notebooks/pdp_ice.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# License \n",
    "***\n",
    "Copyright 2017 J. Patrick Hall, phall@h2o.ai & Navdeep Gill, navdeep@h2o.ai\n",
    "\n",
    "Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the \"Software\"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:\n",
    "\n",
    "The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.\n",
    "\n",
    "THE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Partial Dependence and ICE Plots\n",
    "***\n",
    "\n",
    "Based on:\n",
    "\n",
    "Goldstein, Alex, Kapelner, Adam, Bleich, Justin, and Pitkin, Emil. Peeking inside the black box: Visualizing statistical learning with plots of individual conditional expectation. Journal of Computational and Graphical Statistics, 24(1):44–65, 2015.\n",
    "https://arxiv.org/pdf/1309.6392.pdf\n",
    "\n",
    "Hastie, Trevor, Tibshirani, Robert, and Friedman, Jerome. *The Elements of Statistical Learning.* Springer, 2008.\n",
    "https://statweb.stanford.edu/~tibs/ElemStatLearn/printings/ESLII_print12.pdf"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preliminaries: imports, start h2o, load and clean data "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# imports\n",
    "import h2o \n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from h2o.estimators.gbm import H2OGradientBoostingEstimator\n",
    "\n",
    "# display matplotlib graphics in notebook\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Checking whether there is an H2O instance running at http://localhost:54321..... not found.\n",
      "Attempting to start a local H2O server...\n",
      "  Java Version: java version \"1.8.0_181\"; Java(TM) SE Runtime Environment (build 1.8.0_181-b13); Java HotSpot(TM) 64-Bit Server VM (build 25.181-b13, mixed mode)\n",
      "  Starting server from /home/patrickh/anaconda3/lib/python3.6/site-packages/h2o/backend/bin/h2o.jar\n",
      "  Ice root: /tmp/tmpwep6m3qz\n",
      "  JVM stdout: /tmp/tmpwep6m3qz/h2o_patrickh_started_from_python.out\n",
      "  JVM stderr: /tmp/tmpwep6m3qz/h2o_patrickh_started_from_python.err\n",
      "  Server is running at http://127.0.0.1:54321\n",
      "Connecting to H2O server at http://127.0.0.1:54321... successful.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div style=\"overflow:auto\"><table style=\"width:50%\"><tr><td>H2O cluster uptime:</td>\n",
       "<td>01 secs</td></tr>\n",
       "<tr><td>H2O cluster timezone:</td>\n",
       "<td>America/New_York</td></tr>\n",
       "<tr><td>H2O data parsing timezone:</td>\n",
       "<td>UTC</td></tr>\n",
       "<tr><td>H2O cluster version:</td>\n",
       "<td>3.20.0.4</td></tr>\n",
       "<tr><td>H2O cluster version age:</td>\n",
       "<td>1 month and 20 days </td></tr>\n",
       "<tr><td>H2O cluster name:</td>\n",
       "<td>H2O_from_python_patrickh_bwpt12</td></tr>\n",
       "<tr><td>H2O cluster total nodes:</td>\n",
       "<td>1</td></tr>\n",
       "<tr><td>H2O cluster free memory:</td>\n",
       "<td>3.422 Gb</td></tr>\n",
       "<tr><td>H2O cluster total cores:</td>\n",
       "<td>8</td></tr>\n",
       "<tr><td>H2O cluster allowed cores:</td>\n",
       "<td>8</td></tr>\n",
       "<tr><td>H2O cluster status:</td>\n",
       "<td>accepting new members, healthy</td></tr>\n",
       "<tr><td>H2O connection url:</td>\n",
       "<td>http://127.0.0.1:54321</td></tr>\n",
       "<tr><td>H2O connection proxy:</td>\n",
       "<td>None</td></tr>\n",
       "<tr><td>H2O internal security:</td>\n",
       "<td>False</td></tr>\n",
       "<tr><td>H2O API Extensions:</td>\n",
       "<td>XGBoost, Algos, AutoML, Core V3, Core V4</td></tr>\n",
       "<tr><td>Python version:</td>\n",
       "<td>3.6.4 final</td></tr></table></div>"
      ],
      "text/plain": [
       "--------------------------  ----------------------------------------\n",
       "H2O cluster uptime:         01 secs\n",
       "H2O cluster timezone:       America/New_York\n",
       "H2O data parsing timezone:  UTC\n",
       "H2O cluster version:        3.20.0.4\n",
       "H2O cluster version age:    1 month and 20 days\n",
       "H2O cluster name:           H2O_from_python_patrickh_bwpt12\n",
       "H2O cluster total nodes:    1\n",
       "H2O cluster free memory:    3.422 Gb\n",
       "H2O cluster total cores:    8\n",
       "H2O cluster allowed cores:  8\n",
       "H2O cluster status:         accepting new members, healthy\n",
       "H2O connection url:         http://127.0.0.1:54321\n",
       "H2O connection proxy:\n",
       "H2O internal security:      False\n",
       "H2O API Extensions:         XGBoost, Algos, AutoML, Core V3, Core V4\n",
       "Python version:             3.6.4 final\n",
       "--------------------------  ----------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# start h2o\n",
    "h2o.init()\n",
    "h2o.remove_all()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Load and prepare data for modeling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Parse progress: |█████████████████████████████████████████████████████████| 100%\n"
     ]
    }
   ],
   "source": [
    "# load clean data\n",
    "path = '../data/train.csv'\n",
    "frame = h2o.import_file(path=path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# assign target and inputs\n",
    "y = 'SalePrice'\n",
    "X = [name for name in frame.columns if name not in [y, 'Id']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# determine column types\n",
    "# impute\n",
    "reals, enums = [], []\n",
    "for key, val in frame.types.items():\n",
    "    if key in X:\n",
    "        if val == 'enum':\n",
    "            enums.append(key)\n",
    "        else: \n",
    "            reals.append(key)\n",
    "            \n",
    "_ = frame[reals].impute(method='median')\n",
    "_ = frame[enums].impute(method='mode')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# split into training an validation, and 30% test\n",
    "train, valid = frame.split_frame([0.7])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train a predictive model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "gbm Model Build progress: |███████████████████████████████████████████████| 100%\n"
     ]
    }
   ],
   "source": [
    "# train GBM model\n",
    "model = H2OGradientBoostingEstimator(ntrees=100,\n",
    "                                     max_depth=10,\n",
    "                                     distribution='huber',\n",
    "                                     learn_rate=0.1,\n",
    "                                     stopping_rounds=5,\n",
    "                                     seed=12345)\n",
    "\n",
    "model.train(y=y, x=X, training_frame=train, validation_frame=valid)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fce7a648438>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model.varimp_plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## One-dimensional partial dependence plots\n",
    "* One-dimensional partial depedence plots (PDPs) show us the average behavior of a complex response function w.r.t a single input\n",
    "* They allow us to compare this average behavior to domain knowledge and expected behavior\n",
    "* The average behavior of PDPs can be misleading in the presence of strong interactions or for highly nonlinear response functions\n",
    "* Great to explore most important variables in model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "gbm prediction progress: |████████████████████████████████████████████████| 100%\n",
      "    OverallQual  partial_dependence\n",
      "0           2.0       150173.741376\n",
      "1           2.4       150173.741376\n",
      "2           2.8       150173.741376\n",
      "3           3.2       150173.741376\n",
      "4           3.6       150617.412859\n",
      "5           4.0       150617.412859\n",
      "6           4.4       150617.412859\n",
      "7           4.8       156187.307991\n",
      "8           5.2       156187.307991\n",
      "9           5.6       166655.393408\n",
      "10          6.0       166655.393408\n",
      "11          6.4       166655.393408\n",
      "12          6.8       195692.834893\n",
      "13          7.2       195692.834893\n",
      "14          7.6       206924.036451\n",
      "15          8.0       206924.036451\n",
      "16          8.4       206924.036451\n",
      "17          8.8       210674.282410\n",
      "18          9.2       210674.282410\n",
      "19          9.6       210386.983428\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fce72028438>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# manually calculate 1-D partial dependence\n",
    "# for educational purposes\n",
    "\n",
    "def par_dep(xs, frame, model, resolution=20, bins=None):\n",
    "    \n",
    "    \"\"\" Creates Pandas dataframe containing partial dependence for a single variable.\n",
    "    \n",
    "    Args:\n",
    "        xs: Variable for which to calculate partial dependence.\n",
    "        frame: Data for which to calculate partial dependence.\n",
    "        model: Model for which to calculate partial dependence.\n",
    "        resolution: The number of points across the domain of xs for which to calculate partial dependence.\n",
    "    \n",
    "    Returns:\n",
    "        Pandas dataframe containing partial dependence values.\n",
    "        \n",
    "    \"\"\"\n",
    "    \n",
    "    # init empty Pandas frame w/ correct col names\n",
    "    par_dep_frame = pd.DataFrame(columns=[xs, 'partial_dependence'])\n",
    "    \n",
    "    # cache original data \n",
    "    col_cache = h2o.deep_copy(frame[xs], xid='col_cache')\n",
    "  \n",
    "    # determine values at which to calculate partial dependency\n",
    "    if bins == None:\n",
    "        min_ = frame[xs].min()\n",
    "        max_ = frame[xs].max()\n",
    "        by = (max_ - min_)/resolution\n",
    "        bins = np.arange(min_, max_, by)\n",
    "        \n",
    "    # calculate partial dependency  \n",
    "    # by setting column of interest to constant \n",
    "    for j in bins:\n",
    "        frame[xs] = j\n",
    "        par_dep_i = model.predict(frame)\n",
    "        par_dep_j = par_dep_i.mean()[0]\n",
    "        par_dep_frame = par_dep_frame.append({xs:j,\n",
    "                                              'partial_dependence': par_dep_j}, \n",
    "                                              ignore_index=True)\n",
    "        \n",
    "    # return input frame to original cached state    \n",
    "    frame[xs] = h2o.get_frame('col_cache')\n",
    "\n",
    "    return par_dep_frame\n",
    "\n",
    "# show some output\n",
    "par_dep_OverallQual = par_dep('OverallQual', valid, model)\n",
    "par_dep_OverallQual.plot.line(x='OverallQual', y='partial_dependence')\n",
    "print(par_dep_OverallQual)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PartialDependencePlot progress: |█████████████████████████████████████████| 100%\n",
      "PartialDependence: Partial Dependence Plot of model GBM_model_python_1537556590414_1 on column 'OverallQual'\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div style=\"overflow:auto\"><table style=\"width:50%\"><tr><td><b>overallqual</b></td>\n",
       "<td><b>mean_response</b></td>\n",
       "<td><b>stddev_response</b></td>\n",
       "<td><b>std_error_mean_response</b></td></tr>\n",
       "<tr><td>2.0</td>\n",
       "<td>150173.7413763</td>\n",
       "<td>35218.7018989</td>\n",
       "<td>1694.4605852</td></tr>\n",
       "<tr><td>3.0</td>\n",
       "<td>150173.7413763</td>\n",
       "<td>35218.7018989</td>\n",
       "<td>1694.4605852</td></tr>\n",
       "<tr><td>4.0</td>\n",
       "<td>150617.4128588</td>\n",
       "<td>35176.2738326</td>\n",
       "<td>1692.4192639</td></tr>\n",
       "<tr><td>5.0</td>\n",
       "<td>156187.3079913</td>\n",
       "<td>34987.6147970</td>\n",
       "<td>1683.3424018</td></tr>\n",
       "<tr><td>6.0</td>\n",
       "<td>166655.3934084</td>\n",
       "<td>38493.0438860</td>\n",
       "<td>1851.9974375</td></tr>\n",
       "<tr><td>7.0</td>\n",
       "<td>195692.8348930</td>\n",
       "<td>54407.9260499</td>\n",
       "<td>2617.7025626</td></tr>\n",
       "<tr><td>8.0</td>\n",
       "<td>206924.0364511</td>\n",
       "<td>53089.1614124</td>\n",
       "<td>2554.2534694</td></tr>\n",
       "<tr><td>9.0</td>\n",
       "<td>210674.2824099</td>\n",
       "<td>55413.1094712</td>\n",
       "<td>2666.0644725</td></tr>\n",
       "<tr><td>10.0</td>\n",
       "<td>210386.9834277</td>\n",
       "<td>55430.1936130</td>\n",
       "<td>2666.8864336</td></tr></table></div>"
      ],
      "text/plain": [
       "overallqual    mean_response    stddev_response    std_error_mean_response\n",
       "-------------  ---------------  -----------------  -------------------------\n",
       "2              150174           35218.7            1694.46\n",
       "3              150174           35218.7            1694.46\n",
       "4              150617           35176.3            1692.42\n",
       "5              156187           34987.6            1683.34\n",
       "6              166655           38493              1852\n",
       "7              195693           54407.9            2617.7\n",
       "8              206924           53089.2            2554.25\n",
       "9              210674           55413.1            2666.06\n",
       "10             210387           55430.2            2666.89"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "[]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fce720b95f8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# use h2o to calculate 1-D partial dependence\n",
    "# (easy, fast)\n",
    "model.partial_plot(data=valid, cols=['OverallQual'], plot=True, plot_stddev=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Two-dimensional partial dependence plots\n",
    "* Most machine learning algorithms implicity model high degree interactions \n",
    "* Two-dimensional PDPs enable visualization of two-way interactions in a complex model \n",
    "* A significant drawback of PDPs is that they can only visualize 2nd degree interactions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Investigate one dimensional behavior"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "PartialDependencePlot progress: |█████████████████████████████████████████| 100%\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAf4AAALWCAYAAACnePHjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvNQv5yAAAIABJREFUeJzs3XeYFFXWx/HvISgoAiJBBAR0cRUMIJjDoqALGDChggIiiq6Y87rua0DdNWFes4Iuiq7KqoiBZRmzIChGVEBEWUEFJAxKGDjvH7dGmmFCA9NT3V2/z/P0Mz33VlWf2xNO31u3bpm7IyIiIslQLe4AREREpOoo8YuIiCSIEr+IiEiCKPGLiIgkiBK/iIhIgijxi4iIJIgSvySamV1hZg+lue0wM7su0zFVBTP7xsy6xh1HOvLpfV9fZnafmf017jgkvyjxS1aLEtSvZlZoZj+Y2aNmVmcDj9XZzGanlrn7De5+WiXEeYqZrYriLDSzmVGsO2zssZPAzArMbFn03s0zs+fMrOkGHMfN7Hfl1Jf8ORWa2d0bF/1ax29jZiPN7CczW2xm08zsLjNrXs4+6/xeFnP3M919yHq8/uZRm8ZsSPySDEr8kguOcPc6wO7AHsCV63sAM6tR6VGt690oznpAV+BXYLKZ7VwFr50Pzo7evx2A+sBtGXqdd929Tsrj7PU9QGm/T9EHjgnA90AHd68L7AfMAPZP9zgb6ThgOXBoeR+cqujvQbKUEr/kDHf/H/AysDOAmQ0ws6lmtsTMvjazM4q3Le5FmdllZjYXeDLad5uUnt42Zna1mf0zZb9/mdlcM1tkZm+YWbsNiHOVu89w97OA14GrU46/t5m9Y2YLzewjM+ucUldgZn8zs4nR6z9vZg3WY98hZvZ29H68ZmYNU+r7mtksM5tvZn9JjdfMqpnZ5WY2I6p/uvh1zaxV1Ivub2bfRr3xv6TsWz06XTIjet3JZtYiqtvRzMaa2QIz+9LMjk/z/VsAPEv0cy7JzE43s+nRcV8ws22i8jeiTT6Kfr4npPN6KcetZ2aPRb31WWZ2pZlVi+pOid7b28xsASk/0xRXA2+7+4XuPjtqy4/ufru7j4yOU/L38tEKYvrtNEf0u354Sl2N6Oexe8ou/YH7gI+Bk0oc65vodT8Glkb7b2Nmz0Ztnmlm56Zsv6eZvRv9vs0xs7vNbJM0307JYkr8kjOihNID+DAq+hE4HKgLDABuK/FPcGugAdAS6Ad0B75P6el9X8rLvAy0ARoDHwAjNjLs54ADovibAS8B10VxXQw8a2aNUrbvB5wKbAMUAXeux759CO9DY2CTaBvMrC1wL9A3Ou5WQOrQ87nAUcAfovqfgXtKtGN/4PdAF+D/zGynqPxCoDfh51I3iv0XM9scGAs8EcXTG/hHOh+kog8sx7Lm55xadzDwN+B4oCkwCxgJ4O4HRpvtFv18n6rotUq4izBasx3hvehHeD+L7QV8HbXn+lL270r4wFKR1N/LQesR35OE97HYH4F57v4BgJltC3Qm/M6OiOIvqTdwGGFEZTXwIvAR0Izwsz3fzP4YbbsKuABoCOwT1Z+1HvFKtnJ3PfTI2gfwDVAILCT8k/8HULuMbf8NnBc97wysAGql1HcGZpfY52rgn2Ucrz7gQL3o+2HAdWVsewrwVinl3YCV0fPLgMdL1L8K9I+eFwB/T6lrG7Whepr7XplSdxbwSvT8/4CRKXWbR8ftGn0/FeiSUt8UWAnUAFpF70HzlPqJwInR8y+BnqW0+wTgzRJl9wNXlfH+FQC/RD/n/xESV6OS7zvwMHBTyn51olhbRd878Ltyfp9OIXygWpjy2Dt6j5cDbVO2PQMoSNnv2wp+V4uAbinfnx0dvxB4cH1+L1PqUtv+O2AJsFn0/Qjg/1K2vRKYEj3fhpC4O5T4Wzo15fu9SrYJ+DPwaBmxnA+M2ti/aT3if+g8j+SCo9z9PyULzaw7cBXhnHA1YDPgk5RNfnL3Zem+iJlVJ/TkegGNCD0iCD2eRRsWOs2ABdHzlkAvMzsipb4mMD7l++9Sns+K6humue/clOe/EJIihCTw23HdfamZzU/ZtiUwysxWp5StApqkcewWhHPYJbUE9jKzhSllNYDHS9m22LnuXtEVFtsQRmIAcPfCqC3NCIktHe+5+1rn3M2sCWGUZFZK8azouMVSfzalmU/40FQc293A3dFQfeoIy3r9XqYcb7qZTQWOMLMXgSOBDimb9AMejLb93sxeJwz9p46cpLahJeHUV+rPqDrwJoCFialDgU6Ev60awOT1jVuyj4b6JSeZ2aaEYdVbgCbuXh8YA1jKZiVvPVnRrSj7AD0JQ7b1CL1dShxzfR1N9I+U8E/3cXevn/LY3N3/nrJ9i5Tn2xJ6s/PS3Lcsc1KPa2abEYb7i30HdC9x7Foe5lRU5Dtg+zLKXy9xzDru/qc0jlme7wkJCwiz2AltSSfW8swjvNctU8q2LXHcin5/xgHHpPFaG3NL1OLh/p7A5+4+HcDM9iWcovqzhTkqcwk9+t629kS+1Nf+DphZ4me0hbv3iOrvBb4A2niYqHgFG/e3IFlCiV9y1SbApsBPQFHU+z+0gn1+ALYys3pl1G9BGO6dT+jh3LAhgUUT3lqb2V2EYdxroqp/Enprf4y2qRVN9krtDZ5sZm2j5Hwt8Iy7r0pz37I8AxxuZvtHk7OuZe2//fuA682sZRR/IzPrmWZzHwKGWLiMzcxsVzPbChgN7GBhUmHN6LFHytyADfUEMMDM2kcf/m4AJrj7N1H9D4Rz9Osleo+fJrwPW0TvxYWE9z1dVwMHmNnQaE5G8XyFtNoc/UxTH6Ul2ZGE3/M/Ed6LYv0JcyraAu2jx86E3+PuZbzkRGBxNOGvdvR7tbOZ7RHVbwEsBgrNbMfoNSUPKPFLTnL3JYRJaU8TJqP1AV6oYJ8vCD2mr6OZytuU2OQxwvDu/4DPgffWM6x9zKyQ8M+ygDDZbQ93/yR6/e8IPbUrCB9YvgMuYe2/w8cJ53XnArWiNqa7b6nc/TNgMCFRzCG8X6nXjd9BeO9eM7MlUbv3SrPNQwk/g9eidj9MmIOxhJCgTiT00ucCNxI+rG0wdx8H/JUw2jOHMNpwYsomVwPDo59vWlcRpDgHWEqYwPcW4f16ZD1i+4owX6A54cqCJcDbhPZXtAhPM8Lln6mPdUZS3H0O8C6wL/AUhA8MhMmOd7n73JTHTMLvU/8y4l0FHEH4kDCTMOrxEGG0C8Lk0D6EeQUPFr+e5D5z35hRJxGpLGZWQJhomNZKgiIiG0I9fhERkQRR4hcREUkQDfWLiIgkiHr8IiIiCaLELyIikiBauS/SsGFDb9WqVal1S5cuZfPNN6/agKqQ2pfb1L7cls/ty+e2Qfa1b/LkyfPcvVFF2ynxR1q1asWkSZNKrSsoKKBz585VG1AVUvtym9qX2/K5ffncNsi+9pnZrIq30lC/iIhIoijxi4iIJIgSv4iISIIo8YuIiCSIEr+IiEiCKPGLiIgkiBK/iIhIgijxi4iIJIgSv4iISIIo8YuIiCSIEr+IiEiCKPGLiIgkiBK/iIhIgijxi4iIJIgSv4iISIIo8YuIiCSIEr+IiEiCKPGLiIgkiBK/iIhIgijxi4iIJIgSv4iISIIo8YuIiCSIEr+IiEiCKPGLiIgkiBK/iIhIgijxi4iIJIgSv4iISIIo8YuIiCSIEr+IiEiCKPGLiIgkiBK/iIhIgijxi4iIJIgSv4iISIIo8YuIiCSIEr+IiEiCKPGLiIgkiBK/iIhIgijxi4iIJIgSv4iISIIo8YuIiCSIEr+IiEiCKPGLiIgkiBK/iIhIgijxi4iIJIgSv4iISIIo8YuIiCSIEr+IiEiCKPGLiIgkSMYSv5nVMrOJZvaRmX1mZtdE5a3NbIKZTTOzp8xsk6h80+j76VF9q5Rj/Tkq/9LM/phS3i0qm25ml6eUl/oaIiIiSZfJHv9y4GB33w1oD3Qzs72BG4Hb3L0N8DMwMNp+IPCzu/8OuC3aDjNrC5wItAO6Af8ws+pmVh24B+gOtAV6R9tSzmuIiIgkWsYSvweF0bc1o4cDBwPPROXDgaOi5z2j74nqu5iZReUj3X25u88EpgN7Ro/p7v61u68ARgI9o33Keg0REZFEy+g5/qhnPgX4ERgLzAAWuntRtMlsoFn0vBnwHUBUvwjYKrW8xD5llW9VzmuIiIgkWo1MHtzdVwHtzaw+MArYqbTNoq9WRl1Z5aV9aClv+3WY2SBgEECTJk0oKCgobTMKCwvLrMsHal9uU/tyWz63L5/bBrnbvowm/mLuvtDMCoC9gfpmViPqkTcHvo82mw20AGabWQ2gHrAgpbxY6j6llc8r5zVKxvUA8ABAp06dvHPnzqXGX1BQQFl1+UDty21qX27L5/blc9sgd9uXyVn9jaKePmZWG+gKTAXGA8dFm/UHno+evxB9T1T/X3f3qPzEaNZ/a6ANMBF4H2gTzeDfhDAB8IVon7JeQ0REJNEy2eNvCgyPZt9XA55299Fm9jkw0syuAz4EHo62fxh43MymE3r6JwK4+2dm9jTwOVAEDI5OIWBmZwOvAtWBR9z9s+hYl5XxGiIiIomWscTv7h8DHUop/5owI79k+TKgVxnHuh64vpTyMcCYdF9DREQk6bRyn4iISIIo8YuIiCSIEr+IiEiCKPGLiIgkiBK/iIhIgijxi4iIJIgSv4iISIIo8YuIiCSIEr+IiEiCKPGLiIgkiBK/iIhIgijxi4iIJIgSv4iISIIo8YuIiCSIEr+IiEiCKPGLiIgkiBK/iIhIgijxi4iIJIgSv4iISIIo8YuIiCSIEr+IiEiCKPGLiIgkiBK/iIhIgijxi4iIJIgSv4iISIIo8YuIiCSIEr+ISBVbvhy++QYmT4aJE7fkiSfgrrtg3rxQ/9JLcMgh0KEDbLst1KkDm24KM2aE+ltuCd+XfPz4Y6i/6qrS63/9NdRfdNG6dZtttia+M85Yt75x4zX1ffqsW7/99mvqjzwylB166IG/1e+225r6gw5ad//9919Tv8ce69Z3776mfqed1q0/7rg19c2br1s/YMCa+nr11q0/99xQV1RU+nt3xRWhfuHCNWWp7fvb30L9d9+Vvv9dd4X6qVNLr3/00VA/cWLp9c88E+rHjSu9/pVXyvx1W0eN9DcVEZGyzJ4d/gE3agTTp4d/5PPnh8e8eeHr/ffDPvvAc8+F5BmsyYidOkHDhrByJRQWhgS2226w1VawySZQv37Ybo894MIL142hOHnvt1/p9TWi//gHHRSOl6paSjfw0EOhQYO162vVWvP88MOhZcu16+vVW/P86KOhXTv49tvv2HbbsGHqB4cTToC99157/xYt1jzv2xfmzFm7PvWDxamnwoIFa9e3bbvm+ZlnwtKla9d36LDm+bnnhgSfaq+9wlez0t+7Aw4IXzfddE19avuK999ii9L3L379Bg1Kr9955/B1661Lr99hh/C1ZcvS61u3XresLObu6W+dxzp16uSTJk0qta6goIDOnTtXbUBVSO3LbWpfvAoL4YYb4NZbQ7Lv0wfeeAMOPjj8k2/YMCTuhg3hyiuhY0eYORPGjw/l3377AYccsjtbbRW2r1497hZVnmz/2W2sbGufmU12904Vbacev4jIBli9GkaMgMsuC73Tvn3hwAND3f77w4oVa/eiU7VuvaaHVlCwmB13rJqYRUCJX0RkgwwcCMOGheH5Z58NQ/jFykr4ItlAiV9EJE0//AC1a0PdutC/f+jh9++vRC+5Rb+uIiIVWLEizKRv0waGDAllnTuHmeJK+pJr9CsrIlKOl1+GXXaBSy4JM7tPPz3uiEQ2jhK/iEgZhgyBHj3APVxb/9JLay6rEslVOscvIpJi8eKw0E2TJtCrV7h+/bzz1r3uXSRXqccvIkK4PG/YMPj972Hw4FC2445hiF9JX/KJEr+IJN6ECeFyvAEDoFWrcG2+SL5S4heRRHvssbB87Hffhedvvx2WxBXJV0r8IpI4y5fDt9+G5z16hKV0v/wyrL6ny/Mk3+lXXEQSwx1efDHcEOWYY8J5/YYNw+z9LbaIOzqRqqHELyKJcdll4ZaxNWrA9derdy/JpMv5RCQRpkwJd9Dr3x8efBBq1ow7IpF46POuiCTCXXeF2+DedpuSviSbEr+IJML998P48bDllnFHIhIvDfWLSF5bsgRWrYL69aFdu7ijEYmfevwikteuvhp22iksxSsiSvwiksc++wzuuAOOOALq1o07GpHsoMQvInnJHc45JyT8G26IOxqR7KFz/CKSl/71rzCZ7557wiI9IhKoxy8ieek//4H27eGMM+KORCS7qMcvInnpgQdg4UKoXj3uSESyi3r8IpJXvvkGvv46PK9fP9ZQRLKSEr+I5A13OOss2GcfWLYs7mhEspOG+kUkb7z4Irz8MgwdCrVqxR2NSHZSj19E8sKvv8L550PbtnD22XFHI5K91OMXkbxw880wcyaMG6eb8IiURz1+EckLhYXQuzccfHDckYhkN/X4RSQv3HQTrF4ddxQi2U89fhHJaW+9BW+/HZ5X0380kQqpxy8iOWv5chg4MDz//HMt1iOSDiV+EclZt98OX30VLuFT0hdJjwbGRCQnzZ4NQ4ZAz57QrVvc0YjkDiV+EclJl1wCRUVw221xRyKSWzTULyI5xx322AM6dIDWreOORiS3ZKzHb2YtzGy8mU01s8/M7Lyo/CkzmxI9vjGzKVF5KzP7NaXuvpRjdTSzT8xsupndaWYWlTcws7FmNi36umVUbtF2083sYzPbPVPtFJGqZwYXXgiXXhp3JCK5J5ND/UXARe6+E7A3MNjM2rr7Ce7e3t3bA88Cz6XsM6O4zt3PTCm/FxgEtIkexWf0LgfGuXsbYFz0PUD3lG0HRfuLSB549FEYMSL0+kVk/WUs8bv7HHf/IHq+BJgKNCuuj3rtxwNPlnccM2sK1HX3d93dgceAo6LqnsDw6PnwEuWPefAeUD86jojksLlzw3r8//xn3JGI5K4qmdxnZq2ADsCElOIDgB/cfVpKWWsz+9DMXjezA6KyZsDslG1ms+YDRBN3nwPhgwbQOGWf78rYR0Ry1OWXh5vx3HFHGO4XkfWX8cl9ZlaHMKR/vrsvTqnqzdq9/TnAtu4+38w6Av82s3ZAaX/eFQ3ypbWPmQ0inAqgSZMmFBQUlHqwwsLCMuvygdqX25LSvk8/rcvw4bvTp88svv9+Jt9/H3dklSOff3753DbI4fa5e8YeQE3gVeDCEuU1gB+A5uXsWwB0ApoCX6SU9wbuj55/CTSNnjcFvoye3w/0Ttnnt+3KenTs2NHLMn78+DLr8oHal9uS0L6iIvcOHdybN3dfsiTuiCpXPv/88rlt7tnXPmCSp5GbMzmr34CHganuPrREddcomc9O2b6RmVWPnm9HmJj3tYch/CVmtnd0zH7A89FuLwD9o+f9S5T3i2b37w0sio4jIjmoWjX461/h3nuhTp24oxHJbZkc6t8P6At8UnzJHnCFu48BTmTdSX0HAteaWRGwCjjT3RdEdX8ChgG1gZejB8DfgafNbCDwLdArKh8D9ACmA78AAyq3aSJSlczg6KPjjkIkP2Qs8bv7W5R+rh13P6WUsmcJcwFK234SsHMp5fOBLqWUOzB4/SIWkWx0zz3b8+678Oc/xx2JSH7Qkr0ikrUmTYJnn23Ojz/GHYlI/lDiF5GstHo1nH021K+/kquvjjsakfyhxC8iWemxx2DCBDjjjBnUqxd3NCL5Q4lfRLLO8uXhnP6++8Ihh/wQdzgieUV35xORrLPppjB6dPg6b17c0YjkF/X4RSSrfP55OL/fsSPsvM61PCKysZT4RSQrLF8Of/kL7LorPPBA3NGI5C8N9YtI7CZOhAEDQm+/f3844YS4IxLJX+rxi0ishg6FffaBxYthzBgYNgy23DLuqETylxK/iMTCo/tl7r47DBwIn34K3bvHG5NIEmioX0Sq1NKl4Vz+ppvCjTdC587hISJVQz1+Eaky48eHyXt33BEm8xX3+kWk6ijxi0jGLVkCf/oTHHxwuMXu66/D7beHu+6JSNVS4heRjJs9G4YPhwsvhI8+ggMPjDsikeTSOX4RyYiFC+Hpp2HQINhpJ5g5E5o0iTsqEVGPX0Qq3ejR0K4dnHVWuDYflPRFsoUSv4hUmgULoG9fOOIIaNAA3nsP2raNOyoRSaWhfhGpFKtXwwEHwFdfwf/9H1xxRbhkT0SyixK/iGyU+fOhfn2oXh1uvhm22Qbat487KhEpi4b6RWSDuMNTT8GOO8Ldd4eyHj2U9EWynRK/iKy3efPg2GPhxBOhdWvo2jXuiEQkXRrqF5H1MmECHHcc/PQT3HQTXHAB1NB/EpGcoT9XEVkvv/wCtWvDO++EG+yISG7RUL+IVGjx4rAYD8BBB8Fnnynpi+QqJX4RKdfHH0OnTnDyyfDtt6GsZs14YxKRDafELyJlGjYM9toLCgth3DjYdtu4IxKRjaXELyKlOussGDAA9t0XPvwwLM4jIrlPiV9EStWuHVx5Jbz2mtbZF8knmtUvIr957jmoVg2OOgoGD447GhHJBPX4RYSVK+Gii8KiPPfcE1blE5H8pMQvknD/+1+4RG/oUDj77HBLXbO4oxKRTNFQv0iCzZ0LHTrAr7/CyJFwwglxRyQimabEL5JgW28N554bluDdcce4oxGRqqChfpGEmTcPjj46LMwDYea+kr5IcijxiyTIe++Fof0xY2Dq1LijEZE4KPGLJIA73HFHWISnZk14912dzxdJKiV+kQR47DE4/3zo0QMmT9YNdkSSTJP7RPJYURHUqAF9+oRL9Pr21aV6IkmnHr9InvrqK9htN5g/Pwzv9+unpC8iSvwieWn1ahg0CL7/PlyjLyJSTEP9InnokUfg9dfhwQehefO4oxGRbKIev0iemTMHLr4Y/vAHGDgw7mhEJNso8YvkmVtvhWXL4IEHdE5fRNalxC+SZ264AcaNgx12iDsSEclGOscvkieWLAmT+urVg/32izsaEclW6vGL5InLLoNddoHCwrgjEZFspsQvkgfefhvuvReOPRbq1Ik7GhHJZkr8IjluxQrjtNOgZUsYMiTuaEQk2+kcv0iOe+KJlnzxRbjjnnr7IlIR9fhFcpg7TJtWhz59oHv3uKMRkVygHr9IDjOD6677lL337hx3KCKSI9TjF8lRL70Es2aF5F+7dtzRiEiuUOIXyUGzZ0Pv3nDeeXFHIiK5RolfJMe4w1lnQVERDB0adzQikmuU+EVyzDPPwIsvhkv3ttsu7mhEJNco8YvkkJ9/hnPOgY4dNcwvIhtGiV8kh1SrBkcdBQ8+CDV0TY6IbAD96xDJIfXqwX33xR2FiOQy9fhFcsCvv4Z1+D/4IO5IRCTXKfGL5IAhQ+C558I5fhGRjaHEL5LlPv4Ybr4ZTjkFunSJOxoRyXVK/CJZbNUqOO00aNAAbrkl7mhEJB9ocp9IFnvsMXj/fXjySdhqq7ijEZF8oMQvksVOPhlq1YITTog7EhHJFxrqF8lC7lBYCDVrhjX5zeKOSETyhRK/SBZ68kn4/e9h+vS4IxGRfJOxxG9mLcxsvJlNNbPPzOy8qPxqM/ufmU2JHj1S9vmzmU03sy/N7I8p5d2isulmdnlKeWszm2Bm08zsKTPbJCrfNPp+elTfKlPtFKls8+aF5XhbtIDWreOORkTyTSZ7/EXARe6+E7A3MNjM2kZ1t7l7++gxBiCqOxFoB3QD/mFm1c2sOnAP0B1oC/ROOc6N0bHaAD8DA6PygcDP7v474LZoO5GccNFFsHBhWJa3evW4oxGRfJOxxO/uc9z9g+j5EmAq0KycXXoCI919ubvPBKYDe0aP6e7+tbuvAEYCPc3MgIOBZ6L9hwNHpRxrePT8GaBLtL1IVhs7Nszkv/xy2GWXuKMRkXxUJbP6o6H2DsAEYD/gbDPrB0wijAr8TPhQ8F7KbrNZ80HhuxLlewFbAQvdvaiU7ZsV7+PuRWa2KNp+Xom4BgGDAJo0aUJBQUGp8RcWFpZZlw/Uvuxx55070KJFfQ44YBIFBavT2ieX2rch1L7clc9tg9xtX8YTv5nVAZ4Fznf3xWZ2LzAE8OjrrcCpQGk9cqf0UQkvZ3sqqFtT4P4A8ABAp06dvHPnzqW2oaCggLLq8oHalz3+8Af46Sdo3PjAtPfJpfZtCLUvd+Vz2yB325fRWf1mVpOQ9Ee4+3MA7v6Du69y99XAg4ShfAg99hYpuzcHvi+nfB5Q38xqlChf61hRfT1gQeW2TqTyfP45zJoVLttr3DjuaEQkn2VyVr8BDwNT3X1oSnnTlM2OBj6Nnr8AnBjNyG8NtAEmAu8DbaIZ/JsQJgC+4O4OjAeOi/bvDzyfcqz+0fPjgP9G24tknZUr4aSToGvXsESviEgmZXKofz+gL/CJmU2Jyq4gzMpvTxh6/wY4A8DdPzOzp4HPCVcEDHb3VQBmdjbwKlAdeMTdP4uOdxkw0syuAz4kfNAg+vq4mU0n9PRPzGA7RTbKbbfBlCnwzDOaxS8imZexxO/ub1H6ufYx5exzPXB9KeVjStvP3b9mzamC1PJlQK/1iVckDl9+CVddBT17wjHHxB2NiCSBVu4TicmqVTBgANSuDffeq2V5RaRq6CY9IjFZsSJcqz94MDRtWvH2IiKVQYlfJCa1a8P998cdhYgkjYb6RapY8RD/pElxRyIiSaTEL1LFhg6FYcNg2rS4IxGRJFLiF6lCU6fCX/8KRx8NJ+oiUxGJgRK/SBUpKoJTToE6dTSLX0Tio8l9IlVk2DCYOBFGjoQmTeKORkSSSolfpIr07w9160IvLS0lIjHSUL9IhhUVwc8/Q82acPzxGuIXkXgp8Ytk2E03Qbt2MHdu3JGIiCjxi2TUp5/C1VfD/vvD1lvHHY2IiBK/SMasXBlm8devD/fcE3c0IiKBJveJZMhNN8HkyeF2u40axR2NiEigHr9IBriHJXlPOAGOPTbuaERE1lCPXyQDzOC552DZsrgjERFZm3r8IpXsiSdg5syQ/GvXjjsaEZG1KfGLVKKPPgoT+q69Nu5IRERKp8QvUkmKZ/E3aAC33BJ3NCIipdNNJfkRAAAgAElEQVQ5fpFKcsMNMGUKjBoFW20VdzQiIqVTj1+kEkyZAtddByedBEcdFXc0IiJlU+IXqQTbbQfnnQd33hl3JCIi5dNQv8hGcg933dN5fRHJBerxi2yEDz+EDh3gyy/jjkREJD3q8YtsoBUroH9/mDcPGjeOOxoRkfQo8YtsoCFD4JNP4MUXYcst445GRCQ9GuoX2QCTJ8Pf/gb9+sHhh8cdjYhI+pT4RTbAbbdBkyZw++1xRyIisn401C+yAR59NKzHryF+Eck16vGLrIcvv4QFC6BmTdhhh7ijERFZf+rxi6Rp+XI45hjYfHOYMCHcfU9EJNco8Yuk6eqr4fPPYcwYJX0RyV0a6hdJw8SJcNNNcOqp0L173NGIiGw4JX6RCrjD2WdD06YwdGjc0YiIbBwN9YtUYOlSaNYMTj8d6tWLOxoRkY2jxC9SgTp1YNSouKMQEakcGuoXKcesWfDVV3FHISJSeZT4RcoxZAjsvjsUFsYdiYhI5VDiFynD3Lnw+ONhPf46deKORkSkcijxi5Thrrtg5Uq44IK4IxERqTxK/CKlKCyEe++Fo46CNm3ijkZEpPIo8YuUYuJE+OUXuOSSuCMREalcupxPpBQHHwzffw8NGsQdiYhI5VKPX6SERYvCVyV9EclHSvwiKdzhkEOgb9+4IxERyQwlfpEUb70F778P++4bdyQiIpmhxC+S4pZboGFD6N8/7khERDJDiV8k8uWX8MILMHgwbLZZ3NGIiGSGEr9I5J57oFYtOOusuCMREckcJX6RyN/+BmPGQOPGcUciIpI5Svwikc03h4MOijsKEZHMUuKXxPvlF9hvP3jttbgjERHJPCV+Sbxhw+CddzShT0SSQYlfEm3VKhg6FPbaK/T6RUTynRK/JNrzz8OMGXDxxWAWdzQiIpmnxC+Jdsst0Lo1HH103JGIiFQN3Z1PEssdLrgAqlWD6tXjjkZEpGoo8UtimUGvXnFHISJStTTUL4k0bRpcd92aW/CKiCSFEr8k0tChMGQILFsWdyQiIlVLiV8S56efwrX7/fpBkyZxRyMiUrWU+CVx/vGP0NO/8MK4IxERqXpK/JIov/4Kd98Nhx8OO+0UdzQiIlVPiV8SZf78sErfxRfHHYmISDx0OZ8kSvPmMHp03FGIiMQnYz1+M2thZuPNbKqZfWZm50XlN5vZF2b2sZmNMrP6UXkrM/vVzKZEj/tSjtXRzD4xs+lmdqdZWFzVzBqY2VgzmxZ93TIqt2i76dHr7J6pdkrumDIFZs6MOwoRkXhlcqi/CLjI3XcC9gYGm1lbYCyws7vvCnwF/Dllnxnu3j56nJlSfi8wCGgTPbpF5ZcD49y9DTAu+h6ge8q2g6L9JeHOPhsOOSSs2CciklQZS/zuPsfdP4ieLwGmAs3c/TV3L4o2ew9oXt5xzKwpUNfd33V3Bx4DjoqqewLDo+fDS5Q/5sF7QP3oOJJQ774Lb78N556rm/GISLJVyeQ+M2sFdAAmlKg6FXg55fvWZvahmb1uZgdEZc2A2SnbzI7KAJq4+xwIHzSAxin7fFfGPpJAt94K9evDqafGHYmISLwyPrnPzOoAzwLnu/vilPK/EE4HjIiK5gDbuvt8M+sI/NvM2gGl9c8qGqxNax8zG0Q4FUCTJk0oKCgo9WCFhYVl1uWDfG/ftGmrGTXKOfHEb5k0Kf9O8uf7z0/ty1353DbI3fZlNPGbWU1C0h/h7s+llPcHDge6RMP3uPtyYHn0fLKZzQB2IPTWU08HNAe+j57/YGZN3X1ONJT/Y1Q+G2hRxj6/cfcHgAcAOnXq5J07dy61HQUFBZRVlw/yvX1jx37OZpsZt9zSkqZNW8YdTqXL95+f2pe78rltkLvty+SsfgMeBqa6+9CU8m7AZcCR7v5LSnkjM6sePd+OMDHv62gIf4mZ7R0dsx/wfLTbC0D/6Hn/EuX9otn9ewOLik8JSPIccsiPfP89NNUsDxGRjPb49wP6Ap+Y2ZSo7ArgTmBTYGx0Vd570Qz+A4FrzawIWAWc6e4Lov3+BAwDahPmBBTPC/g78LSZDQS+BYpvsjoG6AFMB34BBmSojZLlfvopfN1ii3jjEBHJFhlL/O7+FqWfax9TxvbPEk4LlFY3Cdi5lPL5QJdSyh0YvD7xSv5Ztgx23hm6dm1NDo7GiYhkhJbslbz1z3/Cjz9Cx44/xx2KiEjWUOKXvLR6dbiEr0MH6NBhYdzhiIhkDa3VL3lpzBj44gsYMUIL9oiIpFKPX/LSffdBixbQq1fF24qIJIl6/JKXnngCpk+HmjXjjkREJLuoxy95xx3q1oXddU9GEZF1KPFLXpk5E9q3h/ffjzsSEZHspMQveeX22+Hzz2GbbeKOREQkOynxS95YsAAefhj69IFmuhejiEiplPglLyxbBuefD0uXwkUXxR2NiEj2UuKXvPDCC/D44/DnP8Ouu8YdjYhI9tLlfJKz3GHqVGjbNlyv/7vfaSa/iEhF1OOXnPTDD9CjB+yxB3z7bVidT0lfRKRiaSV+M6ttZr/PdDAi6XjpJdhlFygogJtuCiv0iYhIeipM/GZ2BDAFeCX6vr2ZvZDpwERKcodzzoHDD4emTWHSJBg8WGvxi4isj3R6/FcDewILAdx9CtAqcyGJlK44wV9wAUyYAO3axRuPiEguSmdyX5G7LzJ1qyQG7nDnnbDvvuF8/p13qocvIrIx0unxf2pmfYDqZtbGzO4C3slwXCLMnRsm8J1/frhUD5T0RUQ2Vjo9/nOAvwDLgSeBV4EhmQxKZPRoOPVUWLIE/vEPOPPMuCMSyaBVq2DRIpg/PzwWLFj3a1FR3FGutx2+/x5Gjow7jIzJ1fZVmPjd/RdC4v+LmVUHNnf3ZRmPTBLr5ZfhiCNgt93C7XXbto07IpE0ucPixaUn7jKS+n4//hg+4bqXfkwzqF8fNtmkattSCRquWJGTcacrV9tXYeI3syeAM4FVwGSgnpkNdfebMx2cJMvy5bDppnDIIXDHHXDGGeF7kaw2fXro9T31VFhRatWqsretWxe22goaNAhfW7fmx2XLaLbLLuH71Lrir/XqQfXqVdeeSvROQQGdO3eOO4yMybr2pXkuNJ2h/rbuvtjMTgLGAJcRPgAo8UulWL06TNq77bZwO93GjeHcc+OOSqQcs2eHRD9yZLiuFOCAA+DSS8tO4FtuCTVrrnOoaQUFNMum5CF5L53EX9PMagJHAXe7+0ozK2NMSmT9zJ0Lp5wCr74ahveraS1JyVY//QTPPBOS/ZtvhqH5jh3hllvghBOgefO4IxRJSzqJ/37gG+Aj4A0zawkszmRQkgyjR8OAAVBYuGYCn2btS1ZZvBhGjQrJfuzYMIy/005wzTVw4onQpk3cEYqst3Qm990J3JlSNMvMDspcSJIUDz0EzZrBk0+G/6UiWeHXX8On0pEjw/rQy5dDq1ZwySXQu3dYL1qfUCWHpTO5b1PgWMJqfanbX5uhmCRPLV8OL74YRkdbt4Zhw6B2bU3gkyywciW89lpI9v/+dxiG2nrrMMO0d2/Yay8le8kb6Qz1Pw8sIkzoW57ZcCTfuMPEiTB8ePif+vPPcPzxYV5U/fpxRyeJtmoVvPFG+MV85plwed2WW4Yh/N694Q9/yNnZ9CLlSSfxN3f3bhmPRPLSIYfAuHFQqxYccwz07w9dusQdlSRW8SfR4svv5syBzTeHnj1Dsj/00Jy8LltkfaST+N8xs13c/ZOMRyM5benSMA/qxRdhxAioUWNN56lXr3AJs0hsliwJw02vvBKSe48e4ZfzsMNC8hdJiHQS//7AKWY2kzDUb4C7+64ZjUxywurV4cqm4cPhX/8Kp0ZbtYJZs2D77eG00+KOUIRw3ehhh8FHH4XL7wYO1LkmSax0En/3jEchOWf16nDN/euvw8EHwxZbhM5Uv35hHRNdjy9Z46uvoFs3+OEHeOGF0NMXSbB0LuebZWa7AQdERW+6+0eZDUuy0aJFoVc/fDjssw/cdBMceGA4XXrEEbDZZnFHKFLChAlw+OHh+fjxsOee8cYjkgUq7JeZ2XnACKBx9PinmZ2T6cAke/z3v9CnT7i66fTTwwJm220X6qpXD4uWKelL1hk9Gg46KEwuefddJX2RSDpD/QOBvdx9KYCZ3Qi8C9yVycAkHqtWhcXK5sxZc1e8Rx8N86EGDAiz8vfcU5c0S5Z76KFwDf7uu4cPAE2axB2RSNZIJ/Eb4c58xVZFZZKFVqyAhQvDo2nTcO7922/D2iQLF4bh+uL6v/4VdtgB3nijIf36hbIlS9Yc66OPYNdd4dZbw/9RLbQjWc89LKd7zTXQvTs8/TTUqRN3VCJZJZ3E/ygwwcxGERJ+T+DhjEYVg+nTw/nqVN26wRVXhOcHHwxFRWvXH3MMnH9+WPSrtGvTTz4ZBg0KyfaII9aUF/eWTz89bDN3briqqGT9OefA0UfD11+H45Ssv+SScNnxhAlw1FEhcS9btma755+HI4+ETz8NrwVh0l29emFC8/z5oWyrrVbQpcua8nr1ws3Eiu850rhx+e+dSFYoKgo3fHj44TA8df/9pd4NTyTp0pncN9TMCgiX9QEMcPcPMxpVTGqUeDdSF+0qWQdrz1xPt95T7mtYnMDdwyz5kvWpZcUJPbV+5crwtVGj8MGifv01ibt+/TDKCdC5c+j1168fOj8lh+nbtVvM4MHrxi+SM5YuDZNNXnoJrrwSrr1W56NEypBOj7+YAavJ02H+3/0uTGIry2uvlV1Xs2b5+9arV35906bhsriybL89vPVW2fXbbQcPPFB2/WabafKd5LGffgoz9ydNgnvvDb1+ESlTOrP6/w8YDmwJNAQeNbMrMx2YiEiFvv4a9tsPPv4YnntOSV8kDen0+HsDHdx9GYCZ/R34ALguk4GJiJRr8uSwGE9RUbghxL77xh2RSE5IZ321b4BaKd9vCszISDQiIul49dVw97zateGdd5T0RdZDOol/OfCZmQ0zs0eBT4FCM7vTzO7MbHgiIiUMHx7O6bdpExbm+f3v445IJKekM9Q/KnoUK8hMKCIi5XCHv/0N/vIX6NoVnn1Wt3wU2QDpXM43vPi5mW0JtHD3jzMalYhIqlWr4Nxz4R//gJNOgkceCbfWFZH1ls6s/gIzq2tmDYCPCLP6h2Y+NBERqLZ8ORx3XEj6l14Kjz2mpC+yEdIZ6q/n7ovN7DTgUXe/yszU4xeRzFuwgN0uugg+/xzuuCP0+kVko6Qzua+GmTUFjgdGZzgeEZFg1izYbz+2+OqrsOa+kr5IpUgn8V8LvArMcPf3zWw7YFpmwxKRRPvoI9hnH5g7l49uvjkM9YtIpUhnct+/gH+lfP81cGwmgxKRBCoqgvffD9foDx0a1rp+6y0W/fRT3JGJ5JV0JvftYGbjzOzT6PtdtWSviFSKmTPDXfSOPRYaNgwL8QwZAp06hWv027WLO0KRvJPO5L4HgUuA+wHc/WMzewIt2Ssi62vJEhg/Ptz16rXXYFp01rBFC+jVK9xnuksXaNAg3jhF8lg6iX8zd59oa9/isqisjUVEfrNqFXzwQUjyr74aevFFReF2kQcdBOecE5L9DjvoNroiVSSdxD/PzLYHHMDMjgPmZDQqEcld3323pkf/n//AggWhvGNHuOSSkOj32Qc23TTeOEUSKp3EPxh4ANjRzP4HzAROymhUIpI7li6F119fk+ynTg3l22wDRx4ZEn3XrtCoUbxxighQQeI3s2pAJ3fvamabA9XcfUnVhCYiWWvJErjvPnjlFXjrLVixItwp7w9/gNNPD8m+bVsN34tkoXITv7uvNrOzgafdfWkVxSQi2WzChLBe/owZsNtucN55IdHvvz/UqlXx/iISq3SG+sea2cXAU8Bvyd/dF2QsKhHJPkVF4e5411wDzZvDm2+GZC8iOSWdxH9q9HVwSpkD21V+OCKSlWbOhL594e23Q2//nnvCAjsiknPSWbmvdXn1ZnaIu4+tvJBEJGu4w4gRcNZZ4Xz9iBHQp0/cUYnIRkhnrf6K3FgJxxCRbLNwYejd9+0L7dvDxx8r6YvkgcpI/Jq2K5Jv3ngjTNz717/g+uvDanstW8YdlYhUgspI/F4JxxCRbLByJfzlL9C5M2yySTinf8UVUL163JGJSCVJZ3KfiCTBtGlhKH/SJBg4EG6/HerUiTsqEalkldHj/6YSjiEicXGHhx4K5/G//hqefTZ8r6QvkpfS6vGb2b5Aq9Tt3f2x6OsxGYlMRDJv/vyw0t6oUeGueMOHQ7NmcUclIhlUYY/fzB4HbgH2B/aIHp3S2K+FmY03s6lm9pmZnReVNzCzsWY2Lfq6ZVRuZnanmU03s4/NbPeUY/WPtp9mZv1Tyjua2SfRPndadAvBsl5DRFKMHQu77AIvvQS33BLW2VfSF8l76Qz1dwL2c/ez3P2c6HFuGvsVARe5+07A3sBgM2sLXA6Mc/c2wLjoe4DuQJvoMQi4F0ISB64C9gL2BK5KSeT3RtsW79ctKi/rNURk+XK46KKwzG79+mEJ3osugmqVceZPRLJdOn/pnwJbr++B3X2Ou38QPV8CTAWaAT2B4dFmw4Gjouc9gcc8eA+ob2ZNgT8CY919gbv/DIwFukV1dd39XXd34LESxyrtNUSS7bPPYM89YehQGDw4TORr3z7uqESkCqVzjr8h8LmZTQSWFxe6+5HpvoiZtQI6ABOAJu4+JzrGHDNrHG3WDPguZbfZUVl55bNLKaec1xBJJne4+2649FKoWxdGj4bDDos7KhGJQTqJ/+qNeQEzqwM8C5zv7out7Nt0llbhG1C+PrENIpwqoEmTJhQUFJS6XWFhYZl1+UDty20Vta/mggXseNNNbDVhAvP32osvLr2UlZtvDjnyniT955fL8rltkMPtc/eMPYCawKvAhSllXwJNo+dNgS+j5/cDvUtuB/QG7k8pvz8qawp8kVL+23ZlvUZ5j44dO3pZxo8fX2ZdPlD7clu57Rs92r1RI/datdzvvtt99eoqi6uyJPrnl+PyuW3u2dc+YJKnkZvTmdW/t5m9b2aFZrbCzFaZ2eI09jPgYWCquw9NqXoBKJ6Z3x94PqW8XzS7f29gkYfh+leBQ81sy2hS36HAq1Hdkig+A/qVOFZpryGSHE88AYcfDttsA5Mnh3P6ZY+4iUhCpDPUfzdwIvAvwgz/foQZ9BXZD+gLfGJmU6KyK4C/A0+b2UDgW6BXVDcG6AFMB34BBgC4+wIzGwK8H213rbsviJ7/CRgG1AZejh6U8xoiyfDqq9C/f1h69+WXoVatuCMSkSyR1gI+7j7dzKq7+yrgUTN7J4193qLsG/h0KWV7BwaXcaxHgEdKKZ8E7FxK+fzSXkMkESZMgGOPhZ13huefV9IXkbWkk/h/MbNNgClmdhMwB9g8s2GJyAb54oswW79Jk9DTr1s37ohEJMukcx1/32i7s4GlQAvg2EwGJSIbYPbssChPjRphFb6t13v5DRFJgAp7/O4+y8xqE2bJX1MFMYnI+lqwAP74R1i0CF5/HbbfPu6IRCRLpTOr/whgCvBK9H17M3sh04GJSHqqLVsWZu/PmBHO6WslPhEpRzpD/VcT1shfCODuUwh36hORuK1cSburrw4T+p54IsziFxEpRzqT+4rcfVE5K+6JSBxWr4aBA9lqwgS4/344RnfIFpGKpXWTHjPrA1Q3szZmdhdQ4eV8IpJhl14Kjz/OzFNPhUGD4o5GRHJEOon/HKAd4QY9TwCLgPMyGZSIVODmm+HWW+Hss5l18slxRyMiOSSdxN82etQAahFueft+uXuISOYMHx56+yeeCHfcoWV4RWS9pHOOfwRwMfApsDqz4YhIuUaPhoED4ZBDwgeAaul8dhcRWSOdxP+Tu7+Y8UhEpHxvvw29ekGHDvDss7DJJnFHJCI5KJ3Ef5WZPQSMI5znB8Ddn8tYVCKytk8/Ddfqb7stjBkDW2wRd0QikqPSSfwDgB2BmqwZ6ndAiV+kKsyaFVbl22yzcNe9Ro3ijkhEclg6iX83d98l45GIyLp++imsv//LL/Dmm9CqVdwRiUiOS2dm0Htm1jbjkYjI2goLw532vv0WXnwx3GZXRGQjpdPj3x/ob2YzCef4DXB33zWjkYkk2YoVYSW+Dz6AUaNg//3jjkhE8kQ6ib9bxqMQkTVWr4b+/WHsWHj0UTjiiLgjEpE8ktZteasiEBEB3OH882HkSLjxRjjllLgjEpE8o9U/RLLJDTfAXXfBhRfCJZfEHY2I5CElfpFs8eCDcOWV0LdvWItfS/GKSAYo8Ytkg+eegzPPhB494OGHtRSviGSM/ruIxO3116FPH9hzT3j6aahZM+6IRCSPKfGLxKmgICzFu/328NJLsPnmcUckInlOiV8kLi+9BN27h/X3x46FBg3ijkhEEkCJXyQOTz0FRx0F7dqFof5ttok7IhFJCCV+kar28MPQuzfssw/897/QsGHcEYlIgijxi1Sl226D004Ld9t75RWoWzfuiEQkYZT4RaqCO1xzTViY57jj4Pnnw212RUSqWDpr9YvIxnCHiy4Kvf1TTgkL9dTQn56IxEM9fpFMWrUKBg0KSf/cc8P5fSV9EYmREr9IpqxYASedBA89FJbivf12rcgnIrFT10MkE379FXr1Ctfq33STbrgjIllDiV+ksi1ZAkccAW+8AffdB2ecEXdEIiK/UeIXqUwLFoTV+CZPhn/+M6zBLyKSRZT4RSrLnDlw6KEwbVq4296RR8YdkYjIOpT4RSrDrFnQtWtI/i+9BF26xB2RiEiplPhFNtaXX4akX1gYbrazzz5xRyQiUiYlfpGNMWVKGN43C7fY3W23uCMSESmXLioW2VDvvgudO8Omm4YZ/Er6IpIDlPhFNsR//gOHHAKNGsFbb8Hvfx93RCIiaVHiF1lfzz8Phx0G220Hb74JLVvGHZGISNqU+EXWx4gRcOyx0L59OKe/9dZxRyQisl6U+EXSdd990LcvHHBAGOpv0CDuiERE1psSv0g6br0V/vQn6NEDxoyBLbaIOyIRkQ2ixC9SkbvvhosvDjfdGTUKateOOyIRkQ2mxC9SnkcegXPOgZ49w/n9mjXjjkhEZKMo8YuUZeRIOO20sEDPU08p6YtIXlDiFynN88/DySeHiXyjRoVFekRE8oASv0hJr74Kxx8PHTvC6NGw2WZxRyQiUmmU+EVSvfEGHH007LQTvPKKZu+LSN5R4hcpNmFCWJGvZUt47TXYcsu4IxIRqXRK/CIAH30E3bpB48ZhcZ7GjeOOSEQkI5T4RaZODTfc2WILGDcOmjWLOyIRkYxR4pdkmzEDunaFatVCT79Vq7gjEhHJqBpxByASm+++gy5dYNkyeP112GGHuCMSEck4JX5JprlzQ9L/+Wf4739h553jjkhEpEoo8UvyzJ8fzun/739h9n7HjnFHJCJSZZT4JVkWLYI//hGmTYOXXoL99os7IhGRKqXEL8lRWBhuq/vxx2EZ3i5d4o5IRKTKKfFLMixbFu6w99574YY7hx0Wd0QiIrFQ4pe8ZytXwnHHwfjxMHx4eC4iklBK/JLfiorY6frrw+V6990HffvGHZGISKy0gI/kr9Wr4dRTafz66zB0KJxxRtwRiYjETolf8pM7DB4Mjz/OzFNPhQsuiDsiEZGsoKF+yT/ucPHFYWj/8suZdeihtI47JhGRLKEev+Sfq64KQ/vnnAM33ABmcUckIpI1Mpb4zewRM/vRzD5NKXvKzKZEj2/MbEpU3srMfk2puy9ln45m9omZTTezO83Cf3Eza2BmY81sWvR1y6jcou2mm9nHZrZ7ptooWejGG2HIEDj1VLj9diV9EZESMtnjHwZ0Sy1w9xPcvb27tweeBZ5LqZ5RXOfuZ6aU3wsMAtpEj+JjXg6Mc/c2wLjoe4DuKdsOivaXJHj4Ybj8cujdGx54INxxT0RE1pKx/4zu/gawoLS6qNd+PPBkeccws6ZAXXd/190deAw4KqruCQyPng8vUf6YB+8B9aPjSD77/HM4++ywBv/w4VC9etwRiYhkpbgm9x0A/ODu01LKWpvZh8Bi4Ep3fxNoBsxO2WZ2VAbQxN3nALj7HDNrHJU3A74rZZ85JYMws0GEUQGaNGlCQUFBqcEWFhaWWZcPcr19tmIFHc86i01q1WLSmWey4u2316rP9fZVRO3LbfncvnxuG+Ru++JK/L1Zu7c/B9jW3eebWUfg32bWDijtBK1XcOy093H3B4AHADp16uSdO3cu9YAFBQWUVZcPcr59F10EM2bAiy+y7+GHr1Od8+2rgNqX2/K5ffncNsjd9lV54jezGsAxwG/3QnX35cDy6PlkM5sB7EDorTdP2b058H30/Aczaxr19psCP0bls4EWZewj+Wbs2DCD/6yzoJSkLyIia4tj9lNX4At3/20I38wamVn16Pl2hIl5X0dD+UvMbO9oXkA/4PlotxeA/tHz/iXK+0Wz+/cGFhWfEpA8M28e9O8PO+0EN98cdzQiIjkhYz1+M3sS6Aw0NLPZwFXu/jBwIutO6jsQuNbMioBVwJnuXjwx8E+EKwRqAy9HD4C/A0+b2UDgW6BXVD4G6AFMB34BBvx/e3ceJkV173/8/WUZQET2HRQwIBEXRCJEjc6ViLjkQqL+LgQjbkHjEo0aQZOrxgU1cYkag0FBIC5oNF65PG5EHVHcQBCCMcqAIINcENlFloHz++OcdnpmehZwZqqr6/N6nnq6+pyq7vOlhvnOOXWqqsaDk+g5BxdcAF9+Cc8/D/vsE3WLRERiodYSv3NuRAXl52QoewZ/eV+m7ecCh2Qo/xIo90D1MPv/kj1srsTNww/Dc8/BXXdB375Rt0ZEJDZ0obPEz8cfwxVXwA9/6F9FRKTalPglXnbsgJ/+FBo39tfr6yY9IiJ7RA/pkXi5/nqYNw+efRY6dYq6NSIisaPuksTHa6/B738Po0fDsGFVby8iIiMCapgAACAASURBVOUo8Us8rFsHP/sZ9Ozpr9sXEZG9oqF+yX7OwYUXwurV8M470LRp1C0SEYktJX7JfpMnw9NPw+23w5FHVrm5iIhUTEP9kt0KC+GyyyA/H66+OurWiIjEnhK/ZK+dO2HkSMjLg6lT9ahdEZEaoKF+yV6/+x289x489RR07Vr19iIiUiX1+CU7zZoF48bBuefCmWdWvb2IiFSLEr9knw0b/KV7PXrAvfdG3RoRkZyioX7JLs7BL34BK1fC7NnQrFnULRIRySlK/JJdHnsMpk2DW26BAQOibo2ISM7RUL9kj6VL4eKL4dhjYezYqFsjIpKTlPglOxQXw1lngRk8+qgu3RMRqSUa6pfscOut8Pbb8PjjcMABUbdGRCRnqccv0XvrLbjpJt/jHzEi6taIiOQ0JX6J1qZNPuEfcAA88EDUrRERyXka6pdoXXopLF8Ob7wB++0XdWtERHKeevwSnSeegL/+Ff77v+Hoo6NujYhIIijxSzSWL/c36hk4EH7726hbIyKSGEr8Uvd27fK35N2929+wp4HOOImI1BX9xpW6d8cd/pz+1Kn+fvwiIlJn1OOXuvX3v8MNN8Dw4X42v4iI1Cklfqkb69b5RH/66XDooTB+vL9Ln4iI1Cklfql906dDnz7w5JNw443w7rvQokXUrRIRSSSd45fas349XH65v2TvsMPghRegb9+oWyUikmjq8UvtmDHD9/Iffxyuvx7mzFHSFxHJAurxS83asAGuuAKmTPHn8mfMgH79om6ViIgE6vFLzXn+ed/Lf/RRf1OeuXOV9EVEsowSv3x7GzbAeefBqadCy5bwzjtw882Qlxd1y0REpAwlfvl2XnwRDjnED+1fdx28/z707x91q0REpAJK/LJ3Nm6ECy6Ak0+G5s19L//WW6FRo6hbJiIilVDilz338su+l//IIzB2rO/lf+97UbdKRESqQbP6pfo2bYKrr4aHHoLeveGtt2DAgKhbJSIie0A9fqmemTP95XkTJ8I118D8+Ur6IiIxpMQvldu8GS66CAYPhiZNYPZs/3S9xo2jbpmIiOwFJX6p2Cuv+F7+hAl+iH/+fBg4MOpWiYjIt6Bz/FLi669hyRJYvNjfjOfhh6FXL3jzTTj66KhbJyIiNUCJP2l27IClS31yD8vh774La9dCURE457erVw+uvBJuucUP8YuISE5Q4s9FxcWwfLlP7J98UirJs2wZ7N5dsm3LltTv0AGOPx569iy9NG8eWQgiIlI7lPjjaPt2f2ndpk3w6aelE/snn/iynTtLtm/WzCfy730PRo4sndxbt2ZeQQH5+fmRhSMiInVHib8uOOfPn2/a5GfJb95cvfWK6tKTekqTJj6RH3oo/OQnfr1XL//arh2Y1X3cIiKSdZT4U+bPh333zVj1g127oH79vftc52DbttLD6xUx823Ybz/fS2/WzK+3a1f6ffr6/vv75N6pkz8vLyIiUgkl/pS2beGsszJWfb5iBV27dt37z27cuHzSzpTE99lHyVtERGqVEn9Kly5w550Zq5YUFNBV58BFRCQHqHspIiKSIEr8IiIiCaLELyIikiBK/CIiIgmixC8iIpIgSvwiIiIJosQvIiKSIEr8IiIiCaLELyIikiBK/CIiIgmixC8iIpIgSvwiIiIJosQvIiKSIEr8IiIiCaLELyIikiBK/CIiIgmixC8iIpIgSvwiIiIJUmuJ38wmmdkaM1uUVnajma00sw/Cckpa3bVmVmhmH5vZSWnlQ0JZoZmNTSvvbmbvmtliM3vSzPJCeaPwvjDUd6utGEVEROKmNnv8k4EhGcrvcc71DcvzAGZ2MDAc6BP2+bOZ1Tez+sADwMnAwcCIsC3AHeGzegLrgfND+fnAeufcd4B7wnYiIiJCLSZ+59wsYF01Nx8KTHPObXfOfQoUAkeFpdA5t9Q5twOYBgw1MwNOAJ4O+08BhqV91pSw/jQwKGwvIiKSeFGc47/UzBaGUwEtQ1lnYEXaNkWhrKLy1sAG51xxmfJSnxXqN4btRUREEq9BHX/feOBmwIXXu4DzgEw9ckfmP0xcJdtTRV0pZjYaGA3Qvn17CgoKMjZ6y5YtFdblAsUXb4ov3nI5vlyODeIbX50mfufc6tS6mT0EzAhvi4CuaZt2AT4P65nK1wItzKxB6NWnb5/6rCIzawA0p4JTDs65CcAEgP79+7v8/PyM7S4oKKCiulyg+OJN8cVbLseXy7FBfOOr06F+M+uY9vbHQGrG/3RgeJiR3x3oCbwHzAF6hhn8efgJgNOdcw54DTgj7D8KeC7ts0aF9TOAV8P2IiIiiVdrPX4zewLIB9qYWRFwA5BvZn3xQ+/LgAsBnHMfmtlTwL+AYuAS59yu8DmXAi8B9YFJzrkPw1eMAaaZ2S3AfGBiKJ8I/NXMCvE9/eG1FaOIiEjc1Frid86NyFA8MUNZavtbgVszlD8PPJ+hfCl+1n/Z8m3AmXvUWBERkYTQnftEREQSRIlfREQkQZT4RUREEkSJX0REJEGU+EVERBJEiV9ERCRBlPhFREQSRIlfREQkQZT4RUREEkSJX0REJEHq+rG8IiJSDbt3w8aNsG4dFBfDQQf58gUL4OuvS2/bvDl897t+fd482LGjdH2rVtCrl1+fMwd27Spd37YtHHigX3/3XSj7WLMOHaBbN9+m994r39ZOnWD//WHnTnj//ZLyf/1rPxo3hq5doXNn2LYNPvig/P7duvnv+Oor+Oc/y9cfeKBv4+bN8OGH5et79fIxbtgA//53+frevaFFC/jyS1i8uHx9nz7QrBmsWQNLl5avP+ww2GcfWLUKli8vH98RR0CjRrByJaxYUX7//v2hQQP47DP4/PPy9UcdBfXqwaefwurVpevMYMAAv15YCGvXlq5v0MB//h5xzmlxjiOPPNJV5LXXXquwLhcovnhTfNlt2zbnvvii5P2bbzo3ebJzd9/t3G9/69zQoUXuuutK6ocNc65VK+fMnPMp2Lmjjy6pP/jgkvLUctJJJfVdu5avP+OMkvqWLcvXn3tuSX2DBuXrL7vM1339dfk6cN+0f82azPW33ebrlyzJXP+nP/n6BQsy10+ZUvJvl6n+mWd8/QsvZK5/+WVf/9RTmevfftvXT5yYuX7RIl9/772Z65cv9/W33JK5/ssvff2YMZnrd+zw9RdfXL6uUaOSY/Ozn5Wvb9OmpB6Y66qR79TjFxHZQ199BUVFJcu6dfCrX/m6G26AZ5+F9et9+dat0LFjSU/v9tthxgy/bgbNmrVj48aSzz7qKN87btXKLy1bQo8eJfV/+Qts2VK6PW3alKxPnep71unaty9Z/9vffM88XefOJeszZpTv8e+/v39t2BBeeKH8v0eqfc2bl65fuHAhhx122DejDR06ZN4/NVrRvXvm+kMP9a8HH5y5vm9f/3rkkZnrjzjCv/7gB5nre/f2r4MHZ65PxT90aMnICZTE17atfz98uG9DWfvu61/PPRfy88vX16/vXy+5BH70o9J19dJOyF99Nfz0p6XrGzYs/3lVMVf2CCdU//793dy5czPWFRQUkJ/paOUIxRdviq9mbd3qh2tXrChJ7CtWwP33Q16e/+V7112l96lXzyfbhg193RtvlCTuVq2gXTu44AK/7aef+sTasqVPlLNm5e7x089m3TKz951zVQ78q8cvInts+3b4xz9gv/2gSxffY8zLi7pV1bN2LcyfXzqpFxXBQw/5OP70JxgzpvQ+bdvCjTf6nvvgwdC6tT9v3aVLyfnrVM/rqqv8UpHu3WstNJFqUeIXkT122mk+8ac7+WR4/nm/fvvtfhg7lRhTfxw0alQ77dm82U9yatLET77629/80PqqVf7188/9EPj3vw8vvwwjR5bs266db+OmTb6Np57qX1Nt79QJGjcu2X7wYL+IxJUSv4hUy9atvlffoAFceSVceKEfqk71mNu1K9n24YdhyZLS+59xhk/I4BNvixY+uaYSbK9ePsmm27wZPvusCQUFfuZ3t26wbBlce23ppP7VV/D44zBihE/8v/61n4XdqZNf+vf3fxQADBoEr79e8R8jffr4RSRXKfGLSJVmzYLzz/fL2LG+d1+ZwkKftFeuLBlOT00gKy6GRYt82fr1Jfv86ldw993+UrVDD/WXNflJbP5apt//3id08JekdeoE/fr50YeOHeHww33dwIH+MrhmzfyoQ1nt25ee7CaSNEr8IlKhzZt9ov/zn/256dT1xNXRrJmfLZ2aMZ3SoIG/Fh18Tz31x0GHDiVlRx3lk3OnTrBhw0cMGvTdb2Z+d+vm/7CoSF5efOYbiERBiV9EMnrjDTjrLN8zv+IKuOUWaNq0Zr+jaVM/xJ9+iVSbNn7YPqWgYDX5+d+t2S8WSTAlfhHJKC/P99pnz/aT4kQkNyjxi8g3nnvO3/L1d7/zw/oLF5a+gYiIxJ/+S4sIX3zhZ8QPGwbTp5fcC15JXyT36L+1SII5B9Om+VuhPvMM3HSTf0hL6tI3Eck9GuoXSbDVq/0len36wKRJcMghUbdIRGqbevwiCeMcvPSSf+3Qwc/ef+stJX2RpFDiF0mQ5cv9zXeGDCm5vW6/fv7aehFJBiV+kQTYvRvGj/e9+jff9A+iqerueyKSm/R3vkgC/PSn8OSTcOKJMGGCv/udiCSTEr9Ijiou9ufxGzb0d+A76SQ455zM968XkeTQUL9IDnr/fX8Dnjvv9O9POw3OPVdJX0SU+EVyyubN/il3Rx3lH1d70EFRt0hEso2G+kVyxGuvwahR/kl3F10E48b5Z96LiKRT4hfJEU2bQqtWfhKfHqojIhVR4heJqV274MEHYdasA8nP98P78+frPL6IVE7n+EViaOFCOOYYuPRSWLq0KTt3+nIlfRGpinr8IjGydat/kM5dd0HLlvDoo9Cp00IaNsyPumkiEhPq8YvEyNq18MADcPbZ8NFHMHKkevkismeU+EWy3OrVcMcd/mY8++8PixfDxInQunXULROROFLiF8lSu3fDQw9B795w/fXw4Ye+vEOHaNslIvGmxC+Shf71Lzj+eBg9Gg4/HBYs0GNzRaRmaHKfSJYpLoZTT4VNm2DSJN1fX0RqlhK/SJZ4801/LX5eHkybBj16QNu2UbdKRHKNhvpFIrZ2re/V/+AH/oY84B+wo6QvIrVBPX6RiMyZ4yfvPfEEbNsG110HP/951K0SkVynxC9Sh7ZuhX328evXXQdvvQX/9V9w1VXQp0+0bRORZFDiF6llzsGsWfDww/A//wMffwydOsH48X44v3nzqFsoIkmixC9SS9av90P5Dz/sb7rTvLm/497u3b7+O9+Jtn0ikkya3CdSg3btgjVr/PpXX/nh/PbtYcoU+Pxzf7vdLl2ibaOIJJt6/CI14LPP4JFH/HX3vXvDSy/5BL9smRK9iGQXJX7JGTNn+qH1Tp188r3oIl++ZUsDdu2C+vVr/jv/8Q+4+2548UX//sQT/d32UpT0RSTbKPFLTti1Cy65BFat8pPpevYsSfxjxhzKj3/s/yDYf3/o2hUGDoRf/tLXf/wxtGoFbdpU7w55ixf7hN6kCcyf72+n+5vfwHnnQffutRejiEhN0Dl+yQlPP+0T8pQpsHkzzJ5dUjds2Equucbf+75BA3j3XT/LPmXQIGjXzl9m16uXf3/vvSX1r74KixbBY49Bfr7f5plnfN2ll8Ly5XDzzUr6IhIP6vFLTti92yfsYcN8rz11rTzAiSeuIT//4Ar3feABfy5+xYqSZe1aX1dc7IfvUzPxe/SAceP8d4Hv9YuIxIkSv+SEESP8sjeGDq24zsyPDqxY4R+He9xxUE/jZCISY0r8EmvOwfTpcMop0LBhzX9+/fpwzDE1/7kiIlFR30Vi7fXX/fD+lClRt0REJB6U+CXWxo3zN8gZOTLqloiIxIMSv8TWnDn+2v0rr9QkOxGR6lLil9i67TZo0aLken0REamaEr/E0tdfw9KlcNllsN9+UbdGRCQ+NKtfYqlJE5g3D3bujLolIiLxoh6/xM4XX8DGjf56+kaNom6NiEi8KPFL7Nxwg79t7tdfR90SEZH4UeKXWFm1yj/6dtgwzeQXEdkbtZb4zWySma0xs0VpZX8ws3+b2UIze9bMWoTybmb2tZl9EJYH0/Y50sz+aWaFZnafmX9+mpm1MrOZZrY4vLYM5Ra2Kwzf06+2YpS6d/fd/rz+NddE3RIRkXiqzR7/ZGBImbKZwCHOucOAT4Br0+qWOOf6hiX9Aq3xwGigZ1hSnzkWeMU51xN4JbwHODlt29Fhf8kB69bB+PEwfDgceGDUrRERiadaS/zOuVnAujJlLzvnisPbd4AulX2GmXUE9nPOve2cc8BUYFioHgqkbtQ6pUz5VOe9A7QInyMxN2MGfPUVjB1b9bYiIpJZlOf4zwNeSHvf3czmm9nrZvaDUNYZKErbpiiUAbR3zq0CCK/t0vZZUcE+EmNnnw2LF8Ohh0bdEhGR+IrkOn4z+w1QDDwWilYB+zvnvjSzI4H/MbM+gGXY3VX18dXdx8xG408H0L59ewoKCjJ+4JYtWyqsywVxiG/Hjnrk5e0GoKioio3LiEN834bii7dcji+XY4P4xlfnid/MRgGnAYPC8D3Oue3A9rD+vpktAXrhe+vppwO6AJ+H9dVm1tE5tyoM5a8J5UVA1wr2KcU5NwGYANC/f3+Xn5+fsc0FBQVUVJcLsj2+bdvgoIPgiivgV7/a8/2zPb5vS/HFWy7Hl8uxQXzjq9OhfjMbAowB/tM5tzWtvK2Z1Q/rPfAT85aGIfzNZjYwzOY/G3gu7DYdGBXWR5UpPzvM7h8IbEydEpB4mjwZPvsMDj886paIiMRfrfX4zewJIB9oY2ZFwA34WfyNgJnhqrx3wgz+44CbzKwY2AVc5JxLTQz8Bf4KgSb4OQGpeQG3A0+Z2fnAZ8CZofx54BSgENgKnFtbMUrtKy6GO+6AAQPgP/4j6taIiMRfrSV+59yIDMUTK9j2GeCZCurmAodkKP8SGJSh3AGX7FFjJWtNmwbLlsG994Jlmr0hIiJ7RHfuk6zlHNx5p5/Ff9ppUbdGRCQ36Ol8krXM4LnnYM0a/0AeERH59pT4JasdcIBfRESkZqgfJVlp5kw46SRYuTLqloiI5Bb1+CUrjRsHn3wCbdpE3RIRkdyixC9Z5623oKDAP4mvUaOoWyMikls01C9Z57bboHVr+PnPo26JiEjuUeKXrLJggX8K3+WXw777Rt0aEZHco6F+ySrduvnz+xddFHVLRERykxK/ZJXmzeHaa6NuhYhI7tJQv2SN226DZzLeuFlERGqKEr9khaIiuOEGeO21qFsiIpLblPglK9x5p783/69/HXVLRERymxK/RO6LL2DCBDjrLN2eV0SktinxS+TuvRe2bYMxY6JuiYhI7lPil8gdcghcdRX07h11S0REcp8u55PIDR/uFxERqX3q8Utktm6F+++HLVuibomISHIo8UtkJk6EX/4S5s2LuiUiIsmhxC+R2LED/vAHOPZYOO64qFsjIpIcOscvkXjsMVixAv7yl6hbIiKSLOrxS53btQtuvx2OOAKGDIm6NSIiyaIev9S5tWuhQwe47DIwi7o1IiLJosQvda59e3j9dX+LXhERqVsa6pc69eqrsGqVX1dvX0Sk7inxS52ZMgVOPBHOOUe9fRGRqCjxS52YMMEn/BNOgGefVW9fRCQqSvxS6+67Dy68EE45Bf73f2GffaJukYhIcinxS63asQMefRR+/GPf02/cOOoWiYgkm2b1S61wDoqLIS8PZs70vfyGDaNulYiIqMcvNc45uO46OPVU2L4dmjdX0hcRyRZK/FKjnIMrr/R35uvRQwlfRCTbKPFLjdm9Gy6+GP74R7j8chg/HurpJ0xEJKvo17LUmF//Gh58EMaMgXvu0SV7IiLZSJP7pMaMGgXt2sE11yjpi4hkK/X45VvZsQMef9yf2z/sMN/bV9IXEcleSvyy17Ztg5/8BEaOhPfei7o1IiJSHRrql72ydSsMG+av0R8/HgYMiLpFIiJSHUr8sse2bIHTToNZs+CRR/w9+EVEJB6U+GWPzZ4Nb78Njz0GI0ZE3RoREdkTSvxSbbt3++vyTzoJliyBLl2ibpGIiOwpTe6Talmzxp/HnzHDv1fSFxGJJ/X4pUqrVsGgQbBsGTRqFHVrRETk21Dil0qtWAEnnAD/93/wwgtw/PFRt0hERL4NJX6p0Nq1cNxxsG4dvPwyfP/7UbdIRES+LZ3jlwq1bg3Dh8Mrryjpi4jkCvX45RsrVvhL9WbP9uf0hw2D226LulUiIlKTlPgTbtcuuPnm73L22T7xAzRtCp07R9suERGpHUr8CbFxI7zzTkmPvl07eOIJqF8fNmzI4+ij4Zhj4Oij4fDDoYF+MkREcpJ+vecg5/wleJ06+ffnnANTp/ryevV8Yk8/Z3/XXQvIz8+PoqkiIlLHlPhzwI4dMH++78m/9ZZ/XbfO9/IbN/Yz83v08D36AQNg332jbrGIiERFiT8mdu7019KvXFmynHkmdOwI99wDY8f67bp39xPzjjnGn78HOO+86NotIiLZRYk/C2zf7u99n57UV66ECy6Afv38jXNOPdUP1ac76CCf+E8/HQ480J+fTw3vi4iIZKLEH2zfDosXly5r1gw6dPDrZesAmjf3k+Scg8LC8vUtW0KbNv7Z9S++WD6xX3wxnHEGfPghHHlk6X1btfI99379oHdvuP56P9M+fWnd2m/7ne/4RUREpCpK/MFHH0GvXqXLRo2CyZP9ep8+frg93aWXwv33+/Ky+wJcey2MG+efX3/66b6sYUPfK+/c2T/tDqBnT3j88ZKE3qkTNGlS8jndu8ONN9ZElCIiknRK/MEBB8BNN5Uu69GjZH3KlJJEnZJK9vXrw6OPlv/MPn38a5s2MG+eT+pt2viZ9emaNdNz7UVEpG4o8QctW8LIkRXXV5aY69evfN969eCII/a+bSIiIjVF9+oXERFJECV+ERGRBFHiFxERSRAlfhERkQRR4hcREUkQJX4REZEEUeIXERFJECV+ERGRBFHiFxERSZBaTfxmNsnM1pjZorSyVmY208wWh9eWodzM7D4zKzSzhWbWL22fUWH7xWY2Kq38SDP7Z9jnPjOzyr5DREQk6Wq7xz8ZGFKmbCzwinOuJ/BKeA9wMtAzLKOB8eCTOHADMAA4CrghLZGPD9um9htSxXeIiIgkWq0mfufcLGBdmeKhwJSwPgUYllY+1XnvAC3MrCNwEjDTObfOObcemAkMCXX7Oefeds45YGqZz8r0HSIiIokWxUN62jvnVgE451aZWbtQ3hlYkbZdUSirrLwoQ3ll31GKmY3GjxjQvn17CgoKMjZ4y5YtFdblAsUXb4ov3nI5vlyODeIbXzY9nc8ylLm9KK8259wEYAJA//79XX5+fsbtCgoKqKguFyi+eFN88ZbL8eVybBDf+KKY1b86DNMTXteE8iKga9p2XYDPqyjvkqG8su8QERFJtCgS/3QgNTN/FPBcWvnZYXb/QGBjGK5/CRhsZi3DpL7BwEuhbrOZDQyz+c8u81mZvkNERCTRanWo38yeAPKBNmZWhJ+dfzvwlJmdD3wGnBk2fx44BSgEtgLnAjjn1pnZzcCcsN1NzrnUhMFf4K8caAK8EBYq+Q4REZFEq9XE75wbUUHVoAzbOuCSCj5nEjApQ/lc4JAM5V9m+g4REZGk0537REREEkSJX0REJEGU+EVERBJEiV9ERCRBlPhFREQSRIlfREQkQZT4RUREEkSJX0REJEGU+EVERBJEiV9ERCRBlPhFREQSxPwt8sXMvgCWV1DdBlhbh82pa4ov3hRfvOVyfLkcG2RffAc459pWtZESfzWY2VznXP+o21FbFF+8Kb54y+X4cjk2iG98GuoXERFJECV+ERGRBFHir54JUTeglim+eFN88ZbL8eVybBDT+HSOX0REJEHU4xcREUkQJX7AzJaZ2T/N7AMzmxvKWpnZTDNbHF5bhnIzs/vMrNDMFppZv2hbX56ZTTKzNWa2KK1sj+Mxs1Fh+8VmNiqKWDKpIL4bzWxlOIYfmNkpaXXXhvg+NrOT0sqHhLJCMxtb13FUxMy6mtlrZvaRmX1oZpeH8pw4hpXElxPH0Mwam9l7ZrYgxPe7UN7dzN4Nx+JJM8sL5Y3C+8JQ3y3tszLGHaVK4ptsZp+mHb++oTxWP58AZlbfzOab2YzwPieO3Tecc4lfgGVAmzJlvwfGhvWxwB1h/RTgBcCAgcC7Ubc/QzzHAf2ARXsbD9AKWBpeW4b1llHHVkl8NwJXZ9j2YGAB0AjoDiwB6odlCdADyAvbHBx1bKHNHYF+Yb0Z8EmIIyeOYSXx5cQxDMdh37DeEHg3HJengOGh/EHgF2H9YuDBsD4ceLKyuLM4vsnAGRm2j9XPZ2jblcDjwIzwPieOXWpRj79iQ4EpYX0KMCytfKrz3gFamFnHKBpYEefcLGBdmeI9jeckYKZzbp1zbj0wExhS+62vWgXxVWQoMM05t9059ylQCBwVlkLn3FLn3A5gWtg2cs65Vc65eWF9M/AR0JkcOYaVxFeRWB3DcBy2hLcNw+KAE4CnQ3nZ45c6rk8Dg8zMqDjuSFUSX0Vi9fNpZl2AU4GHw3sjR45dihK/54CXzex9Mxsdyto751aB/0UFtAvlnYEVafsWUfkvrWyxp/HEMc5Lw1DipNQwODGPLwwdHoHvVeXcMSwTH+TIMQxDxR8Aa/AJbQmwwTlXHDZJb+s3cYT6jUBrYhSfcy51/G4Nx+8eM2sUyuJ2/P4IXAPsDu9bk0PHDpT4U45xzvUDTgYuMbPjKtnWMpTF+dKIiuKJW5zjgQOBvsAq4K5QHtv4zGxf4BngCufcpso2zVCW9TFmiC9njqFzbpdzri/QBd/T+26mzcJr7OMzs0OAa4HewPfww/djwuaxic/MTgPW1CvYmQAABH9JREFUOOfeTy/OsGlsjx0o8QPgnPs8vK4BnsX/R12dGsIPr2vC5kVA17TduwCf111r99qexhOrOJ1zq8Mvo93AQ5QMq8UyPjNriE+Kjznn/h6Kc+YYZoov144hgHNuA1CAP7fdwswahKr0tn4TR6hvjj+VFaf4hoRTOM45tx14hHgev2OA/zSzZfhTRyfgRwBy6tglPvGbWVMza5ZaBwYDi4DpQGqW6SjgubA+HTg7zFQdCGxMDb9muT2N5yVgsJm1DEOug0NZViozz+LH+GMIPr7hYfZtd6An8B4wB+gZZuvm4SfmTK/LNlcknCOcCHzknLs7rSonjmFF8eXKMTSztmbWIqw3AX6In8fwGnBG2Kzs8Usd1zOAV51zjorjjlQF8f077Y9Sw58DTz9+sfj5dM5d65zr4pzrhv95etU5N5IcOXbfqI0Zg3Fa8DOCF4TlQ+A3obw18AqwOLy2CuUGPIA/Z/dPoH/UMWSI6Qn8UOlO/F+e5+9NPMB5+EkphcC5UcdVRXx/De1fiP9P1zFt+9+E+D4GTk4rPwU/o3xJ6rhnwwIcix8WXAh8EJZTcuUYVhJfThxD4DBgfohjEXB9KO+B/+VfCPwNaBTKG4f3haG+R1VxZ2l8r4bjtwh4lJKZ/7H6+UxrWz4ls/pz4tilFt25T0REJEESP9QvIiKSJEr8IiIiCaLELyIikiBK/CIiIgmixC8iIpIgSvwislfMP03v6rB+k5n9MOo2iUjVGlS9iYhIaWl3MQPAOXd9DX1ufefcrpr4LBHJTD1+ESnHzP7bzP5tZjPN7Akzu9rMCsxsnJm9DlxeZvvJZnaGmZ1sZk+lleeb2f+G9fFmNtfSnuEeypeZ2fVm9iZwppkdaGYvhodmvWFmvcN2PzL/zPP5ZvYPM2tfN/8aIrlFPX4RKcXM+gOn45+a1wCYB6QeWtLCOXd82O7GDLvPBP5iZk2dc18B/wU8Gep+45xbZ2b1gVfM7DDn3MJQt805d2z43FeAi5xzi81sAPBn/D3T3wQGOuecmV2Af4LaVTUavEgCKPGLSFnHAs85574GSPXYgycz7+I554rN7EXgR2b2NP655teE6v9n/rHXDYCOwMH4275+87nhiX1HA3/zt3wHIPV41y7Ak+Ge8HnAp3sdoUiCKfGLSFmZHima8lU19n8SuAT/lLI5zrnN4UElVwPfc86tN7PJ+Pucl/3cevhnn/fN8Ln3A3c756abWT5wYzXaIiJl6By/iJT1Jr7H3jj0wE/dw/0LgH7AzykZIdgPn9w3hnPzJ2fa0Tm3CfjUzM4E/6Q3Mzs8VDcHVob1UZn2F5GqKfGLSCnOuTn4p+MtAP4OzAU27sH+u4AZ+OQ+I5QtwD/R7UNgEjC7ko8YCZxvZqknZg4N5TfiTwG8AaytfkQikk5P5xORcsxsX+fcFjPbB5gFjHbOzYu6XSLy7ekcv4hkMsHMDsafh5+ipC+SO9TjFxERSRCd4xcREUkQJX4REZEEUeIXERFJECV+ERGRBFHiFxERSRAlfhERkQT5/2NJewtndCUSAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fce72060748>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "GrLivArea_housingMedianAge = model.partial_plot(data=valid, cols=['GrLivArea'], plot=True, plot_stddev=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Simple function for two-dimensional partial dependence \n",
    "* Using H2O for speed and scalability"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     OverallQual  GrLivArea  partial_dependence\n",
      "0            2.0      438.0       124488.430637\n",
      "1            2.0      631.9       124488.430637\n",
      "2            2.0      825.8       129263.083759\n",
      "3            2.0     1019.7       129605.449116\n",
      "4            2.0     1213.6       139737.571464\n",
      "5            2.0     1407.5       141172.628910\n",
      "6            2.0     1601.4       155845.579325\n",
      "7            2.0     1795.3       159534.718566\n",
      "8            2.0     1989.2       162659.405194\n",
      "9            2.0     2183.1       167161.620196\n",
      "10           2.0     2377.0       168083.763179\n",
      "11           2.0     2570.9       167555.612517\n",
      "12           2.0     2764.8       175281.240802\n",
      "13           2.0     2958.7       178497.631146\n",
      "14           2.0     3152.6       178497.631146\n",
      "15           2.0     3346.5       178497.631146\n",
      "16           2.0     3540.4       178497.631146\n",
      "17           2.0     3734.3       178497.631146\n",
      "18           2.0     3928.2       178497.631146\n",
      "19           2.0     4122.1       178497.631146\n",
      "20           2.4      438.0       124488.430637\n",
      "21           2.4      631.9       124488.430637\n",
      "22           2.4      825.8       129263.083759\n",
      "23           2.4     1019.7       129605.449116\n",
      "24           2.4     1213.6       139737.571464\n",
      "25           2.4     1407.5       141172.628910\n",
      "26           2.4     1601.4       155845.579325\n",
      "27           2.4     1795.3       159534.718566\n",
      "28           2.4     1989.2       162659.405194\n",
      "29           2.4     2183.1       167161.620196\n",
      "..           ...        ...                 ...\n",
      "370          9.2     2377.0       240577.998252\n",
      "371          9.2     2570.9       242690.079911\n",
      "372          9.2     2764.8       250418.134714\n",
      "373          9.2     2958.7       253634.525058\n",
      "374          9.2     3152.6       253634.525058\n",
      "375          9.2     3346.5       253634.525058\n",
      "376          9.2     3540.4       253634.525058\n",
      "377          9.2     3734.3       253634.525058\n",
      "378          9.2     3928.2       253634.525058\n",
      "379          9.2     4122.1       253634.525058\n",
      "380          9.6      438.0       193693.873445\n",
      "381          9.6      631.9       193693.873445\n",
      "382          9.6      825.8       193738.936865\n",
      "383          9.6     1019.7       193918.551684\n",
      "384          9.6     1213.6       194847.438867\n",
      "385          9.6     1407.5       196412.869244\n",
      "386          9.6     1601.4       208569.678494\n",
      "387          9.6     1795.3       214140.240633\n",
      "388          9.6     1989.2       232124.568632\n",
      "389          9.6     2183.1       235720.714678\n",
      "390          9.6     2377.0       240319.736160\n",
      "391          9.6     2570.9       242431.817819\n",
      "392          9.6     2764.8       250159.872622\n",
      "393          9.6     2958.7       253376.262966\n",
      "394          9.6     3152.6       253376.262966\n",
      "395          9.6     3346.5       253376.262966\n",
      "396          9.6     3540.4       253376.262966\n",
      "397          9.6     3734.3       253376.262966\n",
      "398          9.6     3928.2       253376.262966\n",
      "399          9.6     4122.1       253376.262966\n",
      "\n",
      "[400 rows x 3 columns]\n"
     ]
    }
   ],
   "source": [
    "# manually calculate 2-D partial dependence\n",
    "\n",
    "def par_dep_2d(xs1, xs2, frame, model, resolution=20):\n",
    "    \n",
    "    \"\"\" Creates Pandas dataframe containing partial dependence for two variables.\n",
    "    \n",
    "    Args:\n",
    "        xs1: First variable for which to calculate partial dependence.\n",
    "        xs2: Second variable for which to calculate partial dependence.\n",
    "        frame: Data for which to calculate partial dependence.\n",
    "        model: Model for which to calculate partial dependence.\n",
    "        resolution: The number of points across the domain of xs for which to calculate partial dependence.\n",
    "    \n",
    "    Returns:\n",
    "        Pandas dataframe containing partial dependence values.\n",
    "        \n",
    "    \"\"\"\n",
    "    \n",
    "    # init empty Pandas frame w/ correct col names\n",
    "    par_dep_frame = pd.DataFrame(columns=[xs1, xs2, 'partial_dependence'])\n",
    "    \n",
    "    # cache original data \n",
    "    col_cache1 = frame[xs1]\n",
    "    col_cache2 = frame[xs2] \n",
    "    \n",
    "    # determine values at which to calculate partial dependency\n",
    "    # for xs1\n",
    "    min1_ = frame[xs1].min()\n",
    "    max1_ = frame[xs1].max()\n",
    "    by1 = float((max1_ - min1_)/resolution)\n",
    "    range1 = np.arange(min1_, max1_, by1)\n",
    "    \n",
    "    # determine values at which to calculate partial dependency\n",
    "    # for xs2\n",
    "    min2_ = frame[xs2].min()\n",
    "    max2_ = frame[xs2].max()\n",
    "    by2 = float((max2_ - min2_)/resolution)\n",
    "    range2 = np.arange(min2_, max2_, by2)  \n",
    "    \n",
    "    # calculate partial dependency \n",
    "    for j in range1:\n",
    "        for k in range2:\n",
    "            frame[xs1] = j\n",
    "            frame[xs2] = k\n",
    "            par_dep_i = model.predict(frame)\n",
    "            par_dep_j = par_dep_i.mean()[0]\n",
    "            std_j = model.predict(frame).sd()[0]\n",
    "            pos_std, neg_std = par_dep_j + std_j, par_dep_j - std_j\n",
    "            par_dep_frame = par_dep_frame.append({xs1:j,\n",
    "                                                  xs2:k,\n",
    "                                                  'partial_dependence': par_dep_j}, \n",
    "                                                  ignore_index=True)\n",
    "\n",
    "    # return input frame to original cached state    \n",
    "    frame[xs1] = col_cache1\n",
    "    frame[xs2] = col_cache2\n",
    "\n",
    "    return par_dep_frame\n",
    "\n",
    "# calculate 2-D partial dependence\n",
    "h2o.no_progress()\n",
    "resolution = 20\n",
    "par_dep_OverallQual_v_GrLivArea = par_dep_2d('OverallQual', \n",
    "                                             'GrLivArea', \n",
    "                                              valid, \n",
    "                                              model, \n",
    "                                              resolution=resolution)\n",
    "\n",
    "print(par_dep_OverallQual_v_GrLivArea)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Use matplotlib to plot two-dimensional partial dependence"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fce71e766a0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# create 2-D partial dependence plot\n",
    "\n",
    "# imports\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "from matplotlib import cm\n",
    "from matplotlib.ticker import LinearLocator, FormatStrFormatter\n",
    "\n",
    "# create 3-D grid \n",
    "new_shape = (resolution, resolution)\n",
    "x = np.asarray(par_dep_OverallQual_v_GrLivArea['OverallQual']).reshape(new_shape)\n",
    "y = np.asarray(par_dep_OverallQual_v_GrLivArea['GrLivArea']).reshape(new_shape)\n",
    "z = np.asarray(par_dep_OverallQual_v_GrLivArea['partial_dependence']).reshape(new_shape)\n",
    "\n",
    "\n",
    "fig = plt.figure(figsize=(8,6))\n",
    "ax = plt.axes(projection='3d')\n",
    "\n",
    "# set axes labels\n",
    "ax.set_title('Partial Dependence for Sale Price')\n",
    "ax.set_xlabel('OverallQual')\n",
    "ax.set_ylabel('GrLivArea')\n",
    "ax.set_zlabel('\\nSale Price')\n",
    "\n",
    "# axis decorators/details\n",
    "#ax.zaxis.set_major_locator(LinearLocator(10))\n",
    "#ax.zaxis.set_major_formatter(FormatStrFormatter('%.1f'))\n",
    "\n",
    "# surface\n",
    "surf = ax.plot_surface(x, y, z, \n",
    "                       cmap=cm.coolwarm, \n",
    "                       linewidth=0.05, \n",
    "                       rstride=1, \n",
    "                       cstride=1, \n",
    "                       antialiased=True)\n",
    "plt.tight_layout()\n",
    "\n",
    "_ = plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ICE Plots"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Helper function for finding quantile indices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_quantile_dict(y, id_, frame):\n",
    "\n",
    "    \"\"\" Returns the percentiles of a column y as the indices for another column id_.\n",
    "    \n",
    "    Args:\n",
    "        y: Column in which to find percentiles.\n",
    "        id_: Id column that stores indices for percentiles of y.\n",
    "        frame: H2OFrame containing y and id_. \n",
    "    \n",
    "    Returns:\n",
    "        Dictionary of percentile values and index column values.\n",
    "    \n",
    "    \"\"\"\n",
    "    \n",
    "    quantiles_df = frame.as_data_frame()\n",
    "    quantiles_df.sort_values(y, inplace=True)\n",
    "    quantiles_df.reset_index(inplace=True)\n",
    "    \n",
    "    percentiles_dict = {}\n",
    "    percentiles_dict[0] = quantiles_df.loc[0, id_]\n",
    "    percentiles_dict[99] = quantiles_df.loc[quantiles_df.shape[0]-1, id_]\n",
    "    inc = quantiles_df.shape[0]//10\n",
    "    \n",
    "    for i in range(1, 10):\n",
    "        percentiles_dict[i * 10] = quantiles_df.loc[i * inc,  id_]\n",
    "\n",
    "    return percentiles_dict\n",
    "\n",
    "quantile_dict = get_quantile_dict('SalePrice', 'Id', valid)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Find ICE values for each quantile"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>OverallQual</th>\n",
       "      <th>partial_dependence</th>\n",
       "      <th>Percentile_0</th>\n",
       "      <th>Percentile_10</th>\n",
       "      <th>Percentile_20</th>\n",
       "      <th>Percentile_30</th>\n",
       "      <th>Percentile_40</th>\n",
       "      <th>Percentile_50</th>\n",
       "      <th>Percentile_60</th>\n",
       "      <th>Percentile_70</th>\n",
       "      <th>Percentile_80</th>\n",
       "      <th>Percentile_90</th>\n",
       "      <th>Percentile_99</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2.0</td>\n",
       "      <td>150173.741376</td>\n",
       "      <td>86012.877179</td>\n",
       "      <td>86363.236328</td>\n",
       "      <td>111291.918290</td>\n",
       "      <td>138486.191881</td>\n",
       "      <td>102926.359104</td>\n",
       "      <td>159657.800499</td>\n",
       "      <td>141670.846493</td>\n",
       "      <td>167694.218287</td>\n",
       "      <td>176369.894265</td>\n",
       "      <td>214675.671179</td>\n",
       "      <td>288136.035995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2.4</td>\n",
       "      <td>150173.741376</td>\n",
       "      <td>86012.877179</td>\n",
       "      <td>86363.236328</td>\n",
       "      <td>111291.918290</td>\n",
       "      <td>138486.191881</td>\n",
       "      <td>102926.359104</td>\n",
       "      <td>159657.800499</td>\n",
       "      <td>141670.846493</td>\n",
       "      <td>167694.218287</td>\n",
       "      <td>176369.894265</td>\n",
       "      <td>214675.671179</td>\n",
       "      <td>288136.035995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2.8</td>\n",
       "      <td>150173.741376</td>\n",
       "      <td>86012.877179</td>\n",
       "      <td>86363.236328</td>\n",
       "      <td>111291.918290</td>\n",
       "      <td>138486.191881</td>\n",
       "      <td>102926.359104</td>\n",
       "      <td>159657.800499</td>\n",
       "      <td>141670.846493</td>\n",
       "      <td>167694.218287</td>\n",
       "      <td>176369.894265</td>\n",
       "      <td>214675.671179</td>\n",
       "      <td>288136.035995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3.2</td>\n",
       "      <td>150173.741376</td>\n",
       "      <td>86012.877179</td>\n",
       "      <td>86363.236328</td>\n",
       "      <td>111291.918290</td>\n",
       "      <td>138486.191881</td>\n",
       "      <td>102926.359104</td>\n",
       "      <td>159657.800499</td>\n",
       "      <td>141670.846493</td>\n",
       "      <td>167694.218287</td>\n",
       "      <td>176369.894265</td>\n",
       "      <td>214675.671179</td>\n",
       "      <td>288136.035995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3.6</td>\n",
       "      <td>150617.412859</td>\n",
       "      <td>86547.315395</td>\n",
       "      <td>86660.874251</td>\n",
       "      <td>111995.398790</td>\n",
       "      <td>139054.260855</td>\n",
       "      <td>102926.359104</td>\n",
       "      <td>159657.800499</td>\n",
       "      <td>142238.915467</td>\n",
       "      <td>168603.108166</td>\n",
       "      <td>177004.708046</td>\n",
       "      <td>214675.671179</td>\n",
       "      <td>288136.035995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>4.0</td>\n",
       "      <td>150617.412859</td>\n",
       "      <td>86547.315395</td>\n",
       "      <td>86660.874251</td>\n",
       "      <td>111995.398790</td>\n",
       "      <td>139054.260855</td>\n",
       "      <td>102926.359104</td>\n",
       "      <td>159657.800499</td>\n",
       "      <td>142238.915467</td>\n",
       "      <td>168603.108166</td>\n",
       "      <td>177004.708046</td>\n",
       "      <td>214675.671179</td>\n",
       "      <td>288136.035995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>4.4</td>\n",
       "      <td>150617.412859</td>\n",
       "      <td>86547.315395</td>\n",
       "      <td>86660.874251</td>\n",
       "      <td>111995.398790</td>\n",
       "      <td>139054.260855</td>\n",
       "      <td>102926.359104</td>\n",
       "      <td>159657.800499</td>\n",
       "      <td>142238.915467</td>\n",
       "      <td>168603.108166</td>\n",
       "      <td>177004.708046</td>\n",
       "      <td>214675.671179</td>\n",
       "      <td>288136.035995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>4.8</td>\n",
       "      <td>156187.307991</td>\n",
       "      <td>89149.412334</td>\n",
       "      <td>98467.360261</td>\n",
       "      <td>110878.728502</td>\n",
       "      <td>142834.024482</td>\n",
       "      <td>109400.880788</td>\n",
       "      <td>161961.615944</td>\n",
       "      <td>142393.818253</td>\n",
       "      <td>170891.782910</td>\n",
       "      <td>181253.347414</td>\n",
       "      <td>221551.648596</td>\n",
       "      <td>294828.177017</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>5.2</td>\n",
       "      <td>156187.307991</td>\n",
       "      <td>89149.412334</td>\n",
       "      <td>98467.360261</td>\n",
       "      <td>110878.728502</td>\n",
       "      <td>142834.024482</td>\n",
       "      <td>109400.880788</td>\n",
       "      <td>161961.615944</td>\n",
       "      <td>142393.818253</td>\n",
       "      <td>170891.782910</td>\n",
       "      <td>181253.347414</td>\n",
       "      <td>221551.648596</td>\n",
       "      <td>294828.177017</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>5.6</td>\n",
       "      <td>166655.393408</td>\n",
       "      <td>93781.748211</td>\n",
       "      <td>103386.196013</td>\n",
       "      <td>116844.643320</td>\n",
       "      <td>156824.541381</td>\n",
       "      <td>115167.118755</td>\n",
       "      <td>175861.408073</td>\n",
       "      <td>158738.778891</td>\n",
       "      <td>187710.938458</td>\n",
       "      <td>192376.604317</td>\n",
       "      <td>240482.091665</td>\n",
       "      <td>321885.095364</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>6.0</td>\n",
       "      <td>166655.393408</td>\n",
       "      <td>93781.748211</td>\n",
       "      <td>103386.196013</td>\n",
       "      <td>116844.643320</td>\n",
       "      <td>156824.541381</td>\n",
       "      <td>115167.118755</td>\n",
       "      <td>175861.408073</td>\n",
       "      <td>158738.778891</td>\n",
       "      <td>187710.938458</td>\n",
       "      <td>192376.604317</td>\n",
       "      <td>240482.091665</td>\n",
       "      <td>321885.095364</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>6.4</td>\n",
       "      <td>166655.393408</td>\n",
       "      <td>93781.748211</td>\n",
       "      <td>103386.196013</td>\n",
       "      <td>116844.643320</td>\n",
       "      <td>156824.541381</td>\n",
       "      <td>115167.118755</td>\n",
       "      <td>175861.408073</td>\n",
       "      <td>158738.778891</td>\n",
       "      <td>187710.938458</td>\n",
       "      <td>192376.604317</td>\n",
       "      <td>240482.091665</td>\n",
       "      <td>321885.095364</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>6.8</td>\n",
       "      <td>195692.834893</td>\n",
       "      <td>127689.452160</td>\n",
       "      <td>134461.469365</td>\n",
       "      <td>148242.083018</td>\n",
       "      <td>164831.598790</td>\n",
       "      <td>133408.663469</td>\n",
       "      <td>177823.944998</td>\n",
       "      <td>175072.589349</td>\n",
       "      <td>209086.398187</td>\n",
       "      <td>219914.470620</td>\n",
       "      <td>264947.901302</td>\n",
       "      <td>448130.934994</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>7.2</td>\n",
       "      <td>195692.834893</td>\n",
       "      <td>127689.452160</td>\n",
       "      <td>134461.469365</td>\n",
       "      <td>148242.083018</td>\n",
       "      <td>164831.598790</td>\n",
       "      <td>133408.663469</td>\n",
       "      <td>177823.944998</td>\n",
       "      <td>175072.589349</td>\n",
       "      <td>209086.398187</td>\n",
       "      <td>219914.470620</td>\n",
       "      <td>264947.901302</td>\n",
       "      <td>448130.934994</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>7.6</td>\n",
       "      <td>206924.036451</td>\n",
       "      <td>149187.773815</td>\n",
       "      <td>151946.542434</td>\n",
       "      <td>170223.877939</td>\n",
       "      <td>178729.903777</td>\n",
       "      <td>148769.848964</td>\n",
       "      <td>190368.515182</td>\n",
       "      <td>185353.316174</td>\n",
       "      <td>215395.879926</td>\n",
       "      <td>215658.870461</td>\n",
       "      <td>284826.200170</td>\n",
       "      <td>483161.565853</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>8.0</td>\n",
       "      <td>206924.036451</td>\n",
       "      <td>149187.773815</td>\n",
       "      <td>151946.542434</td>\n",
       "      <td>170223.877939</td>\n",
       "      <td>178729.903777</td>\n",
       "      <td>148769.848964</td>\n",
       "      <td>190368.515182</td>\n",
       "      <td>185353.316174</td>\n",
       "      <td>215395.879926</td>\n",
       "      <td>215658.870461</td>\n",
       "      <td>284826.200170</td>\n",
       "      <td>483161.565853</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>8.4</td>\n",
       "      <td>206924.036451</td>\n",
       "      <td>149187.773815</td>\n",
       "      <td>151946.542434</td>\n",
       "      <td>170223.877939</td>\n",
       "      <td>178729.903777</td>\n",
       "      <td>148769.848964</td>\n",
       "      <td>190368.515182</td>\n",
       "      <td>185353.316174</td>\n",
       "      <td>215395.879926</td>\n",
       "      <td>215658.870461</td>\n",
       "      <td>284826.200170</td>\n",
       "      <td>483161.565853</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>8.8</td>\n",
       "      <td>210674.282410</td>\n",
       "      <td>154778.887829</td>\n",
       "      <td>151946.542434</td>\n",
       "      <td>174454.171885</td>\n",
       "      <td>178729.903777</td>\n",
       "      <td>154360.962978</td>\n",
       "      <td>194598.809128</td>\n",
       "      <td>191403.724571</td>\n",
       "      <td>215827.639625</td>\n",
       "      <td>216118.164844</td>\n",
       "      <td>286568.233618</td>\n",
       "      <td>500087.868587</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>9.2</td>\n",
       "      <td>210674.282410</td>\n",
       "      <td>154778.887829</td>\n",
       "      <td>151946.542434</td>\n",
       "      <td>174454.171885</td>\n",
       "      <td>178729.903777</td>\n",
       "      <td>154360.962978</td>\n",
       "      <td>194598.809128</td>\n",
       "      <td>191403.724571</td>\n",
       "      <td>215827.639625</td>\n",
       "      <td>216118.164844</td>\n",
       "      <td>286568.233618</td>\n",
       "      <td>500087.868587</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>9.6</td>\n",
       "      <td>210386.983428</td>\n",
       "      <td>154553.436204</td>\n",
       "      <td>151479.015293</td>\n",
       "      <td>174094.228447</td>\n",
       "      <td>178404.621979</td>\n",
       "      <td>154121.818956</td>\n",
       "      <td>194326.717822</td>\n",
       "      <td>191101.555632</td>\n",
       "      <td>215902.188682</td>\n",
       "      <td>215557.560433</td>\n",
       "      <td>286198.554678</td>\n",
       "      <td>500087.868587</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    OverallQual  partial_dependence   Percentile_0  Percentile_10  \\\n",
       "0           2.0       150173.741376   86012.877179   86363.236328   \n",
       "1           2.4       150173.741376   86012.877179   86363.236328   \n",
       "2           2.8       150173.741376   86012.877179   86363.236328   \n",
       "3           3.2       150173.741376   86012.877179   86363.236328   \n",
       "4           3.6       150617.412859   86547.315395   86660.874251   \n",
       "5           4.0       150617.412859   86547.315395   86660.874251   \n",
       "6           4.4       150617.412859   86547.315395   86660.874251   \n",
       "7           4.8       156187.307991   89149.412334   98467.360261   \n",
       "8           5.2       156187.307991   89149.412334   98467.360261   \n",
       "9           5.6       166655.393408   93781.748211  103386.196013   \n",
       "10          6.0       166655.393408   93781.748211  103386.196013   \n",
       "11          6.4       166655.393408   93781.748211  103386.196013   \n",
       "12          6.8       195692.834893  127689.452160  134461.469365   \n",
       "13          7.2       195692.834893  127689.452160  134461.469365   \n",
       "14          7.6       206924.036451  149187.773815  151946.542434   \n",
       "15          8.0       206924.036451  149187.773815  151946.542434   \n",
       "16          8.4       206924.036451  149187.773815  151946.542434   \n",
       "17          8.8       210674.282410  154778.887829  151946.542434   \n",
       "18          9.2       210674.282410  154778.887829  151946.542434   \n",
       "19          9.6       210386.983428  154553.436204  151479.015293   \n",
       "\n",
       "    Percentile_20  Percentile_30  Percentile_40  Percentile_50  Percentile_60  \\\n",
       "0   111291.918290  138486.191881  102926.359104  159657.800499  141670.846493   \n",
       "1   111291.918290  138486.191881  102926.359104  159657.800499  141670.846493   \n",
       "2   111291.918290  138486.191881  102926.359104  159657.800499  141670.846493   \n",
       "3   111291.918290  138486.191881  102926.359104  159657.800499  141670.846493   \n",
       "4   111995.398790  139054.260855  102926.359104  159657.800499  142238.915467   \n",
       "5   111995.398790  139054.260855  102926.359104  159657.800499  142238.915467   \n",
       "6   111995.398790  139054.260855  102926.359104  159657.800499  142238.915467   \n",
       "7   110878.728502  142834.024482  109400.880788  161961.615944  142393.818253   \n",
       "8   110878.728502  142834.024482  109400.880788  161961.615944  142393.818253   \n",
       "9   116844.643320  156824.541381  115167.118755  175861.408073  158738.778891   \n",
       "10  116844.643320  156824.541381  115167.118755  175861.408073  158738.778891   \n",
       "11  116844.643320  156824.541381  115167.118755  175861.408073  158738.778891   \n",
       "12  148242.083018  164831.598790  133408.663469  177823.944998  175072.589349   \n",
       "13  148242.083018  164831.598790  133408.663469  177823.944998  175072.589349   \n",
       "14  170223.877939  178729.903777  148769.848964  190368.515182  185353.316174   \n",
       "15  170223.877939  178729.903777  148769.848964  190368.515182  185353.316174   \n",
       "16  170223.877939  178729.903777  148769.848964  190368.515182  185353.316174   \n",
       "17  174454.171885  178729.903777  154360.962978  194598.809128  191403.724571   \n",
       "18  174454.171885  178729.903777  154360.962978  194598.809128  191403.724571   \n",
       "19  174094.228447  178404.621979  154121.818956  194326.717822  191101.555632   \n",
       "\n",
       "    Percentile_70  Percentile_80  Percentile_90  Percentile_99  \n",
       "0   167694.218287  176369.894265  214675.671179  288136.035995  \n",
       "1   167694.218287  176369.894265  214675.671179  288136.035995  \n",
       "2   167694.218287  176369.894265  214675.671179  288136.035995  \n",
       "3   167694.218287  176369.894265  214675.671179  288136.035995  \n",
       "4   168603.108166  177004.708046  214675.671179  288136.035995  \n",
       "5   168603.108166  177004.708046  214675.671179  288136.035995  \n",
       "6   168603.108166  177004.708046  214675.671179  288136.035995  \n",
       "7   170891.782910  181253.347414  221551.648596  294828.177017  \n",
       "8   170891.782910  181253.347414  221551.648596  294828.177017  \n",
       "9   187710.938458  192376.604317  240482.091665  321885.095364  \n",
       "10  187710.938458  192376.604317  240482.091665  321885.095364  \n",
       "11  187710.938458  192376.604317  240482.091665  321885.095364  \n",
       "12  209086.398187  219914.470620  264947.901302  448130.934994  \n",
       "13  209086.398187  219914.470620  264947.901302  448130.934994  \n",
       "14  215395.879926  215658.870461  284826.200170  483161.565853  \n",
       "15  215395.879926  215658.870461  284826.200170  483161.565853  \n",
       "16  215395.879926  215658.870461  284826.200170  483161.565853  \n",
       "17  215827.639625  216118.164844  286568.233618  500087.868587  \n",
       "18  215827.639625  216118.164844  286568.233618  500087.868587  \n",
       "19  215902.188682  215557.560433  286198.554678  500087.868587  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bins = list(par_dep_OverallQual['OverallQual'])\n",
    "for i in sorted(quantile_dict.keys()):\n",
    "    col_name = 'Percentile_' + str(i)\n",
    "    par_dep_OverallQual[col_name] = par_dep('OverallQual', \n",
    "                                            valid[valid['Id'] == int(quantile_dict[i])], \n",
    "                                            model, \n",
    "                                            bins=bins)['partial_dependence']\n",
    "par_dep_OverallQual"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Plot partial dependence and ICE for each quantile"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fce71e82ba8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "par_dep_OverallQual.drop('partial_dependence', axis=1).plot(x='OverallQual', colormap='gnuplot', ax=ax)\n",
    "\n",
    "par_dep_OverallQual.plot(title='Partial Dependence and ICE for Sales Price',\n",
    "                         x='OverallQual', \n",
    "                         y='partial_dependence',\n",
    "                         style='r-', \n",
    "                         linewidth=3, \n",
    "                         ax=ax)\n",
    "\n",
    "_ = plt.legend(bbox_to_anchor=(1.05, 0),\n",
    "               loc=3, \n",
    "               borderaxespad=0.)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From this partial dependence and ICE plot, it can be seen that most individual percentiles of predicted sale price are well-represented by partial dependence, except for the most expensive houses, above the 90th percentile. Pricing for these homes appears to behave differently than the average behavior of other homes under the model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Shutdown H2O"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Are you sure you want to shutdown the H2O instance running at http://127.0.0.1:54321 (Y/N)? y\n",
      "H2O session _sid_a219 closed.\n"
     ]
    }
   ],
   "source": [
    "h2o.cluster().shutdown(prompt=True)"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}