dt_shap_lime.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## License \n",
    "\n",
    "Copyright 2019 Patrick Hall, Navdeep Gill, and the H2O.ai team\n",
    "\n",
    "Licensed under the Apache License, Version 2.0 (the \"License\");\n",
    "you may not use this file except in compliance with the License.\n",
    "You may obtain a copy of the License at\n",
    "\n",
    "    http://www.apache.org/licenses/LICENSE-2.0\n",
    "\n",
    "Unless required by applicable law or agreed to in writing, software\n",
    "distributed under the License is distributed on an \"AS IS\" BASIS,\n",
    "WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
    "See the License for the specific language governing permissions and\n",
    "limitations under the License."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**DISCLAIMER:** This notebook is not legal compliance advice."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Single Decision Tree Model, Shapley Feature Importance, and LIME Reason Codes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Python imports\n",
    "In general, NumPy and Pandas will be used for data manipulation purposes and H2o and scikit-learn will be used for modeling tasks."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/patrickh/anaconda3/lib/python3.6/site-packages/sklearn/ensemble/weight_boosting.py:29: DeprecationWarning: numpy.core.umath_tests is an internal NumPy module and should not be imported. It will be removed in a future NumPy release.\n",
      "  from numpy.core.umath_tests import inner1d\n"
     ]
    }
   ],
   "source": [
    "# sklearn for single DT\n",
    "import sklearn\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.metrics import roc_auc_score\n",
    "\n",
    "import h2o\n",
    "from h2o.estimators.glm import H2OGeneralizedLinearEstimator # for LIME\n",
    "\n",
    "import numpy as np   # array, vector, matrix calculations\n",
    "import pandas as pd  # DataFrame handling\n",
    "\n",
    "pd.options.display.max_columns = 999 # enable display of all columns in notebook\n",
    "\n",
    "import shap # Python Shapley value package\n",
    "\n",
    "# system packages for calling external graphviz processes\n",
    "import os\n",
    "import re\n",
    "import subprocess\n",
    "\n",
    "import operator # for sorting dictionaries\n",
    "\n",
    "# in-notebook display\n",
    "from IPython.display import Image\n",
    "from IPython.display import display\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load, explore, and prepare UCI credit card default data\n",
    "\n",
    "UCI credit card default data: https://archive.ics.uci.edu/ml/datasets/default+of+credit+card+clients\n",
    "\n",
    "The UCI credit card default data contains demographic and payment information about credit card customers in Taiwan in the year 2005. The data set contains 23 input variables: \n",
    "\n",
    "* **`LIMIT_BAL`**: Amount of given credit (NT dollar)\n",
    "* **`SEX`**: 1 = male; 2 = female\n",
    "* **`EDUCATION`**: 1 = graduate school; 2 = university; 3 = high school; 4 = others \n",
    "* **`MARRIAGE`**: 1 = married; 2 = single; 3 = others\n",
    "* **`AGE`**: Age in years \n",
    "* **`PAY_0`, `PAY_2` - `PAY_6`**: History of past payment; `PAY_0` = the repayment status in September, 2005; `PAY_2` = the repayment status in August, 2005; ...; `PAY_6` = the repayment status in April, 2005. The measurement scale for the repayment status is: -1 = pay duly; 1 = payment delay for one month; 2 = payment delay for two months; ...; 8 = payment delay for eight months; 9 = payment delay for nine months and above. \n",
    "* **`BILL_AMT1` - `BILL_AMT6`**: Amount of bill statement (NT dollar). `BILL_AMNT1` = amount of bill statement in September, 2005; `BILL_AMT2` = amount of bill statement in August, 2005; ...; `BILL_AMT6` = amount of bill statement in April, 2005. \n",
    "* **`PAY_AMT1` - `PAY_AMT6`**: Amount of previous payment (NT dollar). `PAY_AMT1` = amount paid in September, 2005; `PAY_AMT2` = amount paid in August, 2005; ...; `PAY_AMT6` = amount paid in April, 2005. \n",
    "\n",
    "These input variables are used to predict the target variable, whether or not a customer defaulted on their credit card bill in late 2005."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Import data and clean"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import XLS file\n",
    "path = 'default_of_credit_card_clients.xls'\n",
    "data = pd.read_excel(path,\n",
    "                     skiprows=1)\n",
    "\n",
    "# remove spaces from target column name \n",
    "data = data.rename(columns={'default payment next month': 'DEFAULT_NEXT_MONTH'}) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Assign modeling roles\n",
    "The shorthand name y is assigned to the prediction target. X is assigned to all other input variables in the credit card default data except the row indentifier, ID, and the demographic variables are dropped from the analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "y = DEFAULT_NEXT_MONTH\n",
      "X = ['PAY_0', 'PAY_2', 'PAY_3', 'PAY_4', 'PAY_5', 'PAY_6', 'BILL_AMT1', 'BILL_AMT2', 'BILL_AMT3', 'BILL_AMT4', 'BILL_AMT5', 'BILL_AMT6', 'PAY_AMT1', 'PAY_AMT2', 'PAY_AMT3', 'PAY_AMT4', 'PAY_AMT5', 'PAY_AMT6']\n"
     ]
    }
   ],
   "source": [
    "# assign target and inputs for decision tree\n",
    "y = 'DEFAULT_NEXT_MONTH'\n",
    "X = [name for name in data.columns if name not in [y, 'ID', 'AGE', 'EDUCATION', 'SEX', 'MARRIAGE', 'LIMIT_BAL']]\n",
    "print('y =', y)\n",
    "print('X =', X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Display descriptive statistics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>LIMIT_BAL</th>\n",
       "      <th>SEX</th>\n",
       "      <th>EDUCATION</th>\n",
       "      <th>MARRIAGE</th>\n",
       "      <th>AGE</th>\n",
       "      <th>PAY_0</th>\n",
       "      <th>PAY_2</th>\n",
       "      <th>PAY_3</th>\n",
       "      <th>PAY_4</th>\n",
       "      <th>PAY_5</th>\n",
       "      <th>PAY_6</th>\n",
       "      <th>BILL_AMT1</th>\n",
       "      <th>BILL_AMT2</th>\n",
       "      <th>BILL_AMT3</th>\n",
       "      <th>BILL_AMT4</th>\n",
       "      <th>BILL_AMT5</th>\n",
       "      <th>BILL_AMT6</th>\n",
       "      <th>PAY_AMT1</th>\n",
       "      <th>PAY_AMT2</th>\n",
       "      <th>PAY_AMT3</th>\n",
       "      <th>PAY_AMT4</th>\n",
       "      <th>PAY_AMT5</th>\n",
       "      <th>PAY_AMT6</th>\n",
       "      <th>DEFAULT_NEXT_MONTH</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>3.000000e+04</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>3.000000e+04</td>\n",
       "      <td>30000.00000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "      <td>30000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>15000.500000</td>\n",
       "      <td>167484.322667</td>\n",
       "      <td>1.603733</td>\n",
       "      <td>1.853133</td>\n",
       "      <td>1.551867</td>\n",
       "      <td>35.485500</td>\n",
       "      <td>-0.016700</td>\n",
       "      <td>-0.133767</td>\n",
       "      <td>-0.166200</td>\n",
       "      <td>-0.220667</td>\n",
       "      <td>-0.266200</td>\n",
       "      <td>-0.291100</td>\n",
       "      <td>51223.330900</td>\n",
       "      <td>49179.075167</td>\n",
       "      <td>4.701315e+04</td>\n",
       "      <td>43262.948967</td>\n",
       "      <td>40311.400967</td>\n",
       "      <td>38871.760400</td>\n",
       "      <td>5663.580500</td>\n",
       "      <td>5.921163e+03</td>\n",
       "      <td>5225.68150</td>\n",
       "      <td>4826.076867</td>\n",
       "      <td>4799.387633</td>\n",
       "      <td>5215.502567</td>\n",
       "      <td>0.221200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>8660.398374</td>\n",
       "      <td>129747.661567</td>\n",
       "      <td>0.489129</td>\n",
       "      <td>0.790349</td>\n",
       "      <td>0.521970</td>\n",
       "      <td>9.217904</td>\n",
       "      <td>1.123802</td>\n",
       "      <td>1.197186</td>\n",
       "      <td>1.196868</td>\n",
       "      <td>1.169139</td>\n",
       "      <td>1.133187</td>\n",
       "      <td>1.149988</td>\n",
       "      <td>73635.860576</td>\n",
       "      <td>71173.768783</td>\n",
       "      <td>6.934939e+04</td>\n",
       "      <td>64332.856134</td>\n",
       "      <td>60797.155770</td>\n",
       "      <td>59554.107537</td>\n",
       "      <td>16563.280354</td>\n",
       "      <td>2.304087e+04</td>\n",
       "      <td>17606.96147</td>\n",
       "      <td>15666.159744</td>\n",
       "      <td>15278.305679</td>\n",
       "      <td>17777.465775</td>\n",
       "      <td>0.415062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>21.000000</td>\n",
       "      <td>-2.000000</td>\n",
       "      <td>-2.000000</td>\n",
       "      <td>-2.000000</td>\n",
       "      <td>-2.000000</td>\n",
       "      <td>-2.000000</td>\n",
       "      <td>-2.000000</td>\n",
       "      <td>-165580.000000</td>\n",
       "      <td>-69777.000000</td>\n",
       "      <td>-1.572640e+05</td>\n",
       "      <td>-170000.000000</td>\n",
       "      <td>-81334.000000</td>\n",
       "      <td>-339603.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>7500.750000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>28.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>3558.750000</td>\n",
       "      <td>2984.750000</td>\n",
       "      <td>2.666250e+03</td>\n",
       "      <td>2326.750000</td>\n",
       "      <td>1763.000000</td>\n",
       "      <td>1256.000000</td>\n",
       "      <td>1000.000000</td>\n",
       "      <td>8.330000e+02</td>\n",
       "      <td>390.00000</td>\n",
       "      <td>296.000000</td>\n",
       "      <td>252.500000</td>\n",
       "      <td>117.750000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>15000.500000</td>\n",
       "      <td>140000.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>34.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>22381.500000</td>\n",
       "      <td>21200.000000</td>\n",
       "      <td>2.008850e+04</td>\n",
       "      <td>19052.000000</td>\n",
       "      <td>18104.500000</td>\n",
       "      <td>17071.000000</td>\n",
       "      <td>2100.000000</td>\n",
       "      <td>2.009000e+03</td>\n",
       "      <td>1800.00000</td>\n",
       "      <td>1500.000000</td>\n",
       "      <td>1500.000000</td>\n",
       "      <td>1500.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>22500.250000</td>\n",
       "      <td>240000.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>41.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>67091.000000</td>\n",
       "      <td>64006.250000</td>\n",
       "      <td>6.016475e+04</td>\n",
       "      <td>54506.000000</td>\n",
       "      <td>50190.500000</td>\n",
       "      <td>49198.250000</td>\n",
       "      <td>5006.000000</td>\n",
       "      <td>5.000000e+03</td>\n",
       "      <td>4505.00000</td>\n",
       "      <td>4013.250000</td>\n",
       "      <td>4031.500000</td>\n",
       "      <td>4000.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>30000.000000</td>\n",
       "      <td>1000000.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>79.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>964511.000000</td>\n",
       "      <td>983931.000000</td>\n",
       "      <td>1.664089e+06</td>\n",
       "      <td>891586.000000</td>\n",
       "      <td>927171.000000</td>\n",
       "      <td>961664.000000</td>\n",
       "      <td>873552.000000</td>\n",
       "      <td>1.684259e+06</td>\n",
       "      <td>896040.00000</td>\n",
       "      <td>621000.000000</td>\n",
       "      <td>426529.000000</td>\n",
       "      <td>528666.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 ID       LIMIT_BAL           SEX     EDUCATION      MARRIAGE  \\\n",
       "count  30000.000000    30000.000000  30000.000000  30000.000000  30000.000000   \n",
       "mean   15000.500000   167484.322667      1.603733      1.853133      1.551867   \n",
       "std     8660.398374   129747.661567      0.489129      0.790349      0.521970   \n",
       "min        1.000000    10000.000000      1.000000      0.000000      0.000000   \n",
       "25%     7500.750000    50000.000000      1.000000      1.000000      1.000000   \n",
       "50%    15000.500000   140000.000000      2.000000      2.000000      2.000000   \n",
       "75%    22500.250000   240000.000000      2.000000      2.000000      2.000000   \n",
       "max    30000.000000  1000000.000000      2.000000      6.000000      3.000000   \n",
       "\n",
       "                AGE         PAY_0         PAY_2         PAY_3         PAY_4  \\\n",
       "count  30000.000000  30000.000000  30000.000000  30000.000000  30000.000000   \n",
       "mean      35.485500     -0.016700     -0.133767     -0.166200     -0.220667   \n",
       "std        9.217904      1.123802      1.197186      1.196868      1.169139   \n",
       "min       21.000000     -2.000000     -2.000000     -2.000000     -2.000000   \n",
       "25%       28.000000     -1.000000     -1.000000     -1.000000     -1.000000   \n",
       "50%       34.000000      0.000000      0.000000      0.000000      0.000000   \n",
       "75%       41.000000      0.000000      0.000000      0.000000      0.000000   \n",
       "max       79.000000      8.000000      8.000000      8.000000      8.000000   \n",
       "\n",
       "              PAY_5         PAY_6      BILL_AMT1      BILL_AMT2     BILL_AMT3  \\\n",
       "count  30000.000000  30000.000000   30000.000000   30000.000000  3.000000e+04   \n",
       "mean      -0.266200     -0.291100   51223.330900   49179.075167  4.701315e+04   \n",
       "std        1.133187      1.149988   73635.860576   71173.768783  6.934939e+04   \n",
       "min       -2.000000     -2.000000 -165580.000000  -69777.000000 -1.572640e+05   \n",
       "25%       -1.000000     -1.000000    3558.750000    2984.750000  2.666250e+03   \n",
       "50%        0.000000      0.000000   22381.500000   21200.000000  2.008850e+04   \n",
       "75%        0.000000      0.000000   67091.000000   64006.250000  6.016475e+04   \n",
       "max        8.000000      8.000000  964511.000000  983931.000000  1.664089e+06   \n",
       "\n",
       "           BILL_AMT4      BILL_AMT5      BILL_AMT6       PAY_AMT1  \\\n",
       "count   30000.000000   30000.000000   30000.000000   30000.000000   \n",
       "mean    43262.948967   40311.400967   38871.760400    5663.580500   \n",
       "std     64332.856134   60797.155770   59554.107537   16563.280354   \n",
       "min   -170000.000000  -81334.000000 -339603.000000       0.000000   \n",
       "25%      2326.750000    1763.000000    1256.000000    1000.000000   \n",
       "50%     19052.000000   18104.500000   17071.000000    2100.000000   \n",
       "75%     54506.000000   50190.500000   49198.250000    5006.000000   \n",
       "max    891586.000000  927171.000000  961664.000000  873552.000000   \n",
       "\n",
       "           PAY_AMT2      PAY_AMT3       PAY_AMT4       PAY_AMT5  \\\n",
       "count  3.000000e+04   30000.00000   30000.000000   30000.000000   \n",
       "mean   5.921163e+03    5225.68150    4826.076867    4799.387633   \n",
       "std    2.304087e+04   17606.96147   15666.159744   15278.305679   \n",
       "min    0.000000e+00       0.00000       0.000000       0.000000   \n",
       "25%    8.330000e+02     390.00000     296.000000     252.500000   \n",
       "50%    2.009000e+03    1800.00000    1500.000000    1500.000000   \n",
       "75%    5.000000e+03    4505.00000    4013.250000    4031.500000   \n",
       "max    1.684259e+06  896040.00000  621000.000000  426529.000000   \n",
       "\n",
       "            PAY_AMT6  DEFAULT_NEXT_MONTH  \n",
       "count   30000.000000        30000.000000  \n",
       "mean     5215.502567            0.221200  \n",
       "std     17777.465775            0.415062  \n",
       "min         0.000000            0.000000  \n",
       "25%       117.750000            0.000000  \n",
       "50%      1500.000000            0.000000  \n",
       "75%      4000.000000            0.000000  \n",
       "max    528666.000000            1.000000  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Split data into training and test sets for early stopping"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train data rows = 20946, columns = 25\n",
      "Test data rows = 9054, columns = 25\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(12345) # set random seed for reproducibility\n",
    "split_ratio = 0.7     # 70%/30% train/test split\n",
    "\n",
    "# execute split\n",
    "split = np.random.rand(len(data)) < split_ratio\n",
    "train = data[split]\n",
    "test = data[~split]\n",
    "\n",
    "# summarize split\n",
    "print('Train data rows = %d, columns = %d' % (train.shape[0], train.shape[1]))\n",
    "print('Test data rows = %d, columns = %d' % (test.shape[0], test.shape[1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Train Single Decision Tree"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Train decision tree classifier\n",
    "Several different depths were assesed on validation data and these hyperparameters performed well enough. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "clf = DecisionTreeClassifier(random_state=12345, max_depth=3)\n",
    "clf.fit(train[X].values, train[y].values)\n",
    "sklearn.tree.export_graphviz(clf, 'model.gv', feature_names=X, proportion=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Test AUC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7391279135043209"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "roc_auc_score(test[y].values, clf.predict_proba(test[X].values)[:, 1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Create PNG from GraphViz dot file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Calling external process ...\n",
      "dot -Tpng model.gv -o model.png\n"
     ]
    }
   ],
   "source": [
    "# construct call to generate PNG from \n",
    "# graphviz representation of the tree\n",
    "png_args = str('dot -Tpng model.gv -o model.png')\n",
    "png_args = png_args.split()\n",
    "\n",
    "# call\n",
    "print('Calling external process ...')\n",
    "print(' '.join(png_args))\n",
    "_ = subprocess.call(png_args)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualize and Explain Single Tree"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Display tree\n",
    "This is the entire model, displayed as a directed graph."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# display in-notebook\n",
    "display(Image(('model.png')))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Display global Shapley variable importance\n",
    "Shapley values are a locally-accurate and globally consistent variable importance metric. Instead of relying on traditional single-value variable importance measures, local Shapley values for each input will be calculated and aggregated below to get a more holistic and consisent measurement for the global importance of each input variable."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 98%|===================| 17798/18108 [00:14<00:00]        "
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAh0AAAIMCAYAAABR6h7ZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzs3X+UXlWd5/t3ETtAzII4wShNgyiuhKWGpu/6ggKDRseAP6iFbVRkCoErSYWGgdUs7UvakIWX/NBp6KGdm44/yoRcSUJjhB6pQZoW2iCxSeP3BsI4MDAYsZOO1TRC2giICan7xzllPzw+lTzPU09Oqqj3a61a1Nlnn733yT/1Ye99zukaHBxEkiTpQDvkYA9AkiSND4YOSZJUCUOHJEmqhKFDkiRVwtAhSZIqYeiQJEmVMHRIkqRKGDokSVIlDB2SJKkSrznYA3i16e/vH+zu7j7Yw5AkqSpdzVZ0pkOSJFXC0CFJkiph6JAkSZUwdEiSpEoYOiRJUiUMHZIkqRKGDkmSVAlDhyRJqoShQ5IkVcLQIUmSKmHokCRJlTB0SJKkShg6JElSJQwdkiSpEoYOSZJUia7BwcGDPYZXlfdd9o/+g0qSRqW/W3HcgWi2q9mKznRIkqRKGDokSVIlDB2SJKkShg5JklQJQ4ckSaqEoUOSJFXiNQd7APsSERuA04DdwMvAVmBJZt5WU6cPmAvMysz7yrLDgc3AusxcXFP3GOAR4NLMXL+fvhcAVwJHAj8AejPzqY7dnCRJ48xYmOlYnJmTganALcCtETEdICKOAM4HngV6hy7IzBeBHmBBRJxS1u0CbgLuaCJwXARcBXwImAY8CXw7IsbCv5ckSaPSmPkjmpl7gBXABGBmWdwDvARcAcyJiKk19TcD1wFrImJSWecEitmL/ekFVmTmw5n5PLAAmEEx6yJJktowZkJHREwELqdYatlSFvcCa4H1wC7g4rrLrgd2AOuAJUBPZu5qorvfB/6/oYPM/AXw47JckiS1YSyEjoURsRPYDpwLzMnMJyPiVOBkYFVm7gZuBubVXpiZe4ELgdnA8szctL/OymWY1wL/WndqJ3DESG9GkqTxaiyEjqWZOSUzp2Xm6ZnZX5bPBx7KzIfL45XAjIiYVXtxZm4DnqPYQLpfmTkIPE+xgbTWFOAXbd6DJEnj3qh+emU45QbS84BDImKg5tQgRRjZMMIutgD/B/Dfa/o7gX9b1pEkSS0ak6EDuADYC5wEvFBTfg6wPCKOysxnRtD+14AvRsR/o3hyZRnwOPDACNqUJGlcG6uhoxfoy8yttYURsRpYRLGh9IZ2G8/M/zcifhe4m397T8dHyj0ikiSpDV2Dg4MHewyvKu+77B/9B5UkjUp/t+K4A9FsV7MVx+pMx6h11Qe30N3dfbCHIUnSqDMuQ0dELAKuHub07Mx074YkSR02LkNH+T2WxfutKEmSOmYsvKdDkiS9Chg6JElSJQwdkiSpEoYOSZJUCUOHJEmqhC8H6zBfDibpQDpAL3eSRqLpl4M50yFJkiph6JAkSZUwdEiSpEoYOiRJUiUMHZIkqRKj+tsrEbEBOA3YDbwMbAWWZOZtNXX6gLnArMy8ryw7HNgMrCu/szJU9xjgEeDSzFy/j37/L+B84C3Ai8AG4E8yc1sn70+SpPFkLMx0LM7MycBU4Bbg1oiYDhARR1CEg2eB3qELMvNFoAdYEBGnlHW7gJuAO/YVOEq/A1wOvAGYDvwa+HYnb0qSpPFmLIQOADJzD7ACmADMLIt7gJeAK4A5ETG1pv5m4DpgTURMKuucAFzZRF9LM/PvM/NXmfkL4M+APyhDjiRJasOYCR0RMZFi9mE3sKUs7gXWAuuBXcDFdZddD+wA1gFLgJ7M3NVG9/8BeKoMIJIkqQ1jIXQsjIidwHbgXGBOZj4ZEacCJwOrMnM3cDMwr/bCzNwLXAjMBpZn5qZWO4+IMykCyx+N7DYkSRrfRvVG0tLSzFzSoHw+8FBmPlwerwSuiohZmblhqFJmbouI5yg2kLYkImYBtwOXZObftDxySZL0G2MhdPyWcm/FecAhETFQc2qQIoxs6EAfH6JYlrkoM91EKknSCI2F5ZVGLgD2Au+gWGIZ+ukF/jAijhpJ4xHxCYonZc43cEiS1BljcqaDIlz0ZebW2sKIWA0sothQesMI2v9z4LXA+oioLZ+emTtG0K4kSeOWn7bvMD9tL+lA8tP2GoX8tL0kSRpdxuryyohExCLg6mFOz87MB9pt+6oPbqG7u7vdyyVJetUal6Gj/B7L4v1WlCRJHePyiiRJqoShQ5IkVcLQIUmSKmHokCRJlTB0SJKkShg6JElSJXwjaYf5RlKNZ74tUxqXfCOpJEkaXQwdkiSpEoYOSZJUCUOHJEmqhKFDkiRVoqUPvkXEBuA0YDfwMrAVWJKZt9XU6QPmArMy876y7HBgM7Cu/NjaUN1jgEeASzNzfRP9XwSsBq7JzKV15zYCZwBzMvP2mvIzgI3AjzPzrRHxOHBMeXoiRfD6VU1T04FPAJcD04A9wIPAn2Tmj/Y3RkmS1Fg7Mx2LM3MyMBW4Bbg1IqYDRMQRwPnAs0Dv0AWZ+SLQAyyIiFPKul3ATcAdzQSO0vyy7bkR0WjsjwHz6srmleVDY5mRmZPLe1gGbBg6Ln92AP3AuzLzSOBo4HvAnU2OUZIkNdD28kpm7gFWABOAmWVxD/AScAUwJyKm1tTfDFwHrImISWWdE4Arm+kvImZSzLJcCBwLnNWg2reAd0bEceU1RwIfoZgdaeXefpyZPy8Pu4C9wLHluCVJUhvaDh0RMZFiCWI3sKUs7gXWAuuBXcDFdZddD+wA1gFLgJ7M3NVkl73AQ5l5J3A3xaxHvRcoZl8uKY97gHuAp5vs4zci4j0RsZNi6eXPgC9m5guttiNJkgrthI6F5R/j7cC5FHsonoyIU4GTgVWZuRu4mbqljszcSzFTMRtYnpmbmumwnGH4FLCqLFoJnBMRRzeo3gd8OiImlP33tXqD5Vjvy8wpwL8DPkOxr0OSJLWpndCxNDOnZOa0zDw9M/vL8vkUMxEPl8crgRkRMav24szcBjxHsYG0WecBh1HMokCx5+I5/m1Go7b9h4EB4FpgCvDdFvr5LZn5HPBfgW8M7V2RJEmt68gjs+UG0vOAEyNiICIGgHuBQRovg7Sql2LvyGNl29uAIxl+Q+nXgGuAleXsykgdQvGky1s60JYkSeNSp97TcQHFZst3UCyxDP30An8YEUe123BEnAS8i2Ipp7btdwG/B5zd4LI1FBtNl7fZ5x+Vj/NSjn0F8Evgh+20J0mSWnxPxz70An2ZubW2MCJWA4soNpTe0Gbb84EHM/M7deUDEXF7ef6u2hPlI7r3tNkfwDuBReXTL7uAfwDeX/NEiyRJapGftu8wP22v8cxP20vjkp+2lyRJo0unlldGJCIWAVcPc3p2Zj5Q5XhG4qoPbqG7u/tgD0OSpFFnVISO8nssi/dbUZIkjVkur0iSpEoYOiRJUiUMHZIkqRKGDkmSVAlDhyRJqoShQ5IkVcI3knbYaH4jqW+LlCQdAL6RVJIkjS6GDkmSVAlDhyRJqoShQ5IkVcLQIUmSKjEqPvg2nIjYAJwG7AZeBrYCSzLztpo6fcBcYFZm3leWHQ5sBtaVH5MbqnsM8AhwaWau30e/S4AFwK9qir+UmQs7dGuSJI07Y2GmY3FmTgamArcAt0bEdICIOAI4H3gW6B26IDNfBHqABRFxSlm3C7gJuGNfgaPGPZk5uebHwCFJ0giMhdABQGbuAVYAE4CZZXEP8BJwBTAnIqbW1N8MXAesiYhJZZ0TgCurHLckSSqM6uWVWhExEbicYqllS1ncC6wF1gNfAi4G/rzmsuuBDwDrgPcBZ2Xmria7PCMingF+AXwXWJiZz4zwNiRJGrfGwkzHwojYCWwHzgXmZOaTEXEqcDKwKjN3AzcD82ovzMy9wIXAbGB5Zm5qss+/At4GvB54P3A88NcduBdJksatsTDTsTQzlzQonw88lJkPl8crgasiYlZmbhiqlJnbIuI5ig2kTcnMH9Ucbo2IXuCpiHhTZv609VuQJEljIXT8lnID6XnAIRExUHNqkCKMbOhwl3vL/zb9fnlJkvRKY2F5pZELKILAOyiWWIZ+eoE/jIijRtJ4RMyJiNeXv/8e8GXgwcx8aiTtSpI0no3JmQ6KcNGXmVtrCyNiNbCIYkPpDSNo/+PAVyLitcDPgbsp3gUiSZLa5KftO8xP20uSxhk/bS9JkkaXsbq8MiIRsQi4epjTszPzgXbbvuqDW+ju7m73ckmSXrXGZegov8eyeL8VJUlSx7i8IkmSKmHokCRJlTB0SJKkShg6JElSJQwdkiSpEoYOSZJUCd9I2mGj7Y2kvoVUknSA+UZSSZI0uhg6JElSJQwdkiSpEoYOSZJUCUOHJEmqxKj+4FtEbABOA3YDLwNbgSWZeVtNnT5gLjArM+8ryw4HNgPryo+7DdU9BngEuDQz1++n7wD+M/BO4NfAhsz8aOfuTpKk8WUszHQszszJwFTgFuDWiJgOEBFHAOcDzwK9Qxdk5otAD7AgIk4p63YBNwF3NBE43g7cC/wVMA14A7Csw/clSdK4MhZCBwCZuQdYAUwAZpbFPcBLwBXAnIiYWlN/M3AdsCYiJpV1TgCubKK7zwP9mdmXmS9k5u7MzI7djCRJ49CYCR0RMRG4nGKpZUtZ3AusBdYDu4CL6y67HtgBrAOWAD2ZuauJ7t4L/DIiNkbEzyPiHyLi/SO/C0mSxq+xEDoWRsROYDtwLjAnM5+MiFOBk4FVmbkbuBmYV3thZu4FLgRmA8szc9P+OiuXYaZSLs8AbwS+AtwREcd37K4kSRpnxkLoWJqZUzJzWmaenpn9Zfl84KHMfLg8XgnMiIhZtRdn5jbgOYoNpPuVmYPAL4HbMnNjubRyE8Um1rM6cD+SJI1Lo/rpleGUG0jPAw6JiIGaU4MUYWTDCLt4uGyr3qj6rookSWPJmAwdwAXAXuAk4IWa8nOA5RFxVGY+M4L2VwBfjYivAg9SLLUcD/ztCNqUJGlcG6uhoxfoy8yttYURsRpYRLGh9IZ2G8/MWyJiGvBNYArwGPDhzPxpu21KkjTe+Wn7DvPT9pKkccZP20uSpNFlrC6vjEhELAKuHub07Mx8oN22r/rgFrq7u9u9XJKkV61xGTrK77Es3m9FSZLUMS6vSJKkShg6JElSJQwdkiSpEoYOSZJUCUOHJEmqhKFDkiRVwjeSdtjBfCOpbx+VJB0EvpFUkiSNLoYOSZJUCUOHJEmqhKFDkiRVwtAhSZIqsd8PvkXEBuA0YHdZNAAsz8y/KM8/BVyTmWsi4njgJ8Cxmbm9QVuDwJmZubHVgUbEcWXb38/M99ad+zxwLfDlzLyspvwwYAfwOuDNwALggvL0IcDhwPM1Tc0HNgArgJOB44BPZeaaVscrSZJeqdmZjsWZOTkzJ1P80V4aEbMP4LgamQvsBGZFxPQG558APhkRk2rKPkYRkgDIzEtr7uOssmxyzc9aYC/wt8B/BH4rOEmSpPa0vLySmZuAR4GZnR9OYxExAbgE+ALwI6C3QbVtwCbgEzVl84C+VvrKzJ9l5l9m5g+Al9sbsSRJqtdS6IiIrog4AzgReODADKmhc4BpwM3AKuCiiDi0Qb0+iqBBRMygGOe3qxqkJEkaXrOhY2FE7KTY/7ARWAs8eMBG9dvmA3dm5j9TBI8jgI82qNcPvCUi3k4RPr4B/LqyUUqSpGE1GzqWZuaUzJwEHAu8jWLG4YCLiDcBZw/1l5nPAHdQBJFXyMw9wGrgcuBCWlxakSRJB047ezq2A9+k8UzDgTCXYpxfj4iBiBigCCHvKZdQ6vVRBJJHM/OJisYoSZL2Y7+PzNaLiDcCHwe27KPaoeXjqkNezsyhR24n1p3bm5kNl0Ai4jUUG0i/CHyp7vT3KDaUfqa2MDO3RsS7gZ/t92aGUTO+LuB3yuM95UyKJElqQ7OhY1FELCh/fx64D/jsPuo/WXd8N/CB8vd76849TrHhs5Fuinds3JiZT9eeiIgbgWUR8bn6i8onT0bixZrfV5U//zfw+RG2K0nSuOWn7TvMT9tLksYZP20vSZJGl5b3dHRaRJwJ3DXM6WWZuazK8YzUVR/cQnd398EehiRJo85BDx2ZeT8w+WCPQ5IkHVgur0iSpEoYOiRJUiUMHZIkqRKGDkmSVAlDhyRJqoQvB+uw4V4O5ou7JEmvUr4cTJIkjS6GDkmSVAlDhyRJqoShQ5IkVcLQIUmSKmHokCRJlWjpg28RsQE4DdgNvAxsBZZk5m01dfqAucCszLyvLDsc2Aysy8zFNXWPAR4BLs3M9U30fxGwGrgmM5fWndsInAHMyczba8rPADYCP87Mt0bE48Ax5emJFMHrVzVNTQfeASwATgZeBxydmQP7G58kSRpeOzMdizNzMjAVuAW4NSKmA0TEEcD5wLNA79AFmfki0AMsiIhTyrpdwE3AHc0EjtL8su25EdFo7I8B8+rK5pXlQ2OZkZmTy3tYBmwYOi5/dgC/pAg3Fzc5LkmStB9tL69k5h5gBTABmFkW9wAvAVcAcyJiak39zcB1wJqImFTWOQG4spn+ImImxSzLhcCxwFkNqn0LeGdEHFdecyTwEYoA0cq9/X1mfgN4tJXrJEnS8NoOHRExEbicYqllS1ncC6wF1gO7+O2ZguuBHcA6YAnQk5m7muyyF3goM+8E7qaY9aj3AsXsyyXlcQ9wD/B0k31IkqQDpJ3QsTAidgLbgXMp9lA8GRGnUuyBWJWZu4GbqVvqyMy9FDMVs4HlmbmpmQ7LmZFPAavKopXAORFxdIPqfcCnI2JC2X9fqzcoSZI6r53QsTQzp2TmtMw8PTP7y/L5FDMRD5fHK4EZETGr9uLM3AY8R7GBtFnnAYdRzKIA9JdtXFJfsex/ALgWmAJ8t4V+JEnSAdKRR2bLDaTnASdGxEBEDAD3AoM0XgZpVS/F3pHHyra3AUcy/IbSrwHXACvL2RVJknSQtfTI7D5cAOwFTqLYVzHkHGB5RByVmc+003BEnAS8C/gwxWO3Q44GfgicDdxVd9ka4CdAttnnIRSP004siw6NiMOAlzLTz/JKktSGToWOXqAvM7fWFkbEamARxYbSG9psez7wYGZ+p658ICJuL8+/InSUj+je02Z/AO/jlcsyT5X/PZPinR+SJKlFXYOD/o97J73vsn9s+A/6dyuOq3ookiRVoavZir4GXZIkVaJTyysjEhGLgKuHOT07Mx+ocjwjcdUHt9Dd3X2whyFJ0qgzKkJH+T2WxfutKEmSxiyXVyRJUiUMHZIkqRKGDkmSVAlDhyRJqoShQ5IkVcLQIUmSKmHokCRJlTB0SJKkShg6JElSJQwdkiSpEoYOSZJUCUOHJEmqRFMffIuIDcBpwO6yaABYnpl/UZ5/CrgmM9dExPHAT4BjM3N7g7YGgTMzc2Org42I48q2v5+Z760793ngWuDLmXlZTflhwA7gdcCbgQXABeXpQ4DDgedrmpoPPAd8FjgJmAD8CPhcZt7f6pglSVKhlZmOxZk5OTMnU/zRXhoRsw/QuIYzF9gJzIqI6Q3OPwF8MiIm1ZR9jCIkAZCZl9bcx1ll2eSan7UUAeX/Ad4KvB5YB9wVEccekLuSJGkcaGt5JTM3AY8CMzs7nOFFxATgEuALFDMPvQ2qbQM2AZ+oKZsH9LXSV2auzcy/zsydmbknM78M/BI4pa3BS5Kk1kNHRHRFxBnAicADnR/SsM4BpgE3A6uAiyLi0Ab1+iiCBhExg2Kc3x5JxxExEzgK+B8jaUeSpPGsldCxMCJ2Uux/2AisBR48IKNqbD5wZ2b+M0XwOAL4aIN6/cBbIuLtFOHjG8Cv2+00IqYBtwE3ZOb/brcdSZLGu1ZCx9LMnJKZk4BjgbdRzDgccBHxJuDsof4y8xngDoog8gqZuQdYDVwOXEiLSyt1/f4u8D3gb4E/bbcdSZLU/p6O7cA3aTzTcCDMpRjr1yNiICIGKELIe8ollHp9FIHk0cx8op0Oy6dw7gfuysz/lJmD7Q1dkiRBk4/M1ouINwIfB7bso9qh5eOqQ17OzKFHbifWndubmQ2XQCLiNRQbSL8IfKnu9PcoNpR+prYwM7dGxLuBn+33Zhr3eSJwD7A6M69ppw1JkvRKrYSORRGxoPz9eeA+indZDOfJuuO7gQ+Uv99bd+5xig2fjXRTPMJ6Y2Y+XXsiIm4ElkXE5+ovyswf7GNs+3M1cAzwxxHxxzXl88tHaiVJUou6BgddNeik/v7+we7u7oM9DEmSqtLVbEVfgy5JkirR1p6OTouIM4G7hjm9LDOXVTkeSZLUeaMidJTfNJl8sMchSZIOHJdXJElSJQwdkiSpEoYOSZJUCUOHJEmqhKFDkiRVwtAhSZIqYeiQJEmVMHRIkqRKGDokSVIlDB2SJKkShg5JklQJQ4ckSarEqPjg23AiYgNwGrAbeBnYCizJzNtq6vQBc4FZmXlfWXY4sBlYl5mLa+oeAzwCXJqZ6/fR7x8DlwPTgD3Ag8CfZOaPOnqDkiSNI2NhpmNxZk4GpgK3ALdGxHSAiDgCOB94FugduiAzXwR6gAURcUpZtwu4CbhjX4Gj1A+8KzOPBI4Gvgfc2dG7kiRpnBkLoQOAzNwDrAAmADPL4h7gJeAKYE5ETK2pvxm4DlgTEZPKOicAVzbR148z8+flYRewFzi2bEeSJLVhVC+v1IqIiRRLHruBLWVxL7AWWA98CbgY+POay64HPgCsA94HnJWZu5rs7z3At4EjgUHgi5n5wohvRJKkcWoszHQsjIidwHbgXGBOZj4ZEacCJwOrMnM3cDMwr/bCzNwLXAjMBpZn5qZmO83M+zJzCvDvgM9Q7OuQJEltGguhY2lmTsnMaZl5emb2l+XzgYcy8+HyeCUwIyJm1V6cmduA5yg2kLYsM58D/ivwjaG9JJIkqXVjZnmlVrmB9DzgkIgYqDk1SBFGNnS4y0OAicBbgCc63LYkSePCWJjpaOQCis2d76BYYhn66QX+MCKOGknjEfFH5eO1lG2tAH4J/HAk7UqSNJ6NyZkOinDRl5lbawsjYjWwiGJD6Q0jaP+dwKKIOBLYBfwD8P6aJ1okSVKLugYHBw/2GF5V+vv7B7u7uw/2MCRJqkpXsxXH6vKKJEkaY8bq8sqIRMQi4OphTs/OzAeqHI8kSePBuAwd5fdYFu+3oiRJ6hiXVyRJUiUMHZIkqRKGDkmSVAlDhyRJqoShQ5IkVcLQIUmSKmHokCRJlTB0SJKkShg6JElSJQwdkiSpEoYOSZJUCUOHJEmqxKj+4FtEbABOA3YDLwNbgSWZeVtNnT5gLjArM+8ryw4HNgPryo+7DdU9BngEuDQz1zc5hm8Bc4DTMnNTJ+5LkqTxaCzMdCzOzMnAVOAW4NaImA4QEUcA5wPPAr1DF2Tmi0APsCAiTinrdgE3AXe0EDjOA6Z08F4kSRq3xkLoACAz9wArgAnAzLK4B3gJuAKYExFTa+pvBq4D1kTEpLLOCcCVzfQXEa8HvgDM79Q9SJI0no2Z0BERE4HLKZZatpTFvcBaYD2wC7i47rLrgR3AOmAJ0JOZu5rscgVwI/DTEQ1ckiQBYyN0LIyIncB24FxgTmY+GRGnAicDqzJzN3AzMK/2wszcC1wIzAaWN7snIyI+DhwD/GXnbkOSpPFtLISOpZk5JTOnZebpmdlfls8HHsrMh8vjlcCMiJhVe3FmbgOeo9hAul/lEs1/AS4pQ4skSeqAUf30ynDKDaTnAYdExEDNqUGKMLJhBM3/AfBG4P6IqC3/m4hYnpnXjKBtSZLGrTEZOoALgL3AScALNeXnAMsj4qjMfKbNtu8H3lxzPAF4CvgU8P0225Qkadwbq6GjF+jLzK21hRGxGlhEsaH0hnYazsyXKPaPDLU59G/0L5n5r+20KUmSoGtwcPBgj+FVpb+/f7C7u/tgD0OSpKp0NVtxLGwklSRJrwJjdXllRCJiEXD1MKdnZ+YDVY5HkqTxYFyGjvJ7LIv3W1GSJHWMyyuSJKkShg5JklQJQ4ckSaqEoUOSJFXC0CFJkiph6JAkSZUwdEiSpEoYOiRJUiUMHZIkqRKGDkmSVAlDhyRJqkRL316JiA3AacBu4GVgK7AkM2+rqdMHzAVmZeZ9ZdnhwGZgXfndk6G6xwCPAJdm5vom+r8IWA1ck5lL685tBM4A5mTm7TXlZwAbgR9n5lsj4nHgmPL0RIrg9auapqZn5o6ICOA/A+8Efg1syMyP7m+MkiSpsXZmOhZn5mRgKnALcGtETAeIiCOA84Fngd6hCzLzRaAHWBARp5R1u4CbgDuaCRyl+WXbcyOi0dgfA+bVlc0ry4fGMiMzJ5f3sIwiTEyu+dkREW8H7gX+CpgGvKGsK0mS2tT28kpm7gFWABOAmWVxD/AScAUwJyKm1tTfDFwHrImISWWdE4Arm+kvImZSzLJcCBwLnNWg2reAd0bEceU1RwIfoZgdacXngf7M7MvMFzJzd2Zmi21IkqQabYeOiJgIXE6x1LKlLO4F1gLrgV3AxXWXXQ/sANYBS4CezNzVZJe9wEOZeSdwN8WsR70XKGZfLimPe4B7gKeb7GPIe4FfRsTGiPh5RPxDRLy/xTYkSVKNdkLHwojYCWwHzqXYQ/FkRJwKnAysyszdwM3ULXVk5l6KmYrZwPLM3NRMh+XMyKeAVWXRSuCciDi6QfU+4NMRMaHsv6+VmyuXfaZSLgcBbwS+AtwREce30pYkSfo37YSOpZk5JTOnZebpmdlfls+nmIl4uDxeCcyIiFm1F2fmNuA5ig2kzToPOIxiFgWgv2zjkvqKZf8DwLXAFOC7LfRDZg4CvwRuy8yN5dLKTRSbZhst6UiSpCZ05JHZcgPpecCJETEQEQMUGzEHabwM0qpeir0jj5VtbwOOZPgNpV8DrgFWlrMrrXqYYuz1GpVJkqQmtPTI7D5cAOwFTqLYVzHkHGB5RByVmc+003BEnAS8C/gwxWO3Q44GfgicDdxVd9ka4CdAu5s/VwBfjYivAg9SLLUcD/xtm+1JkjTudSp09AJ9mbm1tjAiVgNnL7pUAAAgAElEQVSLKDaU3tBm2/OBBzPzO3XlAxFxe3n+FaGjfET3njb7IzNviYhpwDcplmgeAz6cmT9tt01Jksa7rsFBVww6qb+/f7C7u/tgD0OSpKp0NVvR16BLkqRKdGp5ZUQiYhFw9TCnZ2fmA1WOR5Ikdd6oCB3l91gW77eiJEkas1xekSRJlTB0SJKkShg6JElSJQwdkiSpEoYOSZJUCUOHJEmqhKFDkiRVwtAhSZIqYeiQJEmVMHRIkqRKGDokSVIlDB2SJKkSLX3wLSI2AKcBu4GXga3Aksy8raZOHzAXmJWZ95VlhwObgXXlx92G6h4DPAJcmpnrm+j/ImA1cE1mLq07txE4A5iTmbfXlJ8BbAR+nJlvjYjHgWPK0xMpgtevapqaDlwGLKgr/1JmLtzfGCVJUmPtzHQszszJwFTgFuDWiJgOEBFHAOcDzwK9Qxdk5otAD7AgIk4p63YBNwF3NBM4SvPLtudGRKOxPwbMqyubV5YPjWVGZk4u72EZsGHouPzZUVa9p67cwCFJ0gi0vbySmXuAFcAEYGZZ3AO8BFwBzImIqTX1NwPXAWsiYlJZ5wTgymb6i4iZFLMsFwLHAmc1qPYt4J0RcVx5zZHARyhmRyRJ0kHUduiIiInA5RRLLVvK4l5gLbAe2AVcXHfZ9cAOYB2wBOjJzF1NdtkLPJSZdwJ3U8x61HuBYvblkvK4B7gHeLrJPmqdERHPRMTWiPhqRBzVRhuSJKnUTuhYGBE7ge3AuRR7KJ6MiFOBk4FVmbkbuJm6pY7M3EsxUzEbWJ6Zm5rpsJwZ+RSwqixaCZwTEUc3qN4HfDoiJpT997V6g8BfAW8DXg+8Hzge+Os22pEkSaV2QsfSzJySmdMy8/TM7C/L51PMRDxcHq8EZkTErNqLM3Mb8BzFBtJmnQccRjGLAtBftnFJfcWy/wHgWmAK8N0W+hlq40eZuS0zBzNzK8Usy7+PiDe12pYkSSp05JHZcgPpecCJETEQEQPAvcAgjZdBWtVLsXfksbLtbcCRDL+h9GvANcDKcnZlpIba6OpAW5IkjUstPTK7DxdQ/GE+iWJfxZBzgOURcVRmPtNOwxFxEvAu4MMUj90OORr4IXA2cFfdZWuAnwDZZp9zgO9n5r9ExO8BXwYezMyn2mlPkiR1LnT0An3lUsRvRMRqYBHFhtIb2mx7PsUf/O/UlQ9ExO3l+VeEjvIR3Xva7A/g48BXIuK1wM8pNq7OHUF7kiSNe12Dg4MHewyvKv39/YPd3d0HexiSJFWl6a0HvgZdkiRVolPLKyMSEYuAq4c5PTszH6hyPJIkqfNGRegov8eyeL8VJUnSmOXyiiRJqoShQ5IkVcLQIUmSKmHokCRJlTB0SJKkShg6JElSJQwdkiSpEoYOSZJUCUOHJEmqhKFDkiRVwtAhSZIqYeiQJEmVaOqDbxGxATgN2F0WDQDLM/MvyvNPAddk5pqIOB74CXBsZm5v0NYgcGZmbmx1sBFxXNn29zPzvXXnPg9cC3w5My+rKT8M2AG8DngzsAC4oDx9CHA48HxNU/Mzc23N9X8ErAAWZeaSVscsSZIKrcx0LM7MyZk5meKP9tKImH2AxjWcucBOYFZETG9w/gngkxExqabsYxQhCYDMvLTmPs4qyybX/NQGjjcBnwH+xwG4F0mSxpW2llcycxPwKDCzs8MZXkRMAC4BvgD8COhtUG0bsAn4RE3ZPKCvzW5XAguBZ9u8XpIklVoOHRHRFRFnACcCD3R+SMM6B5gG3AysAi6KiEMb1OujCBpExAyKcX671c4iYj7wfGbe2vaIJUnSb7QSOhZGxE6K/Q8bgbXAgwdkVI3NB+7MzH+mCB5HAB9tUK8feEtEvJ0ifHwD+HUrHZV7R64BLttfXUmS1JxWQsfSzJySmZOAY4G3Ucw4HHDl3oqzh/rLzGeAOyiCyCtk5h5gNXA5cCHtLa18HViSmf/U5pAlSVKddvd0bAe+SeOZhgNhLsVYvx4RAxExQBFC3lMuodTrowgkj2bmE230NxtYFhHPRMQzwBnAn0bE/W2OX5Kkca+pR2brRcQbgY8DW/ZR7dDycdUhL2fm0CO3E+vO7c3MhksgEfEaig2kXwS+VHf6exQbSj9TW5iZWyPi3cDP9nszjR1bd7weuB/48zbbkyRp3GsldCyKiAXl788D9wGf3Uf9J+uO7wY+UP5+b925xyk2fDbSTfGOjRsz8+naExFxI8WMxOfqL8rMH+xjbPtU/36RiHgJ+EW5n0SSJLWha3Bw8GCP4VWlv79/sLu7+2APQ5KkqnQ1W9HXoEuSpEq0taej0yLiTOCuYU4vy8xlVY5HkiR13qgIHZl5PzD5YI9DkiQdOC6vSJKkShg6JElSJQwdkiSpEoYOSZJUCUOHJEmqhKFDkiRVwtAhSZIqYeiQJEmVMHRIkqRKGDokSVIlDB2SJKkShg5JklSJpj74FhEbgNOA3WXRALA8M/+iPP8UcE1mromI44GfAMdm5vYGbQ0CZ2bmxlYHGxHHlW1/PzPfW3fu88C1wJcz87Ka8sOAHcDrgDcDC4ALytOHAIcDz9c0NR/4J+B7deWPZObprY5ZkiQVWvnK7OLMXAIQEe8C7o2I/5mZ3z0wQ2toLrATmBUR0zPzibrzTwCfjIjPZuYLZdnHKELS6wAy81LgUoCI+PfA/Zn5ii/cRsQs4OX6ckmS1L62llcycxPwKDCzs8MZXkRMAC4BvgD8COhtUG0bsAn4RE3ZPKDvgA9QkiTtU8uhIyK6IuIM4ETggc4PaVjnANOAm4FVwEURcWiDen0UQYOImEExzm+30d+EiNgWEQMRcWdE/H6b45YkSbQWOhZGxE6KfQ4bgbXAgwdkVI3NB+7MzH+mCB5HAB9tUK8feEtEvJ0ifHwD+HWLff0v4GSKPSAnAo8AfxcRv9vm2CVJGvdaCR1LM3NKZk4CjgXeRjHjcMBFxJuAs4f6y8xngDsogsgrZOYeYDVwOXAhbSytZOZAZm7JzD2ZuTMz/xR4Fvhg2zchSdI41+6eju3AN2k803AgzKUY69fL5Y4BihDynnIJpV4fRSB5tMFm03btBbo61JYkSeNOK0+v/EZEvBH4OLBlH9UOLR9XHfJyZg49cjux7tzezGy4BBIRr6HYQPpF4Et1p79HsaH0M7WFmbk1It4N/Gy/N9O4z/cB/whsBSYBnwXeANzdTnuSJKm10LEoIhaUvz8P3Efxx3g4T9Yd3w18oPz93rpzj1PsnWikm+Jx1xsz8+naExFxI7AsIj5Xf1Fm/mAfY9uf3wduAo6iuNfNwOzM3DaCNiVJGte6BgcHD/YYXlX6+/sHu7u7D/YwJEmqStNbD3wNuiRJqkRbezo6LSLOBO4a5vSyzFxW5XgkSVLnjYrQkZn3A75yXJKkVzGXVyRJUiUMHZIkqRKGDkmSVAlDhyRJqoShQ5IkVcLQIUmSKmHokCRJlTB0SJKkShg6JElSJQwdkiSpEoYOSZJUCUOHJEmqRFMffIuIDcBpwO6yaABYnpl/UZ5/CrgmM9dExPHAT4BjM3N7g7YGgTMzc2Org42I48q2v5+Z760793ngWuDLmXlZTflhwA7gdcCbgQXABeXpQ4DDgedrmpqfmWsjYhpwPXAO8DvAVuBDmbmj1XFLkqTWZjoWZ+bkzJxM8Ud7aUTMPkDjGs5cYCcwKyKmNzj/BPDJiJhUU/YxipAEQGZeWnMfZ5Vlk2t+1pZB5V7g18AMYArQA/zygNyVJEnjQFvLK5m5CXgUmNnZ4QwvIiYAlwBfAH4E9Daotg3YBHyipmwe0NdidxdRBI3LMvOZzNybmf8zM3/R+sglSRK0EToioisizgBOBB7o/JCGdQ4wDbgZWAVcFBGHNqjXRxE0iIgZFOP8dot9vRf438DqiPh5RPyviLiq7ZFLkqSWQsfCiNhJsf9hI7AWePCAjKqx+cCdmfnPFMHjCOCjDer1A2+JiLdThI9vUCyTtOIoiuDxIHA0xXLSwojoaXPskiSNe62EjqWZOSUzJwHHAm+jmHE44CLiTcDZQ/1l5jPAHRRB5BUycw+wGrgcuJDWl1YAdgH/lJlfysxfZ2YCa4Bz27oBSZLU9p6O7cA3aTzTcCDMpRjr1yNiICIGKELIe8ollHp9FIHk0cx8oo3+HgYGG5Q3KpMkSU1o6pHZehHxRuDjwJZ9VDu0fApkyMuZOfTI7cS6c3szs+ESSES8hmID6ReBL9Wd/h7FhtLP1BZm5taIeDfws/3eTGOrgasj4nLgK8A7KJ5e+U9ttidJ0rjXSuhYFBELyt+fB+4DPruP+k/WHd8NfKD8/d66c49TbPhspJviHRs3ZubTtSci4kZgWUR8rv6izPzBPsa2T5n504j4EHAj8GcU7/n4fGbe2m6bkiSNd12Dg64YdFJ/f/9gd3f3wR6GJElV6Wq2oq9BlyRJlWhrT0enRcSZwF3DnF6WmcuqHI8kSeq8URE6MvN+YPLBHockSTpwXF6RJEmVMHRIkqRKGDokSVIlDB2SJKkShg5JklQJQ4ckSaqEoUOSJFXC0CFJkiph6JAkSZUwdEiSpEoYOiRJUiUMHZIkqRItffAtIjYApwG7gZeBrcCSzLytpk4fMBeYlZn3lWWHA5uBdZm5uKbuMcAjwKWZub6J/i8CVgPXZObSunMbgTOAOZl5e035GcBG4MeZ+daIeBw4pjw9kSJ4/aqmqenABcD5wFuAF4ENwJ9k5rb9jVGSJDXWzkzH4sycDEwFbgFujYjpABFxBMUf62eB3qELMvNFoAdYEBGnlHW7gJuAO5oJHKX5ZdtzI6LR2B8D5tWVzSvLh8YyIzMnl/ewDNgwdFz+7AB+B7gceANFCPk18O0mxyhJkhpoe3klM/cAK4AJwMyyuAd4CbgCmBMRU2vqbwauA9ZExKSyzgnAlc30FxEzKWZZLgSOBc5qUO1bwDsj4rjymiOBj1DMjrRyb0sz8+8z81eZ+Qvgz4A/KEOVJElqQ9uhIyImUswG7Aa2lMW9wFpgPbALuLjusuuBHcA6YAnQk5m7muyyF3goM+8E7qaY9aj3AsXsyyXlcQ9wD/B0k30M5z8AT5UBRJIktaGd0LEwInYC24FzKfZQPBkRpwInA6syczdwM3VLHZm5l2KmYjawPDM3NdNhOTPyKWBVWbQSOCcijm5QvQ/4dERMKPvva/UG6/o+kyIg/dFI2pEkabxrJ3QszcwpmTktM0/PzP6yfD7FTMTD5fFKYEZEzKq9uNyM+RzFBtJmnQccRjGLAtBftnFJfcWy/wHgWmAK8N0W+nmFcuzfBi7JzL9ptx1JktShR2bLvQ7nASdGxEBEDAD3AoM0XgZpVS/F3pHHyra3AUcy/IbSrwHXACvL2ZWWRcSHgP8G/J+Z+c32hi1Jkoa09MjsPlwA7AVOothXMeQcYHlEHJWZz7TTcEScBLwL+DDFY7dDjgZ+CJwN3FV32RrgJ0C22ecnKJZlPpmZ9W1LkqQ2dCp09AJ9mbm1tjAiVgOLKDaU3tBm2/OBBzPzO3XlAxFxe3n+FcGgfET3njb7A/hz4LXA+oioLZ9ePlIrSZJa1DU4OHiwx/Cq0t/fP9jd3X2whyFJUlW6mq3oa9AlSVIlOrW8MiIRsQi4epjTszPzgSrHI0mSOm9UhI7yeyyL91tRkiSNWS6vSJKkShg6JElSJQwdkiSpEoYOSZJUCUOHJEmqhKFDkiRVwtAhSZIqYeiQJEmVMHRIkqRKGDokSVIlDB2SJKkShg5JklSJlj74FhEbgNOA3cDLwFZgSWbeVlOnD5gLzMrM+8qyw4HNwLry425DdY8BHgEuzcz1TfR/EbAauCYzl9ad2wicAczJzNtrys8ANgI/zsy3RsTjwDHl6YkUwetXNU1Nz8wdNdd/C5gDnJaZm/Y3RkmS1Fg7Mx2LM3MyMBW4Bbg1IqYDRMQRwPnAs0Dv0AWZ+SLQAyyIiFPKul3ATcAdzQSO0vyy7bkR0WjsjwHz6srmleVDY5mRmZPLe1gGbBg6Ln9qA8d5wJQmxyZJkvah7eWVzNwDrAAmADPL4h7gJeAKYE5ETK2pvxm4DlgTEZPKOicAVzbTX0TMpJhluRA4FjirQbVvAe+MiOPKa44EPkIxO9KSiHg98AWKoCNJkkao7dAREROByymWWraUxb3AWmA9sAu4uO6y64EdwDpgCdCTmbua7LIXeCgz7wTupnEYeIFi9uWS8rgHuAd4usk+aq0AbgR+2sa1kiSpTjuhY2FE7AS2A+dS7KF4MiJOBU4GVmXmbuBm6pY6MnMvxUzFbGB5s3skypmRTwGryqKVwDkRcXSD6n3ApyNiQtl/X6s3GBEfp9j38ZetXitJkhprJ3QszcwpmTktM0/PzP6yfD7FTMTD5fFKYEZEzKq9ODO3Ac9RbCBt1nnAYRSzKAD9ZRuX1Fcs+x8ArqXYj/HdFvqhXBL6L8AlZUiSJEkd0JFHZssNpOcBJ0bEQEQMAPcCg3RmT0Qvxd6Rx8q2twFHMvyG0q8B1wAr2wgOfwC8Ebg/Ip6hCDAAfxMRS9oavSRJau2R2X24ANgLnESxr2LIOcDyiDgqM59pp+GIOAl4F/BhisduhxwN/BA4G7ir7rI1wE+AbKPL+4E31xxPAJ6iWN75fhvtSZIkOhc6eoG+zNxaWxgRq4FFFBtKb2iz7fnAg5n5nbrygYi4vTz/itBRPqJ7TzudZeZLFPtVAIiIoX+jf8nMf22nTUmSBF2Dg4MHewyvKv39/YPd3d0HexiSJFWlq9mKvgZdkiRVolPLKyMSEYvg/2/vzKPtqKo8/G0SAsQYoSFRJIQkMosRycEWZYgDg4oKCjLIJFNwWCoOLY2AaUFmRWlBEIUwNYioQAAJiNIKS6GPoIIgkUAIU0yAJAZQIHD6j30uVIp777vvvbr1Hi+/b627blWdae8z1T5DVfHVFs7bxxh/V6c8QgghhKieQWF05O+xHNujRyGEEEK8atHyihBCCCFqQUaHEEIIIWpBRocQQgghakFGhxBCCCFqQUaHEEIIIWpBRocQQgghakFGhxBCCCFqQUaHEEIIIWpBRocQQgghakFGhxBCCCFqQUaHEEIIIWpBRocQQgghaqGjD76FEG4CtgKez5fmA9+LMX4nu88FjooxXhRCmAA8AKwbY3y4SVwJ2CbGeHNvhQ0hjM9x/ybG+O6S23Tg68D3Y4yfLlxfFXgUWAOYCBwB7JOdVwJWA54uRDUNmAd8F5gADAPmAMfFGH/WW5mFEEII4fRmpuPYGOOoGOMo/Kb9zRDC9l2SqxUHA4uBqSGEDZu4zwb2DCGMLFzbDTeSAIgxHlbQY4d8bVThdzFwL7ArsCawOvAF4KIQwiZd0UoIIYRYAejT8kqM8ffA3cBbqhWnNSGEYcBBwAnAXcChTbw9BPwe+Hjh2iHAOb1JK8a4IMb4YIwxAQa8iOfV+n0QXQghhBD0wegIIVgI4V3AxsDvqhepJTsDY4ELgXOB/UMIqzTxdw5uaBBC2AiX88q+JBhCWAw8C/wWuBW4vi/xCCGEEKJ3RsfX8k34aeBm4GLgtq5I1ZxpwDUxxr/jhsdo4KNN/M0EJoUQ3owbHxcAz/UlwRjj6sAofKnlWmBZX+IRQgghRO+Mjm/GGFePMY4E1gU2xWccuk4IYT1gx0Z6McbHgatwQ2Q5YozLgBnAZ4D96OXSSpP4no0xXgFsh+8pEUIIIUQf6OuejoeBy2g+09ANDsZl/WEIYX4IYT5uhGyXl1DKnIMbJHfHGGdXJMNwYIOK4hJCCCFWODp6ZLZMCOENwO7An9p4WyU/rtrghRhj45HbESW3F2OMTZdAQgjD8Q2kJ+KPsRb5Nb6h9EvFizHG+0MI2wKP9ahM8zQ/hj8Jcw+eR/sC7wFO6Ut8QgghhOid0XF0COGIfPw08L/Al9v4v690PgvYKR/fWHK7F9/w2YwP4e/YOC3GuKDoEEI4DTg+hHBkOVCM8ZY2svXE2riRsza+H+ReYK8Y4w39iFMIIYRYobGU0kDLMKSYOXNm+tCHPjTQYgghhBB1YZ161GvQhRBCCFELfdrTUTUhhG2AX7RwPj7GeHyd8gghhBCiegaF0RFj/C3+PgwhhBBCDFG0vCKEEEKIWpDRIYQQQohakNEhhBBCiFqQ0SGEEEKIWpDRIYQQQohakNEhhBBCiFqQ0SGEEEKIWpDRIYQQQohakNEhhBBCiFqQ0SGEEEKIWpDRIYQQQohakNEhhBBCiFqQ0SGEEEKIWpDRIYQQQohasJTSQMswpFhllVXueu655/410HJ0i+HDh6+1bNmyxwdajm4y1HUc6vrB0NdxqOsHQ1/HIabf4ymlnTrymVLSr8LflClT4kDLIP2k44qs34qg41DXb0XQcajr1+qn5RUhhBBC1IKMDiGEEELUgoyO6vnBQAvQZYa6fjD0dRzq+sHQ13Go6wdDX8ehrl9TtJFUCCGEELWgmQ4hhBBC1IKMDiGEEELUwvCBFuDVQghhQ+B8YE3gCWC/GOPfSn6GAacDOwEJODHG+MOe3AYDFei3A3A88Bbgv2OMX65R/I6oQMejgT2BF4DngSNjjLPq06A9Fej3SeBw4EVgGHBOjPH0+jTomf7qWPCzEXAHcOZgqqsVlOF04NPAo9n7LTHGz9Qjfc9UUX4hhI8DRwOW3d8XY/x7PRr0TAVleAEwueB9MrBLjPGqGsTvOprp6JyzgDNijBsCZwBnN/HzCWB9YANgK2B6CGFCB26Dgf7qdz9wMHBK90XtM/3V8TZgyxjjZOBA4MchhNW6LnXn9Fe/nwJvjTFuDrwT+FIIYXKTOAaS/urY6PDPBq7ourS9p9/6ARfEGDfPv0FjcGT6pV8IIQDTge1jjJsBWwNLui92r+iXjjHG/RrlB+wPLAIGzeCmv8jo6IAQwlhgC+CSfOkSYIsQwpiS1z3w0eGLMcaFeKe2ewduA0oV+sUY74sx/hFYVpPYvaIiHWfFGJ/J/v6Mj7TW7LrwHVCRfv+IMTZ2lo8EVsZHYYOCitohwBHA1cDsLovcKyrUb1BSkX6HA6fGGOcDxBiXxBgHzRugu1CGBwEXxxif7ZbMdSOjozPWBR6JMb4AkP8fzdeLjAceLJzPK/hp5zbQVKHfYKdqHfcD5sQYH+6CrH2hEv1CCB8OIfwl+zklxnhnV6XuHf3WMYTwVmBH4LSuS9t7qqqje4YQ/hxCuD6EsFU3Be4lVei3KTAphPCbEMLtIYSjQgjWZbl7Q2X9TAhhBLA3cG7XpB0AZHQI0UtCCNsBxwJ7DbQsVRNjvCrG+GZgQ2DfvPdhSBBCWBl/N8JhjZvCEOQsYGJeAjwFuDKEMChm4ypiGL7HYXtgO+D9wL4DKlH32AWYl2eQhwwyOjrjIWCdvBbcWBN+Y75eZB6wXuF8fMFPO7eBpgr9BjuV6JhHjhfhG7vu7arEvaPSMowxzsP3sOzcFWn7Rn91XBt4E3BtCGEu8AXgkBDCYHlJU7/LMMY4P8b4fD6+IV/frMtyd0pV/ejlMcZnY4xLgSuBt3dV6t5RZTs8kCE2ywEyOjoixrgA+CMvj2z3Au7Ia3FFfoJ3YivlNbxdgMs7cBtQKtJvUFOFjiGELYEfA7vFGG+vR/LOqEi/TRqeQghrAe8GBs3ySn91jDHOizGuFWOcEGOcAHwHX1c/tCYV2lJRGa7T8BRC2ByYAAwK47iifuZ/gB1CCJZnrt4L/Kn70ndGVX1pCGEcsA1wcfelrhc9Mts5hwHnhxCOwXcT7wcQQrgWOCbGGIELgX8HGo9HfSPG+EA+buc2GOiXfiGErYFLgdGAhRD2BA4aTI+U0v8yPBNYDTjbN9EDsO8g2vfQX/0OzY8+P49vkv1ejPH6OhXogP7qONjpr37HhxCm4I91P4fXz/l1KtAD/dXvUiAAd+OPds8CflSf+B1RRR3dH5gZY1xUn9j1oNegCyGEEKIWtLwihBBCiFqQ0SGEEEKIWpDRIYQQQohakNEhhBBCiFqQ0SGEEEKIWpDRITCzHc3st4XzqWY2dwBFqg0zm2FmlX3t18wmmFkqnI8xswfNbK0Owh5mZhdWJcurATPbxswWD7QcKyJmtk9v2nnVbUW0p1ttow/lfqKZHVtV+jI6VnDMzPDvUHy9B3+fMrO7zOwfZrbIzKKZ7VFwn2tm+zQJ94rr5szOcY0quU01s2RmT+Xfo2Z2npn9W/80HRhSSgvxFxr1lL+vAb6Bf0FzhSGl9NuU0uoDLUcrzGy6mf1yoOVYEehWXpvZTWZ2VNXxdpty2xjAungS8BkzW6dHnx0go0PsAIwAft3Kg5nthd80DwJeh7/W93D8xTd94d3AJPzlPs2+X/JCSmlUSmkU/unqrfC3R75aORf4pJmNbuNnH+DOlNKcmmRaDjMbZmbqD4QQy5FSWgT8AphWRXzqZGokj/qPMrNf51H8nWY22cz2MrP7zGyJmf3QzIYXwow3s8vNbL6ZPWZmPzCz1xbcjzez+3N8c8zsCwW3CXnWYF8zu9vMlprZ9Wa2dkGsXYBfpvZviXsn8JuU0q3J+We2wvv6tsppwHX4W/naVuSU0v34Z8jfVnYzs+E5T3YpXZ9hZufl4/ea2a15dmahmV1qZmNbpZfza+vC+VQzW1ZK88g8U7PYzG4xs9A8tpd0+BvwOPC+Nt52AW4oyfJ5M/trLrd5ZnaCmQ3LbqeY2RUl/1Oz39fk883MbFbWuxF+5ezWqBsHmdndwDPAWDPb08z+lGehHjOzsxvx5XBvMLOZua7OzuGTmU0o+Dkkz4otMbM7zGyHVko3yd8ZZnahmZ2b8/eR3D42N7P/y/r92szeWAgz18yOMbObczuIZrZlwb1tHTCzlXOZ3pvjn2Nmu5nP5B0JTLWXZ94mtdBju5zGklxm0wpuU81smZntkeNeYmaXFdtxk/j60ldMNrNfZT3vz+GHFdzfnvPmKTO7GTf8i2mONLNTzcrpAFUAAAraSURBVOwBM3vSzK4zs/VbydhE5jXN7ALzvmq+mZ1vhRlKK816FurguFZ5bWYHZH2/muvjAjP7VpN6PK4Q7wFmdl8+/h7+OvGjc5xNXwlvPotwo5mdlOvIE2b2RTNbL+fpUjP7g5ltUgjTr7ZSqOvnFOr6K+pNPm6bPyVdllsGq6jcb8D7qP6TUtKvph8wF3/t7SbAyviHw+bgX758Df7RnwXAJ7L/VYH78Gn31YA1gGuBcwtx7oPPPBjwHuCfwI7ZbQKQ8Jv2Wvgrym8BzimEvxX4XEnOqcDcwvnuwL+A4/BvHazeQrd9eroOjAGeBT6KGxIJmFJKe1nhfH382xHntsjTk4ErCuejgKeAbfL51sCW+Cv/3wD8Brik4H8G8MPCeQK2biPPN3OeTcK/eHkQblCsUczzJnLOBI5rUzf+Dny4dO1jwMRctm/LfqZlt03x11yPKfg/H/hRPh4LPIEbdSOAdYAIHFOqGzfmfBmR9Xk/8GZ8QLI+/rrpEwpp3Aj8NNelscBNOZ4J2f0QvM6+NcfxgVwe67fQu5y/M/A6/MEc/rAc/ipgHDAS+BXL1+G5+OfDp2Q9jgAWAqM7rAMnZT0n57weB0zObtNxo7xdu56YZT4gp/EO4Elg94KOCX9d9yjg9Xg/8LUK+4rX5fpxNLBKDnc/8JWC+xM5b0bk/JjP8u38YryveH3281/AX4GVm7WVJjJfh9fzNfLvGuCaNn3BhJwv41rldc7T54Ez8D7wTcBs4MhmcRTC3Fc4vwk4qocynJ7TOZiX28ELwC9LZXBDIUx/28oMvN58OMfx0SzDei3aRqv8ua907aVyqqLcs58p+Mz0iHb52Mmv1pvuiv7Lje4rhfMP5EpYvHFcBpyWj3cD5pTimILftIe1SONy4OR83GiQWxbcPwPcUTifDRxQimNqsVLmazsDP8M7thfw5ZjNSro9DSwu/V5k+Y7mP/DOstGR3Q6cXUo75bCLgAfwz3W/wtDJ/jfBb75j8/mBwOw2ZbAzsKBw/lIDzectjQ78hrQU2LYU550NHWltdFwMnNlGrueAqT3Un1OBywrntwKH5+PX5vx/Vz7/MvCrUviPkTuoQt3Ytoc0Pwvclo/H5TCTCu7vZfmO9C5gv1IcM2nR6dPc6CjeqEbm+HcvXPs0y9fhucCxhXPDv+K5d091IPt9CvhgC7/T6dnoOBK4pXTtBGBWqU4X2/kpwM/bxDmX3vUVe+NfKbWC+zTg3nz8iZwnRfdvkts5PihJwPiC+0rAEnJ7oI3RgQ98ErBB4dpG+draBZ36YnQ8C4wsXDuY3MbLcRTC9MXo+Evp2oImZbCowrYyg0Jdz9cWAh9p0TZa5U87o6Pf5Z6vbZD9jW2Xj5389MG3+nmscPwMvn9hYelaY9p1IjDeXrmDOeEjtkfM7HP46HIc3oGuhm9cbJXm04X4wW/s7fYaeIIpXY1bw5jZxvjHz642s4kp10p8FH5RMZwVdkmbmWVZL0opPZ8v/wg40cy+nFJamq+9kDrcXJhSusfMbsdnfL4NfBI4r5DmFOB4fOQ9Es+jUU2i6oS1ctiZVnhCBR8FjWse5CVG4wZUK15RDuZ7ab6Iz6oMx0chvy94OQ/4FL4R+OPAwymlW7LbROBdpbpj+CiuyNxSmtsDxwAb4yPmYXjnCz5bAt6JNXiwFN9E4AwzO71wbTjwMJ3zUn1NKT3j1eYV7aa8NDG3ECaZ2TxymfRQB8bgMwezeyFfmXV5ZdnOAT5SOC+383I7bEZv+op1gQcLbbEhw7r5eFwT96LME/P/n3N+N1i5EEc7Gn6Kcc4puD1G31mQUnqmcD6XnttbXyjL+Axt6l0FbaVZmp3Ui95QVbmP5uXBYL/Qno7BzYO4Rb966bdqSukRM3sXPjU8DVgr36hn4p1qp9yBT9V3TErpr/iNbj18GrVT3oNPQx7YWPfFp/JG4SO1vnIecEBeh3wHcEHB7VJ8NmXDlNJomm9cLfIUfhNq8MbC8eN4p/C+Unm8JqV0Yg/xbobndSuWKwczWxefzj0OHym+Dp9iLpbtpcCGZrYFPuI5r+D2ID4qKsr5uuSbc4u8WEhzBHBFjnd8zq+vFtJ8JP+PL4QvHjfSPbCU7qiU0qfa6F4FExoH2bgdz8uGTrs6sBC/mWzQIt4XW1wv8lAx/cykfL0uHgLWs+XvHEUZHmniPqFw3LghblAqu5EppUs6TL8c56SS21Jaty1onddjzWxkSe5G2TYGKn2Jt89U1FZ6SzM9ynkKy+tfVblvhs8EPddH2V9CRsfg5mpghPkmt9eas46Z7ZrdR+NLHQuBZGYfxNcZe8MV+LRfS8zsQDPb3fK7JvKmrcOAu1NKT/YirWn4evrGwOb5txl+szy0l3IXuRQ3Zk7H11wfKbiNxqcKl5rZeHxtsx1/APY3sxF5w9cXGw55tPBd4FQz2wDAzEaZv+ek3NG9RDaGxuDrw624guU3mo7C2+dC4HkzewewbzFASmkx8HPcMHkHvqejwQVAyGW3qpmtlDee7dRGhhH4iG1RSumfZrYpPmXcSO9hfKr6xFwfxwDlRxFPA6abb/w0M1vNzLbOs2Pd5EAz28J8g+FX8BmNa7JbyzqQy/RM4GTzjbdmvrFxcvYyH59tHNEm7UuAKWa2n/lG47fjdb3OT65fg5fdkbnuboTfBBsyXI3Xqa+Yb5zdAt+PBEBKaQE+Q3qm5UcjzWx1M9vVSo+1NyOl9ChwPfCtHG4N4FvAL1JKjdH8H4C9cpsZg+8/KdIqr1cCTsp1aRK+dHh+TvcJsqFr/gTWW/DZ1HK8HW+I7ZAq2kpvaZY/f8SNsp1zG98V2LbgXlW5b4/3Uf1GRscgJk8pvgcfAf8V7zhvxG/WALPwm8tt+Ch8N/wm1BtmAcvMbGobP4vwafx7zOxpfC/BYnxtvCPMnxbYBTg1pTS/+MNna95mPTwF0oqU0hJc7/fjj6cWORRfA16K70n5SQ/RfRbvoJ7E18xnlNy/DlwJXGlm/8A3+x1G+7Z0IDAjy9mKC4G35k6VlNI9hbQW4zfKZiPO83C9ZxU6d3K+vhvP87l4Gf6c0s71Iimlp/ByPtnMnsJnVspLdXvjN/SH8U3Jjfx8NsdxDr6597yc5jz85rJyG92r4Ae40bkI2APfo9HI757qwNfwsr4i+7mJl29SP8FH6vPNnzCYWApLSukBfL3/s/imvQuBo1NKl1WlXE9kXXfADde/83Lf8O3svhjfnLsHnkenA98vRXMIvmn7JjNbiu9V2h2fVu+EffD8uxfvrxYD+xXcj8IHSY/heXxpKXyrvH4Qr28P4H3PdXgda7A/3hctyfqWjb3TcAN8sZn9pUNd2lJFW+kDr8if5I/Yfx6v/08CO+GbVxty9rvczWx1vH6f1Ue5l8OWX+oRKyJ59HtkSmnbfD4Vv0lOGEi5Xo3k2ZEHUkqWz8fgT42E0np8s7CH4RtB923nbzBhZjvihtFqaYA6E/N9Q0eV9xOJVz9mdgBetlXPVNTOYGgrfcHMTsD3E1XygjVtJBWklK7DRw+iYrKhsV6Hfs+iotFEtzCzzfG15TvxTWjHAT9+NXWiQtTBUGkrKaX/rDI+La+IZszl1f0G0IFkMb45dqiyBr5E8RRwM/BnfHpXCLE8aitN0PKKEEIIIWpBMx1CCCGEqAUZHUIIIYSoBRkdQgghhKgFGR1CCCGEqAUZHUIIIYSohf8HTytfOzTZHOcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ff6a175ccf8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "explainer = shap.TreeExplainer(clf, test[X], model_output='probability', feature_dependence='independent')\n",
    "shap_values = explainer.shap_values(test[X].values)\n",
    "shap.summary_plot(shap_values[1], X, plot_type='bar', color='royalblue')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Find riskiest customer\n",
    "The riskiest customer in this data set is selected from all the customers who have a probability of default of ~0.77."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>4445</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>ID</th>\n",
       "      <td>14563.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LIMIT_BAL</th>\n",
       "      <td>240000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SEX</th>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EDUCATION</th>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MARRIAGE</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AGE</th>\n",
       "      <td>39.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAY_0</th>\n",
       "      <td>4.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAY_2</th>\n",
       "      <td>3.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAY_3</th>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAY_4</th>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAY_5</th>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAY_6</th>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BILL_AMT1</th>\n",
       "      <td>47739.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BILL_AMT2</th>\n",
       "      <td>46712.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BILL_AMT3</th>\n",
       "      <td>45660.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BILL_AMT4</th>\n",
       "      <td>48905.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BILL_AMT5</th>\n",
       "      <td>47993.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BILL_AMT6</th>\n",
       "      <td>52015.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAY_AMT1</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAY_AMT2</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAY_AMT3</th>\n",
       "      <td>4000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAY_AMT4</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAY_AMT5</th>\n",
       "      <td>5000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PAY_AMT6</th>\n",
       "      <td>2000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DEFAULT_NEXT_MONTH</th>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>p_DEFAULT_NEXT_MONTH</th>\n",
       "      <td>0.777108</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                               4445\n",
       "ID                     14563.000000\n",
       "LIMIT_BAL             240000.000000\n",
       "SEX                        2.000000\n",
       "EDUCATION                  2.000000\n",
       "MARRIAGE                   1.000000\n",
       "AGE                       39.000000\n",
       "PAY_0                      4.000000\n",
       "PAY_2                      3.000000\n",
       "PAY_3                      2.000000\n",
       "PAY_4                      2.000000\n",
       "PAY_5                      2.000000\n",
       "PAY_6                      2.000000\n",
       "BILL_AMT1              47739.000000\n",
       "BILL_AMT2              46712.000000\n",
       "BILL_AMT3              45660.000000\n",
       "BILL_AMT4              48905.000000\n",
       "BILL_AMT5              47993.000000\n",
       "BILL_AMT6              52015.000000\n",
       "PAY_AMT1                   0.000000\n",
       "PAY_AMT2                   0.000000\n",
       "PAY_AMT3                4000.000000\n",
       "PAY_AMT4                   0.000000\n",
       "PAY_AMT5                5000.000000\n",
       "PAY_AMT6                2000.000000\n",
       "DEFAULT_NEXT_MONTH         1.000000\n",
       "p_DEFAULT_NEXT_MONTH       0.777108"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.options.mode.chained_assignment = None\n",
    "yhat = 'p_DEFAULT_NEXT_MONTH'\n",
    "test[yhat] = clf.predict_proba(test[X].values)[:, 1]\n",
    "test.reset_index(inplace=True, drop=True)\n",
    "test.sort_values(by=yhat, inplace=True)\n",
    "id_ = test.iloc[-1, :].loc['ID']\n",
    "idx = test.iloc[-1, :].name\n",
    "pd.DataFrame(test.iloc[-1, :])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Display local Shapley values for the riskiest customer predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ff6a24af160>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "s_df = pd.DataFrame(shap_values[1][idx, :], index=X)\n",
    "s_df.sort_values(by=0, inplace=True)\n",
    "_ = s_df[s_df[0] != 0].plot(kind='barh' , color='royalblue', legend=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The Shapley values combine the customer's data values, the rules of the displayed decision tree, and crucially, the rules of many similar perturbed decision trees to create locally accurate and globally consistent feature importance values for each customer in the dataset. Note that these values consider variables *not* on the customer's decision path. This is because Shapley values take other encodings of the signal in the dataset into account when creating local feature importance. Shapley values are one the few explanation methods that account for the so-called Rashomon effect in machine learning."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Check Shapley sum"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It can be seen that the Shapley values are locally accurate. i.e. they sum to the model prediction for this customer. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7771084197639089"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_df.sum()[0] + explainer.expected_value[1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Approximate LIME trends\n",
    "LIME uses linear models fit to a more complex machine learning function to explain that machine learning function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Checking whether there is an H2O instance running at http://localhost:54321 . connected.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div style=\"overflow:auto\"><table style=\"width:50%\"><tr><td>H2O cluster uptime:</td>\n",
       "<td>1 hour 13 mins</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>3.26.0.3</td></tr>\n",
       "<tr><td>H2O cluster version age:</td>\n",
       "<td>15 days </td></tr>\n",
       "<tr><td>H2O cluster name:</td>\n",
       "<td>H2O_from_python_patrickh_s01bpc</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>1.515 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>locked, healthy</td></tr>\n",
       "<tr><td>H2O connection url:</td>\n",
       "<td>http://localhost: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>Amazon S3, 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:         1 hour 13 mins\n",
       "H2O cluster timezone:       America/New_York\n",
       "H2O data parsing timezone:  UTC\n",
       "H2O cluster version:        3.26.0.3\n",
       "H2O cluster version age:    15 days\n",
       "H2O cluster name:           H2O_from_python_patrickh_s01bpc\n",
       "H2O cluster total nodes:    1\n",
       "H2O cluster free memory:    1.515 Gb\n",
       "H2O cluster total cores:    8\n",
       "H2O cluster allowed cores:  8\n",
       "H2O cluster status:         locked, healthy\n",
       "H2O connection url:         http://localhost:54321\n",
       "H2O connection proxy:\n",
       "H2O internal security:      False\n",
       "H2O API Extensions:         Amazon S3, XGBoost, Algos, AutoML, Core V3, Core V4\n",
       "Python version:             3.6.4 final\n",
       "--------------------------  ---------------------------------------------------"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Parse progress: |█████████████████████████████████████████████████████████| 100%\n",
      "glm Model Build progress: |███████████████████████████████████████████████| 100%\n",
      "\n",
      "Local GLM Coefficients:\n",
      "PAY_6.-1:                 -0.18178449655001785\n",
      "PAY_2.-1:                 -0.14299724067347283\n",
      "PAY_6.-2:                 -0.12477348565333289\n",
      "PAY_6.0:                  -0.11555612541709302\n",
      "PAY_5.-1:                 -0.02764077237891427\n",
      "PAY_5.-2:                 -0.019940748847883594\n",
      "PAY_2.2:                  -0.0024667856635693115\n",
      "PAY_0.3:                  -0.002358539123788783\n",
      "PAY_5.2:                  -0.0005538556971602666\n",
      "PAY_AMT3:                 -5.86716289798128e-07\n",
      "BILL_AMT1:                3.433893789444963e-08\n",
      "PAY_AMT1:                 4.806161776083594e-07\n",
      "PAY_0.4:                  0.0008928696129156811\n",
      "PAY_2.4:                  0.0010890792598825059\n",
      "PAY_5.3:                  0.0017246840049273081\n",
      "PAY_6.2:                  0.0035605853689740837\n",
      "PAY_0.5:                  0.004290344886079893\n",
      "PAY_6.3:                  0.006104123002309118\n",
      "PAY_2.3:                  0.006493682999310938\n",
      "PAY_6.4:                  0.007078038951467809\n",
      "PAY_2.0:                  0.008450709441869802\n",
      "PAY_5.7:                  0.008767743869587124\n",
      "Intercept:                0.7699043700639103\n",
      "\n",
      "Local GLM R-square:\n",
      "0.732572\n"
     ]
    }
   ],
   "source": [
    "# select riskiest customers, those who are more than 1 month late on their most recent payment\n",
    "local = test[(test['PAY_0'] > 1.5)] \n",
    "\n",
    "# variables with non-zero Global Shapley importance\n",
    "local_X = s_df[s_df[0] != 0].index\n",
    "\n",
    "h2o.init() # use h2o to train a linear model in the region of interest\n",
    "\n",
    "# convert data into H2o frame\n",
    "# treat categorical variables as such\n",
    "local = h2o.H2OFrame(local)\n",
    "local['PAY_0'] = local['PAY_0'].asfactor()\n",
    "local['PAY_2'] = local['PAY_2'].asfactor()\n",
    "local['PAY_5'] = local['PAY_5'].asfactor()\n",
    "local['PAY_6'] = local['PAY_6'].asfactor()\n",
    "\n",
    "# initialize\n",
    "local_glm = H2OGeneralizedLinearEstimator(lambda_search=True, seed=12345)\n",
    "\n",
    "# train \n",
    "local_glm.train(x=list(local_X), y=yhat, training_frame=local)\n",
    "\n",
    "# coefs\n",
    "print('\\nLocal GLM Coefficients:')\n",
    "for c_name, c_val in sorted(local_glm.coef().items(), key=operator.itemgetter(1)):\n",
    "    if c_val != 0.0:\n",
    "        print('%s %s' % (str(c_name + ':').ljust(25), c_val))\n",
    "        \n",
    "# r2\n",
    "print('\\nLocal GLM R-square:\\n%.6f' % local_glm.r2())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Because the intercept of these LIMEs is logical and the LIMEs have a good R<sup>2</sup>, this information can be taken as a rough, linear description of the tree model's response surface in the region of interest, i.e. an average description for risky customers under the model.    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Assess LIME prediction accuracy for row of interest\n",
    "It can also be seen that the LIMES have very good local accuracy for the riskiest customer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "glm prediction progress: |████████████████████████████████████████████████| 100%\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead>\n",
       "<tr><th style=\"text-align: right;\">  predict</th></tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr><td style=\"text-align: right;\">  0.77959</td></tr>\n",
       "</tbody>\n",
       "</table>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": []
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "local_glm.predict(local[local['ID'] == id_])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Visually assess LIME model fit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "glm prediction progress: |████████████████████████████████████████████████| 100%\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7ff6a45bb198>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# ranked predictions plot\n",
    "pred_frame = local_glm.predict(local).cbind(local)\\\n",
    "                       .as_data_frame()[['predict', yhat]]\n",
    "pred_frame.columns = ['LIME Preds.', 'DT Preds.']\n",
    "pred_frame.sort_values(by='DT Preds.', inplace=True)\n",
    "pred_frame.reset_index(inplace=True, drop=True)\n",
    "_ = pred_frame.plot(title='DT vs. LIME Ranked Predictions Plot')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This ranked predictions plot shows that the LIMEs follows the decision tree predictions for the most part and provides more evidence that the LIMES have decent fidelity to the decision tree model."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Conclusion**: Pairing a directly interpretable model and Shapley values enabled visualization of an entire predictive model and accurate local feature importance values for any individual in the model. Combining Shapley values and LIME allows for comparison of the average behavior of customers in a region of interest and anyone single customer in the dataset."
   ]
  }
 ],
 "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
}