let's find the most frequent category for FirePlaceQu\n",
    "\n",
    "X_train['FireplaceQu'].mode()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'FirePlaceQu')"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaQAAAEICAYAAAAQkoCgAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzt3Xl8VNX5+PHPk4QkJGENAYEAAUQRNCxGFhFcqKCAii0q1AWty1crVWu11dpvq/3ZBb9WrXWvuBQXUFCk1YqiAi7sqCibCWELhCVAgIQl2/P7496BEGaSSTJr8rxfr3nNzLnnnnPuTJIn59xzzxVVxRhjjAm3mHA3wBhjjAELSMYYYyKEBSRjjDERwQKSMcaYiGAByRhjTESwgGSMMSYiWEAyxhgTESwgGRNgItJZRIpEJDYIZZ8nInmBLteYSGAByZh6EJGNInLIDUBFIlIElKlqiqqW17HM60Wk3C1vv4h8IyJjAtz0mtrQS0Rmi8g+ETkgIp+KyKBQtsE0PhaQjKm/S9wA5Hls85VRHP783i1U1RSgJTAFeEtEWgeqwdURke7Al8B3QFegAzAL+FhEBoSiDaZxsoBkTICJSIaIqIjEue/nicifRORL4CDQTURaiMgUEckXka0i8rC3IT5VrQBeApoC3bzUdZ+IrHd7MatF5PIq228WkTWVtvd30zuIyEwR2SUiG0Tkjkq7PYgTEB9Q1T2qekBVnwReAya7+58wdOj2Fn9U90/ONHYWkIwJjWuBW4BmwCbgVaAMOBnoB4wAbqq6kxvUbgKKgGwv5a4HhgItgIeA10SkvbvvFTjB5TqgOXApsNvtof0b+BboCAwH7hKRkW6ZFwJve6nrLWCoiCTW7tCN8Y8FJGPqb5aIFLqPWT7yvKKqq1S1DGgNXAzcparFqroTeBwYXyn/IBEpBLYDE4DLVXVf1UJV9W1V3aaqFao6HSdoeYbVbgIeUdWl6shR1U3AWUCaqv5RVUtUNRf4Z6X62wD5Xo4hH4h1229MwMWFuwHGNABjVXWu542IZHjJs6XS6y5AEyBfRDxpMVXyLFLVc2qqWESuA+4GPHWm4AQUgE44PaiqugAd3IDnEQt87r4uANp72a89oO72U2pqmzG1ZQHJmNCofJ+XLcARoI3bY6oTEemC07MZjnPOp1xEvgE8UW4L0N3LrluADaraw0fRc4ErgJerpF+JEyhLRKQYSKrUllggra7HYgzYkJ0xIaeq+cBHwN9EpLmIxIhIdxE5t5ZFJeMEul0AInIDcHql7S8C94jIme7svpPdILYE2C8ivxGRpiISKyKni8hZ7n4PAWe7EzFai0gzEfkFcAPwBzfPD0CiiIwWkSbA74CE2n8axhxjAcmY8LgOiAdWA3uBGXgfJvNJVVcDfwMWAjuAM3Cma3u2vw38CXgDOIAzdbu1e33UJUBfYAPOENyLOBMjUNVs4BygD7ARKAT+H855rI/dPPuAn7v7bQWKAbtg19SL2B1jjTHVEZF0YBHwB1WdEu72mIbLekjGmGqpah7OrMD2IpIS7vaYhst6SMYYYyKC9ZCMMcZEBJv2XQtt2rTRjIyMcDfDGGOiyvLlywtUtcbLAiwg1UJGRgbLli0LdzOMMSaqiMgmf/LZkJ0xxpiIYAHJGGNMRLCAZIwxJiLYOSRjTEQoLS0lLy+Pw4cPh7sppo4SExNJT0+nSZMmddrfApIxJiLk5eXRrFkzMjIyqLQKuokSqsru3bvJy8uja9eudSrDhuyMMRHh8OHDpKamWjCKUiJCampqvXq4FpCMMRHDglF0q+/3ZwHJNAzZ2fD003DkSLhbYoypIwtIJvqVlMDw4TBpEvzhDzXnN8aHlJTj14595ZVXmDRpUphaUzvLli3jjjvuAGDevHl89dVXR7ddf/31zJgxo8YyduzYwU9/+lO6devGmWeeyeDBg3n33Xe95j3vvPMCvlCABSQT/ebOhS1boEkTmDIFKirC3SJjQi4rK4snn3wSODEg+UNVGTt2LMOGDSM3N5fly5czbdo08vJCd5srC0gm+n30ESQmwj/+AQUFsHp1uFtkGqBNmzYxfPhwMjMzGT58OJs3bwZO7H14eln5+fkMGzaMvn37cvrpp/P5558D8NFHHzF48GD69+/PFVdcQVFREQD33XcfvXr1IjMzk3vuueeE+s844wwKCwtRVVJTU/nXv/4FwLXXXsvcuXOZN28eY8aMYePGjTz33HM8/vjj9O3b92i9CxYs4Oyzz6Zbt25ee0uffvop8fHx3HrrrUfTunTpwi9+8QsADh06xPjx48nMzOSqq67i0KFD9f5Mq7Jp3yb6zZ8PQ4bAiBHO+wUL4PTTq9/HRLSH/r2K1dv2B7TMXh2a84dLeleb59ChQ/Tt2/fo+z179nDppZcCMGnSJK677jomTpzISy+9xB133MGsWbN8lvXGG28wcuRIHnjgAcrLyzl48CAFBQU8/PDDzJ07l+TkZCZPnsxjjz3GpEmTePfdd1m7di0iQmFh4QnlDRkyhC+//JIuXbrQrVs3Pv/8c6677joWLVrEs88+e3T4LCMjg1tvvZWUlJSjgW3KlCnk5+fzxRdfsHbtWi699FLGjRt3XPmrVq2if//+Po/n2WefJSkpiZUrV7Jy5cpq89aVBSQT3UpLnR7RnXdCRga0agUrV4a7VSZKNW3alG+++ebo+1deeeXoH/qFCxfyzjvvAE6v5Ne//nW1ZZ111ln87Gc/o7S0lLFjx9K3b1/mz5/P6tWrGTJkCAAlJSUMHjyY5s2bk5iYyE033cTo0aMZM2bMCeUNHTqUBQsW0KVLF2677TZeeOEFtm7dSuvWrU849+XN2LFjiYmJoVevXuzYsaPG/LfffjtffPEF8fHxLF26lAULFhw9R5WZmUlmZmaNZdSWBSQT3bKznUkNmZkgAr1725BdA1BTTyYSeKY4x8XFUeGet1RVSkpKABg2bBgLFizg/fff59prr+Xee++lVatWXHjhhbz55psnlLdkyRI++eQTpk2bxlNPPcWnn3563PZhw4bx9NNPs3nzZv70pz/x7rvvMmPGDIYOHepXexMSEo6+9nZj1t69ezNz5syj759++mkKCgrIyso64ZiDxc4hmej23XfO8xlnOM+9esGqVWB3QjYBdvbZZzNt2jQAXn/9dc455xzAGSJbvnw5AO+99x6lpaWAc86pbdu23Hzzzdx4442sWLGCQYMG8eWXX5KTkwPAwYMH+eGHHygqKmLfvn2MGjWKJ5544rhemkenTp0oKCggOzubbt26cc455/Doo496DUjNmjXjwIEDtTq+Cy64gMOHD/Pss88eTTt48ODR18OGDeP1118H4Pvvv2dlEEYiLCCZ6LZqFcTGQs+ezvtTT4U9e8DLGLwx9fHkk0/y8ssvk5mZydSpU/n73/8OwM0338z8+fMZMGAAixcvJjk5GXBmuvXt25d+/foxc+ZM7rzzTtLS0njllVeYMGECmZmZDBo0iLVr13LgwAHGjBlDZmYm5557Lo8//rjXNgwcOJBTTjkFcIbwtm7dejQwVnbJJZfw7rvvHjepoSYiwqxZs5g/fz5du3ZlwIABTJw4kcmTJwNw2223UVRURGZmJo888ggDBgyo9WdYYxu8dd2Md1lZWWo36Isw11wDX34JGzY47995B37yE1ixAvr1C2/bTK2sWbOG0047LdzNMPXk7XsUkeWqmuVjl6Osh2SiW24uVF7I0XOL+Y0bw9EaY0w9WEAy0W3DBgtIxjQQFpBM9Dp0CLZvh27djqW1agUpKRaQjIlCFpBM9PIEnco9JBGnl2QByZioYwHJRC/PRIaqNwNLT4etW0PfHmNMvVhAMtHL0wvynDfyaN/eGcozxkSVoAYkEblIRNaJSI6I3Odle4KITHe3LxaRjErb7nfT14nIyJrKFJGubhnZbpnxbvowEVkhImUiMq5K/RPd/NkiMjEYn4EJovx85xqktm2PT2/fHnbssFW/Ta019ttPiAi/+tWvjr5/9NFHefDBB4/L06dPHyZMmBCYRlcRtIAkIrHA08DFQC9ggoj0qpLtRmCvqp4MPA5MdvftBYwHegMXAc+ISGwNZU4GHlfVHsBet2yAzcD1wBtV2tca+AMwEBgA/EFEWgXm6E1I5Oc7wSg29vj0k06CsjJn5W9jGon63n4CnOWF3nnnHQp8/O6sWbOGiooKFixYQHFxcb3a600we0gDgBxVzVXVEmAacFmVPJcBr7qvZwDDxVks6TJgmqoeUdUNQI5bntcy3X0ucMvALXMsgKpuVNWVQNV/l0cCH6vqHlXdC3yME/xMtNi+3Qk+VbVvf2y7MQHS0G8/Ac66fLfccovPlSLeeOMNrr32WkaMGMHs2bPr8jFWK5iLq3YEtlR6n4fTG/GaR1XLRGQfkOqmL6qyb0f3tbcyU4FCVS3zkr827atpHxNJ8vOPBZ/KPGn5+c6iqyb63HUXeFnPrV769oUnnqg2S2O+/YTH7bffTmZmptfVzKdPn87HH3/MunXreOqppwI+dBfMgORtWdiq6xT5yuMr3VuPrrr81fFrHxG5BbgFoHPnzjUUaUJq+3bnj0xVnl5Tfn5o22Oint1+Apo3b851113Hk08+SdOmTY+mL126lLS0NLp06UJ6ejo/+9nP2Lt3L61aBe5MRzADUh7QqdL7dGCbjzx5IhIHtAD21LCvt/QCoKWIxLm9JG91eWvfeVXKmlc1k6q+ALwAzlp2NZRpQqW83Jm4UF0PyYbsolcNPZlI0NBuP1HZXXfdRf/+/bnhhhuOpr355pusXbuWDHdW6/79+5k5cyY33XSTX/X7I5jnkJYCPdzZb/E4kxSqDjrOBjyz28YBn6rzSc0Gxruz8LoCPYAlvsp09/nMLQO3zPdqaN8cYISItHInM4xw00w02L3bCUreziElJUHz5tZDMgHV0G8/UVnr1q258sormTJlCgAVFRW8/fbbrFy5ko0bN7Jx40bee+89r4G1PoIWkNyeyiScP/JrgLdUdZWI/FFELnWzTQFSRSQHuBu4z913FfAWsBr4ELhdVct9lemW9RvgbresVLdsROQsEckDrgCeF5FVbh17gP+HE+SWAn9000w08AQbbz0kcAKV9ZBMADX0209U9atf/erobLsFCxbQsWNHOnY8dpp92LBhrF69mvwA/uNnt5+oBbv9RASZMwcuugi++ALc8fjjnHMOJCTAJ5+Evm2mTuz2Ew2D3X7CND419ZDatLHrkIyJMhaQTHTyDMd5O4cEkJYGu3aFrj3GmHqzgGSiU36+M3EhKcn7dk8PyYako4qdQohu9f3+LCCZ6ORrlQaPNm2gtBT27w9dm0y9JCYmsnv3bgtKUUpV2b17N4mJiXUuI5jXIRkTPL5WafBIS3OeCwqgRYvQtMnUS3p6Onl5eeyyodaolZiYSHp6ep33t4BkotP27dC/v+/tbdo4z7t2QffuoWmTqZcmTZrQteq9rUyjYkN2JjrVpodkjIkKFpBM9Ckqch41nUMCm2lnTBSxgGSij2fKt/WQjGlQLCCZ6FPTNUgAycnOSg3WQzImalhAMtHHs0pDdQFJxFZrMCbKWEAy0cefITtwhu0sIBkTNSwgmeiTnw9xcZCaWn2+Nm1syM6YKGIByUSf7duhXTuIqeHH13pIxkQVC0gm+uTnV3/+yMN6SMZEFQtIJvps317z+SNwAtK+fc6adsaYiGcByUQff3tInnNMe+xGwMZEAwtIJrqUlzvDcP72kAB27w5um4wxAWEByUSXnTuhoqJ2PSSb2GBMVLCAZKKLv9cgwbEekgUkY6KCBSQTXfxZpcHDhuyMiSoWkEx0qU0PyYbsjIkqFpBMdKlND6lpU0hKsoBkTJSwgGSiy/bt0LIlJCb6l79NGxuyMyZKWEAy0cXfa5A8UlOth2RMlLCAZKKLv6s0eNgtKIyJGhaQTHSpSw/JhuyMiQpBDUgicpGIrBORHBG5z8v2BBGZ7m5fLCIZlbbd76avE5GRNZUpIl3dMrLdMuOrq0NEmojIqyLynYisEZH7g/dJmIBQdXpItQlI1kMyJmoELSCJSCzwNHAx0AuYICK9qmS7EdirqicDjwOT3X17AeOB3sBFwDMiEltDmZOBx1W1B7DXLdtnHcAVQIKqngGcCfxP5YBoItCBA3DwYO2H7AoLoawseO0yxgREMHtIA4AcVc1V1RJgGnBZlTyXAa+6r2cAw0VE3PRpqnpEVTcAOW55Xst097nALQO3zLE11KFAsojEAU2BEmB/4A7fBJznGqTaDtmBLbBqTBQIZkDqCGyp9D7PTfOaR1XLgH1AajX7+kpPBQrdMqrW5auOGUAxkA9sBh5V1RP+aonILSKyTESW7bJ764SX5xqk2vaQwIbtjIkCwQxI4iVN/cwTqPTq6hgAlAMdgK7Ar0Sk2wkZVV9Q1SxVzUpLS/NSlAmZugQkTw/JJjYYE/GCGZDygE6V3qcD23zlcYfOWgB7qtnXV3oB0NIto2pdvur4KfChqpaq6k7gSyCrjsdqQsF6SMY0aMEMSEuBHu7st3icSQqzq+SZDUx0X48DPlVVddPHuzPkugI9gCW+ynT3+cwtA7fM92qoYzNwgTiSgUHA2gAevwm0bdsgIQFatfJ/HwtIxkSNuJqz1I2qlonIJGAOEAu8pKqrROSPwDJVnQ1MAaaKSA5Or2W8u+8qEXkLWA2UAberajmAtzLdKn8DTBORh4Gv3bLxVQfObL2Xge9xhvVeVtWVQfo4TCDk5zu9I/E2CuuDDdkZEzXE6SwYf2RlZemyZcvC3YzGa/hwOHQIvvqqdvslJcHPfw6PPhqcdhljqiUiy1W1xlMitlKDiR75+dChQ+33swVWjYkKFpBM9Ni2rXYTGjxsgVVjooIFJBMdDh2CffvqFpBs+SBjooIFJBMd6jLl28MWWDUmKlhAMtHBE5Dqeg7JekjGRDwLSCY61KeH1KYN7N1rC6waE+EsIJnosM1deKOuQ3bgBCVjTMSygGSiQ34+NGlyLLjUhq3WYExUsIBkooPnTrExdfiR9QQkm9hgTESzgGSig2fZoLrw9Kqsh2RMRLOAZKJDXVdpABuyMyZKWEAy0WHr1roHJFtg1ZioYAHJRL6DB51bkHfqVHNeb5KSIDHRekjGRDgLSCbybXHvWl/XgCRiF8caEwUsIJnI5wlI6el1L8OWDzIm4llAMpGvvj0ksB6SMVHAApKJfJ6A1LFj3cuweyIZE/EsIJnIt2ULtGsHCQl1L8PuiWRMxLOAZCLfli31G66DYwuslpcHpk3GmICzgGQiXyACUmoqqNoCq8ZEMAtIJvIFqocENmxnTASzgGQi2759cOCABSRjGgELSCayBWLKN9jyQcZEAb8CkojMFJHRImIBzIRWoAKSp4e0a1f9yjHGBI2/AeZZ4KdAtoj8VUR6BrFNxhyzYYPz3KVL/cpp18553r69fuUYY4LGr4CkqnNV9WqgP7AR+FhEvhKRG0SkSTAbaBq57GxncdS6rvTtkZgIrVo5t7EwxkQkv4fgRCQVuB64Cfga+DtOgPo4KC0zBpyAdPLJzgKp9dW+vQUkYyKYv+eQ3gE+B5KAS1T1UlWdrqq/AFKq2e8iEVknIjkicp+X7QkiMt3dvlhEMiptu99NXyciI2sqU0S6umVku2XG+1FHpogsFJFVIvKdiCT683mYEMrOhh49AlOWBSRjIpq/PaQXVbWXqv5FVfPB+UMPoKpZ3nYQkVjgaeBioBcwQUR6Vcl2I7BXVU8GHgcmu/v2AsYDvYGLgGdEJLaGMicDj6tqD2CvW3Z1dcQBrwG3qmpv4Dyg1M/Pw4RCWZlzDskCkjGNgr8B6WEvaQtr2GcAkKOquapaAkwDLquS5zLgVff1DGC4iIibPk1Vj6jqBiDHLc9rme4+F7hl4JY5toY6RgArVfVbAFXdraq2rkwk2bwZSkudIbtA8AQk1cCUZ4wJqLjqNorISUBHoKmI9AM8A/nNcYbvqtMR2FLpfR4w0FceVS0TkX1Aqpu+qMq+nqWevZWZChSqapmX/L7qOAVQEZkDpOEEwEe8fAa3ALcAdO7cuYZDNgGVne08B7KHVFLiLB/UunVgyjTGBEy1AQkYiTORIR14rFL6AeC3Nezr7Sx01X9NfeXxle6tR1dd/urqiAPOAc4CDgKfiMhyVf3kuIyqLwAvAGRlZdm/1qEUjIAETi/JApIxEafagKSqrwKvishPVHVmLcvOAypfzZgObPORJ889p9MC2FPDvt7SC4CWIhLn9pIq56+ujvmqWgAgIh/gzBo8LiCZMMrJgZQUOOmkwJRXOSD17h2YMo0xAVPtOSQRucZ9mSEid1d91FD2UqCHO/stHmeSwuwqeWYDE93X44BPVVXd9PHuDLmuQA9gia8y3X0+c8vALfO9GuqYA2SKSJIbqM4FVtdwTCaUAjnlG44PSMaYiFPTkF2y++xzarcv7vmaSTh/+GOBl1R1lYj8EVimqrOBKcBUEcnB6bWMd/ddJSJv4QSIMuB2z4QDb2W6Vf4GmCYiD+NcJzXFTfdVx14ReQwnyCnwgaq+X9vjNEGUnQ19+gSuPE9A2la1o26MiQSiNuPIb1lZWbps2bJwN6NxKCuDpk3h3nvhz38OXLktW8I118BTTwWuTGNMtdzz814vEarM3wtjHxGR5iLSREQ+EZGCSsN5xgTepk1OUArUhAaPjAynbGNMxPH3OqQRqrofGIMzGeAU4N6gtcqYQM+w8+jSBTZuDGyZxpiA8DcgeRZQHQW8qap7gtQeYxyegBSoi2I9unRxekg2VG1MxPE3IP1bRNYCWTjX66QBh4PXLNPoZWc7U749t40IlIwM5w60hYWBLdcYU2/+3n7iPmAwkKWqpUAxJy4DZEzg5OQ4w3WBmvLt4bmvkg3bGRNxapr2XdlpONcjVd7nXwFujzGO7Gzo3z/w5XoC0qZN0K9f4Ms3xtSZXwFJRKYC3YFvAM8CpIoFJBMMpaXOKt9XXhn4sjMynGebaWdMxPG3h5QF9FK7aMmEwsaNUF4e+Bl2AKmpkJxsQ3bGRCB/JzV8DwRoQTFjapCT4zwHIyCJOOWuWxf4so0x9eJvD6kNsFpElgBHPImqemlQWmUat2Bdg+TRsycsXhycso0xdeZvQHowmI0w5jjZ2dC8OaSlBaxIVWVX0RGS4+NI7tkTpk+HQ4ec5YmMMRHBr4CkqvNFpAvQQ1XnikgSzuKmxgReAFf5VlVmrtjK4x//wNbCQ8TGCPfubcKtqk49mZkBaLAxJhD8XcvuZpzbfz/vJnUEZgWrUaaR81yDVE+qyp8/WMM9b39Lu+YJ/OGSXtwyrBtztRUA381dWO86jDGB4++Q3e3AAGAxgKpmi0jboLXKNF6lpc4MuAkT6l3Ua4s28c/PNzBxcBf+cElvYmKcHtfOfu2oeOF25r+7gKLLxjG4e2q96zLG1J+/s+yOqGqJ5417caxNATeBt2FDQKZ8bygo5v/9Zw3nn5p2XDACaNuuFXrqqZy5ZwO3TF3Glj0H69tqY0wA+BuQ5ovIb4GmInIh8Dbw7+A1yzRaAVhUVVV56N+riI+LYfJPMo8LRh6xWVmctWcjKNw57WvKyivqXJ8xJjD8DUj3AbuA74D/AT4AfhesRplGLADXIC1cv5t563Zx5/AetG2e6D3TmWcStz2fR4e1Y8XmQp7+bH2d6zPGBIa/s+wqRGQWMEtVdwW5TaYxy86GFi2gTZs6F/HMvPW0bZbAdWd38Z3JXSdv5OGtXNqnA09/lsOYPu3pnpZS53qNMfVTbQ9JHA+KSAGwFlgnIrtE5PehaZ5pdLKz67XK93d5+/gip4Abz+lKQlw1Vyb07w+xsfDVV/xuzGkkNInhf2d9j62OZUz41DRkdxcwBDhLVVNVtTUwEBgiIr8MeutM4+MJSHU0ddFGkuNj+enAztVnTEmBs86Czz6jbbNEfnNRT75av5tZ32ytc93GmPqpKSBdB0xQ1Q2eBFXNBa5xtxkTOCUlzircdZzQUHSkjP+szGdMZgeaJTapeYfzzoOlS6GoiJ8O6EzfTi350/tr2HeotE71G2Pqp6aA1ERVC6omuueR/PiNN6YWNm6Eioo6B6T3V27jYEk5V57Vyb8dzj8fysrgyy+JiREeHns6e4pLeOwjW3jVmHCoKSCV1HGbMbWXm+s8d+9ep93fWpZH97Rk+ndu6d8OQ4ZAkybw8ccAnN6xBdcO6sLURZv4fuu+OrXBGFN3NQWkPiKy38vjAHBGKBpoGhFPQOrWrda75u09yPJNe/nJmemIvxMikpOdXtLs2eBOZrh7xKm0To7nd7O+p6LCJjgYE0rVBiRVjVXV5l4ezVTVhuxMYOXmQmIinFT7W299+P12AEaf0b52O44d60ykWLsWgBZNm3D/xafxzZZC3lq2pdbtMMbUnb8XxhoTfOvXO72jOkz5/uC7fHq1b06X1OTa7Xipe0uvWcfWCv5x/44MyGjN5A/XsrfYRqaNCRULSCZy5ObW6fzR9n2HWbG5kFFn1OGmxh07QlYWvPPO0SQR4Y9je7P/cBmPzLEJDsaESlADkohcJCLrRCRHRO7zsj1BRKa72xeLSEalbfe76etEZGRNZYpIV7eMbLfM+JrqcLd3FpEiEbkn8J+A8ZuqE5DqcP5ozipnuO6i02s5XOdx1VWwbNmxdfSAnic152dDMnhzyWYW/GCLkxgTCkELSCISCzwNXAz0AiaISK8q2W4E9qrqycDjwGR3317AeKA3cBHwjIjE1lDmZOBxVe0B7HXL9llHJY8D/w3MUZs6KyiAoqI6BaRP1+6kW5tkTm5bx2V/xo93hgnffPO45F+NOJUebVP41dvfsseG7owJumD2kAYAOaqa6966YhpwWZU8lwGvuq9nAMPFmSJ1GTBNVY+4F+XmuOV5LdPd5wK3DNwyx9ZQByIyFsgFVgXwuE1d1HGG3aGSchbm7ua8U+txe670dDj3XHj99aOz7QASm8Ty9/H92HewlHve/pZym3VnTFAFMyB1BCpPU8pz07zmUdUyYB+QWs2+vtJTgUK3jKp1ea1DRJKB3wAPVXcQInKLiCwTkWW7dtnQTdCsd1fbruU5pEW5uykpq+C8U9PqV//VV8OehJoHAAAezUlEQVQPP8CKFccl9+rQnP8dcxqfrt3JXz5YU786jDHVCmZA8jZVquq/mL7yBCq9ujoewhniK/Ky/VhG1RdUNUtVs9LS6vlHz/jm6SFlZNRqt8/W7aRpk1gGdG1dv/p/8hOIj4c33jhh07WDM7j+7Axe/GIDT3+WU796jDE+BTMg5QGV13BJB7b5yuPehbYFsKeafX2lFwAt3TKq1uWrjoHAIyKyEWcR2d+KyKS6Haqpt9xc6NABmjb1exdV5bN1Ozm7eyqJTapZ2dsfrVrBqFHOeaTy8hM2/270aYzt24H/m7OOh/+z2m7oZ0wQBDMgLQV6uLPf4nEmKcyukmc2MNF9PQ74VJ31/2cD490Zcl2BHsASX2W6+3zmloFb5nvV1aGqQ1U1Q1UzgCeAP6vqU4H8AEwt1GGGXW5BMVv2HOK8nvU4f1TZ1VdDfj588skJm+JiY3jsyr5cN7gLL36xgateWET2jgOBqdcYAwQxILnnayYBc4A1wFuqukpE/igi7tWITME5n5MD3I1zZ1pUdRXwFrAa+BC4XVXLfZXplvUb4G63rFS3bJ91mAjjuSi2Fr7Mcdb9PbdHgIZSL7nE6Sm98orXzTExwh8vO50nJ/Tjhx0HGPnEAu6c9jWLc3fbfZSMCQCxXyT/ZWVl6bJly8LdjIbn8GFISoIHH4Tf+3/vx9teW87KvH188Zvz/V+/riaTJsGUKU5PqaXvRVr3Fpfw9Gc5TFu6haIjZaS3asrozPaMOaMDp3dsHrj2GNMAiMhyVc2qKZ+t1GDCb9MmZ7p1LXpIFRXK4g17GNQtNbB//G+4wQmQ06ZVm61Vcjy/G9OLJQ8M57Er+3By2xSmfL6BS576gvMenccjH65l9bb9gWuXMY1AXM1ZjAmyOlyD9MPOA+wpLmFQt3rOrquqf3844wx4+WW49dYasyfFx/Hj/un8uH86hQdL+GjVDv69chvPL8jlmXnrGdm7HQ+M6kXn1KTAttOYBsh6SCb8PNcg1SIgLVq/G4BB3VID2xYRp5e0ZAmsXl2rXVsmxXPlWZ2YeuNAlj7wI+4ZcQqfZxcw6snP+eC7/MC205gGyAKSCb/cXOccUrt2fu+yMHc36a2a0ql1EHoe11wDcXFOL6mOWifHM+mCHnx897n0aJfCz19fwfPz1wewkcY0PBaQTPh5pnz7eS7Ic/5ocKB7Rx5paTBmDEyd6tzivB46tmzK9FsGMyazPX/571oLSsZUwwKSCb9aXoO0dvsBCg+WBn64rrLrr4cdO+DDD+tdVHxcDE9c1fdoUPqvDd8Z45UFJBNedbjtxKJc9/xR9yAGpFGjoG3beg3bVRYXG8OjV/ShX+eW3P3Wt6zbbhfVGlOVBSQTXjt3QnFxrRZVXZi7m86tk+jY0v9lhmqtSRNn5YZ//xt27w5IkYlNYnn+2jNJTojlzmlfc6TsxCWKjGnMLCCZ8KrllO+KCmVJMM8fVTZxIpSW1nhNUm20bZbII+MyWbv9AH/76IeAlWtMQ2AByYRXLQPS6vz97DtUyqDuAb7+yJs+fZzHq6/WnLcWLujZjgkDOvPi57l8v3VfQMs2JppZQDLhtX69M7vOz9tOHD1/FIoeEji9pKVLa31NUk3uu6gnrZPj+d2s76mwG/8ZA1hAMuGWmwsdO0Jiol/ZF+XuJiM1ifYtgnj+qLKf/hRiYwPeS2qR1ITfjjqNb7YUMm3plpp3MKYRsIBkwqsWM+zKPdcfBXN2XVXt2sHFF8Nrr3m9T1J9XN6vIwMyWvO3j9ZRdKR+1zsZ0xBYQDLhtX49dO3qV9bV2/Zz4HBZ6IbrPK67DrZtg3nzAlqsiPDb0aexu7iEF+yCWWMsIJkwOnTI+UN/8sl+ZQ/5+SOP0aOdpY1mzgx40X07tWRMZnv++fkGduw/HPDyjYkmFpBM+Hhm2Pl5DdLC3N10a5NMu+b+nW8KmKQkJyi9807Ah+0A7h15KmUVFTwx16aBm8bNApIJH88q334EpLLyCpZu2BPc1RmqM26cs5TQF18EvOguqclcM6gL05duIWenreBgGi8LSCZ8cnKcZz8C0qpt+zlwJAznjzxGjXJmAs6YEZTiJ51/MknxcTw6x3pJpvGygGTCZ/165zbhrWu+yPXY+aMQXBDrTUqKM9tu5kyoqAh48akpCdw8tBsfrtrO15v3Brx8Y6KBBSQTPuvXO70jP247sTB3N93TkmnbLMTnjyobOxby82HFiqAUf+PQrqQmxzP5w7Wo2sWypvGxgGTCJyfHr+G6Uvf8UUivP/Lm4oud4Pn++0EpPiUhjl9ccDKLcvfweXZBUOowJpJZQDLhUVYGmzb5NeX7u637KC4pZ3C3NiFoWDXS0mDgwKAFJIAJAzuT3qopkz9ca0sKmUbHApIJj82bnaDkRw9p4fownz+qbPRoZ227HTuCUnxCXCx3X3gKq7bt5327kZ9pZCwgmfCoxZTvhet30/OkZqSmJAS5UX4YPdp5/u9/g1bFZX070vOkZvzto3WUlgd+AoUxkcoCkgkPP6d8HykrZ9mmPeGb7l1V377QoUNQh+1iY4R7R57Kxt0HmW4Lr5pGxAKSCY+cHOe6ng4dqs32zeZCDpdWcHa4JzR4iDi9pI8+cm7eFyQX9GzLWRmt+Psn2RwqsTvLmsbBApIJjzVr4NRTIab6H8GFubsRgYFdIyQgAYwcCfv3O+eSgkRE+PVFPdl14Agvf7UhaPUYE0mCGpBE5CIRWSciOSJyn5ftCSIy3d2+WEQyKm27301fJyIjaypTRLq6ZWS7ZcZXV4eIXCgiy0XkO/f5guB9EuYEa9bAaafVmG3h+t307tCcFklNQtAoP11wgdNT+vjjoFZzVkZrhvdsy7Pz1lN4sCSodRkTCYIWkEQkFngauBjoBUwQkV5Vst0I7FXVk4HHgcnuvr2A8UBv4CLgGRGJraHMycDjqtoD2OuW7bMOoAC4RFXPACYCUwN5/KYaxcXOlO9eVX8cjne4tJyvNxdydvcwT/euqlUryMoKekACuPeiUyk6UsbfP8kOel3GhFswe0gDgBxVzVXVEmAacFmVPJcBnltxzgCGi4i46dNU9YiqbgBy3PK8lunuc4FbBm6ZY6urQ1W/VtVtbvoqIFFEImAaVyOwbh2o1thDWr5pLyXlFQyOlAkNlV14ISxa5AzdBVHPk5pz9cDOvPrVRr7fui+odRkTbsEMSB2BylOE8tw0r3lUtQzYB6RWs6+v9FSg0C2jal2+6qjsJ8DXqnqk6kGIyC0iskxElu3atauGQzZ+WbPGea4hIH21voDYGOGsrhFw/VFVF17o3IoiwDft8+bekT1pnRzPA7O+p9wuljUNWDADkrcFyqr+NvnKE6j0GtshIr1xhvH+x0s+VPUFVc1S1ay0tDRvWUxtrVkDsbHQo0e12Rb8UED/zi1JSYgLUcNqYfBg5z5JIRi2a9G0Cb8b3YtvtxTyxuJNQa/PmHAJZkDKAzpVep8ObPOVR0TigBbAnmr29ZVeALR0y6hal686EJF04F3gOlW1e0iHyurVzpJB8fE+sxQUHeG7rfs495QI/ScgIQHOPRfmzg1JdZf17cDQHm348wdr2VBQHJI6jQm1YAakpUAPd/ZbPM4khdlV8szGmVAAMA74VJ1ljmcD490Zcl2BHsASX2W6+3zmloFb5nvV1SEiLYH3gftV9cuAHrmp3qpVNU5o+DzbGR4995S2oWhR3fzoR7B2LeTlBb0qEeGRcZk0iRV+Of0bymwFB9MABS0guedrJgFzgDXAW6q6SkT+KCKXutmmAKkikgPcDdzn7rsKeAtYDXwI3K6q5b7KdMv6DXC3W1aqW7bPOtxyTgb+V0S+cR8R/NevgThwALKzoV+/arMt+KGA1OR4endoHqKG1cGFFzrPIRi2A2jfoil/uvwMvtlSyKMf2Y38TMMjdt8V/2VlZemyZcvC3YzotmCBM9T1/vvOXVi9qKhQzvrTXIb2aMMT46sPXGGl6qw0cf758MYbIav2t+9+xxuLN/OPCf24pE/1K10YEwlEZLmqZtWUz1ZqMKHlubld//4+s6zO38/u4hKGRer5Iw8RZ9hu7tyg3EXWlwcv6U1Wl1bcO+NbvtlSGLJ6jQk2C0gmtJYvd3oVJ53kM8v8H5zzR0N7RHhAAhgxAnbtgm+/DVmV8XExPHvNmbRtlsjEl5awdntwr4UyJlQsIJnQWrGi2t4RwCdrdnBGxxakNYuC65R/9CPn+aOPQlptWrMEXr9pIE2bxHLNi4tZtc0umjXRzwKSCZ09e5xrkAYM8Jll5/7DrNhcyIhe7ULYsHpo3x4yM0MekAA6tU7itZsGEh8bw5XPLWTBD3bhtoluFpBM6HzxhTMR4NxzfWb5aLVzJ9YRvX0P6UWcESOcYysO/fVBJ7dN4Z2fD6FT6ySuf3kJT8z9wVZzMFHLApIJnfnznQtKq+khfbR6BxmpSZzSLiWEDaunESOgpMSZQRgGJ7VIZMZtZzO2b0eemJvNhBcWkbOzKCxtMaY+LCCZ0Jk/HwYNcm7M58X+w6UsXF/AyN4n4ayXGyWGDnWOKQzDdh4pCXE8dlVf/nZFH9Zu38/Ff1/AX/67hqIjZTXvbEyEsIBkQmPPHvj662qH6z5bu5PScmVE7yg5f+SRmOgc15w54W4JPzkznU/vOY+xfTvy/Pxchj3yGS9+nsvhUrvrrIl8FpBMaHzwgXOtzujRPrP8+9ttnNQ8kX6dWoWwYQEyYoQzYWPLlprzBlmblAT+74o+vHf7EHp3aM7D76/h3P/7jKmLNlFSZksOmchlAcmExuzZzoy0LO8Xa+8pLmHeul1c1rcDMTFRNFznMWKE8xyiZYT80adTS6beOJBptwyiU6sk/nfW9/zosfl88F0+tkKLiUQWkEzwFRfDf/8Ll1wCMd5/5N5fuY2yCmVsv6q3zIoSvXs7F/x+8EG4W3KCQd1SefvWwbx8w1kkxcfy89dXMO65hazYvDfcTTPmOBaQTPDNmAFFRXDttT6zvPv1Vnqe1IzT2kfwYqrVEYFLL4UPP4RDh8LdmhOICOef2pb37xjKX398Bpv3HOTHz3zF79/7nmKb+GAihAUkE3wvv+zcjG/IEK+bs3ccYMXmQi6P1t6Rx49/7PQGI2jYrqrYGGH8gM7Mu+c8fjakK1MXbWLE4wv4Kqcg3E0zxgKSCbKVK53p3jfc4PQivHht0SbiY2MYd2Z6iBsXYOedBy1bwsyZ4W5JjZIT4vj9Jb14+38GkxAXw9VTFvO3j9bZRbUmrCwgmeD6618hJQVuvdXr5uIjZcxcsZXRme1JTYmCteuq06SJM2w3e7ZzoWwUyMpozft3DOWKM9P5x6c5XP3iInbuPxzuZplGygKSCZ5Vq2D6dLjtNmjlfSr3O19vpehIGdcO7hLixgXJlVdCYWFETm7wpWl8LI+M68OjV/Thmy2FjHryc760ITwTBhaQTHCowi9/Cc2bw69/7TVLaXkFLyxYT5/0FvTr1DLEDQySkSOd6e0vvRTultTauDPTmT3pHFomxXPtlMX845NsKmwIz4SQBSQTHO+845zcf+ghaNPGa5b3vtnGlj2H+MUFPaJrqaDqxMXBxIlODyk/P9ytqbVT2jXjvduHcGmfDvzt4x+44ZWl7CmOjuFHE/0sIJnA277dOWfUv78zXOdFWXkFz3yWQ6/2zRl+WtsQNzDIbrgBysvh2WfD3ZI6SU6I4/Gr+vKny09n4frdjH7yc5ZvsmuWTPBZQDKBpQo33+xcdzR1qnOi34vXFm0it6CYu37UgHpHHqecAmPHwj/+AQcOhLs1dSIiXD2wC+/8/GziYoWrnl/IE3N/4EiZrYlngscCkgmsF1+E//wHJk+GXr28ZtlddITHPv6BoT3acGG03Iivtu6/35nc8Mwz4W5JvZzesQX/+cVQRp3RnifmZjP6yS9YnLs73M0yDZQFJBM4q1fDXXc5t/WeNMlntoffX8PBknL+cEmvhtc78hgwAEaNgj//OSrPJVXWomkTnpzQj5evP4tDJeVc9cIirn95Cd9uKQx300wDYwHJBEZxMVxxhXPN0b/+5XPNuve+2cq7X2/l9vNP5uS2zULcyBB74gk4fBjuuSfcLQmI83u25eO7h3HfxT35Zkshlz39JZc99QVvLtnMrgNHwt080wCIrfrrv6ysLF22bFm4mxGZrr/eCUQffeT0kLz4fus+rnx+Iae1b870WwYRF9sI/h968EFnpuHUqXDNNeFuTcAcOFzKjOV5vLF4M9k7ixCBzPSWDOrWmr7pLcns1JIOLRIbbg/Y1IqILFdV70v9V85nAcl/FpB8+Oc/4ZZb4Pe/d/74epGzs4if/nMRcTHCrNuH0La597vGNjhlZTB8OCxb5iyh5OP2G9FKVVm1bT+frd3Jp+t28v3WfZSWO39TkuNj6ZaWQre0ZLq1SaF722S6p6XQtU0yiU1iw9xyE0oWkILAApIXc+fCxRc7f3Tffx9iT/xDs2zjHm59bTkAb9w8iFPaNfChuqry852FZfftcz6vfv3C3aKgOVJWzpr8A3yXV8j6XcWs31VE7q5ithYeWwE9RqBT6yROTkuhe9uUY89tU2jR1PusTBPdLCAFgQWkKubPhzFjoGtX+OILZ1WGSg6XlvP8/Fye/DSb9FZNeen6s+ielhKmxoZZbq6z+Oru3c4qDldd5f+++fnwySewfLlzrq5FC+je3QlyvXv7PF8XSQ6VlJNbUMT6XcXk7Cxi/c4iJ1gVFB93F9s2KfF0a+P0qrp7eldpKbRvkWi9qigWEQFJRC4C/g7EAi+q6l+rbE8A/gWcCewGrlLVje62+4EbgXLgDlWdU12ZItIVmAa0BlYA16pqSV3q8MUCkksVXnjBmVHXrZuzIkOHDkc3FxQdYebyPF79aiPb9h3mkj4d+NPlp9M8sZH/97t9O1x+OSxa5CzC+tBD0Lfvifn27YN585wgNHeuc2t0gKQkJ+jv3QtH3EkE7do5s/lGjYILL3SCVRQpr1C27DnI+l1F5Ox0elO5Bc7z7iorRDRLjCMtJYE2KQkkJcSSEBdDfJzzHBcjxMQIsSLExggxIsTG4CXt2KNF0ya0SUmgTUo8ac2cci3oBUfYA5KIxAI/ABcCecBSYIKqrq6U5+dApqreKiLjgctV9SoR6QW8CQwAOgBzgVPc3byWKSJvAe+o6jQReQ74VlWfrW0dqurzyr9GHZBKSpz/8ufNg+eeg2+/pfT8C8h/9iV2JDRjw65isnceYMnGvXyXV0iFwoCM1vzywlMY3D013K2PHKWlzuy7hx5yejunneacV2rRwgk0K1fC9987QT8pCYYOdYZDhw93gldMjLNt40bnu5gzx3kUFjrLFg0deiw4ZWQ4ASxKJxYUHixh/a5iNhQUs2P/YXYdOMKuoiMUHDjC4dJyjpRVOI/ScsoqlApVyiucR4XiPh9L92dZvmYJcU5wapZAWkqCG6icgJXWLIHU5ASSE+JIToglKT6OpPhYmjSGyTn1FAkBaTDwoKqOdN/fD6Cqf6mUZ46bZ6GIxAHbgTTgvsp5Pfnc3U4oE/grsAs4SVXLKtdd2zpUdaGvY6prQFq7fT+T3vgaz2d99BPXY6+rbrtj1pNkZa84miKqoIp4XgPCsbSjeVBEFalUiagieE9z9quUdrRs3LKd90klh4lVZ2hlfZtOPD1wHO/2Ph+VY7+M8bExnJHegiHdU7m0b4eGP627PvbudWYlzpnjBKADB5wV0U85BQYPhvPPh4EDIcGPW3KUlcFXXznr573/vlOeR1ISJCc75SQmej3H5zVg+ZtWF2lpTjANMdVjgaq8Qik8VMKuA0coKDriBLsDRygoKjku8O0qOsKBw9XfUTc+Nob4uBhiBOJiY472zmJFiI11emgxNX129dtc42zGQHxz552axgOjvV/sXhN/A1JcnUr3T0dgS6X3ecBAX3ncQLIPSHXTF1XZ13M7UW9lpgKFqlrmJX9d6jhKRG4BbgHo3LlztQfsS2JcLKd6TuTLcU+ISKXXx7bFZXRhb0UxihzNrCJuJqn02tlRBSDGySscl0cr5QM5ll+OZoYYcety2nN0XwRihJKmSew9qTP5p5xOQXo3OiU24X+bNqFlUhNaJcfTrU0y6a2SiI2Jzv/GQ65VK7jzTudRX3FxMGyY8/jrX2HTJli4EPLyYNs255bqR44410RVVBy/r7d/SP1NqysftyIJNhEhVjj6M9o0vintWzStcb/DpeVHg9buohKKS8o4VFJOcUk5B4+UUVxSTklZxdGeWFmFUlGhlKvzXFahVPfp1dQpqPGTryFD9bX7r10IZsYGMyB5+8tU9ZPxlcdXure+cXX561LH8QmqLwAvgNND8rJPjTLaJPP01f1rt9P4hjsTywRZly7OwwREYpNY0lslkd4qKdxNafCCOfiZB3Sq9D4d2OYrjzuc1gLYU82+vtILgJZuGVXrqm0dxhhjwiCYAWkp0ENEuopIPDAemF0lz2xgovt6HPCpOv3X2cB4EUlwZ8/1AJb4KtPd5zO3DNwy36tjHcYYY8IgaEN27vmaScAcnCnaL6nqKhH5I7BMVWcDU4CpIpKD02sZ7+67yp01txooA273zH7zVqZb5W+AaSLyMPC1WzZ1qcMYY0zo2YWxtdCop30bY0wd+TvLzibQG2OMiQgWkIwxxkQEC0jGGGMiggUkY4wxEcEmNdSCiOwCinGue2ps2mDH3ZjYcTcuwT7uLqqaVlMmC0i1JCLL/Jkt0tDYcTcudtyNS6Qctw3ZGWOMiQgWkIwxxkQEC0i190K4GxAmdtyNix134xIRx23nkIwxxkQE6yEZY4yJCBaQjDHGRIRGGZBE5P9EZK2IrBSRd0WkZaVt94tIjoisE5GRldIvctNyROS+SuldRWSxiGSLyHT3thi4t7WY7uZfLCIZNdURiXwdd6QTkU4i8pmIrBGRVSJyp5veWkQ+dr+vj0WklZsuIvKke5wrRaR/pbImuvmzRWRipfQzReQ7d58nxb2PtK86Qnz8sSLytYj8x30fsJ/T2v4uhIqItBSRGe7v9hoRGdwYvm8R+aX7M/69iLwpIolR+32raqN7ACOAOPf1ZGCy+7oX8C2QAHQF1uPc5iLWfd0NiHfz9HL3eQsY775+DrjNff1z4Dn39XhgenV1hPsz8fE5+TzuSH8A7YH+7utmwA/uZ/8IcJ+bfl+l734U8F+cOwkPAha76a2BXPe5lfu6lbttCTDY3ee/wMVuutc6Qnz8dwNvAP8J5M9pXX4XQnjMrwI3ua/jgZYN/fsGOgIbgKaVvoPro/X7DvsfjnA/gMuB193X9wP3V9o2x/0BHAzMqZR+v/sQnKubPcHtaD7Pvu7rODef+Koj3J+Dj8/G63GHu111PJb3gAuBdUB7N609sM59/TwwoVL+de72CcDzldKfd9PaA2srpR/N56uOEB5rOvAJcAHwn0D+nNbldyFEx9wc5w+zVElv0N83TkDaghNA49zve2S0ft+Ncsiuip/h/LcDx75cjzw3zVd6KlCoqmVV0o8ry92+z83vq6xIFE1t9ckdlugHLAbaqWo+gPvc1s1W2+++o/u6ajrV1BEqTwC/Birc94H8Oa3L70IodAN2AS+7Q5UvikgyDfz7VtWtwKPAZiAf5/tbTpR+3w02IInIXHdMterjskp5HsC5W+zrniQvRWkd0utSViSKprZ6JSIpwEzgLlXdX11WL2l1/e7DRkTGADtVdXnlZC9Z6/pzGqmfRxzQH3hWVfvhrDlZ3TnPaDs+r9zzVZfhDLN1AJKBi71kjYrvO2i3MA83Vf1Rddvdk5VjgOHq9jlxonynStnSgW3ua2/pBUBLEYlz/1OonN9TVp6IxAEtcG6hXl0dkSaa2noCEWmCE4xeV9V33OQdItJeVfNFpD2w0033dax5wHlV0ue56ele8ldXRygMAS4VkVFAIs5Q1hME9ue0tr8LoZAH5KnqYvf9DJyA1NC/7x8BG1R1F4CIvAOcTZR+3w22h1QdEbkI+A1wqaoerLRpNjDenYnSFeiBcyJzKdDDnVUSj3MycLYbyD4Dxrn7T8Q5V+EpyzNDZxzwqZvfVx2RyOtxh7lNfnFnQE0B1qjqY5U2Vf5eqn5f17mzrwYB+9zhlznACBFp5f43OgJnrDwfOCAig9y6rsP7d1+5jqBT1ftVNV1VM3C+r09V9WoC93Nal9+FoFPV7cAWETnVTRoOrKaBf984Q3WDRCTJbZfnuKPz+w7VybdIegA5OOOi37iP5yptewBnVsk63Fk0bvoonJla64EHKqV3c7+4HOBtIMFNT3Tf57jbu9VURyQ+fB13pD+Ac3CGEFZW+p5H4Yx9fwJku8+t3fwCPO0e53dAVqWyfuZ+jznADZXSs4Dv3X2e4tjKJ17rCMNncB7HZtkF7Oe0tr8LITzevsAy9zufhTNLrsF/38BDwFq3bVNxZspF5fdtSwcZY4yJCI1yyM4YY0zksYBkjDEmIlhAMsYYExEsIBljjIkIFpCMMcZEBAtIxhhjIoIFJGOMMRHh/wO0Mq//cUKsuQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# let's plot the distribution of the target for the houses that\n",
    "# show the most frequent category and those that show missing data\n",
    "\n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "# houses with the most frequent label for FirePlaceQu\n",
    "X_train[X_train['FireplaceQu']=='Gd']['SalePrice'].plot(kind='kde', ax=ax)\n",
    "\n",
    "# houses with missing data in FirePlaceQu\n",
    "X_train[X_train['FireplaceQu'].isnull()]['SalePrice'].plot(kind='kde', ax=ax, color='red')\n",
    "\n",
    "# add the legend\n",
    "lines, labels = ax.get_legend_handles_labels()\n",
    "labels = ['Houses with Gd', 'Houses with NA']\n",
    "ax.legend(lines, labels, loc='best')\n",
    "\n",
    "plt.title('FirePlaceQu')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The distributions are quite different, houses with missing data cost overall less than those with the label Gd."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# I keep the original distribution of SalePrice for houses\n",
    "# that show Gd in FirePlaceQu, for later.\n",
    "\n",
    "tmp = X_train[X_train['FireplaceQu']=='Gd']['SalePrice']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Let's fill na in both train and test\n",
    "\n",
    "X_train['FireplaceQu'].fillna('Gd', inplace=True)\n",
    "X_test['FireplaceQu'].fillna('Gd', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Number of houses')"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# let's inspect the number of observations per category\n",
    "\n",
    "X_train['FireplaceQu'].value_counts().sort_values(ascending=False).plot.bar()\n",
    "plt.xlabel('FireplaceQu')\n",
    "plt.ylabel('Number of houses')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this case, we see a big distortion in the original distribution of the variable. Now we have a lot more houses that contain the variable Gd, respect to the houses that show the other variables. Originally, half of the observartions in FirePlaceQu were missing. Now, all of those values contain Gd. That is why, we observe a big increase in Gd in the bar plot. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'FirePlaceQu')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaQAAAEICAYAAAAQkoCgAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzs3Xl4VNX5wPHvmz2QECAEZA9L2HfCJqKCstQFrBtoq2hd61ItXcS2ttRqK7Wtv9Z9warVyqJVse6KiCsQKvuWsCasCWFJCGR9f3/cGwzJTGaSzGR9P88zz8zce+57zkwgb869554jqooxxhhT10LqugHGGGMMWEIyxhhTT1hCMsYYUy9YQjLGGFMvWEIyxhhTL1hCMsYYUy9YQjLGGFMvWEIyJsBEpIuI5IpIaBBinysiGYGOa0x9YAnJmBoQkZ0icsJNQLkikgsUqWqMqhZXM+Z1IlLsxjsmIqtF5KIAN91XG/qJyGIROSoiOSKyRERG12YbTNNjCcmYmrvYTUClj73eCorDn/93X6tqDNASmAcsFJHWgWpwZUSkB/AlsA7oBnQA3gQ+EpGRtdEG0zRZQjImwEQkUURURMLc90tF5EER+RLIA7qLSJyIzBORfSKyR0Qe8HSKT1VLgOeBaKC7h7pmi8g2txezUUS+X27/TSKyqcz+Ye72DiLyuohkisgOEflJmcPm4CTEX6tqtqrmqOo/gJeBue7xFU4dur3F86v/zZmmzhKSMbXjGuBmIBbYBbwIFAE9gaHAJODG8ge5Se1GIBdI9RB3GzAOiAN+D7wsIu3dY6/ASS7XAi2AqcAht4f2NrAG6AicB9wtIpPdmBOBRR7qWgiME5Goqn10Y/xjCcmYmntTRI64jze9lHlBVTeoahHQGvgecLeqHlfVg8AjwIwy5UeLyBFgP3AV8H1VPVo+qKouUtW9qlqiqgtwklbpabUbgT+r6kp1pKnqLmAEkKCq96tqgapuB54tU38bYJ+Hz7APCHXbb0zAhdV1A4xpBC5R1Y9L34hIoocy6WVedwXCgX0iUrotpFyZb1T1LF8Vi8i1wCygtM4YnIQC0BmnB1VeV6CDm/BKhQKfu6+zgPYejmsPqLu/l6+2GVNVlpCMqR1l13lJB/KBNm6PqVpEpCtOz+Y8nGs+xSKyGijNculADw+HpgM7VDXJS+iPgSuAf5bbfiVOoiwQkeNAszJtCQUSqvtZjAE7ZWdMrVPVfcCHwF9FpIWIhIhIDxE5p4qhmuMkukwAEbkeGFBm/3PAz0VkuDu6r6ebxFYAx0TkHhGJFpFQERkgIiPc434PnOkOxGgtIrEicidwPfA7t8xWIEpELhSRcOA3QGTVvw1jvmMJyZi6cS0QAWwEDgOv4fk0mVequhH4K/A1cAAYiDNcu3T/IuBB4N9ADs7Q7dbu/VEXA0OAHTin4J7DGRiBqqYCZwGDgZ3AEeAPONexPnLLHAVuc4/bAxwH7IZdUyNiK8YaYyojIp2Ab4Dfqeq8um6Pabysh2SMqZSqZuCMCmwvIjF13R7TeFkPyRhjTL1gPSRjjDH1gg37roI2bdpoYmJiXTfDGGMalFWrVmWpqs/bAiwhVUFiYiIpKSl13QxjjGlQRGSXP+XslJ0xxph6wRKSMcaYeiGoCUlEpojIFhFJE5HZHvZHisgCd//ysnOAici97vYtZWYh9hpTRLq5MVLdmBGV1SEi4SLyooisc6fnvzd434QxxhhfgnYNyZ3b6nGcqewzgJUisti9u7zUDcBhVe0pIjNw1lqZLiL9cGYe7o+zONjHIlI6maO3mHOBR1R1vog85cZ+0lsdOHN1RarqQBFpBmwUkVdVdWewvhNjmrrCwkIyMjI4efJkXTfFBEFUVBSdOnUiPDy8WscHc1DDSCDNndoeEZkPTMOZKqXUNJz1WsCZOuUxcaY/ngbMV9V8YIeIpPHdlPoVYorIJmACcLVb5kU37pOV1KFAc3e9mWigADgWqA9vjKkoIyOD2NhYEhMTKTPTuWkEVJVDhw6RkZFBt27dqhUjmKfsOnL6dPoZ7jaPZdxZj48C8ZUc6217PHCkzMzJZevyVsdrOPNv7QN2A39R1ezyH0JEbhaRFBFJyczM9PezG2M8OHnyJPHx8ZaMGiERIT4+vka932AmJE//4spPC+GtTKC2V1bHSKAY55RgN+BnIlJhiWhVfUZVk1U1OSHBZtc3pqYsGTVeNf3ZBjMhZeAsEFaqE7DXWxn31FkckF3Jsd62ZwEt3Rjl6/JWx9XA+6pa6K7Y+SWQXM3Pahqy/Hx44glYtqyuW2JMkxbMhLQSSHJHv0XgDFJYXK7MYmCm+/pyYIk6k+stBma4I+S6AUk4a7h4jOke86kbAzfmWz7q2A1McNeJaQ6MBjYH8PObhuIXv4Dbb4cJE2DdurpujQmyjIwMpk2bRlJSEj169OCuu+6ioKDAY9m9e/dy+eWXe9xX1gUXXMCRI0d8lvNkzpw5/OUvf6nWsZVZvHgxDz30ULXq3rlzJwMGDPBwRHAFLSG512vuAD4ANgELVXWDiNwvIlPdYvOAeHfQwixgtnvsBmAhzgCI94HbVbXYW0w31j3ALDdWvBvbax04o/VigPU4ie6fqro2CF+Fqc8OHHB6R5dfDrGxcP/9dd0iE0SqyqWXXsoll1xCamoqW7duJTc3l1//+tcVyhYVFdGhQwdee+01n3HfffddWrZsGYwmV0tRURFTp05l9uwKd9vUa0G9D0lV31XVXqraQ1UfdLf9VlUXu69PquoVqtpTVUeWjp5z9z3oHtdbVd+rLKa7fbsbo6cbM7+yOlQ1193eX1X7qerDwfwuTD21YAEUF8Pvfw/XXANvvw05OXXdKhMkS5YsISoqiuuvvx6A0NBQHnnkEZ5//nny8vJ44YUXuOKKK7j44ouZNGnSaT2FvLw8rrzySgYNGsT06dMZNWrUqanEEhMTycrKYufOnfTt25ebbrqJ/v37M2nSJE6cOAHAs88+y4gRIxg8eDCXXXYZeXl5Xtt59OhREhMTKSkpOVV3586dKSws9BrnuuuuY9asWYwfP5577rmHF154gTvuuAOAt99+m1GjRjF06FDOP/98Dhw4cKquNWvWMGHCBJKSknj22WcrtKW4uJhf/OIXjBgxgkGDBvH000/X9Mfglc1lZ5q2Dz6A3r2hXz+47DJ49FH48EPntQmq37+9gY17A3unRb8OLfjdxf297t+wYQPDhw8/bVuLFi3o0qULaWlpAHz99desXbuW1q1bs3PnzlPlnnjiCVq1asXatWtZv349Q4YM8VhHamoqr776Ks8++yxXXnklr7/+Oj/84Q+59NJLuemmmwD4zW9+w7x587jzzjs9xoiLi2Pw4MF89tlnjB8/nrfffpvJkycTHh5eaZytW7fy8ccfExoaygsvvHAq3llnncU333yDiPDcc8/x5z//mb/+9a8ArF27lm+++Ybjx48zdOhQLrzwwtPaMm/ePOLi4li5ciX5+fmMHTuWSZMmVXtod2Vs6iDTdBUVweefw/jxzvsxYyA62gY3NGKq6nEkWNntEydOpHXr1hXKfPHFF8yYMQOAAQMGMGjQII91dOvW7VSyGj58+Kmktn79esaNG8fAgQN55ZVX2LBhg8fjS02fPp0FCxYAMH/+fKZPn+4zzhVXXEFoaGiFWBkZGUyePJmBAwfy8MMPn3bMtGnTiI6Opk2bNowfP54VK1acduyHH37ISy+9xJAhQxg1ahSHDh0iNTW10rZXl/WQTNO1YYNzem7cOOd9RISTlCwh1YrKejLB0r9/f15//fXTth07doz09HR69OjBqlWraN68ucdj/V3MNDIy8tTr0NDQU6fsrrvuOt58800GDx7MCy+8wNKlSyuNM3XqVO69916ys7NZtWoVEyZM8BnHW9vvvPNOZs2axdSpU1m6dClz5sw5ta98gi7/XlV59NFHmTx5MsFmPSTTdK1Z4zwPG/bdtjFjnJF2NrVNo3TeeeeRl5fHSy+9BDjXR372s59x3XXX0axZs0qPPeuss1i4cCEAGzduZF0VR2Tm5OTQvn17CgsLeeWVV3yWj4mJYeTIkdx1111cdNFFp3o+VY0DzjWpjh2duQJefPHF0/a99dZbnDx5kkOHDrF06VJGjBhx2v7Jkyfz5JNPUlhYCDinBY8fP+5XvVVlCck0XWvWQFQU9Oz53bYhQ5xBDj5Op5iGSUR44403WLRoEUlJSfTq1YuoqCj++Mc/+jz2tttuIzMzk0GDBjF37lwGDRpEXFyc33X/4Q9/YNSoUUycOJE+ffr4dcz06dN5+eWXT52uq26cOXPmcMUVVzBu3DjatGlz2r6RI0dy4YUXMnr0aO677z46dOhw2v4bb7yRfv36MWzYMAYMGMAtt9xCUVERwSD+dkMNJCcnqy3Q14hMnAhHjsDKld9tS02FXr3guefghhvqrm2N1KZNm+jbt29dN6NaiouLKSwsJCoqim3btnHeeeexdetWIiIi6rpp9Yqnn7GIrFJVnxMP2DUk0zSpOj2kiy8+fXuPHtC8+Xen84xx5eXlMX78eAoLC1FVnnzySUtGAWYJyTRNWVmQmQkDB56+PSQEBg2C1avrpl2m3oqNjcXOkASXXUMyTZN7zwlJSRX3DRgAmzbVbnuMMZaQTBO1bZvz3KNHxX29ejk9qMOHa7dNxjRxlpBM07RtG4hAYmLFfb3cxYmDdPOfMcYzS0imadq2DTp2dIZ9l1d6Gm/r1tptkzFNnCUk0zRt2+b5dB1A9+7O4AZLSI1STExM0Ovw576m8spOhhpIKSkp/OQnP6l23bXxfZWyhGSapsoSUmQkdO1qp+xMtVUnIQVDUVERycnJ/OMf/6jrpvjFEpJpenJznXWQvCUkcK4jWQ+pUVu6dCnnnHMOV155Jb169WL27Nm88sorjBw5koEDB7LNHfhy3XXXceuttzJu3Dh69erFf//7X6Bir+Kiiy5i6dKlzJ49mxMnTjBkyBB+8IMfAPDyyy8zcuRIhgwZwi233EJxcTEA//znP+nVqxfnnHMOX375ZYU2lpSUkJiYeNrifz179uTAgQNel5SYM2cON998M5MmTeLaa69l6dKlXHTRRQCsWLGCM888k6FDh3LmmWeyZcuWU3HT09OZMmUKvXv35ve//73H7+zhhx8+tQzF7373u2p/997YfUim6dnuLrtVWUJKSoKvvnJuoPUwO7QJgLvvDvz9XkOGwP/9n9/F16xZw6ZNm2jdujXdu3fnxhtvZMWKFfz973/n0Ucf5f/cWDt37uSzzz5j27ZtjB8//tRSFZ489NBDPPbYY6x2P9umTZtYsGABX375JeHh4dx222288sorTJw4kd/97nesWrWKuLg4xo8fz9ChQ0+LFRISwrRp03jjjTe4/vrrWb58OYmJibRr167SJSVWrVrFF198QXR09GmTr/bp04dly5YRFhbGxx9/zK9+9atTk82uWLGC9evX06xZM0aMGMGFF15IcvJ3kyt8+OGHpKamsmLFClSVqVOnsmzZMs4++2y/v29fgtpDEpEpIrJFRNJEpMLShe4S5Qvc/ctFJLHMvnvd7VtEZLKvmO6y5stFJNWNGVFZHSLyAxFZXeZRIiKeFzgxjUvpGjfdu3svk5TkzAR+8GCtNMnUjREjRtC+fXsiIyPp0aMHkyZNAmDgwIGnrYV05ZVXEhISQlJSEt27d2fz5s1+1/HJJ5+watUqRowYwZAhQ/jkk0/Yvn07y5cv59xzzyUhIYGIiIjT5qsry9syFJUtKTF16lSio6MrxDp69ChXXHEFAwYM4Kc//elpx0ycOJH4+Hiio6O59NJL+eKLL0479sMPP+TDDz9k6NChDBs2jM2bNwd8GYqg9ZBEJBRnmfCJQAawUkQWq+rGMsVuAA6rak8RmQHMBaaLSD9gBtAf6AB8LCLuWFyvMecCj6jqfBF5yo39pLc6VPUV4BW3rQOBt1TVbs9vCjIynOfOnb2XKV18bNcuaNcu+G1qiqrQkwmWsktFhISEnHofEhJy2gSinpZoCAsLO7WiK8BJLzPEqyozZ87kT3/602nb33zzTY9rM5U3ZswY0tLSyMzM5M033+Q3v/kNUPmSEt6WobjvvvsYP348b7zxBjt37uTcc8+t9DOW/xz33nsvt9xyi882V1cwe0gjgTR3afECYD4wrVyZaUDpXOivAeeJ8y1MA+arar6q7gDS3HgeY7rHTHBj4Ma8xEcdZV0FvFrjT2wahowMCAuDtm29lym9P6nMX8mm6Vq0aBElJSVs27aN7du307t3bxITE1m9ejUlJSWkp6eftrBdeHj4qeUazjvvPF577TUOur3t7Oxsdu3axahRo1i6dCmHDh2isLCQRYsWeaxbRPj+97/PrFmz6Nu3L/Hx8UDlS0p4U/aYsivKAnz00UdkZ2dz4sQJ3nzzTcaOHXva/smTJ/P888+Tm5sLwJ49e059pkAJ5jWkjkB6mfcZwChvZVS1SESOAvHu9m/KHdvRfe0pZjxwRFWLPJT3VkdWmTjTqZgsARCRm4GbAbp06eL905qGIyPDuQcppJK/x7p2dZ4tIRmgd+/enHPOORw4cICnnnqKqKgoxo4dS7du3Rg4cCADBgxgWJl1tW6++WYGDRrEsGHDeOWVV3jggQeYNGkSJSUlhIeH8/jjjzN69GjmzJnDmDFjaN++PcOGDTs12KG86dOnM2LEiNOSSOmSEh07dmT06NHs2LHD5+f45S9/ycyZM/nb3/52asG/UmeddRbXXHMNaWlpXH311addPwKYNGkSmzZtYsyYMYAzHPzll1+mbWV/2FWVqgblAVwBPFfm/TXAo+XKbAA6lXm/DSdZPA78sMz2ecBl3mICCTg9p9LtnYF1ldVR5v2o0rK+HsOHD1fTCIwfrzp2rO9yrVur/vjHwW9PE7Jx48a6bkKVzZw5UxctWlTXzWgwPP2MgRT143dsME/ZZbiJoVQnYK+3MiISBsQB2ZUc6217FtDSjVG+Lm91lJqBna5rWjIyoFMn3+USE62HZEwtCmZCWgkkuaPfInB+8S8uV2YxMNN9fTmwxM2mi4EZ7gi5bkASsMJbTPeYT90YuDHf8lEHIhKC0+uaH8DPbeozVUtIpkpeeOEFLr/8ct8FTY0FLSGpcz3nDuADYBOwUFU3iMj9IjLVLTYPiBeRNGAWMNs9dgOwENgIvA/crqrF3mK6se4BZrmx4t3YXutwnQ1kqOr2wH8Dpl46fBhOnKhaQrJVlQNK7ftstGr6sw3qjbGq+i7wbrltvy3z+iROD8XTsQ8CD/oT092+HWcUXvntldWxFBhd2WcwjUzpkG9/E9KJE85CfoG8cNuERUVFcejQIeLj4/0a8mwaDlXl0KFDRHmasNhPNlODaVqqkpDKjrSzhBQQnTp1IiMjg8zMzLpuigmCqKgoOvnzf8sLS0imaalqDwmchDSyQufbVEN4eDjdSm86NqYcm1zVNC0ZGc79R2ec4btsaQ9p167gtskYA1hCMk1Nejq0b+/M1OBLXBy0amUj7YypJZaQTNPi75DvUjb025haYwnJNC1790KHDv6XT0wEP6ZkMcbUnCUk07Ts3++csvNXYqJzDcnunTEm6CwhmaYjPx+ys6uWkLp2hbw8OHQoeO0yxgCWkExTUjpVvj8j7ErZrN/G1BpLSKbp2L/fea5OQrKh38YEnSUk03RYQjKmXrOEZJqOffuc56okpFatIDbWEpIxtcASkmk6SntIVZmXTsTpJdk1JGOCzhKSaTr274f4eIiIqNpxXbtaD8mYWmAJyTQd+/dX7XRdKUtIxtQKS0im6ahuQkpMhCNH4NixgDfJGPOdoCYkEZkiIltEJE1EZnvYHykiC9z9y0Ukscy+e93tW0Rksq+Y7rLmy0Uk1Y0Z4Ucdg0TkaxHZICLrRKT6K0uZ+q8mPSSwXpIxQRa0hCQiocDjwPeAfsBVItKvXLEbgMOq2hN4BJjrHtsPmAH0B6YAT4hIqI+Yc4FHVDUJOOzGrqyOMOBl4FZV7Q+cCxQG9Esw9YdqzROSDWwwJqiC2UMaCaSp6nZVLQDmA9PKlZkGvOi+fg04T5x1jacB81U1X1V3AGluPI8x3WMmuDFwY17io45JwFpVXQOgqodUtTiAn9/UJzk5znLk1kMypt4KZkLqCKSXeZ/hbvNYRlWLgKNAfCXHetseDxxxY5Svy1sdvQAVkQ9E5H8i8ktPH0JEbhaRFBFJsWWXG7Dq3BRbql07iIqyhGRMkAUzIYmHbeWnTPZWJlDbK6sjDDgL+IH7/H0ROa9CQdVnVDVZVZMTEhI8hDINQk0Skgh06WIJyZggC2ZCygA6l3nfCdjrrYx7TScOyK7kWG/bs4CWbozydVVWx2eqmqWqecC7wLBqflZT39UkIYHdHGtMLQhmQloJJLmj3yJwBiksLldmMTDTfX05sERV1d0+wx0h1w1IAlZ4i+ke86kbAzfmWz7q+AAYJCLN3ER1DrAxgJ/f1CeBSEjWQzImqMJ8F6keVS0SkTtwfvGHAs+r6gYRuR9IUdXFwDzgXyKShtNrmeEeu0FEFuIkiCLg9tIBB55iulXeA8wXkQeAb93YVFLHYRH5G06SU+BdVX0nWN+HqWP790NYGLRuXb3ju3Z1lq84cQKiowPbNmMMAKK2EqbfkpOTNSUlpa6bYarjRz+CDz+EjIzqHf/yy3DNNbB5M/TuHdi2GdPIicgqVU32Vc5majBNQ3XvQSpl9yIZE3SWkEzTEKiEZNeRjAkaS0imaahpQurQAUJDLSEZE0SWkEzjV1zsDEioSUIKC4POnS0hGRNElpBM43fokJOUapKQwO5FMibILCGZxq86S5d7YvciGRNUlpBM41fTm2JLde0Ke/dCoU0Kb0wwWEIyjV+gElJiIpSUVP9eJmNMpSwhmcYvkD0ksOtIxgSJJSTT+O3fD82bQ0xMzeLYvUjGBJUlJNP41fQepFKd3YnmLSEZExSWkEzjF6iEFBkJnTrB9u01j2WMqcASkmn8ApWQAHr2hLS0wMQyxpzGEpJp/Pbvh/btAxOrRw9LSMYEiSUk07idPAlHjgS2h3TwIOTkBCaeMeYUS0imcTtwwHkOZEIC2LYtMPGMMacENSGJyBQR2SIiaSIy28P+SBFZ4O5fLiKJZfbd627fIiKTfcV0lzVfLiKpbsyIyuoQkUQROSEiq93HU8H7JkydCdQ9SKV69HCe7bSdMQEXtIQkIqHA48D3gH7AVSLSr1yxG4DDqtoTeASY6x7bD2ep8f7AFOAJEQn1EXMu8IiqJgGH3dhe63BtU9Uh7uPWAH58U1/UMCGVlCgZh/M4WVjsbChNSNZDMibggtlDGgmkqep2VS0A5gPTypWZBrzovn4NOE9ExN0+X1XzVXUHkObG8xjTPWaCGwM35iU+6jBNQQ0S0pb9OVzwj885a+6nJD/wMa8s3wUtWkDbttZDMiYIgpmQOgLpZd5nuNs8llHVIuAoEF/Jsd62xwNH3Bjl6/JWB0A3EflWRD4TkXGePoSI3CwiKSKSkpmZ6c/nNvVJaUJq27Zqhx09yTXzlnPoeAG/vagfQ7u05NdvrOfFr3ba0G9jgiSYCclTL0T9LBOo7ZXVsQ/ooqpDgVnAv0WkRYWCqs+oarKqJickJHgIZeq1ffugTRsID6/SYX/470aOnijk5RtG8aOzuvHC9SM5v29b7v/vRo6072IJyZggCGZCygA6l3nfCdjrrYyIhAFxQHYlx3rbngW0dGOUr8tjHe7pwEMAqroK2Ab0quZnNfVVNW6KXZ1+hHfW7eP28T3pfUYsAKEhwl+uGExcdDjv5zd3Zvw+cSIYLTamyQpmQloJJLmj3yJwBiksLldmMTDTfX05sERV1d0+wx0h1w1IAlZ4i+ke86kbAzfmW5XVISIJ7iAJRKS7W4fNCdPYVCMhPbNsGy2iwvjRWd1O296yWQT3TOnNV9LK2bBjR6BaaYwhiAnJvV5zB/ABsAlYqKobROR+EZnqFpsHxItIGs5ps9nusRuAhcBG4H3gdlUt9hbTjXUPMMuNFe/G9loHcDawVkTW4Ax2uFVVs4PxXZg6VMWElJWbzwcbDnDVyC7ERIZV2H/ZsE7kd3USlW7dGrBmGmOg4v+4AFLVd4F3y237bZnXJ4ErvBz7IPCgPzHd7dtxRuGV3+6xDlV9HXjd54cwDZdqlRPSf9fspbhEuWx4J4/7w0JDmDhtHDwGO776lu6XXOKxnDGm6mymBtN4HT0K+flVSkhvrt5L3/Yt6NUu1muZi8/tx6GYVmR8/W0gWmmMcVlCMo1XFe9B2nf0BKvTjzB1cIdKy0WGhZLXPYnmO1JJz86raSuNMS5LSKbxKk1Ifs70vXSLc5/ZeX1937PUevggehzK4NXltlifMYFiCck0XlXsIX26+SAd4qJIaut7qfPmQwbS8mQuH3+2joKikpq00hjjsoRkGq8qJKSCohK+TMvinN5t8WtmqT59AGi9exsfbTxQk1YaY1x+JSQReV1ELhQRS2Cm4di/35mhoVUrn0XX7TnC8YJizk5q41/svn0BGJZ3gAUp6T4KG2P84W+CeRK4GkgVkYdEpE8Q22RMYJQO+fajx7Nix2EARnRr7V/sTp0gJobzJZvPUzPJOGyDG4ypKb8Skqp+rKo/AIYBO4GPROQrEbleRKo2SZgxtaUK9yCt3JlN94TmtImJ9C+2CPTpQ78jewBYlJJR3VYaY1x+n4ITkXjgOuBG4Fvg7zgJ6qOgtMyYmtq3z6+EVFKipOzMZmSin72jUn36ELUtlbN6tmFRSjrFJeXnDjbGVIW/15D+A3wONAMuVtWpqrpAVe8EfA9JMqYu+NlD2nowh2MnixhR1YTUty+kp/OD/vHsPXqSz1NteRJjasLfHtJzqtpPVf+kqvvAWRocQFWTg9Y6Y6qrqAgyM/1KSCt3utePqpOQgAmaRevmESy0wQ3G1Ii/CekBD9u+DmRDjAmozExnLjs/EtKa9CO0iYmgc+voqtUxcCAAEZs2cunQjny08QBZufnVaa0xBh8JSUTOEJHhQLSIDBWRYe7jXJzTd8bUT1W4B2n9nqMM6Bjn3/1HZXXvDs2bw9q1TB/RmcJi5Y3/7alGY432h3JkAAAgAElEQVQx4Hu278k4Axk6AX8rsz0H+FWQ2mRMzfmZkE4UFLP1QA6T+rWreh0hIU4vac0aktrFMrxrK+av3M2N47pVPbkZYyrvIanqi6o6HrhOVceXeUxV1f/UUhuNqTo/E9LGfccoURjQMa569QwaBGvXgirTR3RmW+ZxVu06XL1YxjRxvk7Z/dB9mSgis8o/aqF9xlSPnwlpXcYRAAZ1alm9egYNguxs2LuXCwe2JyYyjPkrbXCDMdXha1BDc/c5Boj18KiUiEwRkS0ikiYisz3sjxSRBe7+5SKSWGbfve72LSIy2VdMd1nz5SKS6saM8FWHu7+LiOSKyM99fR7TgOzfD7Gx0KzyS53r9hyjTUwk7Vr4eUNseYMGOc9r19I8MoyLB3fgnbX7OJpXWL14xjRhvk7ZPe0+/97To7JjRSQUeBz4HtAPuEpE+pUrdgNwWFV7Ao8Ac91j+wEzgP7AFOAJEQn1EXMu8IiqJgGH3dhe6yjjEeC9yj6LaYD27oUOla9rBM4cdgM7tqj+NR93pB1r1gBw7ZiunCgs5sWvd1YvnjFNmL83xv5ZRFqISLiIfCIiWWVO53kzEkhT1e2qWgDMB6aVKzMNeNF9/Rpwnji/GaYB81U1X1V3AGluPI8x3WMmuDFwY17iow5E5BJgO7DBn+/BNCB79vhMSHkFRaQdzGVgdU/XAbRsCV27OteRgL7tW3B+33Y8/+UOcvOLqh/XmCbI3/uQJqnqMeAiIAPoBfzCxzEdgbIn0zPcbR7LqGoRcBSIr+RYb9vjgSNujPJ1eaxDRJoD9wCV9vRMA7V3L3Qs/8/tdGkHcylR6Nfe59nnypUObHDdMaEnR/IKefkbW7zPmKrwNyGVTqB6AfCqqmb7cYyncyDlJ/vyViZQ2yur4/c4p/hyPez/roEiN4tIioikZGba1DANgqpfp+xSDzg/+qR2AUhImzfDyZMADOnckrN7JfDk0m1kHy+oWWxjmhB/E9LbIrIZSAY+EZEE4KSPYzKAzmXedwL2eisjImFAHJBdybHetmcBLd0Y5evyVsco4M8ishO4G/iViNxR/kOo6jOqmqyqyQkJCT4+sqkXsrKgsNBnDyn1YC7hoULX1jW8x3vYMCguPq2XdN+FfTmeX8TDH2ypWWxjmhB/l5+YDYwBklW1EDhOxetB5a0EktzRbxE4gxQWlyuzGJjpvr4cWKKq6m6f4Y6Q6wYkASu8xXSP+dSNgRvzrcrqUNVxqpqoqonA/wF/VNXH/Pk+TD231/1bxEcPKe1gDt3bxBAWWsN1J5Pd6RxTUk5tSmoXy8wzE5m/cjdfpmXVLL4xTURV/if2BaaLyLU4v9gnVVbYvV5zB/ABsAlYqKobROR+EZnqFpuHcz0nDZgFzHaP3QAsBDYC7wO3q2qxt5hurHuAWW6seDe21zpMI1aakPzoIfVsG4DJ6jt3hoQEWLnytM0/m9SLHgkx3DV/NQeP+TqhYIwRp3Pho5DIv4AewGqg2N2sqvqTILat3klOTtaUMn8Fm3rquefgpptg505nBJwHJwuL6fvb9/nJhCR+OrFXzeu84AJIT4d1607bvGV/Dpc8/iVd45sx/+bRtGwWUfO6jGlgRGSVPytD+NtDSgbGquptqnqn+2hSycg0IKU9pPbtvRbZnnkcVUhqF6DlvEaMgI0b4fjx0zb3PiOWZ64dzvas41zy+Jds2Z8TmPqMaYT8TUjrAf/Wgjamru3d65xCi/DeG0k96CSGpLY1HGFXKjkZSkpg9eoKu8YlJfDKjaM4XlDMxY9+wUPvbeboCZvJwZjy/E1IbYCNIvKBiCwufQSzYcZUmx83xaYdzCU0REhsE6BVVIYPd569nNIdkdiad38yjosHd+Cpz7Yxbu4SHv80jeN286wxp/hafqLUnGA2wpiA8uOm2NQDuXSNb0ZkWGhg6uzQwXlUco0xITaSv145mB+dlcgjH23l4Q+2MO+LHfz4nB5cM6YrUeEBaosxDZS/w74/A3YC4e7rlcD/gtguY6rPn5tiD+bQMyFA149KJSfDihU+i/XvEMdzM0fwxm1n0r9DCx58dxMTH/nMri+ZJs/fuexuwpkH7ml3U0fgzWA1yphqKyyEAwcq7SEVFJWw81Be4AY0lBo9GrZudW7M9cPQLq341w2j+PdNo8gvLOGyJ79i1S5/JkExpnHy9xrS7cBY4BiAqqYCbYPVKGOq7cABZ+qgSnpIOw8dp7hEAzegodTYsc7z119X6bAze7ThrTvGkhAbyY9eSGFbZqWzWRnTaPmbkPLd2bWBU1Pw+L6ByZjatmeP81xJQko76PzCD8hNsWWNGAFhYfDll1U+tH1cNC/9aCShIcLtr/yPk4XFvg8yppHxNyF9JiK/AqJFZCKwCHg7eM0yppr8mKUh9UAuItAj0NeQoqOdee2++qpah3du3Yy/XjGYzftz+PsnqYFtmzENgL8JaTaQCawDbgHeBX4TrEYZU21+zGOXejCHzq2aER0RhFFtY8c6UwgVVG+W7/F92nLZsE489/l2ttupO9PE+DvKrgRnEMNtqnq5qj6r/sw5ZExt27PHOW1WyczsaQdzSQr06bpSY8c6y1D8r/qDUGd/rw9RYaE88M6mADbMmPqv0oQkjjkikgVsBraISKaI/LZ2mmdMFe3d60wZFOL5n3ZRcQnbM48H/vpRqTPPdJ6redoOnPuVfjy+B0s2H2RN+pEANcyY+s9XD+lunNF1I1Q1XlVb46wjNFZEfhr01hlTVT5madidnUdBcUnwElL79tC9O3zxRY3CXDsmkZbNwvmHXUsyTYivhHQtcJWq7ijdoKrbgR+6+4ypX9LToUsXr7tTDwZoldjKnHsuLF3qzG1XTTGRYdwwthufbD7I5v3HAtY0Y+ozXwkpXFUr3OWnqpl8t6y5MfWDKuze7axP5EXQhnyXNWECHD4Ma9bUKMw1Y7oSGRbCS1/vClDDjKnffCWkyoYKVW8YkTHBkp0NJ074TEgd4qKIifR3GsdqGD/eeV6ypEZhWjaLYOrgDrz57R6OnbTZwU3j5yshDRaRYx4eOcBAX8FFZIqIbBGRNBGpsFKru0T5Anf/chFJLLPvXnf7FhGZ7Cumu6z5chFJdWNGVFaHiIwUkdXuY42IfN/X5zH1XHq681zpKbscegbzdB0417D69IFPPqlxqGvHJJJXUMzrqzIC0DBj6rdKE5KqhqpqCw+PWFWt9JSdiIQCjwPfA/oBV4lIv3LFbgAOq2pP4BFgrntsP2AG0B+YAjwhIqE+Ys4FHlHVJOCwG9trHThrPCWr6hC3jqfdGShMQ7V7t/PspYdUUqLBHfJd1oQJsGyZM7deDQzsFMeQzi359/Ld2J0WprHz98bY6hgJpKnqdnfaofnAtHJlpgEvuq9fA84TEXG3z1fVfHdARZobz2NM95gJbgzcmJdUVoeq5qlq6WI0UdhUSA2fjx7SniMnOFkYxBF2ZU2Y4Kweu3JljUNdmdyZ1IO5rN9jgxtM4xbMhNQRSC/zPsPd5rGMmxyOAvGVHOttezxwpEyCKVuXtzoQkVEisgFnBopbyxxvGqLdu51VYr3cFPvdKrG1kJDOPdd5ruF1JIALB7YnIiyE1/9np+1M4xbMhCQetpXvhXgrE6jtlbZDVZeran9gBHCviESVLygiN4tIioikZGZmeghl6o30dOjUyetNsakHamGEXan4eBgyJCDXkeKahTOxbzsWr9lLYXH1h5IbU98FMyFlAGVP5ncC9nor416/iQOyKznW2/YsoGWZa0Bl6/JWxymqugk4Dgwo/yFU9RlVTVbV5IRKpqMx9cDu3ZUOaEg7mEtCbCQtm0XUTnsmTXJm/j5W81Nt3x/akezjBXy2xf4oMo1XMBPSSiDJHf0WgTNIYXG5MouBme7ry4El7hx5i4EZ7gi5bkASsMJbTPeYT90YuDHfqqwON0YYgIh0BXrjrIprGqr09EqHfKfW1oCGUhdc4AxqCEAv6ZzeCcQ3j+CNb/cEoGHG1E9BS0ju9Zg7gA+ATcBCVd0gIveLyFS32DwgXkTSgFk4s4qjqhuAhcBG4H3gdlUt9hbTjXUPMMuNFe/G9loHcBawRkRWA2/gTBzr31Kfpv4pLnamDfLSQ1KtxRF2pc48E+Li4J13ahwqPDSECwa255PNB8grsEudpnEK6jBnVX0XZ6mKstt+W+b1SeAKL8c+CDzoT0x3+3acUXjlt3usQ1X/BfzL54cwDcO+fU5S8tJD2n/sJLn5RbVz/ahUeLhz2u7dd51ZJMTT5Uz/XTCwPf/6ZhdLNh/kokHe5+szpqEK5ik7Y2pP6ZBvLwnpuwENQb4ptrwLLnCS5erVNQ41sltr2sRE8s7afQFomDH1jyUk0ziU3hTr5ZTdd5Oq1mIPCWDKFOf53Qqd+ioLDREuGHgGn245yPF8O21nGh9LSKZx8NFDSjuYS6tm4cQ3r6URdqXOOAOGDw/IdSRw7kk6WVjCks0HAxLPmPrEEpJpHHbvhthYZxCBB2kHc0hqG4vU8DpOtVx0EXzzDRw4UONQyYmtSYi103amcbKEZBqH7dudhfE8UFW2HsilZ22frit16aXOoIY336xxqNAQ4YIBdtrONE6WkEzjsGMHdOvmcVdWbgFHTxTW7pDvsgYOhB494PXXAxLuwkEdyC8q4RM7bWcaGUtIpuFThZ07vSak0jnsanXId1kicNll8OmnzppNNZTctRVtYyN5Z235iU+MadgsIZmG7+BByMvzmpBKV4lNqu0h32VddhkUFcHbb9c4VEiIcMHA9ny6JZMcW7jPNCKWkEzDt3278+yth3Qgl9jIMNq1iKzFRpUzYoQz8WuATttdPLg9BUUlfLyp5gMljKkvLCGZhm/HDufZy6CGtIPOgIY6GWFXSsQZ3PDhh5CTU+NwQzu3okNcFG+vsdF2pvGwhGQavtKElJjocXetT6rqzWWXQX5+QO5JCgkRLhrcgc9TMzmSVxCAxhlT9ywhmYZvxw5o1w6aNauw6/DxArJy8+v2+lGpsWOhY0f4978DEu7iQR0oLFY+2LA/IPGMqWuWkEzDt317JSPsanFRPl9CQ+Hqq+G99yCr5hPLD+jYgsT4ZnbazjQalpBMw7djh9frR1sPONdrep1RD3pIAD/8oTPabuHCGocSES4a1IGvtmWRmZMfgMYZU7csIZmGrajImcfO6wi7HGIiw+gQV2F1+roxaJBzo+zLLwck3MWDO1Ci8P566yWZhs8SkmnY0tOddZC8JKStB3Lp2baOR9iVd8018PXXsG1bjUP1PiOWXu1iWLzGbpI1DZ8lJNOw+boH6WAOvepqDjtvrrrKGQb+0ksBCTd1cAdW7jzM7kN5AYlnTF0JakISkSkiskVE0kRktof9kSKywN2/XEQSy+y7192+RUQm+4opIt3cGKluzIjK6hCRiSKySkTWuc8TgvdNmKCpJCEdys0nK7eAXu3qyfWjUp06OeskzZvnnHKsoUuHdUIEXluVHoDGGVN3gpaQRCQUeBz4HtAPuEpE+pUrdgNwWFV7Ao8Ac91j+wEzgP7AFOAJEQn1EXMu8IiqJgGH3dhe6wCygItVdSAwE1vOvGHauhUiIz0uzLfVXSW23iUkgFtvhT174L//rXGoDi2jOTspgddWZVBcogFonDF1I5g9pJFAmqpuV9UCYD4wrVyZacCL7uvXgPPEOdk/DZivqvmqugNIc+N5jOkeM8GNgRvzksrqUNVvVbX0xPsGIEpE6nBuGVMtW7dCz57OkOpySidVrZcJ6YILnJ7SU08FJNyVyZ3Ze/QkX6TVfDi5MXUlmAmpI1D2HEKGu81jGVUtAo4C8ZUc6217PHDEjVG+Lm91lHUZ8K2qVhg7KyI3i0iKiKRkZmb6+Mim1m3dCr16ed51IIfYqDqew86bsDC46Sb44IOADG44v19bWjULZ2GKnbYzDVcwE5KnYU3lzyd4KxOo7T7bISL9cU7j3eKhHKr6jKomq2pyQkKCpyKmrhQVOb/MvSakXHq1q6NVYv1xww1Ozy4AvaTIsFAuGdqRjzYcIPu4TSVkGqZgJqQMoHOZ952A8mNTT5URkTAgDsiu5Fhv27OAlm6M8nV5qwMR6QS8AVyrqjX/M9XUrl27oLDQY0JyVomthyPsyurY0Znf7pln4OjRGoe7emQXCopL+PfyXQFonDG1L5gJaSWQ5I5+i8AZpLC4XJnFOAMKAC4HlqiquttnuCPkugFJwApvMd1jPnVj4MZ8q7I6RKQl8A5wr6p+GdBPbmrH1q3Oc+/eFXZl5uZzJK+wfl4/KuuXv4Rjx+Dpp2scKqldLGf3SuClr3dRUFQSgMYZU7uClpDc6zV3AB8Am4CFqrpBRO4XkalusXlAvIikAbOA2e6xG4CFwEbgfeB2VS32FtONdQ8wy40V78b2Wocbpydwn4isdh9tg/JlmODYssV59tBDSq3PI+zKGj4czj8f/u//nJnAa+iGs7pxMCef/9pqsqYBEqdzYfyRnJysKSkpdd0MU+q22+DVV51lwctdJ5r3xQ7+8N+NrPz1+STE1sNBDWV9/DFMnOj0km6+uUahVJVJjywjPDSEd35yVv29fmaaFBFZparJvsrZTA2m4SodYefhl+7GvcdIiI2s/8kI4LzzYPRo+MMf4MSJGoUSEW4a152N+47x6ZaDAWqgMbXDEpJpuLZs8TrCbuO+Y/Rr36KWG1RNIvCnP0FGBjzxRI3DfX9YR7rGN+MvH2ylxG6UNQ2IJSTTMB096vwC79+/wq6CohLSDubQr0MDSUgA554LkyY5iamGI+7CQ0O4+/wkNu47xnvrbfE+03BYQjIN0/r1zvPAgRV2pR7MobBYG04PqdQf/wiHDsHDD9c41NTBHUlqG8Pc9zdzsrA4AI0zJvgsIZmGad0653nAgAq7Nu49BtCwekjgjLi7+mr4y1+cRQdrIDREmDO1P7uz83hyqd1iZxoGS0imYVq/Hlq08Dip6oa9x4gODyUxvnkdNKyG5s51Zm/4+c9rHGpszzZcPLgDT362je2ZuQFonDHBZQnJNEzr1jm9I08j7PYdo2/7WEJDGuCQ506d4N574T//gSVLahzuNxf2JTo8lLsXrLabZU29ZwnJNDyqTg/Jw+k6VWXT3mMN73RdWT//ubO+01131Xi9pHYtoph72UDWZhzlrx9uCVADjQkOS0im4dm3z7kZ1sOAhozDJ8jJL6Jf+7g6aFiAREXBX//qJN0ATLw6ZUB7rh7VhaeXbeet1XsC0EBjgsMSkml4SkfYeeghrdvjDJlu0D0kgEsucW6Yve8+yKr5Gke/u7gfI7u15heL1vL1tkMBaKAxgWcJyTQ8a9Y4zx56SKvTjxARGkLf9vV8DjtfRODvf4ecHPjVr2ocLjIslGeuGU7n1tHc8OJKS0qmXrKEZBqelBRITIT48usswurdR+jXoQWRYRVXkG1w+veHn/wEnnsOVq6scbiWzSJ49ebRdGwZzXX/XMGSzQcC0EhjAscSkml4Vq6EESMqbC4qLmHdnqMM6dyyDhoVJHPmQLt2cMcdUFLzUXJtY6OYf/NoeraN4cYXU3ju8+3YBMumvrCEZBqWrCznptHkihMHbzmQw4nCYoZ2aUQJqUULZ+aGFSvg+ecDEjI+JpJFt45hYr92PPDOJu79zzobEm7qBUtIpmFZtcp59tBDWp1+BIChnVvVZouC7wc/gHHjYPZsZ3RhADSLCOPJHwznjvE9mb8ynR/OW25Ln5s6ZwnJNCyl11KGDauwa/XuI7RuHkHn1tG13KggE4HHHoPDh+E3vwlY2JAQ4eeTe/P3GUNYnX6EqY99web9xwIW35iqCmpCEpEpIrJFRNJEZLaH/ZEissDdv1xEEsvsu9fdvkVEJvuK6S5rvlxEUt2YEZXVISLxIvKpiOSKyGPB+xZMQC1f7ixZHlfxPqNv048wpHPLxrko3aBBcPvtzn1J//tfQENPG9KRhbeMoaCohMue+IqPNtpgB1M3gpaQRCQUeBz4HtAPuEpE+pUrdgNwWFV7Ao8Ac91j+wEzgP7AFOAJEQn1EXMu8IiqJgGH3dhe6wBOAvcBNZ80zNSO4mL4/HM4++wKuw7l5pN2MJfhXRvZ6bqy7r8fEhKcxBSAAQ5lDencksV3nEWPtjHc8q8UXluVEdD4xvgjmD2kkUCaqm5X1QJgPjCtXJlpwIvu69eA88T583YaMF9V81V1B5DmxvMY0z1mghsDN+YlldWhqsdV9QucxGQagvXrnbWCPCSkFTucayuju1ccCt5otGzpTL76zTfw0ksBD39GXBQLbh7DmT3a8IvX1vDqit0Br8OYygQzIXUE0su8z3C3eSyjqkXAUSC+kmO9bY8HjrgxytflrQ6/iMjNIpIiIimZmZn+HmaCYdky59lDQlq+I5vo8FAGdWrAUwb549prYcwY+OUv4ciRgIePjgjluZnJnNMrgXv/s47XradkalEwE5KnE/nlb3jwViZQ2/1th1eq+oyqJqtqckJCgr+HmWBYtgy6dvW45MQ32w+RnNiK8NBGPk4nJAQef9xZyG92hcuyAREVHsrT1wznzB7x3PP6WpZuORiUeowpL5j/ezOAzmXedwL2eisjImFAHJBdybHetmcBLd0Y5evyVodpSEpKYOlSj72j7OMFbN6f07hP15U1dCjMmgVPP+0sUxEEkWFOUkpqF8ttr/yPNemB740ZU14wE9JKIMkd/RaBM0hhcbkyi4GZ7uvLgSXq3Da+GJjhjpDrBiQBK7zFdI/51I2BG/MtH3WYhiQlxbkpdsqUCrtK52Ub3b11bbeq7jz4oHMv1o9+VOPVZb2JjQrnxetH0Lp5BDe8uJL07Lyg1GNMqaAlJPd6zR3AB8AmYKGqbhCR+0VkqltsHhAvImnALGC2e+wGYCGwEXgfuF1Vi73FdGPdA8xyY8W7sb3WASAiO4G/AdeJSIaHUYCmvnj3Xed+nMmTK+xasvkgcdHhDO7UiGZo8CUiAhYscL6TCy5wTuEFQdsWUbxw/Qjyi0r40QsrOXayMCj1GAMg1lnwX3JysqakpNR1M5qmkSMhLAy++uq0zSUlysg/fsyYHm149KqhddS4OrRsGUyaBIMHw/vvQ6vgDHv/Ki2La59fwZge8Tx/3YjGf63OBJSIrFLVivN9lWP/qkz9t2+fM0PDBRdU2LVuz1GycguY0KeJDjg5+2ynp7R6tTO90O7gDNU+s2cb/njpQD5PzeK3b623CVlNUFhCMvXfggXO8+WXV9j16ZaDiMA5vdrWcqPqkWnT4L33nGQ0bBi8805QqrkyuTO3j+/BqyvSeWbZ9qDUYZo2S0im/nv1VWdkWZ8+FXa9t24/w7u0onXziDpoWD0yYYIz8KNzZ7joIrjrLjh+PODV/Gxiby4a1J4/vbeZ99btC3h807RZQjL127ZtztILV11VYdfm/cfYciCHqUM61EHD6qFeveDrr51F/f7xD2eJ948/DmgVISHCX64YzPCurbh7wWq+3X04oPFN02YJydRvzz4LoaEeE9Li1XsJDREuGNi+DhpWT0VFOUufL1sG4eEwcSLceKMzU3igqgh3lkNv1yKKm15KseHgJmAsIZn668QJJyFdcgl06nTarpIS5e21ezmzRzxtYiLrqIH12LhxsGYN3HMPvPCC03t6+mlngtoAiI+J5PnrRlDgDgc/kmdrKZmas4Rk6q9XX3UWpLvjjgq7vkjLIj37BJcN6+ThQANAdDQ89JBzbalfP7j1VhgyxJmYNT+/xuF7to3hqWuGs+tQHpc9+ZX1lEyNWUIy9VNhoTMbwdChcM45FXa/9PUu4ptH8L2BZ9RB4xqYIUOcaZcWLXKmYJo5E9q3d06D/u1vzgi9LVucHmkVndmjDf+6YSSZOflc+uRXp2bNMKY6LCGZ+unFF2H7dmcNoHIL7u3MOs6SzQeYPqIzkWGhddTABkbEGTa/fj18+KEzVPyzz+BnP3Pu7+rTB5o1cxLVmDFwyy3wwQd+neIb1T2e1398JjGRYVz93Df86d1N5OYX+TzOmPJspoYqsJkaasmxY84vyC5dnFFj5RLSzxet4e01e/n8nvG0jY2qo0Y2EllZsHkz7Nz53WPHDmdkY24uJCXBfffBD39Y4edQ3vH8Iv7w343MX5lOQmwkt5zdnStHdKZFVHhtfBJTj/k7U4MlpCqwhFRL7roLHn3U+aWYfPq/4W2ZuUx6ZBkzxyTy24tt6sGgOXkSFi92rkF9+61z2vTpp53l4334dvdhHnpvM8t3ZNMsIpSJ/doxpf8ZnJXUhlhLTk2SJaQgsIRUCz7+2Jmb7bbb4LHHTtulqlz7/ApWpx9hyc/OJSHWRtcFXUkJPP88/OIXTpJ64AG4+25nKL4P6zKO8vI3u/hw434O5xUSItC/QxyjurVmpPto2ayJ39DcRFhCCgJLSEGWkeH0iFq3duaua978tN2LUtL5xWtrmXNxP64b262OGtlE7d8PP/4xvPkmjB4N//ynx5kzPCkqLmHlzsN8vS2L5Tuy+Tb9CAVFJQD0OSP2VHIamdiati3sFGxjZAkpCCwhBVF2tnPvTHq6M6P3gAGn7d68/xjff/wrBneO45UbRxMaUvn1DBMEqs5Q/DvvdKYleuAB+OlP/eotlXWysJg16UdYuTOb5TuyWbXrMHkFzuCJbm2aM6pba0Z3j2dU99a0j4sOxicxtcwSUhBYQgqSffuc+dfWr3eGIE+YcNruXYeOc8VTXwPw9p1n0c7+iq5b+/c79zS99ZYzLP+Xv3RuXo6q3s+lqLiEDXuPsWJHNst3HGLFjmyOnXRG6XWNb3YqQY1IbE2nVtGIj8EVpv6xhBQElpCC4PPPnRFchw45s3pfeOFpu79Ky+KOV7+lRJUFN4+h9xmxddRQcxpV5+f1299CaqpzenXcOOcG3I4dnSHk0dHOQoLh4d89IiIgMtI5LZuQAPHxFXpYxSXKpi9KoH8AAAr9SURBVH3HWL4jm2+2Ownq6AlnYcDYyDB6touhZ0IMbVtEEt88ktioMMJDQwgNEZ8956jwENrERBIfE0mbmAi7baCW1IuEJCJTgL8DocBzqvpQuf2RwEvAcOAQMF1Vd7r77gVuAIqBn6jqB5XFdJc6nw+0Bv4HXKOqBdWpwxtLSAGUlgZ/+pNzwbxbN3jtNWfphNLdB3N5Ymka//nfHrq3ac5zM5PpnhBThw02HpWUwEcfOb2lr76CrVurdoOtyHfJqV07Z7byzp2dIf/uc0nHTmw+EcKq3YfZeiCX1IM57Mg6TlZuAcUl1f/9JQId4qLp1qY5iW2a0a1NDN3bNKdbm+Z0ahVNmC1CGDB1npBEJBTYCkwEMoCVwFWqurFMmduAQap6q4jMAL6vqtPdpcRfBUYCHYCPgV7uYR5jishC4D+qOl9EngLWqOqTVa1DVb3eCWgJqZry8uDAAScJpaQ4p+W++AIiIii+9cdk//LXZEkEaQdz2bTvGMtSM1m/5xiRYSFcP7Ybd5+fRFS4/SXbIJSUOPcv5eU5j4ICZ9aN0kdBgTNt0aFDkJl5+mP/fuca4p49UFTuxtrY2NMTVZs2lERFkR8aTn5evnNNK+84HM8lJPc4ITnHCM3NISQ3l5CcHEKO56AIJRGRFIWHUxAZTU5sKzKbxbEnIpad0ox9EbEcio4ju1kLjsa0pFmHM2jT+QwS28bSJb6Z07NqHkF8TCStmoXTLCKMyLAQQux6pk/+JqSwILZhJJCmqtvdBs0HpgEby5SZBsxxX78GPCbOCeJpwHxVzQd2iEiaGw9PMUVkEzABuNot86Ib98lq1PF1oL6AU9atg+nTndelfwCU/UOg/LbK9nnYpqocOHYSADlV7Lvy4pYTFC3z3md5D3+slI1Vvj1SrgxAWEkRzQpOnhYjtW0iH539AxYM+x67IuLgH8tP7QsNEYZ0bsns7/Xh8uGdbOLUhiYkBFq0cB7VVVz8XXLavbvic0oKHD5MSHEx0UA0OKf9mjeHmBin7thY57lje+c5xu1d5+cTkZ9Ps5wcWmZm0jlzO8MOHoSjRz02pUSEgtBwCkL/v72zi7HqquL4789cZviqDExbMx3AQkuMfbLYRFAfGou0kKaNCQ80JqU2pokfiR8PFsKTxpcqMU2jEYjVmAYVrNVOSBpSCz5KS1NLWwGZispYamlJwcIUnZnlw1535tzLPRfm9l7umcv6JTt3n7XX2fuss/aZde4+686ZyQX//F9XibOCs4AhJJBmZH43LJC3AVZuyLYXkGtmleir916xtWth69aWHkMrA9IAcCKzPQx8Mk/HzEYlnQH6XP6nqn0HvF6rzz7gXTMbraHfyBgTSHoIeAhgyZIldQ3OZfbsyqyxiQmqfFm9thqyvx8/PdFkqrwQVENWvmhEunDSNZLfvwm/vFxWfXFN6Kniwhzv6uL8/AW819vHmetu4M2bPsb78+YDsHqG6J09k9653Syc083Sa+dy0/VzY13/aqerKz2HGhhIKeZ5jI6m5cHu7lQ+SLLDhQvpv1ZUfWub8c479IyMMP7eecbOjTB+boTxkff57+g442PGmBnjY+OMmTE2bsjSTR82efNnVt4u9vP6GfN66Fs4J19h4KI/j02nlQGp1uyo9kieTp681qJuPf1GxqgUmO0AdkBasquxz6W5+WbYvbuhXS8HAXUu2yDoTEql9E2oGfT0TAbBKgTM8RK0llY+tRsGFme2FwFv5OlIKgHzgdN19s2Tvw30eh/VY011jCAIgqANtDIgvQAsl7RUUjewARis0hkENnp9PbDPUpbFILBBUo9nzy0Hns/r0/fZ733gfT7d4BhBEARBG2jZkp0/r/kasJeUov0zM3tN0neBg2Y2CDwOPOEJBadJAQbX201KgBgFvlrOfqvVpw/5MPBrSd8DXvK+aWSMIAiC4MoTP4ydApH2HQRBMHUuN+07fvkVBEEQFIIISEEQBEEhiIAUBEEQFIIISEEQBEEhiKSGKSDpFPCPKzDUtaTfVnUqnWxfJ9sGYd90pp22fcTMrruUUgSkAiLp4OVkpExXOtm+TrYNwr7pzHSwLZbsgiAIgkIQASkIgiAoBBGQismOdh9Ai+lk+zrZNgj7pjOFty2eIQVBEASFIL4hBUEQBIUgAlIQBEFQCCIgtRBJP5B0RNIhSb+T1Jtp2yxpSNJRSXdm5He5bEjSpox8qaQDko5J2uWv38Bfn7HL9Q9IuvFSY7SLPNuKgKTFkvZLOizpNUlfd/lCSc/6eX9W0gKXS9JjbsshSSsyfW10/WOSNmbkn5D0iu/zmJRecZo3Rovs7JL0kqQ9vt20eTXVudtku3olPenX22FJqzrJd5K+6fPyVUm/kjSrU3xXgZlFaVEB1gAlrz8CPOL1W4CXgR5gKfA66XUaXV5fBnS7zi2+z25gg9e3AV/2+leAbV7fAOyqN0Ybz0WubUUoQD+wwuvXAH/1c/h9YJPLN2V8uA54hskX9h5w+ULgb/65wOsLvO15YJXv8wyw1uU1x2iRnd8Cfgnsaea8amTuNtmuXwBf8no30NspvgMGgOPA7Mz5fKBTfFdhays7j1IxqT4P7PT6ZmBzpm2vT/ZVwN6MfLMXkX5hXQ5uE3rlfb1ecj3ljdFG+2va1m6/1Dnep4HPAUeBfpf1A0e9vh24L6N/1NvvA7Zn5Ntd1g8cycgn9PLGaIFNi4DngM8Ce5o5rxqZu02060OkP9iqkneE70gB6QQpUJbcd3d2gu+qSyzZXTkeJN1ZweQEKzPssjx5H/CumY1WySv68vYzrp/XV7so2vHk4ksctwIHgA+b2UkA/7ze1abqwwGvV8upM0azeRT4NjDu282cV43M3WaxDDgF/NyXI38qaS4d4jsz+xewFfgncJLkixfpDN9VEAHpAyLpD76uW13uzehsIb2VdmdZVKMra0DeSF/tomjHUxNJ84DfAt8ws7P1VGvIGvVhy5F0N/CWmb2YFddQbXRetdPuErAC+ImZ3QqcIy2f5VFEG3Lx51L3kpbZbgDmAmvrHNN08l0FLXuF+dWCma2u1+4PRu8G7jD/3ku601icUVsEvOH1WvK3gV5JJb9byeqX+xqWVALmk17VXm+MdlC047kISTNJwWinmT3l4n9L6jezk5L6gbdcnmfPMHB7lfyPLl9UQ7/eGM3k08A9ktYBs0jLXI/S3Hk11bnbLIaBYTM74NtPkgJSp/huNXDczE4BSHoK+BSd4bsK4htSC5F0F/AwcI+Znc80DQIbPBtmKbCc9ND0BWC5Z7Z0kx5IDnog2w+s9/03kp5xlPsqZwOtB/a5ft4Y7aKmbW08ngo8a+px4LCZ/TDTlD2/1ef9fs/YWgmc8SWbvcAaSQv8znYNad39JPAfSSt9rPup7cPsGE3DzDab2SIzu5F07veZ2Rdo3rxqZO42y7Y3gROSPuqiO4C/0CG+Iy3VrZQ0x8cv2zftfXcRrXxAdbUXYIi0NvtnL9sybVtImS1H8Ywdl68jZXi9DmzJyJf55BkCfgP0uHyWbw95+7JLjdHG81HTtiIU4DOk5YhDGX+tI62jPwcc88+Fri/gx27LK8Btmb4edH8MAV/MyG8DXvV9fsTkf0qpOUYLbb2dySy7ps2rqc7dJtv0ceCg++/3pCy5jvEd8B3giB/DE6RMuY7wXbbEvw4KgiAICkEs2QVBEASFIAJSEARBUAgiIAVBEASFIAJSEARBUAgiIAVBEASFIAJSEARBUAgiIAVBEASF4P+nGkkGVgi5oQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# let's plot the distribution of the target for the original variable\n",
    "# fand the imputed variable\n",
    "\n",
    "fig = plt.figure()\n",
    "ax = fig.add_subplot(111)\n",
    "\n",
    "# original distribution of the variable\n",
    "# remember I captured this a few cells up in the notebook\n",
    "tmp.plot(kind='kde', ax=ax)\n",
    "\n",
    "# distribution of the variable after imputation\n",
    "X_train[X_train['FireplaceQu'] == 'Gd']['SalePrice'].plot(\n",
    "    kind='kde', ax=ax, color='red')\n",
    "\n",
    "# add the legend\n",
    "lines, labels = ax.get_legend_handles_labels()\n",
    "labels = ['Original variable', 'Imputed variable']\n",
    "ax.legend(lines, labels, loc='best')\n",
    "\n",
    "plt.title('FirePlaceQu')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The average House value for the most frequent category has decreased quite dramatically after adding the house values of the NA, following re-categorisation of NA as the most frequent label.\n",
    "\n",
    "Thus, the importance of this particular label for prediction of house price, is masked or changed by the addition of missing values. \n",
    "\n",
    "In cases like this, when the percentage of missing values is so high (~50%) we are better off creating a new category to enclose the observations with NA. I will show this in the next lecture."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**That is all for this demonstration. In later sections I will show you how to do mode imputation with scikit-learn and feature engine as part of a machine learning pipeline.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  },
  "toc": {
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": "block",
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}