{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Using Shapash with Lime explainer - Titanic\n",
    "\n",
    "You can compute your local contributions with the [Lime](https://github.com/marcotcr/lime) library and summarize them with Shapash\n",
    "\n",
    "Contents:\n",
    "- Build a Binary Classifier (Random Forest)\n",
    "- Create Explainer using Lime\n",
    "- Compile Shapash SmartExplainer\n",
    "- Display local_plot\n",
    "- to_pandas export\n",
    "\n",
    "Data from Kaggle [Titanic](https://www.kaggle.com/c/titanic)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from category_encoders import OrdinalEncoder\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.model_selection import train_test_split\n",
    "import lime.lime_tabular"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from shapash.data.data_loader import data_loading"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "titan_df, titan_dict = data_loading('titanic')\n",
    "del titan_df['Name']"
   ]
  },
  {
   "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>Survived</th>\n",
       "      <th>Pclass</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>Title</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PassengerId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>Third class</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>7.25</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>Mr</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>First class</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>71.28</td>\n",
       "      <td>Cherbourg</td>\n",
       "      <td>Mrs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>Third class</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7.92</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>Miss</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>First class</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>53.10</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>Mrs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0</td>\n",
       "      <td>Third class</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8.05</td>\n",
       "      <td>Southampton</td>\n",
       "      <td>Mr</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Survived       Pclass     Sex   Age  SibSp  Parch   Fare  \\\n",
       "PassengerId                                                             \n",
       "1                   0  Third class    male  22.0      1      0   7.25   \n",
       "2                   1  First class  female  38.0      1      0  71.28   \n",
       "3                   1  Third class  female  26.0      0      0   7.92   \n",
       "4                   1  First class  female  35.0      1      0  53.10   \n",
       "5                   0  Third class    male  35.0      0      0   8.05   \n",
       "\n",
       "                Embarked Title  \n",
       "PassengerId                     \n",
       "1            Southampton    Mr  \n",
       "2              Cherbourg   Mrs  \n",
       "3            Southampton  Miss  \n",
       "4            Southampton   Mrs  \n",
       "5            Southampton    Mr  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "titan_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create Classification Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = titan_df['Survived']\n",
    "X = titan_df.drop('Survived', axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "varcat=['Pclass','Sex','Embarked','Title']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "categ_encoding = OrdinalEncoder(cols=varcat, \\\n",
    "                                handle_unknown='ignore', \\\n",
    "                                return_df=True).fit(X)\n",
    "X = categ_encoding.transform(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Train Test split + Random Forest fit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(bootstrap=True, ccp_alpha=0.0, class_weight=None,\n",
       "                       criterion='gini', max_depth=None, max_features='auto',\n",
       "                       max_leaf_nodes=None, max_samples=None,\n",
       "                       min_impurity_decrease=0.0, min_impurity_split=None,\n",
       "                       min_samples_leaf=3, min_samples_split=2,\n",
       "                       min_weight_fraction_leaf=0.0, n_estimators=100,\n",
       "                       n_jobs=None, oob_score=False, random_state=None,\n",
       "                       verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Xtrain, Xtest, ytrain, ytest = train_test_split(X, y, train_size=0.75, random_state=1)\n",
    "\n",
    "rf = RandomForestClassifier(n_estimators=100,min_samples_leaf=3)\n",
    "rf.fit(Xtrain, ytrain)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create Lime Explainer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Training Tabular Explainer\n",
    "explainer = lime.lime_tabular.LimeTabularExplainer(Xtrain.values, \n",
    "                                                   mode='classification',\n",
    "                                                   feature_names=Xtrain.columns,\n",
    "                                                   class_names=ytrain)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Apply Explainer to Test Sample And Preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Function features_check Extract feature names from Lime Output to be used by shapash\n",
    "def features_check(s):\n",
    "    for w in list(Xtest.columns):\n",
    "        if f' {w} ' in f' {s} ' :\n",
    "            feat = w\n",
    "    return feat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 57.8 s, sys: 7.34 s, total: 1min 5s\n",
      "Wall time: 10.9 s\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "# Compute local Lime Explanation for each row in Test Sample\n",
    "contrib_l=[]\n",
    "for ind in Xtest.index:\n",
    "    exp = explainer.explain_instance(Xtest.loc[ind].values, rf.predict_proba, num_features=Xtest.shape[1])\n",
    "    contrib_l.append(dict([[features_check(elem[0]),elem[1]] for elem in exp.as_list()]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "contribution_df =pd.DataFrame(contrib_l,index=Xtest.index)\n",
    "# sorting the columns as in the original dataset\n",
    "contribution_df = contribution_df[list(Xtest.columns)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "ypred=pd.DataFrame(rf.predict(Xtest),columns=['pred'],index=Xtest.index)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Use Shapash With Lime Contributions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "from shapash import SmartExplainer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "xpl = SmartExplainer(\n",
    "    model=rf,\n",
    "    preprocessing=categ_encoding,\n",
    "    features_dict=titan_dict\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Use contributions parameter of compile method to declare Lime contributions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "xpl.compile(\n",
    "    contributions=contribution_df, # Lime Contribution pd.DataFrame\n",
    "    y_pred=ypred,\n",
    "    y_target=ytest, # Optional: allows to display True Values vs Predicted Values\n",
    "    x=Xtest\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "xpl.plot.local_plot(index=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>pred</th>\n",
       "      <th>proba</th>\n",
       "      <th>feature_1</th>\n",
       "      <th>value_1</th>\n",
       "      <th>contribution_1</th>\n",
       "      <th>feature_2</th>\n",
       "      <th>value_2</th>\n",
       "      <th>contribution_2</th>\n",
       "      <th>feature_3</th>\n",
       "      <th>value_3</th>\n",
       "      <th>contribution_3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>863</th>\n",
       "      <td>1</td>\n",
       "      <td>0.801675</td>\n",
       "      <td>Sex</td>\n",
       "      <td>female</td>\n",
       "      <td>0.257817</td>\n",
       "      <td>Title of passenger</td>\n",
       "      <td>Mrs</td>\n",
       "      <td>0.188714</td>\n",
       "      <td>Ticket class</td>\n",
       "      <td>First class</td>\n",
       "      <td>0.0880992</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>224</th>\n",
       "      <td>0</td>\n",
       "      <td>0.965208</td>\n",
       "      <td>Sex</td>\n",
       "      <td>male</td>\n",
       "      <td>0.248462</td>\n",
       "      <td>Title of passenger</td>\n",
       "      <td>Mr</td>\n",
       "      <td>0.199544</td>\n",
       "      <td>Ticket class</td>\n",
       "      <td>Third class</td>\n",
       "      <td>0.0838383</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>85</th>\n",
       "      <td>1</td>\n",
       "      <td>0.799397</td>\n",
       "      <td>Sex</td>\n",
       "      <td>female</td>\n",
       "      <td>0.25465</td>\n",
       "      <td>Title of passenger</td>\n",
       "      <td>Miss</td>\n",
       "      <td>0.193198</td>\n",
       "      <td>Age</td>\n",
       "      <td>17</td>\n",
       "      <td>0.0981314</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>681</th>\n",
       "      <td>1</td>\n",
       "      <td>0.786956</td>\n",
       "      <td>Sex</td>\n",
       "      <td>female</td>\n",
       "      <td>0.252464</td>\n",
       "      <td>Title of passenger</td>\n",
       "      <td>Miss</td>\n",
       "      <td>0.187045</td>\n",
       "      <td>Relatives such as brother or wife</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0522808</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>536</th>\n",
       "      <td>1</td>\n",
       "      <td>0.936170</td>\n",
       "      <td>Sex</td>\n",
       "      <td>female</td>\n",
       "      <td>0.250703</td>\n",
       "      <td>Title of passenger</td>\n",
       "      <td>Miss</td>\n",
       "      <td>0.193096</td>\n",
       "      <td>Age</td>\n",
       "      <td>7</td>\n",
       "      <td>0.104632</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     pred     proba feature_1 value_1 contribution_1           feature_2  \\\n",
       "863     1  0.801675       Sex  female       0.257817  Title of passenger   \n",
       "224     0  0.965208       Sex    male       0.248462  Title of passenger   \n",
       "85      1  0.799397       Sex  female        0.25465  Title of passenger   \n",
       "681     1  0.786956       Sex  female       0.252464  Title of passenger   \n",
       "536     1  0.936170       Sex  female       0.250703  Title of passenger   \n",
       "\n",
       "    value_2 contribution_2                          feature_3      value_3  \\\n",
       "863     Mrs       0.188714                       Ticket class  First class   \n",
       "224      Mr       0.199544                       Ticket class  Third class   \n",
       "85     Miss       0.193198                                Age           17   \n",
       "681    Miss       0.187045  Relatives such as brother or wife            0   \n",
       "536    Miss       0.193096                                Age            7   \n",
       "\n",
       "    contribution_3  \n",
       "863      0.0880992  \n",
       "224      0.0838383  \n",
       "85       0.0981314  \n",
       "681      0.0522808  \n",
       "536       0.104632  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "summary_df = xpl.to_pandas(max_contrib=3,positive=True,proba=True)\n",
    "summary_df.head()"
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Aucun(e)",
  "hide_input": false,
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}