Explain Your Model with Microsoft's InterpretML-Github.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Explain Your Model with Microsoft's InterpretML\n",
    "* pip install -U interpret"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from interpret.glassbox import ExplainableBoostingClassifier\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "np.random.seed(0)\n",
    "import matplotlib.pyplot as plt\n",
    "df = pd.read_csv('winequality-red.csv') \n",
    "df['quality'] = df['quality'].astype(int)\n",
    "\n",
    "Y = df['quality']\n",
    "X =  df[['fixed acidity', 'volatile acidity', 'citric acid', 'residual sugar',\n",
    "       'chlorides', 'free sulfur dioxide', 'total sulfur dioxide', 'density',\n",
    "       'pH', 'sulphates', 'alcohol']]\n",
    "X_featurenames = X.columns\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size = 0.2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## (A) Explore the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<!-- http://127.0.0.1:7517/112311418832/ -->\n",
       "<iframe src=\"http://127.0.0.1:7517/112311418832/\" width=100% height=800 frameBorder=\"0\"></iframe>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from interpret import show\n",
    "from interpret.data import Marginal\n",
    "\n",
    "marginal = Marginal().explain_data(X_train, Y_train, name = 'Train Data')\n",
    "show(marginal)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## (B) Train the Explainable Boosting Machine (EBM)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ExplainableBoostingRegressor(binning_strategy='uniform', data_n_episodes=2000,\n",
       "                             early_stopping_run_length=50,\n",
       "                             early_stopping_tolerance=1e-05,\n",
       "                             feature_names=['fixed acidity', 'volatile acidity',\n",
       "                                            'citric acid', 'residual sugar',\n",
       "                                            'chlorides', 'free sulfur dioxide',\n",
       "                                            'total sulfur dioxide', 'density',\n",
       "                                            'pH', 'sulphates', 'alcohol'],\n",
       "                             feature_step_n_inner_bags=0,\n",
       "                             feature_types=['c...ntinuous',\n",
       "                                            'continuous', 'continuous',\n",
       "                                            'continuous', 'continuous',\n",
       "                                            'continuous', 'continuous',\n",
       "                                            'continuous', 'continuous',\n",
       "                                            'continuous'],\n",
       "                             holdout_size=0.15, holdout_split=0.15,\n",
       "                             interactions=0, learning_rate=0.01,\n",
       "                             main_attr='all', max_tree_splits=2,\n",
       "                             min_cases_for_splits=2, n_estimators=16, n_jobs=-2,\n",
       "                             random_state=1234, schema=None, scoring=None,\n",
       "                             training_step_episodes=1)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from interpret.glassbox import ExplainableBoostingRegressor, LinearRegression, RegressionTree\n",
    "\n",
    "lr = LinearRegression(random_state=seed)\n",
    "lr.fit(X_train, Y_train)\n",
    "\n",
    "rt = RegressionTree(random_state=seed)\n",
    "rt.fit(X_train, Y_train)\n",
    "\n",
    "ebm = ExplainableBoostingRegressor(random_state=seed)\n",
    "ebm.fit(X_train, Y_train)  \n",
    "# For Classifier, use ebm = ExplainableBoostingClassifier()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## (C) How Does the EBM Model Perform?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<!-- http://127.0.0.1:7517/112288704656/ -->\n",
       "<iframe src=\"http://127.0.0.1:7517/112288704656/\" width=100% height=800 frameBorder=\"0\"></iframe>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<!-- http://127.0.0.1:7517/112288706128/ -->\n",
       "<iframe src=\"http://127.0.0.1:7517/112288706128/\" width=100% height=800 frameBorder=\"0\"></iframe>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<!-- http://127.0.0.1:7517/112311510544/ -->\n",
       "<iframe src=\"http://127.0.0.1:7517/112311510544/\" width=100% height=800 frameBorder=\"0\"></iframe>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from interpret import show\n",
    "from interpret.perf import RegressionPerf\n",
    "\n",
    "ebm_perf = RegressionPerf(ebm.predict).explain_perf(X_test, Y_test, name='EBM')\n",
    "lr_perf = RegressionPerf(lr.predict).explain_perf(X_test, Y_test, name='Linear Regression')\n",
    "rt_perf = RegressionPerf(rt.predict).explain_perf(X_test, Y_test, name='Regression Tree')\n",
    "show(ebm_perf)\n",
    "show(lr_perf)\n",
    "show(rt_perf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## (D) Global Interpretability - What the Model Says for All Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<!-- http://127.0.0.1:7517/112267717904/ -->\n",
       "<iframe src=\"http://127.0.0.1:7517/112267717904/\" width=100% height=800 frameBorder=\"0\"></iframe>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ebm_global = ebm.explain_global(name='EBM')\n",
    "show(ebm_global)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## (E) Local Interpretability - What the Model Says for Individual Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<!-- http://127.0.0.1:7517/112295927696/ -->\n",
       "<iframe src=\"http://127.0.0.1:7517/112295927696/\" width=100% height=800 frameBorder=\"0\"></iframe>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ebm_local = ebm.explain_local(X_test[:5], Y_test[:5], name='EBM')\n",
    "show(ebm_local)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## (F) Put All in a Dashboard - This is the Best"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<!-- http://127.0.0.1:7517/112301524064/ -->\n",
       "<a href=\"http://127.0.0.1:7517/112301524064/\" target=\"_new\">Open in new window</a><iframe src=\"http://127.0.0.1:7517/112301524064/\" width=100% height=800 frameBorder=\"0\"></iframe>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "show([marginal, lr_global, lr_perf, rt_global, rt_perf, ebm_perf, ebm_global, ebm_local])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}