HEPML_HandsOn_BDT.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "ZT7AoY8sKwp1"
   },
   "source": [
    "# Hands on : introduction to BDT on HEP dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "eT7-MMpfrlHR"
   },
   "source": [
    "## Standard Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "QdmaHjz-xCm-"
   },
   "outputs": [],
   "source": [
    "COLAB=True #if running on https://colab.research.google.com/notebooks/welcome.ipynb\n",
    "#COLAB=False #if running on local anaconda installation https://docs.anaconda.com/anaconda/install/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "xy-_72FJKwp6"
   },
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "#from IPython import display USEFUL ???\n",
    "from IPython.display import display, HTML\n",
    "%matplotlib inline\n",
    "import time\n",
    "pd.set_option('display.max_columns', 100) # to see more columns of df.head()\n",
    "np.random.seed(31415) # set the np random seed for the reproducibility\n",
    "#from IPython.core.interactiveshell import InteractiveShell\n",
    "#InteractiveShell.ast_node_interactivity = \"all\" # instead of \"last_expr\", allow all cell output to appear \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 259
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 3785,
     "status": "ok",
     "timestamp": 1590131858341,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "oF367-yfKwp9",
    "outputId": "c8c445f2-06fa-437f-da6f-e446f779c761"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "running .zshenv\n",
      "Requirement already satisfied: xgboost in /opt/anaconda3/lib/python3.8/site-packages (1.3.3)\n",
      "Requirement already satisfied: numpy in /opt/anaconda3/lib/python3.8/site-packages (from xgboost) (1.19.2)\n",
      "Requirement already satisfied: scipy in /opt/anaconda3/lib/python3.8/site-packages (from xgboost) (1.5.2)\n",
      "running .zshenv\n",
      "Requirement already satisfied: lightgbm in /opt/anaconda3/lib/python3.8/site-packages (3.1.1)\n",
      "Requirement already satisfied: wheel in /opt/anaconda3/lib/python3.8/site-packages (from lightgbm) (0.35.1)\n",
      "Requirement already satisfied: scipy in /opt/anaconda3/lib/python3.8/site-packages (from lightgbm) (1.5.2)\n",
      "Requirement already satisfied: numpy in /opt/anaconda3/lib/python3.8/site-packages (from lightgbm) (1.19.2)\n",
      "Requirement already satisfied: scikit-learn!=0.22.0 in /opt/anaconda3/lib/python3.8/site-packages (from lightgbm) (0.23.2)\n",
      "Requirement already satisfied: threadpoolctl>=2.0.0 in /opt/anaconda3/lib/python3.8/site-packages (from scikit-learn!=0.22.0->lightgbm) (2.1.0)\n",
      "Requirement already satisfied: joblib>=0.11 in /opt/anaconda3/lib/python3.8/site-packages (from scikit-learn!=0.22.0->lightgbm) (0.17.0)\n",
      "1.3.3\n",
      "3.1.1\n"
     ]
    }
   ],
   "source": [
    "# Fix environment if necessary\n",
    "if COLAB:\n",
    "    !pip install xgboost --upgrade #  Colab has 0.9.0, not good enough\n",
    "    pass\n",
    "else:    \n",
    "    # install xgboost and lighgbm, two popular Boosted Decision Tree packages\n",
    "    # normally it is just a matter of a few pip install but there might be some glitches\n",
    "    # note : I have better experience with pip install than conda install. \n",
    "    # the following need to be done only once. To be commented out later\n",
    "    !pip install xgboost    \n",
    "    !pip install lightgbm \n",
    "    pass\n",
    "\n",
    "import xgboost\n",
    "print (xgboost.__version__) # Tested with 1.0.2, version above 1 is recommended.  \n",
    "import lightgbm\n",
    "print (lightgbm.__version__) # Tested with 2.3.1\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 279
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 6824,
     "status": "ok",
     "timestamp": 1590131861387,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "IsChk6u-r7fk",
    "outputId": "7a328e52-52d5-4e0b-f410-9f833be6ca50",
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "if COLAB:\n",
    "    #### Reading files from Google Drive\n",
    "    # one need a google account to be identified\n",
    "    # select a google account, then cut and paste the long password in the pop up field\n",
    "    !pip install PyDrive\n",
    "    import os\n",
    "    from pydrive.auth import GoogleAuth\n",
    "    from pydrive.drive import GoogleDrive\n",
    "    from google.colab import auth\n",
    "    from oauth2client.client import GoogleCredentials\n",
    "    auth.authenticate_user()\n",
    "    gauth = GoogleAuth()\n",
    "    gauth.credentials = GoogleCredentials.get_application_default()\n",
    "    drive = GoogleDrive(gauth)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "g14yIjbPKwqC"
   },
   "source": [
    "# Load events"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 104
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 14965,
     "status": "ok",
     "timestamp": 1590131869535,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "kl-4W4Ifs8EV",
    "outputId": "0377c86f-4593-415d-ed9e-7eb8f8368e3d"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "running .zshenv\n",
      "total 2888\n",
      "-rw-r--r--@ 1 rousseau  staff   40958 12 jan 11:22 Copy_of_Welcome_To_Colaboratory.ipynb\n",
      "-rw-r--r--+ 1 rousseau  staff    3881 13 jan 15:27 requirements.txt\n",
      "-rw-r--r--@ 1 rousseau  staff   63621 20 jan 10:54 SOS2021_practice_Basic_statistics_from_github.ipynb\n",
      "-rw-r--r--@ 1 rousseau  staff  103055 23 jan 20:08 HEPML_HandsOn_NN_veryold_immersion.ipynb\n",
      "-rw-------@ 1 rousseau  staff   18810 24 jan 12:37 HEPML_HandsOn_install_check.ipynb\n",
      "-rw-r--r--+ 1 rousseau  staff    6475 24 jan 20:40 html_original.html\n",
      "-rw-r--r--@ 1 rousseau  staff  188022 24 jan 21:46 Notebook_separation_maury_permutation_importance.ipynb\n",
      "-rw-r--r--+ 1 rousseau  staff  143607 24 jan 23:24 XGBoost.model\n",
      "-rw-r--r--+ 1 rousseau  staff  361110 24 jan 23:24 LightGBM.model\n",
      "-rw-r--r--+ 1 rousseau  staff   16228 24 jan 23:24 Score_BDT_XGBoost_Hist.pdf\n",
      "-rw-r--r--+ 1 rousseau  staff   17086 24 jan 23:24 Score_BDT_LightGBM.pdf\n",
      "-rw-r--r--+ 1 rousseau  staff   48661 24 jan 23:24 ROC_comparing.pdf\n",
      "-rw-r--r--+ 1 rousseau  staff   15438 24 jan 23:24 Significance_comparing.pdf\n",
      "-rw-------@ 1 rousseau  staff  351970 24 jan 23:24 HEPML_HandsOn_BDT.ipynb\n",
      "running .zshenv\n",
      "total 70344\n",
      "-rwxr-xr-x+ 1 rousseau  staff  35905611 13 mai  2020 \u001b[31mdataWW_d1_600k.csv.gz\u001b[m\u001b[m\n",
      "Will take data from :  /Users/rousseau/Google Drive/GD_openData/dataWW_ATLAS_openData13TeV_filtered\n",
      "now : 2021-01-24 23:25:05.047115\n",
      "File loaded with  600000  events \n"
     ]
    }
   ],
   "source": [
    "if COLAB:\n",
    "\n",
    "    #attach dataset from google drive \n",
    "    download = drive.CreateFile({'id': '1nlXp7P-xq_jip4aPE0j0mnPhYnIOcBv4'})\n",
    "    download.GetContentFile(\"dataWW_d1_600k.csv.gz\")\n",
    "\n",
    "\n",
    "    datapath=\"\"\n",
    "\n",
    "\n",
    "    !ls -lrt\n",
    "else :\n",
    "    # make sure the file is available locally. \n",
    "    #Should be downloaded from https://drive.google.com/open?id=1nlXp7P-xq_jip4aPE0j0mnPhYnIOcBv4\n",
    "    !ls -lrt # what is in the local directory\n",
    "    datapath=\"/Users/rousseau/Google\\ Drive/GD_openData/dataWW_ATLAS_openData13TeV_filtered/\"\n",
    "\n",
    "    !ls -lrt {datapath} # what is in the data directory\n",
    "    datapath=os.path.abspath(datapath).replace(\"\\ \", \" \")  # try to normalise the path (annoyance with the space)\n",
    "    print (\"Will take data from : \",datapath)\n",
    "\n",
    "filename=os.path.join(datapath,\"dataWW_d1_600k.csv.gz\")\n",
    "#load data\n",
    "# data was created from ATLAS Open Data see doc\n",
    "# http://opendata.atlas.cern/release/2020/documentation/datasets/intro.html\n",
    "dfall = pd.read_csv(filename) \n",
    "\n",
    "#shuffle the events, already done but just to be safe\n",
    "dfall = dfall.sample(frac=1).reset_index(drop=True)\n",
    "from datetime import datetime\n",
    "print (\"now :\",datetime.now())\n",
    "print (\"File loaded with \",dfall.shape[0], \" events \")\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "Uf9Gt8g8KwqF"
   },
   "source": [
    "At this point, it should tell you \"File Loaded with XXX events\". If not, it could not access the datafile, no point going further !"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "67cCYsYIKwqG"
   },
   "source": [
    "# Examine dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 138
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 14960,
     "status": "ok",
     "timestamp": 1590131869536,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "sx1_PB22KwqG",
    "outputId": "775c6de0-7462-4524-b1fe-070bf41e12bc"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['eventNumber', 'label', 'met_et', 'met_phi', 'lep_n', 'lep_pt_0',\n",
       "       'lep_pt_1', 'lep_eta_0', 'lep_eta_1', 'lep_phi_0', 'lep_phi_1',\n",
       "       'lep_E_0', 'lep_E_1', 'lep_charge_0', 'lep_charge_1', 'lep_type_0',\n",
       "       'lep_type_1', 'jet_n', 'jet_pt_0', 'jet_pt_1', 'jet_eta_0', 'jet_eta_1',\n",
       "       'jet_phi_0', 'jet_phi_1', 'jet_E_0', 'jet_E_1', 'mcWeight', 'runNumber',\n",
       "       'channelNumber'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#dump list of feature\n",
    "dfall.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 218
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 14954,
     "status": "ok",
     "timestamp": 1590131869537,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "j9l7wkkorlHe",
    "outputId": "8ddd33ec-4467-426c-fc2b-7685e25aa8bf"
   },
   "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>eventNumber</th>\n",
       "      <th>label</th>\n",
       "      <th>met_et</th>\n",
       "      <th>met_phi</th>\n",
       "      <th>lep_n</th>\n",
       "      <th>lep_pt_0</th>\n",
       "      <th>lep_pt_1</th>\n",
       "      <th>lep_eta_0</th>\n",
       "      <th>lep_eta_1</th>\n",
       "      <th>lep_phi_0</th>\n",
       "      <th>lep_phi_1</th>\n",
       "      <th>lep_E_0</th>\n",
       "      <th>lep_E_1</th>\n",
       "      <th>lep_charge_0</th>\n",
       "      <th>lep_charge_1</th>\n",
       "      <th>lep_type_0</th>\n",
       "      <th>lep_type_1</th>\n",
       "      <th>jet_n</th>\n",
       "      <th>jet_pt_0</th>\n",
       "      <th>jet_pt_1</th>\n",
       "      <th>jet_eta_0</th>\n",
       "      <th>jet_eta_1</th>\n",
       "      <th>jet_phi_0</th>\n",
       "      <th>jet_phi_1</th>\n",
       "      <th>jet_E_0</th>\n",
       "      <th>jet_E_1</th>\n",
       "      <th>mcWeight</th>\n",
       "      <th>runNumber</th>\n",
       "      <th>channelNumber</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>402756</td>\n",
       "      <td>1</td>\n",
       "      <td>25.609</td>\n",
       "      <td>0.42452</td>\n",
       "      <td>2</td>\n",
       "      <td>48.295</td>\n",
       "      <td>15.214</td>\n",
       "      <td>0.73991</td>\n",
       "      <td>2.27420</td>\n",
       "      <td>-2.316400</td>\n",
       "      <td>-1.39410</td>\n",
       "      <td>62129.0</td>\n",
       "      <td>74721.0</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>26.32</td>\n",
       "      <td>20.064</td>\n",
       "      <td>-1.1350</td>\n",
       "      <td>-2.07540</td>\n",
       "      <td>-2.8092</td>\n",
       "      <td>2.0740</td>\n",
       "      <td>45397.0</td>\n",
       "      <td>81304.0</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>284500</td>\n",
       "      <td>345323</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>101274</td>\n",
       "      <td>0</td>\n",
       "      <td>196.560</td>\n",
       "      <td>1.31140</td>\n",
       "      <td>2</td>\n",
       "      <td>69.459</td>\n",
       "      <td>21.081</td>\n",
       "      <td>-0.52666</td>\n",
       "      <td>0.22380</td>\n",
       "      <td>0.023132</td>\n",
       "      <td>-0.67855</td>\n",
       "      <td>79317.0</td>\n",
       "      <td>21611.0</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>13</td>\n",
       "      <td>2</td>\n",
       "      <td>192.12</td>\n",
       "      <td>36.217</td>\n",
       "      <td>-1.2643</td>\n",
       "      <td>-0.69746</td>\n",
       "      <td>-2.2147</td>\n",
       "      <td>-1.9152</td>\n",
       "      <td>367670.0</td>\n",
       "      <td>45733.0</td>\n",
       "      <td>0.000288</td>\n",
       "      <td>284500</td>\n",
       "      <td>363492</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>468437</td>\n",
       "      <td>1</td>\n",
       "      <td>45.653</td>\n",
       "      <td>-2.76860</td>\n",
       "      <td>2</td>\n",
       "      <td>45.927</td>\n",
       "      <td>22.822</td>\n",
       "      <td>-1.61910</td>\n",
       "      <td>-2.00770</td>\n",
       "      <td>1.901700</td>\n",
       "      <td>-0.11248</td>\n",
       "      <td>120480.0</td>\n",
       "      <td>86498.0</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.00</td>\n",
       "      <td>-7.000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.00000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.0</td>\n",
       "      <td>-7.0</td>\n",
       "      <td>0.000029</td>\n",
       "      <td>284500</td>\n",
       "      <td>345324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>272337</td>\n",
       "      <td>1</td>\n",
       "      <td>49.415</td>\n",
       "      <td>-0.57805</td>\n",
       "      <td>2</td>\n",
       "      <td>45.929</td>\n",
       "      <td>14.263</td>\n",
       "      <td>1.55280</td>\n",
       "      <td>0.12809</td>\n",
       "      <td>2.254700</td>\n",
       "      <td>-2.54810</td>\n",
       "      <td>113360.0</td>\n",
       "      <td>14380.0</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.00</td>\n",
       "      <td>-7.000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.00000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.0</td>\n",
       "      <td>-7.0</td>\n",
       "      <td>0.000029</td>\n",
       "      <td>284500</td>\n",
       "      <td>345324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>354546</td>\n",
       "      <td>1</td>\n",
       "      <td>71.988</td>\n",
       "      <td>-2.60390</td>\n",
       "      <td>2</td>\n",
       "      <td>62.029</td>\n",
       "      <td>21.453</td>\n",
       "      <td>-0.51082</td>\n",
       "      <td>-0.38177</td>\n",
       "      <td>0.236200</td>\n",
       "      <td>1.40820</td>\n",
       "      <td>70300.0</td>\n",
       "      <td>23036.0</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.00</td>\n",
       "      <td>-7.000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.00000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.0</td>\n",
       "      <td>-7.0</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>284500</td>\n",
       "      <td>345323</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   eventNumber  label   met_et  met_phi  lep_n  lep_pt_0  lep_pt_1  lep_eta_0  \\\n",
       "0       402756      1   25.609  0.42452      2    48.295    15.214    0.73991   \n",
       "1       101274      0  196.560  1.31140      2    69.459    21.081   -0.52666   \n",
       "2       468437      1   45.653 -2.76860      2    45.927    22.822   -1.61910   \n",
       "3       272337      1   49.415 -0.57805      2    45.929    14.263    1.55280   \n",
       "4       354546      1   71.988 -2.60390      2    62.029    21.453   -0.51082   \n",
       "\n",
       "   lep_eta_1  lep_phi_0  lep_phi_1   lep_E_0  lep_E_1  lep_charge_0  \\\n",
       "0    2.27420  -2.316400   -1.39410   62129.0  74721.0            -1   \n",
       "1    0.22380   0.023132   -0.67855   79317.0  21611.0            -1   \n",
       "2   -2.00770   1.901700   -0.11248  120480.0  86498.0            -1   \n",
       "3    0.12809   2.254700   -2.54810  113360.0  14380.0            -1   \n",
       "4   -0.38177   0.236200    1.40820   70300.0  23036.0            -1   \n",
       "\n",
       "   lep_charge_1  lep_type_0  lep_type_1  jet_n  jet_pt_0  jet_pt_1  jet_eta_0  \\\n",
       "0             1          13          11      2     26.32    20.064    -1.1350   \n",
       "1             1          13          13      2    192.12    36.217    -1.2643   \n",
       "2             1          11          11      0     -7.00    -7.000    -7.0000   \n",
       "3             1          13          13      0     -7.00    -7.000    -7.0000   \n",
       "4             1          11          13      0     -7.00    -7.000    -7.0000   \n",
       "\n",
       "   jet_eta_1  jet_phi_0  jet_phi_1   jet_E_0  jet_E_1  mcWeight  runNumber  \\\n",
       "0   -2.07540    -2.8092     2.0740   45397.0  81304.0  0.000002     284500   \n",
       "1   -0.69746    -2.2147    -1.9152  367670.0  45733.0  0.000288     284500   \n",
       "2   -7.00000    -7.0000    -7.0000      -7.0     -7.0  0.000029     284500   \n",
       "3   -7.00000    -7.0000    -7.0000      -7.0     -7.0  0.000029     284500   \n",
       "4   -7.00000    -7.0000    -7.0000      -7.0     -7.0  0.000002     284500   \n",
       "\n",
       "   channelNumber  \n",
       "0         345323  \n",
       "1         363492  \n",
       "2         345324  \n",
       "3         345324  \n",
       "4         345323  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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>eventNumber</th>\n",
       "      <th>label</th>\n",
       "      <th>met_et</th>\n",
       "      <th>met_phi</th>\n",
       "      <th>lep_n</th>\n",
       "      <th>lep_pt_0</th>\n",
       "      <th>lep_pt_1</th>\n",
       "      <th>lep_eta_0</th>\n",
       "      <th>lep_eta_1</th>\n",
       "      <th>lep_phi_0</th>\n",
       "      <th>lep_phi_1</th>\n",
       "      <th>lep_E_0</th>\n",
       "      <th>lep_E_1</th>\n",
       "      <th>lep_charge_0</th>\n",
       "      <th>lep_charge_1</th>\n",
       "      <th>lep_type_0</th>\n",
       "      <th>lep_type_1</th>\n",
       "      <th>jet_n</th>\n",
       "      <th>jet_pt_0</th>\n",
       "      <th>jet_pt_1</th>\n",
       "      <th>jet_eta_0</th>\n",
       "      <th>jet_eta_1</th>\n",
       "      <th>jet_phi_0</th>\n",
       "      <th>jet_phi_1</th>\n",
       "      <th>jet_E_0</th>\n",
       "      <th>jet_E_1</th>\n",
       "      <th>mcWeight</th>\n",
       "      <th>runNumber</th>\n",
       "      <th>channelNumber</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>402756</td>\n",
       "      <td>1</td>\n",
       "      <td>25.609</td>\n",
       "      <td>0.42452</td>\n",
       "      <td>2</td>\n",
       "      <td>48.295</td>\n",
       "      <td>15.214</td>\n",
       "      <td>0.73991</td>\n",
       "      <td>2.27420</td>\n",
       "      <td>-2.316400</td>\n",
       "      <td>-1.39410</td>\n",
       "      <td>62129.0</td>\n",
       "      <td>74721.0</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>26.32</td>\n",
       "      <td>20.064</td>\n",
       "      <td>-1.1350</td>\n",
       "      <td>-2.07540</td>\n",
       "      <td>-2.8092</td>\n",
       "      <td>2.0740</td>\n",
       "      <td>45397.0</td>\n",
       "      <td>81304.0</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>284500</td>\n",
       "      <td>345323</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>101274</td>\n",
       "      <td>0</td>\n",
       "      <td>196.560</td>\n",
       "      <td>1.31140</td>\n",
       "      <td>2</td>\n",
       "      <td>69.459</td>\n",
       "      <td>21.081</td>\n",
       "      <td>-0.52666</td>\n",
       "      <td>0.22380</td>\n",
       "      <td>0.023132</td>\n",
       "      <td>-0.67855</td>\n",
       "      <td>79317.0</td>\n",
       "      <td>21611.0</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>13</td>\n",
       "      <td>2</td>\n",
       "      <td>192.12</td>\n",
       "      <td>36.217</td>\n",
       "      <td>-1.2643</td>\n",
       "      <td>-0.69746</td>\n",
       "      <td>-2.2147</td>\n",
       "      <td>-1.9152</td>\n",
       "      <td>367670.0</td>\n",
       "      <td>45733.0</td>\n",
       "      <td>0.000288</td>\n",
       "      <td>284500</td>\n",
       "      <td>363492</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>468437</td>\n",
       "      <td>1</td>\n",
       "      <td>45.653</td>\n",
       "      <td>-2.76860</td>\n",
       "      <td>2</td>\n",
       "      <td>45.927</td>\n",
       "      <td>22.822</td>\n",
       "      <td>-1.61910</td>\n",
       "      <td>-2.00770</td>\n",
       "      <td>1.901700</td>\n",
       "      <td>-0.11248</td>\n",
       "      <td>120480.0</td>\n",
       "      <td>86498.0</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.00</td>\n",
       "      <td>-7.000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.00000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.0</td>\n",
       "      <td>-7.0</td>\n",
       "      <td>0.000029</td>\n",
       "      <td>284500</td>\n",
       "      <td>345324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>272337</td>\n",
       "      <td>1</td>\n",
       "      <td>49.415</td>\n",
       "      <td>-0.57805</td>\n",
       "      <td>2</td>\n",
       "      <td>45.929</td>\n",
       "      <td>14.263</td>\n",
       "      <td>1.55280</td>\n",
       "      <td>0.12809</td>\n",
       "      <td>2.254700</td>\n",
       "      <td>-2.54810</td>\n",
       "      <td>113360.0</td>\n",
       "      <td>14380.0</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>13</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.00</td>\n",
       "      <td>-7.000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.00000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.0</td>\n",
       "      <td>-7.0</td>\n",
       "      <td>0.000029</td>\n",
       "      <td>284500</td>\n",
       "      <td>345324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>354546</td>\n",
       "      <td>1</td>\n",
       "      <td>71.988</td>\n",
       "      <td>-2.60390</td>\n",
       "      <td>2</td>\n",
       "      <td>62.029</td>\n",
       "      <td>21.453</td>\n",
       "      <td>-0.51082</td>\n",
       "      <td>-0.38177</td>\n",
       "      <td>0.236200</td>\n",
       "      <td>1.40820</td>\n",
       "      <td>70300.0</td>\n",
       "      <td>23036.0</td>\n",
       "      <td>-1</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>-7.00</td>\n",
       "      <td>-7.000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.00000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.0000</td>\n",
       "      <td>-7.0</td>\n",
       "      <td>-7.0</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>284500</td>\n",
       "      <td>345323</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   eventNumber  label   met_et  met_phi  lep_n  lep_pt_0  lep_pt_1  lep_eta_0  \\\n",
       "0       402756      1   25.609  0.42452      2    48.295    15.214    0.73991   \n",
       "1       101274      0  196.560  1.31140      2    69.459    21.081   -0.52666   \n",
       "2       468437      1   45.653 -2.76860      2    45.927    22.822   -1.61910   \n",
       "3       272337      1   49.415 -0.57805      2    45.929    14.263    1.55280   \n",
       "4       354546      1   71.988 -2.60390      2    62.029    21.453   -0.51082   \n",
       "\n",
       "   lep_eta_1  lep_phi_0  lep_phi_1   lep_E_0  lep_E_1  lep_charge_0  \\\n",
       "0    2.27420  -2.316400   -1.39410   62129.0  74721.0            -1   \n",
       "1    0.22380   0.023132   -0.67855   79317.0  21611.0            -1   \n",
       "2   -2.00770   1.901700   -0.11248  120480.0  86498.0            -1   \n",
       "3    0.12809   2.254700   -2.54810  113360.0  14380.0            -1   \n",
       "4   -0.38177   0.236200    1.40820   70300.0  23036.0            -1   \n",
       "\n",
       "   lep_charge_1  lep_type_0  lep_type_1  jet_n  jet_pt_0  jet_pt_1  jet_eta_0  \\\n",
       "0             1          13          11      2     26.32    20.064    -1.1350   \n",
       "1             1          13          13      2    192.12    36.217    -1.2643   \n",
       "2             1          11          11      0     -7.00    -7.000    -7.0000   \n",
       "3             1          13          13      0     -7.00    -7.000    -7.0000   \n",
       "4             1          11          13      0     -7.00    -7.000    -7.0000   \n",
       "\n",
       "   jet_eta_1  jet_phi_0  jet_phi_1   jet_E_0  jet_E_1  mcWeight  runNumber  \\\n",
       "0   -2.07540    -2.8092     2.0740   45397.0  81304.0  0.000002     284500   \n",
       "1   -0.69746    -2.2147    -1.9152  367670.0  45733.0  0.000288     284500   \n",
       "2   -7.00000    -7.0000    -7.0000      -7.0     -7.0  0.000029     284500   \n",
       "3   -7.00000    -7.0000    -7.0000      -7.0     -7.0  0.000029     284500   \n",
       "4   -7.00000    -7.0000    -7.0000      -7.0     -7.0  0.000002     284500   \n",
       "\n",
       "   channelNumber  \n",
       "0         345323  \n",
       "1         363492  \n",
       "2         345324  \n",
       "3         345324  \n",
       "4         345323  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#examine first few events\n",
    "display(dfall.head())\n",
    "display(dfall.head())\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 308
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 16072,
     "status": "ok",
     "timestamp": 1590131870662,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "Oz-lWJhgrlHg",
    "outputId": "d71ee65f-03a9-435b-fb55-7be4a7ecaeb0"
   },
   "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>eventNumber</th>\n",
       "      <th>label</th>\n",
       "      <th>met_et</th>\n",
       "      <th>met_phi</th>\n",
       "      <th>lep_n</th>\n",
       "      <th>lep_pt_0</th>\n",
       "      <th>lep_pt_1</th>\n",
       "      <th>lep_eta_0</th>\n",
       "      <th>lep_eta_1</th>\n",
       "      <th>lep_phi_0</th>\n",
       "      <th>lep_phi_1</th>\n",
       "      <th>lep_E_0</th>\n",
       "      <th>lep_E_1</th>\n",
       "      <th>lep_charge_0</th>\n",
       "      <th>lep_charge_1</th>\n",
       "      <th>lep_type_0</th>\n",
       "      <th>lep_type_1</th>\n",
       "      <th>jet_n</th>\n",
       "      <th>jet_pt_0</th>\n",
       "      <th>jet_pt_1</th>\n",
       "      <th>jet_eta_0</th>\n",
       "      <th>jet_eta_1</th>\n",
       "      <th>jet_phi_0</th>\n",
       "      <th>jet_phi_1</th>\n",
       "      <th>jet_E_0</th>\n",
       "      <th>jet_E_1</th>\n",
       "      <th>mcWeight</th>\n",
       "      <th>runNumber</th>\n",
       "      <th>channelNumber</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>6.000000e+05</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>6.000000e+05</td>\n",
       "      <td>6.000000e+05</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>6.000000e+05</td>\n",
       "      <td>6.000000e+05</td>\n",
       "      <td>600000.000000</td>\n",
       "      <td>600000.0</td>\n",
       "      <td>600000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>7.029775e+05</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>60.365356</td>\n",
       "      <td>-0.008556</td>\n",
       "      <td>2.005748</td>\n",
       "      <td>60.270328</td>\n",
       "      <td>28.133970</td>\n",
       "      <td>0.000269</td>\n",
       "      <td>-0.000664</td>\n",
       "      <td>0.005219</td>\n",
       "      <td>0.008524</td>\n",
       "      <td>1.164008e+05</td>\n",
       "      <td>5.630948e+04</td>\n",
       "      <td>-0.027027</td>\n",
       "      <td>0.027037</td>\n",
       "      <td>11.907273</td>\n",
       "      <td>11.958010</td>\n",
       "      <td>1.334185</td>\n",
       "      <td>59.606114</td>\n",
       "      <td>13.521164</td>\n",
       "      <td>-1.934521</td>\n",
       "      <td>-4.402337</td>\n",
       "      <td>-1.939645</td>\n",
       "      <td>-4.401941</td>\n",
       "      <td>1.375410e+05</td>\n",
       "      <td>4.125468e+04</td>\n",
       "      <td>0.000218</td>\n",
       "      <td>284500.0</td>\n",
       "      <td>351379.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>4.568961e+05</td>\n",
       "      <td>0.471405</td>\n",
       "      <td>48.153210</td>\n",
       "      <td>1.812778</td>\n",
       "      <td>0.076061</td>\n",
       "      <td>42.676858</td>\n",
       "      <td>20.352272</td>\n",
       "      <td>1.210614</td>\n",
       "      <td>1.235890</td>\n",
       "      <td>1.811177</td>\n",
       "      <td>1.814205</td>\n",
       "      <td>1.083341e+05</td>\n",
       "      <td>5.748834e+04</td>\n",
       "      <td>0.999636</td>\n",
       "      <td>0.999635</td>\n",
       "      <td>0.995692</td>\n",
       "      <td>0.999119</td>\n",
       "      <td>1.232632</td>\n",
       "      <td>79.586143</td>\n",
       "      <td>38.074439</td>\n",
       "      <td>3.349404</td>\n",
       "      <td>3.491304</td>\n",
       "      <td>3.489197</td>\n",
       "      <td>3.559421</td>\n",
       "      <td>1.980717e+05</td>\n",
       "      <td>8.990068e+04</td>\n",
       "      <td>0.000736</td>\n",
       "      <td>0.0</td>\n",
       "      <td>8564.720183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.044211</td>\n",
       "      <td>-3.141600</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>25.000000</td>\n",
       "      <td>7.000300</td>\n",
       "      <td>-2.700000</td>\n",
       "      <td>-2.699900</td>\n",
       "      <td>-3.141600</td>\n",
       "      <td>-3.141600</td>\n",
       "      <td>2.501300e+04</td>\n",
       "      <td>7.003800e+03</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>11.000000</td>\n",
       "      <td>11.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-7.000000</td>\n",
       "      <td>-7.000000</td>\n",
       "      <td>-7.000000</td>\n",
       "      <td>-7.000000</td>\n",
       "      <td>-7.000000</td>\n",
       "      <td>-7.000000</td>\n",
       "      <td>-7.000000e+00</td>\n",
       "      <td>-7.000000e+00</td>\n",
       "      <td>-0.045557</td>\n",
       "      <td>284500.0</td>\n",
       "      <td>345323.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>3.498568e+05</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>33.196000</td>\n",
       "      <td>-1.577600</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>36.624750</td>\n",
       "      <td>15.316000</td>\n",
       "      <td>-0.899650</td>\n",
       "      <td>-0.946043</td>\n",
       "      <td>-1.560200</td>\n",
       "      <td>-1.562900</td>\n",
       "      <td>5.221675e+04</td>\n",
       "      <td>2.329100e+04</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>11.000000</td>\n",
       "      <td>11.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-7.000000</td>\n",
       "      <td>-7.000000</td>\n",
       "      <td>-7.000000</td>\n",
       "      <td>-7.000000</td>\n",
       "      <td>-7.000000</td>\n",
       "      <td>-7.000000</td>\n",
       "      <td>-7.000000e+00</td>\n",
       "      <td>-7.000000e+00</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>284500.0</td>\n",
       "      <td>345323.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>6.678815e+05</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>50.082000</td>\n",
       "      <td>-0.018694</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>48.229500</td>\n",
       "      <td>23.644000</td>\n",
       "      <td>0.003413</td>\n",
       "      <td>0.002974</td>\n",
       "      <td>0.014583</td>\n",
       "      <td>0.018823</td>\n",
       "      <td>8.184450e+04</td>\n",
       "      <td>3.824900e+04</td>\n",
       "      <td>-1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>11.000000</td>\n",
       "      <td>11.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>38.744000</td>\n",
       "      <td>-7.000000</td>\n",
       "      <td>-0.891010</td>\n",
       "      <td>-7.000000</td>\n",
       "      <td>-1.210600</td>\n",
       "      <td>-7.000000</td>\n",
       "      <td>7.501700e+04</td>\n",
       "      <td>-7.000000e+00</td>\n",
       "      <td>0.000029</td>\n",
       "      <td>284500.0</td>\n",
       "      <td>345324.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>1.001273e+06</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>73.032250</td>\n",
       "      <td>1.559300</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>68.979000</td>\n",
       "      <td>34.149000</td>\n",
       "      <td>0.901950</td>\n",
       "      <td>0.942210</td>\n",
       "      <td>1.573000</td>\n",
       "      <td>1.577300</td>\n",
       "      <td>1.412400e+05</td>\n",
       "      <td>6.730825e+04</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>84.291250</td>\n",
       "      <td>26.881000</td>\n",
       "      <td>0.724705</td>\n",
       "      <td>-0.821005</td>\n",
       "      <td>0.967753</td>\n",
       "      <td>-1.088575</td>\n",
       "      <td>1.766200e+05</td>\n",
       "      <td>5.060500e+04</td>\n",
       "      <td>0.000197</td>\n",
       "      <td>284500.0</td>\n",
       "      <td>363492.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>1.997995e+06</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>7062.200000</td>\n",
       "      <td>3.141600</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>7041.400000</td>\n",
       "      <td>744.900000</td>\n",
       "      <td>2.699800</td>\n",
       "      <td>2.700000</td>\n",
       "      <td>3.141500</td>\n",
       "      <td>3.141600</td>\n",
       "      <td>7.041600e+06</td>\n",
       "      <td>2.107900e+06</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>1658.200000</td>\n",
       "      <td>1429.800000</td>\n",
       "      <td>2.500000</td>\n",
       "      <td>2.500000</td>\n",
       "      <td>3.141600</td>\n",
       "      <td>3.141600</td>\n",
       "      <td>5.638800e+06</td>\n",
       "      <td>2.784300e+06</td>\n",
       "      <td>0.052420</td>\n",
       "      <td>284500.0</td>\n",
       "      <td>363492.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        eventNumber          label         met_et        met_phi  \\\n",
       "count  6.000000e+05  600000.000000  600000.000000  600000.000000   \n",
       "mean   7.029775e+05       0.666667      60.365356      -0.008556   \n",
       "std    4.568961e+05       0.471405      48.153210       1.812778   \n",
       "min    1.000000e+00       0.000000       0.044211      -3.141600   \n",
       "25%    3.498568e+05       0.000000      33.196000      -1.577600   \n",
       "50%    6.678815e+05       1.000000      50.082000      -0.018694   \n",
       "75%    1.001273e+06       1.000000      73.032250       1.559300   \n",
       "max    1.997995e+06       1.000000    7062.200000       3.141600   \n",
       "\n",
       "               lep_n       lep_pt_0       lep_pt_1      lep_eta_0  \\\n",
       "count  600000.000000  600000.000000  600000.000000  600000.000000   \n",
       "mean        2.005748      60.270328      28.133970       0.000269   \n",
       "std         0.076061      42.676858      20.352272       1.210614   \n",
       "min         2.000000      25.000000       7.000300      -2.700000   \n",
       "25%         2.000000      36.624750      15.316000      -0.899650   \n",
       "50%         2.000000      48.229500      23.644000       0.003413   \n",
       "75%         2.000000      68.979000      34.149000       0.901950   \n",
       "max         5.000000    7041.400000     744.900000       2.699800   \n",
       "\n",
       "           lep_eta_1      lep_phi_0      lep_phi_1       lep_E_0  \\\n",
       "count  600000.000000  600000.000000  600000.000000  6.000000e+05   \n",
       "mean       -0.000664       0.005219       0.008524  1.164008e+05   \n",
       "std         1.235890       1.811177       1.814205  1.083341e+05   \n",
       "min        -2.699900      -3.141600      -3.141600  2.501300e+04   \n",
       "25%        -0.946043      -1.560200      -1.562900  5.221675e+04   \n",
       "50%         0.002974       0.014583       0.018823  8.184450e+04   \n",
       "75%         0.942210       1.573000       1.577300  1.412400e+05   \n",
       "max         2.700000       3.141500       3.141600  7.041600e+06   \n",
       "\n",
       "            lep_E_1   lep_charge_0   lep_charge_1     lep_type_0  \\\n",
       "count  6.000000e+05  600000.000000  600000.000000  600000.000000   \n",
       "mean   5.630948e+04      -0.027027       0.027037      11.907273   \n",
       "std    5.748834e+04       0.999636       0.999635       0.995692   \n",
       "min    7.003800e+03      -1.000000      -1.000000      11.000000   \n",
       "25%    2.329100e+04      -1.000000      -1.000000      11.000000   \n",
       "50%    3.824900e+04      -1.000000       1.000000      11.000000   \n",
       "75%    6.730825e+04       1.000000       1.000000      13.000000   \n",
       "max    2.107900e+06       1.000000       1.000000      13.000000   \n",
       "\n",
       "          lep_type_1          jet_n       jet_pt_0       jet_pt_1  \\\n",
       "count  600000.000000  600000.000000  600000.000000  600000.000000   \n",
       "mean       11.958010       1.334185      59.606114      13.521164   \n",
       "std         0.999119       1.232632      79.586143      38.074439   \n",
       "min        11.000000       0.000000      -7.000000      -7.000000   \n",
       "25%        11.000000       0.000000      -7.000000      -7.000000   \n",
       "50%        11.000000       1.000000      38.744000      -7.000000   \n",
       "75%        13.000000       2.000000      84.291250      26.881000   \n",
       "max        13.000000      13.000000    1658.200000    1429.800000   \n",
       "\n",
       "           jet_eta_0      jet_eta_1      jet_phi_0      jet_phi_1  \\\n",
       "count  600000.000000  600000.000000  600000.000000  600000.000000   \n",
       "mean       -1.934521      -4.402337      -1.939645      -4.401941   \n",
       "std         3.349404       3.491304       3.489197       3.559421   \n",
       "min        -7.000000      -7.000000      -7.000000      -7.000000   \n",
       "25%        -7.000000      -7.000000      -7.000000      -7.000000   \n",
       "50%        -0.891010      -7.000000      -1.210600      -7.000000   \n",
       "75%         0.724705      -0.821005       0.967753      -1.088575   \n",
       "max         2.500000       2.500000       3.141600       3.141600   \n",
       "\n",
       "            jet_E_0       jet_E_1       mcWeight  runNumber  channelNumber  \n",
       "count  6.000000e+05  6.000000e+05  600000.000000   600000.0  600000.000000  \n",
       "mean   1.375410e+05  4.125468e+04       0.000218   284500.0  351379.666667  \n",
       "std    1.980717e+05  8.990068e+04       0.000736        0.0    8564.720183  \n",
       "min   -7.000000e+00 -7.000000e+00      -0.045557   284500.0  345323.000000  \n",
       "25%   -7.000000e+00 -7.000000e+00       0.000002   284500.0  345323.000000  \n",
       "50%    7.501700e+04 -7.000000e+00       0.000029   284500.0  345324.000000  \n",
       "75%    1.766200e+05  5.060500e+04       0.000197   284500.0  363492.000000  \n",
       "max    5.638800e+06  2.784300e+06       0.052420   284500.0  363492.000000  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#examine feature distribution\n",
    "dfall.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 52
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 16064,
     "status": "ok",
     "timestamp": 1590131870663,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "lQsalTmorlHj",
    "outputId": "98db6ff1-2f63-48ee-9246-b7ef46c66e5c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "total label weights  B S = (124.6622944642, 6.2000533976)\n",
      "total class number of events B S (200000, 400000)\n"
     ]
    }
   ],
   "source": [
    "label_weights = (dfall[dfall.label==0].mcWeight.sum(), dfall[dfall.label==1].mcWeight.sum() ) \n",
    "print(\"total label weights  B S =\",label_weights)\n",
    "\n",
    "\n",
    "label_nevents = (dfall[dfall.label==0].shape[0], dfall[dfall.label==1].shape[0] )\n",
    "print (\"total class number of events B S\",label_nevents)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "rtI5u5GErlHq"
   },
   "source": [
    "## Event selection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 52
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 16058,
     "status": "ok",
     "timestamp": 1590131870664,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "kaO2JM1hrlHr",
    "outputId": "d2eb784d-c8fe-40cc-ce1d-6ae7f92aece9"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Df shape before selection : (600000, 29)\n",
      "Df shape after selection : (577357, 29)\n"
     ]
    }
   ],
   "source": [
    "print (\"Df shape before selection :\", dfall.shape)\n",
    "# only keep events with exactly two leptons\n",
    "# only keep events with positive weight. This is in principle wrong. Many DS tool choke on negative weight.\n",
    "fulldata=dfall[ (dfall.lep_n==2) & (dfall.mcWeight > 0)]  \n",
    "\n",
    "\n",
    "print (\"Df shape after selection :\",fulldata.shape)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "nviyIMgerlH3"
   },
   "source": [
    "\n",
    "# DO NOT MODIFY ANYTHING ABOVE\n",
    "... and always rerun from this cell whenever you change something below"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 215
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 16498,
     "status": "ok",
     "timestamp": 1590131871120,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "6e0Hlpv6rlH4",
    "outputId": "f495b400-8875-40c5-e869-ccc92b7b0459",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Df shape of dataset to be used : (577357, 6)\n"
     ]
    },
    {
     "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>met_et</th>\n",
       "      <th>met_phi</th>\n",
       "      <th>lep_pt_0</th>\n",
       "      <th>lep_pt_1</th>\n",
       "      <th>lep_phi_0</th>\n",
       "      <th>lep_phi_1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>25.609</td>\n",
       "      <td>0.42452</td>\n",
       "      <td>48.295</td>\n",
       "      <td>15.214</td>\n",
       "      <td>-2.316400</td>\n",
       "      <td>-1.39410</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>196.560</td>\n",
       "      <td>1.31140</td>\n",
       "      <td>69.459</td>\n",
       "      <td>21.081</td>\n",
       "      <td>0.023132</td>\n",
       "      <td>-0.67855</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>45.653</td>\n",
       "      <td>-2.76860</td>\n",
       "      <td>45.927</td>\n",
       "      <td>22.822</td>\n",
       "      <td>1.901700</td>\n",
       "      <td>-0.11248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>49.415</td>\n",
       "      <td>-0.57805</td>\n",
       "      <td>45.929</td>\n",
       "      <td>14.263</td>\n",
       "      <td>2.254700</td>\n",
       "      <td>-2.54810</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>71.988</td>\n",
       "      <td>-2.60390</td>\n",
       "      <td>62.029</td>\n",
       "      <td>21.453</td>\n",
       "      <td>0.236200</td>\n",
       "      <td>1.40820</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    met_et  met_phi  lep_pt_0  lep_pt_1  lep_phi_0  lep_phi_1\n",
       "0   25.609  0.42452    48.295    15.214  -2.316400   -1.39410\n",
       "1  196.560  1.31140    69.459    21.081   0.023132   -0.67855\n",
       "2   45.653 -2.76860    45.927    22.822   1.901700   -0.11248\n",
       "3   49.415 -0.57805    45.929    14.263   2.254700   -2.54810\n",
       "4   71.988 -2.60390    62.029    21.453   0.236200    1.40820"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0    1\n",
       "1    0\n",
       "2    1\n",
       "3    1\n",
       "4    1\n",
       "Name: label, dtype: int64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0    0.000002\n",
       "1    0.000288\n",
       "2    0.000029\n",
       "3    0.000029\n",
       "4    0.000002\n",
       "Name: mcWeight, dtype: float64"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#hide label and weights in separate vectors\n",
    "#they are not real features\n",
    "\n",
    "#WARNING : there should be no selection nor shuffling later on ! (otherwise misalignement)\n",
    "target = fulldata[\"label\"]\n",
    "weights = fulldata[\"mcWeight\"]\n",
    "\n",
    "\n",
    "# for simplicity of the exercise only keep some features\n",
    "# this is actually making a deep copy from fulldata\n",
    "data=pd.DataFrame(fulldata, columns=[\"met_et\",\"met_phi\",\"lep_pt_0\",\"lep_pt_1\",'lep_phi_0', 'lep_phi_1'])\n",
    "#data=pd.DataFrame(fulldata, columns=[\"met_et\",\"met_phi\",\"lep_pt_0\",\"lep_pt_1\",'lep_eta_0', 'lep_eta_1', 'lep_phi_0', 'lep_phi_1','jet_n','jet_pt_0',\n",
    "#       'jet_pt_1', 'jet_eta_0', 'jet_eta_1', 'jet_phi_0', 'jet_phi_1']\n",
    "\n",
    "print (\"Df shape of dataset to be used :\",data.shape)\n",
    "display(data.head())\n",
    "display(target.head())\n",
    "display(weights.head())\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:title={'center':'Missing Transverse Energy for large lepton Pt'}, ylabel='Frequency'>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#data[data.met_et<1000]['met_et'].plot.hist(title='Missing Transverse Energy')\n",
    "data[data.lep_pt_0+data.lep_pt_1>1000]['met_et'].plot.hist(bins=np.linspace(0,400,100),title='Missing Transverse Energy for large lepton Pt')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig=plt.figure()\n",
    "ax=data[target==0].plot.scatter(x='met_et', y='lep_pt_0',color=\"b\",label=\"B\")\n",
    "data[target==1].plot.scatter(x='met_et', y='lep_pt_0',color=\"r\",label=\"S\",ax=ax)\n",
    "#plt.legend(loc='best')\n",
    "#ax.set_xlabel('weight*1000')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "scrolled": true
   },
   "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>met_et</th>\n",
       "      <th>met_phi</th>\n",
       "      <th>lep_pt_0</th>\n",
       "      <th>lep_pt_1</th>\n",
       "      <th>lep_phi_0</th>\n",
       "      <th>lep_phi_1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>308053</th>\n",
       "      <td>2755.4</td>\n",
       "      <td>1.8538</td>\n",
       "      <td>2678.7</td>\n",
       "      <td>42.131</td>\n",
       "      <td>-1.2958</td>\n",
       "      <td>0.13151</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>552671</th>\n",
       "      <td>2145.7</td>\n",
       "      <td>0.3056</td>\n",
       "      <td>2180.0</td>\n",
       "      <td>30.531</td>\n",
       "      <td>-2.8257</td>\n",
       "      <td>1.50800</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        met_et  met_phi  lep_pt_0  lep_pt_1  lep_phi_0  lep_phi_1\n",
       "308053  2755.4   1.8538    2678.7    42.131    -1.2958    0.13151\n",
       "552671  2145.7   0.3056    2180.0    30.531    -2.8257    1.50800"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[data.lep_pt_0+data.lep_pt_1>2000].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "bVHrJtodL2wo"
   },
   "source": [
    "## Examine the weights"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 265
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 16493,
     "status": "ok",
     "timestamp": 1590131871121,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "cFD4s616LS0F",
    "outputId": "cb7c7295-9e44-433c-aa36-b67626e635a9"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig,ax=plt.subplots()\n",
    "#fig=plt.figure()\n",
    "\n",
    "bins=np.linspace(-1,1,101)\n",
    "plt.hist(weights[target==0]*1000,bins=bins,color='b',alpha=0.5,density=True,label='B ackground')\n",
    "plt.hist(weights[target==1]*10000,bins=bins,color='r',alpha=0.5,density=True,label='S ignal (Wx10)')\n",
    "plt.legend(loc='best')\n",
    "ax.set_xlabel('weight*1000')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Some weight studies \n",
    "\n",
    "$s=\\sum w$ for signal dataset : predicted number of signal events (luminosity, cross section, efficiencies etc... already includded in the weights). Ditto for b background. \n",
    "\n",
    "\n",
    "Effective number of events fraction : $\\frac{N_{eff}}{N}= \\frac{1}{1+\\frac{Var(w)}{<w>^2}}$ . Example : if 0.2 it means the precision achieved with this dataset is the one which would be achieved with an unweighted dataset of 0.2 x N events (this is a rough estimate, only true for a simple counting)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 121
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 16486,
     "status": "ok",
     "timestamp": 1590131871121,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "YdpzI4-6MVIK",
    "outputId": "96d931cf-302f-48de-e3cc-77eead2ba346"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Weights quantities for background (target==0) and signal (target==1)\n",
      "Weights sum [159.6884919    6.17939825]\n",
      "N events [180381. 396976.]\n",
      "Weights mean [8.85284436e-04 1.55661759e-05]\n",
      "Weights std [8.53093648e-04 1.35576170e-05]\n",
      "Weights Neff fraction [0.51851137 0.56863978]\n"
     ]
    }
   ],
   "source": [
    "label_n_weights=np.zeros(2)\n",
    "label_sum_weights=np.zeros(2)\n",
    "label_mean_weights=np.zeros(2)\n",
    "label_std_weights=np.zeros(2)\n",
    "label_neff_fraction=np.zeros(2)\n",
    "\n",
    "for i in range(2):\n",
    "  label_n_weights[i]=weights[target==i].size\n",
    "  label_mean_weights[i]=weights[target==i].mean()\n",
    "  label_std_weights[i]=weights[target==i].std()\n",
    "  label_sum_weights[i]=weights[target==i].sum()\n",
    "  label_neff_fraction[i]=1/(1+(label_std_weights[i]/label_mean_weights[i])**2)\n",
    "\n",
    "print (\"Weights quantities for background (target==0) and signal (target==1)\")\n",
    "print (\"Weights sum\",label_sum_weights)\n",
    "print (\"N events\",label_n_weights)\n",
    "print (\"Weights mean\",label_mean_weights)\n",
    "print (\"Weights std\",label_std_weights)\n",
    "print (\"Weights Neff fraction\",label_neff_fraction)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "F77Cx9LhKwqc"
   },
   "source": [
    "# Feature engineering\n",
    "To be switched on in a second iteration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "4oe_9ZEEKwqc"
   },
   "outputs": [],
   "source": [
    "if False: \n",
    "    data[\"lep_deltaphi\"]=np.abs(np.mod(data.lep_phi_1-data.lep_phi_0+3*np.pi,2*np.pi)-np.pi)\n",
    "    #data[\"lep_deltaphi\"]=data.lep_phi_1-data.lep_phi_0\n",
    "\n",
    "\n",
    "    print (data.shape)\n",
    "    display(data.head())\n",
    "   \n",
    "\n",
    "    \n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Plot the features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 734
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 17877,
     "status": "ok",
     "timestamp": 1590131872516,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "o2mf1bLVrlH7",
    "outputId": "135a52d1-8d7b-4c37-c430-52300696407c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x864 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure()\n",
    "\n",
    "ax=data[target==0].hist(weights=weights[target==0],figsize=(15,12),color='b',alpha=0.5,density=True,label=\"B\")\n",
    "ax=ax.flatten()[:data.shape[1]] # to avoid error if holes in the grid of plots (like if 7 or 8 features)\n",
    "data[target==1].hist(weights=weights[target==1],figsize=(15,12),color='r',alpha=0.5,density=True,ax=ax,label=\"S\")\n",
    "\n",
    "\n",
    "plt.legend(loc=\"best\")\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Features correlation matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Background feature correlation matrix\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Signal feature correlation matrix\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import seaborn as sn # seaborn for nice plot quicker\n",
    "print (\"Background feature correlation matrix\")\n",
    "corrMatrix = data[target==0].corr()\n",
    "sn.heatmap(corrMatrix, annot=True)\n",
    "plt.show()\n",
    "\n",
    "print (\"Signal feature correlation matrix\")\n",
    "corrMatrix = data[target==1].corr()\n",
    "sn.heatmap(corrMatrix, annot=True)\n",
    "plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "4kowHjX4rlIC"
   },
   "source": [
    "## Split dataset and transform the features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 207
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 18101,
     "status": "ok",
     "timestamp": 1590131872742,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "Y9j5hdrmrlID",
    "outputId": "a4de829a-842d-4fff-d93c-bd16ff6d51a2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(433017, 6)\n",
      "(433017,)\n",
      "(433017,)\n",
      "(144340, 6)\n",
      "(144340,)\n",
      "(144340,)\n",
      "Orig : total weight sig 6.179398249600002\n",
      "Orig : total weight bkg 159.68849189959997\n",
      "Test : total weight sig 6.179443329600001\n",
      "Test : total weight bkg 158.92470712440002\n",
      "Train : total weight sig 119.95731511850002\n",
      "Train : total weight bkg 119.95731511849999\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(31415) # set the random seed (used for the train/test splitting)\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "train_size = 0.75 # fraction of sample used for training\n",
    "\n",
    "X_train, X_test, y_train, y_test, weights_train, weights_test = \\\n",
    "    train_test_split(data, target, weights, train_size=train_size)\n",
    "#reset index for dataseries, not needed for ndarray (X_train, X_test)\n",
    "y_train, y_test, weights_train, weights_test = \\\n",
    "    y_train.reset_index(drop=True),y_test.reset_index(drop=True), \\\n",
    "    weights_train.reset_index(drop=True), weights_test.reset_index(drop=True)\n",
    "\n",
    "print (X_train.shape)\n",
    "print (y_train.shape)\n",
    "print (weights_train.shape)\n",
    "print (X_test.shape)\n",
    "print (y_test.shape)\n",
    "print (weights_test.shape)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "#scale to mean 0 and variance 1\n",
    "\n",
    "scaler = StandardScaler()\n",
    "X_train = scaler.fit_transform(X_train)\n",
    "X_test = scaler.transform(X_test)  # applied the transformation calculated the line above\n",
    "\n",
    "\n",
    "class_weights_train = (weights_train[y_train == 0].sum(), weights_train[y_train == 1].sum())\n",
    "\n",
    "for i in range(len(class_weights_train)):\n",
    "    #training dataset: equalize number of background and signal\n",
    "    weights_train[y_train == i] *= max(class_weights_train)/ class_weights_train[i] \n",
    "    #test dataset : increase test weight to compensate for sampling\n",
    "    weights_test[y_test == i] *= 1/(1-train_size) \n",
    "    \n",
    "print (\"Orig : total weight sig\", weights[target == 1].sum())\n",
    "print (\"Orig : total weight bkg\", weights[target == 0].sum())\n",
    "\n",
    "\n",
    "print (\"Test : total weight sig\", weights_test[y_test == 1].sum())\n",
    "print (\"Test : total weight bkg\", weights_test[y_test == 0].sum())\n",
    "print (\"Train : total weight sig\", weights_train[y_train == 1].sum())\n",
    "print (\"Train : total weight bkg\", weights_train[y_train == 0].sum())\n",
    "\n",
    "\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "yxybCOi-rlIM"
   },
   "source": [
    "# Testing BDT"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "AIsMSGl-Kwql"
   },
   "source": [
    "## Load significance function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "qubY3CMNKwql"
   },
   "outputs": [],
   "source": [
    "from math import sqrt\n",
    "from math import log\n",
    "def amsasimov(s,b): # asimov significance arXiv:1007.1727 eq. 97\n",
    "        if b<=0 or s<=0:\n",
    "            return 0\n",
    "        try:\n",
    "            return sqrt(2*((s+b)*log(1+float(s)/b)-s))\n",
    "        except ValueError:\n",
    "            print(1+float(s)/b)\n",
    "            print (2*((s+b)*log(1+float(s)/b)-s))\n",
    "        #return s/sqrt(s+b)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "F4DfF0ISrlIN"
   },
   "source": [
    "## XGBoost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 86
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 27562,
     "status": "ok",
     "timestamp": 1590131882207,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "nqMCgvbkrlIN",
    "outputId": "a99b92a1-fd10-459d-fb59-0876c527574d",
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training time: 2.837488889694214\n",
      "auc test: 0.8635099181960473\n",
      "auc train: 0.8642336729914726\n",
      "Z: 0.8015178701442962\n"
     ]
    }
   ],
   "source": [
    "np.random.seed(31415) # set the random seed\n",
    "\n",
    "from xgboost import XGBClassifier\n",
    "from sklearn.metrics import roc_auc_score # for binary classification if x > 0.5 -> 1 else -> 0\n",
    "# tree_method=\"hist\" is 10 times faster, however less robust against awkwards features (not a bad idea to double check without it)\n",
    "# can even try tree_method=\"gpu_hist\" if proper GPU installation\n",
    "# use_label_encoder and eval_metric to silence warning in 1.3.0\n",
    "xgb = XGBClassifier(tree_method=\"hist\",use_label_encoder=False,eval_metric='logloss')\n",
    "# HPO (==Hyper Parameter Optimization), check on the web https://xgboost.readthedocs.io/ for other parameters\n",
    "#xgb = XGBClassifier(tree_method=\"hist\",use_label_encoder=False,max_depth=10,n_estimators=100) \n",
    "\n",
    "\n",
    "\n",
    "starting_time = time.time()\n",
    "\n",
    "xgb.fit(X_train, y_train.values, sample_weight=weights_train.values) # note that XGB 1.3.X requires positive weight\n",
    "\n",
    "        \n",
    "training_time = time.time( ) - starting_time\n",
    "print(\"Training time:\",training_time)\n",
    "\n",
    "y_pred_xgb = xgb.predict_proba(X_test)[:,1]\n",
    "y_pred_xgb = y_pred_xgb.ravel()\n",
    "y_pred_train_xgb = xgb.predict_proba(X_train)[:,1].ravel()\n",
    "auc_test_xgb = roc_auc_score(y_true=y_test, y_score=y_pred_xgb,sample_weight=weights_test)\n",
    "print(\"auc test:\",auc_test_xgb)\n",
    "print (\"auc train:\",roc_auc_score(y_true=y_train.values, y_score=y_pred_train_xgb,))\n",
    "int_pred_test_sig_xgb = [weights_test[(y_test ==1) & (y_pred_xgb > th_cut)].sum() for th_cut in np.linspace(0,1,num=50)]\n",
    "int_pred_test_bkg_xgb = [weights_test[(y_test ==0) & (y_pred_xgb > th_cut)].sum() for th_cut in np.linspace(0,1,num=50)]\n",
    "\n",
    "vamsasimov_xgb = [amsasimov(sumsig,sumbkg) for (sumsig,sumbkg) in zip(int_pred_test_sig_xgb,int_pred_test_bkg_xgb)]\n",
    "significance_xgb = max(vamsasimov_xgb)\n",
    "Z = significance_xgb\n",
    "print(\"Z:\",Z)\n",
    "# To save model\n",
    "xgb.save_model(\"XGBoost.model\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Hyper Parameter Optimisation\n",
    "Can be done by hand, with [random search](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.RandomizedSearchCV.html) or [grid search](https://scikit-learn.org/stable/modules/generated/sklearn.model_selection.GridSearchCV.html) ."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "#RandomSearchCV for advanced HPO \n",
    "import scipy.stats as stats\n",
    "if False:\n",
    "    from sklearn.model_selection import RandomizedSearchCV\n",
    "\n",
    "    # specify parameters and distributions to sample from\n",
    "    param_dist_XGB = {'max_depth': stats.randint(2, 10), # default 6\n",
    "                      'n_estimators': stats.randint(10, 500), #default 100\n",
    "                      'learning_rate': stats.uniform(0.1, 0.5)} #def 0.3 \n",
    "\n",
    "    # default CV is 5 fold, reduce to 2 for speed concern\n",
    "    gsearch = RandomizedSearchCV(estimator = XGBClassifier(tree_method=\"hist\",use_label_encoder=False,eval_metric='logloss'), \n",
    "                        param_distributions = param_dist_XGB, \n",
    "                        scoring='roc_auc',n_iter=10,cv=2)\n",
    "    gsearch.fit(X_train,y_train, sample_weight=weights_train)\n",
    "\n",
    "    print (\"Best parameters : \",gsearch.best_params_)\n",
    "    print (\"Best score (on train dataset CV) : \",gsearch.best_score_)\n",
    "\n",
    "\n",
    "    y_pred_gs = gsearch.predict_proba(X_test)[:,1]\n",
    "    print(\"... corresponding score on test dataset : \",roc_auc_score(y_true=y_test, y_score=y_pred_gs, sample_weight=weights_test))\n",
    "    dfsearch=pd.DataFrame.from_dict(gsearch.cv_results_)\n",
    "    display(dfsearch)\n",
    "    dfsearch.plot.scatter(\"param_n_estimators\",\"mean_test_score\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "NmadcFRJrlIT"
   },
   "source": [
    "## LightGBM"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 104
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 9268,
     "status": "ok",
     "timestamp": 1590131954285,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "9YmfuiwUrlIU",
    "outputId": "5105d03a-1fac-4fb6-e3ae-21ebd16c688c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training time: 3.3410937786102295\n",
      "auc test: 0.8430003035698995\n",
      "auc train: 0.8477427479273085\n",
      "Z: 0.7534538457691546\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<lightgbm.basic.Booster at 0x7fb2a7915be0>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import lightgbm as lgb\n",
    "from sklearn.metrics import roc_auc_score # for binary classification if x > 0.5 -> 1 else -> 0\n",
    "#gbm = lgb.LGBMClassifier()\n",
    "gbm = lgb.LGBMClassifier()\n",
    "# gbm = lgb.LGBMClassifier(max_depth=12) # HPO, check on the web https://lightgbm.readthedocs.io/ for other parameters\n",
    "\n",
    "\n",
    "starting_time = time.time( )\n",
    "\n",
    "gbm.fit(X_train, y_train.values,sample_weight=weights_train.values)\n",
    "#gbm.fit(X_train, y_train.values) #ma\n",
    "\n",
    "\n",
    "training_time = time.time( ) - starting_time\n",
    "print(\"Training time:\",training_time)\n",
    "\n",
    "y_pred_gbm = gbm.predict_proba(X_test)[:,1]\n",
    "y_pred_gbm = y_pred_gbm.ravel()\n",
    "y_pred_train_gbm = gbm.predict_proba(X_train)[:,1].ravel()\n",
    "auc_test_gbm = roc_auc_score(y_true=y_test, y_score=y_pred_gbm)\n",
    "print(\"auc test:\",auc_test_gbm)\n",
    "print (\"auc train:\",roc_auc_score(y_true=y_train.values, y_score=y_pred_train_gbm,))\n",
    "\n",
    "int_pred_test_sig_gbm = [weights_test[(y_test ==1) & (y_pred_gbm > th_cut)].sum() for th_cut in np.linspace(0,1,num=50)]\n",
    "int_pred_test_bkg_gbm = [weights_test[(y_test ==0) & (y_pred_gbm > th_cut)].sum() for th_cut in np.linspace(0,1,num=50)]\n",
    "\n",
    "vamsasimov_gbm = [amsasimov(sumsig,sumbkg) for (sumsig,sumbkg) in zip(int_pred_test_sig_gbm,int_pred_test_bkg_gbm)]\n",
    "significance_gbm = max(vamsasimov_gbm)\n",
    "Z = significance_gbm\n",
    "print(\"Z:\",Z)\n",
    "# To save model\n",
    "gbm.booster_.save_model(\"LightGBM.model\")\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 121
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 727,
     "status": "ok",
     "timestamp": 1590131967829,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "zmJxYTUmrlIZ",
    "outputId": "3f0f7d15-d86c-4cc0-af5a-2a5eb94157ba"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best significance found are:\n",
      "LightGBM:  0.7534538457691546\n",
      "XGBoost :  0.8015178701442962\n",
      "Best auc test found are:\n",
      "LightGBM:  0.8430003035698995\n",
      "XGBoost:  0.8629082861653745\n"
     ]
    }
   ],
   "source": [
    "print('Best significance found are:')\n",
    "print('LightGBM: ', significance_gbm)\n",
    "print('XGBoost : ', significance_xgb)\n",
    "print('Best auc test found are:')\n",
    "print('LightGBM: ', roc_auc_score(y_true=y_test.values, y_score=y_pred_gbm,))\n",
    "print('XGBoost: ', roc_auc_score(y_true=y_test.values, y_score=y_pred_xgb,)) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "T-LB9cbErlIb"
   },
   "source": [
    "## Some nice plots "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "_nxVz32hKwqz"
   },
   "source": [
    "## load score plotting function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "colab": {},
    "colab_type": "code",
    "id": "Ay8QTiKPKwqz"
   },
   "outputs": [],
   "source": [
    "# some utilities\n",
    "from math import sqrt\n",
    "from math import log\n",
    "\n",
    "# Plot score for signal and background, comparing training and testing\n",
    "def compare_train_test(y_pred_train, y_train, y_pred, y_test, high_low=(0,1), bins=30, xlabel=\"\", ylabel=\"Arbitrary units\", title=\"\", weights_train=np.array([]), weights_test=np.array([])):\n",
    "    if weights_train.size != 0:\n",
    "        weights_train_signal = weights_train[y_train == 1]\n",
    "        weights_train_background = weights_train[y_train == 0]\n",
    "    else:\n",
    "        weights_train_signal = None\n",
    "        weights_train_background = None\n",
    "    plt.hist(y_pred_train[y_train == 1],\n",
    "                 color='r', alpha=0.5, range=high_low, bins=bins,\n",
    "                 histtype='stepfilled', density=True,\n",
    "                 label='S (train)', weights=weights_train_signal) # alpha is transparancy\n",
    "    plt.hist(y_pred_train[y_train == 0],\n",
    "                 color='b', alpha=0.5, range=high_low, bins=bins,\n",
    "                 histtype='stepfilled', density=True,\n",
    "                 label='B (train)', weights=weights_train_background)\n",
    "\n",
    "    if weights_test.size != 0:\n",
    "        weights_test_signal = weights_test[y_test == 1]\n",
    "        weights_test_background = weights_test[y_test == 0]\n",
    "    else:\n",
    "        weights_test_signal = None\n",
    "        weights_test_background = None\n",
    "    hist, bins = np.histogram(y_pred[y_test == 1],\n",
    "                                  bins=bins, range=high_low, density=True, weights=weights_test_signal)\n",
    "    scale = len(y_pred[y_test == 1]) / sum(hist)\n",
    "    err = np.sqrt(hist * scale) / scale\n",
    "\n",
    "    #width = (bins[1] - bins[0])\n",
    "    center = (bins[:-1] + bins[1:]) / 2\n",
    "    plt.errorbar(center, hist, yerr=err, fmt='o', c='r', label='S (test)')\n",
    "\n",
    "    hist, bins = np.histogram(y_pred[y_test == 0],\n",
    "                                  bins=bins, range=high_low, density=True, weights=weights_test_background)\n",
    "    scale = len(y_pred[y_test == 0]) / sum(hist)\n",
    "    err = np.sqrt(hist * scale) / scale\n",
    "\n",
    "    #width = (bins[1] - bins[0])\n",
    "    center = (bins[:-1] + bins[1:]) / 2\n",
    "    plt.errorbar(center, hist, yerr=err, fmt='o', c='b', label='B (test)')\n",
    "    plt.title(title)\n",
    "    plt.xlabel(xlabel)\n",
    "    plt.ylabel(ylabel)\n",
    "    plt.legend(loc='best')\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 573
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 1314,
     "status": "ok",
     "timestamp": 1590131974115,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "kEQ342s-rlIe",
    "outputId": "286cfcc6-f4e0-40f5-9a07-b4f91db18ff6"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "compare_train_test(y_pred_train_xgb, y_train, y_pred_xgb, y_test, xlabel=\"XGboost score\", title=\"XGboost\", weights_train=weights_train.values, weights_test=weights_test.values)\n",
    "plt.savefig(\"Score_BDT_XGBoost_Hist.pdf\")\n",
    "plt.show()\n",
    "compare_train_test(y_pred_train_gbm, y_train, y_pred_gbm, y_test, xlabel=\"LightGBM score\", title=\"LightGBM\", weights_train=weights_train.values, weights_test=weights_test.values)\n",
    "plt.savefig(\"Score_BDT_LightGBM.pdf\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 295
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 883,
     "status": "ok",
     "timestamp": 1590131979941,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "tUKIWu2orlIg",
    "outputId": "426e9f5f-2ff4-477f-d977-94be70d6c445"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import roc_curve\n",
    "lw = 2\n",
    "\n",
    "fpr_gbm,tpr_gbm,_ = roc_curve(y_true=y_test, y_score=y_pred_gbm,sample_weight=weights_test.values)\n",
    "fpr_xgb,tpr_xgb,_ = roc_curve(y_true=y_test, y_score=y_pred_xgb,sample_weight=weights_test.values)\n",
    "plt.plot(fpr_gbm, tpr_gbm, color='darkorange',lw=lw, label='LightGBM (AUC  = {})'.format(np.round(auc_test_gbm,decimals=2)))\n",
    "plt.plot(fpr_xgb, tpr_xgb, color='darkgreen',lw=lw, label='XGBoost (AUC  = {})'.format(np.round(auc_test_xgb,decimals=2)))\n",
    "plt.plot([0, 1], [0, 1], color='navy', lw=lw, linestyle='--')\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('Receiver Operating Characteristic')\n",
    "plt.legend(loc=\"lower right\")\n",
    "#import os\n",
    "#new_dir = \"Plots/Comparing\" \n",
    "#if not os.path.isdir(new_dir):\n",
    "#    os.mkdir(new_dir)\n",
    "plt.savefig(\"ROC_comparing.pdf\")\n",
    "plt.show() # blue line = random classification -> maximize true positive rate while miniize false positive rate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 295
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 835,
     "status": "ok",
     "timestamp": 1590131983880,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "CGF3k0KJrlIi",
    "outputId": "daf7acec-b3af-4941-d924-3b3d1be4dd65"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(np.linspace(0,1,num=50),vamsasimov_gbm, label='LightGBM (Z = {})'.format(np.round(significance_gbm,decimals=2)))\n",
    "plt.plot(np.linspace(0,1,num=50),vamsasimov_xgb, label='XGBoost (Z = {})'.format(np.round(significance_xgb,decimals=2)))\n",
    "\n",
    "plt.title(\"BDT Significance\")\n",
    "plt.xlabel(\"Threshold\")\n",
    "plt.ylabel(\"Significance\")\n",
    "plt.legend()\n",
    "plt.savefig(\"Significance_comparing.pdf\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feature importance\n",
    "Feature importance allows to display the importance of each feature without rerunnning the training. It is obtained from internal algorithm quantities, like number of time a feature is used to definea leaf. Magnitude is arbitrary. It can be used as a not very reliable indication of which feature is the most discriminant."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 619
    },
    "colab_type": "code",
    "executionInfo": {
     "elapsed": 855,
     "status": "ok",
     "timestamp": 1590131986197,
     "user": {
      "displayName": "David Rousseau",
      "photoUrl": "",
      "userId": "13371246477126231971"
     },
     "user_tz": -120
    },
    "id": "QN2h9y5JrlIm",
    "outputId": "5d160a28-f3a1-41d8-cdcc-daa1e5b2b7a7"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXcAAAEtCAYAAADz1SBvAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAAc40lEQVR4nO3dfbxdVX3n8c+XgMAgCJQkDQka1AwaHAEngyh0ZEQFpBjqiIaOGiiKjvjQqnWC7YxomxrnpY61lU7xqVHUND4Si63SKLaKEoKCGgISIZKQmFxBEB+IDX77x17XnNyce++5956bc8/K9/163dc5Z+19zv6t+/A9+6y99r6yTURE1GW/XhcQERHdl3CPiKhQwj0iokIJ94iICiXcIyIqlHCPiKhQwj16QtKbJX2g13XURtI6Sad3uO5GSc+a3IqiVxLufaj8Uf5S0s9avo7uwmvutT90239h+2V7a3sjkXS5pKt6XUenRqrX9vG2r+vCNk6XtLlN+zxJKyQNSPqppDsk/ZWkOS3P+3XL7+U9kt465DUsaZuk/Vva9pe0XVJOvOmShHv/Otf2I1u+tvSymNY/1H7Sr3X3gqTHAzcAW4CTbB8GnAr8ADitZdUtg7+Xpf1iSecNebn7gbNbHj8X+Mkklb5PSrhXRNKjJH1Q0tayx/TnkqaVZY+T9GVJ90r6saSPSTq8LPso8Gjg82Vv603t9txa9+7L3uOnJF0l6afAhSNtv02tv9n7lDS37M1dJGmTpJ9IeqWk/yLpO5Lul/TXLc+9UNLXyx7jA5Juk3RGy/KjJa2SdJ+kDZJePmS7rXW/Engz8KLS91vKehdJWi/pQUl3SnpFy2ucLmmzpDeUvc2tki5qWX6wpHdJ+mGp72uSDi7LTpF0fenTLa1DKKVfd5Zt3iXpf4zpF2DPn9HBkpaX7+f68nMdujd+YvkePyDp7yUdJOkQ4B+Bo7X7J8PLga/bfr3tzQC2t9t+j+0V7eqxfRdwPTB/yKKPAi9tefxS4CNj7W+MwHa++uwL2Ag8q03754C/BQ4BZgBrgFeUZY8Hng0cCEwH/gV4z3CvCZwObB5uuzR/6P8GnEezk3DwSNtvU+vlwFXl/lzAwP8HDgKeAzxUXm8GMBvYDjyjrH8hsBP4I+AA4EXAA8CRZflXgSvKa50IDABnjFD3b2ppqe8c4HGAgGcAvwCe0vK92Qm8rWz/uWX5EWX5+4DrSt3TgKeX7/ts4N6y/n7l53Fv+XkcAvwUOK68xizg+NG+dyP9bgDLyvfiCGAO8J3Wn2lZdw1wNHAksB545Qg//x8BF47yu3n6kG3MA+4BntnSZuBJwDbg8PK1rbS5139ftXxlz71/fa7s/d0v6XOSZtJ8zP1D2z+3vR34f8AiANsbbF9re4ftAeDdNKE1Ed+w/TnbvwYOG2n7Hfoz2w/Z/hLwc+ATbvYM7wH+FTipZd3tNG9O/2b774HbgXMkHUMzFPC/ymvdDHwAeEm7um3/sl0htq+x/QM3vgp8CfidllX+DXhb2f4XgJ8Bx0naD/gD4HW277H9sO3rbe8AXgx8wfYXyravBdbShD3Ar4EnSTrY9lbb68bwvWvnhcBf2P6Jmz3t97ZZ5722t9i+D/g8zZvhcI6iCXgAJL26/P79TNL7W9Y7urT/FPg+zVDO14a81kNley+i+R1ZVdqiSxLu/es824eXr/OAx9DsRW4dDH2avegZAJJmqDkQdk/5o7uK5o91Ija13B9x+x3a1nL/l20eP7Ll8T22Ww++/ZBmD/Ro4D7bDw5ZNnuYutuSdLakb5ahnftpArj1+3Wv7Z0tj39R6juK5hPDD9q87GOA81velO+neSOaZfvnNEH3Sprv4TWSnjBanaM4mt372q7fP2q5P9iH4dxL84kCANt/bftw4D00P/tBW8rv5WE0e+W/BJa3eb2P0AzHZEhmEiTc67EJ2AEc1RL6h9k+vix/O83H4SeXP7oX0ww5DBo6S+HnwH8YfFDGzqcPWaf1OaNtv9tmS2qt/9E0B/q2AEdKOnTIsnuGqXuPx5IOBD4NvBOYWQLsC+z+/RrOj2n2QB/XZtkm4KMt35/DbR9iexmA7S/afjZNgN4GvL/Na4zFVprhmEHHjOG57WatrAaeP5YCbD8AfBw4t83if6Xp60z23LOPCUq4V8L2Vpqhg3dJOkzSfmoOog4OvRxKM3Rwv6TZwB8PeYltwGNbHn8fOEjSOZIOAP6UZtx4vNvvthnAayUdIOl84Ik0Qx6baA7gvb0cHHwycDHwsRFeaxswtwypADyCpq8DwE5JZ9McBxhVGaL6EPDucmB3mqSnlTeMq4BzJZ1Z2g8qB2fnSJop6XnlYOYOmp/VwyNsar/y/MGvdj+blcBlko4oP/NXd9KHYhvwW5Ie1dJ2OfA7kt5dXg9JR9F879uS9EiaYZc9hpjKJ69zgecN+RQWXZBwr8tLaYLpVpppZZ9i18fotwJPoTnweA3wmSHPfTvwp2W44I1lj+tVNOPV99Dsye8x73kM2++2G2gO1v0YWAq8wPa9ZdkFNAdptwCfBd5SxreH88lye6+kb5UhndfShONPgN+nGRPu1BuB7wI3AvcB7wD2K288C2lm5wzQ7Mn/Mc3f4X7AG0rN99EcD3nVCNu4gGa4Y/Cr3TDQ22h+ZncB/0zz89jRSQds3wZ8Ariz/E4cbfv7wCk0nwZukfQg8PVS8/9uefpvZtnQDIkdCbSd+WN7XReOLUQbyhtm9BtJFwIvs33aaOvGLpL+J7DI9mR9moopJHvuEZWSNEvSqWWI7DiaTwaf7XVdsXfk7LyIej2CZsbSsTRnhK6gmf8f+4AMy0REVCjDMhERFZoSwzJHHXWU586d2+syIiL6yk033fRj20PPPwGmSLjPnTuXtWvX9rqMiIi+IumHwy3LsExERIUS7hERFUq4R0RUKOEeEVGhhHtERIUS7hERFUq4R0RUKOEeEVGhhHtERIWmxBmqEzV3yTW9LqEjG5ed0+sSImIfkT33iIgKJdwjIiqUcI+IqFDCPSKiQgn3iIgKJdwjIiqUcI+IqFDCPSKiQgn3iIgKdRTukg6X9ClJt0laL+lpko6UdK2kO8rtES3rXyZpg6TbJZ05eeVHREQ7ne65/yXwT7afAJwArAeWAKttzwNWl8dImg8sAo4HzgKukDSt24VHRMTwRg13SYcB/xX4IIDtX9m+H1gILC+rLQfOK/cXAits77B9F7ABOLm7ZUdExEg62XN/LDAAfFjStyV9QNIhwEzbWwHK7Yyy/mxgU8vzN5e23Ui6RNJaSWsHBgYm1ImIiNhdJ+G+P/AU4G9snwT8nDIEMwy1afMeDfaVthfYXjB9+vSOio2IiM50Eu6bgc22byiPP0UT9tskzQIot9tb1j+m5flzgC3dKTciIjoxarjb/hGwSdJxpekM4FZgFbC4tC0Gri73VwGLJB0o6VhgHrCmq1VHRMSIOv1nHa8BPibpEcCdwEU0bwwrJV0M3A2cD2B7naSVNG8AO4FLbT/c9cojImJYHYW77ZuBBW0WnTHM+kuBpeMvKyIiJiJnqEZEVCjhHhFRoYR7RESFEu4RERVKuEdEVCjhHhFRoYR7RESFOj2JKWLc5i65ptcldGTjsnN6XUJE1yTcI8Yhb1gx1WVYJiKiQgn3iIgKJdwjIiqUcI+IqFDCPSKiQgn3iIgKJdwjIiqUcI+IqFDCPSKiQgn3iIgKJdwjIiqUcI+IqFDCPSKiQgn3iIgKJdwjIirUUbhL2ijpu5JulrS2tB0p6VpJd5TbI1rWv0zSBkm3SzpzsoqPiIj2xrLn/t9sn2h7QXm8BFhtex6wujxG0nxgEXA8cBZwhaRpXaw5IiJGMZFhmYXA8nJ/OXBeS/sK2zts3wVsAE6ewHYiImKMOg13A1+SdJOkS0rbTNtbAcrtjNI+G9jU8tzNpW03ki6RtFbS2oGBgfFVHxERbXX6P1RPtb1F0gzgWkm3jbCu2rR5jwb7SuBKgAULFuyxPCIixq+jPXfbW8rtduCzNMMs2yTNAii328vqm4FjWp4+B9jSrYIjImJ0o4a7pEMkHTp4H3gO8D1gFbC4rLYYuLrcXwUsknSgpGOBecCabhceERHD62RYZibwWUmD63/c9j9JuhFYKeli4G7gfADb6yStBG4FdgKX2n54UqqPiIi2Rg1323cCJ7Rpvxc4Y5jnLAWWTri6iIgYl04PqEZExeYuuabXJXRk47Jzel1C38jlByIiKpRwj4ioUMI9IqJCGXOPiCrt68cRsuceEVGhhHtERIUS7hERFUq4R0RUKOEeEVGhhHtERIUS7hERFUq4R0RUKOEeEVGhhHtERIUS7hERFUq4R0RUKOEeEVGhhHtERIUS7hERFUq4R0RUKOEeEVGhhHtERIU6DndJ0yR9W9I/lMdHSrpW0h3l9oiWdS+TtEHS7ZLOnIzCIyJieGPZc38dsL7l8RJgte15wOryGEnzgUXA8cBZwBWSpnWn3IiI6ERH4S5pDnAO8IGW5oXA8nJ/OXBeS/sK2zts3wVsAE7uSrUREdGRTvfc3wO8Cfh1S9tM21sByu2M0j4b2NSy3ubSthtJl0haK2ntwMDAWOuOiIgRjBrukn4X2G77pg5fU23avEeDfaXtBbYXTJ8+vcOXjoiITuzfwTqnAs+T9FzgIOAwSVcB2yTNsr1V0ixge1l/M3BMy/PnAFu6WXRERIxs1D1325fZnmN7Ls2B0i/bfjGwClhcVlsMXF3urwIWSTpQ0rHAPGBN1yuPiIhhdbLnPpxlwEpJFwN3A+cD2F4naSVwK7ATuNT2wxOuNCIiOjamcLd9HXBduX8vcMYw6y0Flk6wtoiIGKecoRoRUaGEe0REhRLuEREVSrhHRFQo4R4RUaGEe0REhRLuEREVSrhHRFQo4R4RUaGEe0REhRLuEREVSrhHRFQo4R4RUaGEe0REhRLuEREVSrhHRFQo4R4RUaGEe0REhRLuEREVSrhHRFQo4R4RUaGEe0REhRLuEREVGjXcJR0kaY2kWyStk/TW0n6kpGsl3VFuj2h5zmWSNki6XdKZk9mBiIjYUyd77juAZ9o+ATgROEvSKcASYLXtecDq8hhJ84FFwPHAWcAVkqZNQu0RETGMUcPdjZ+VhweULwMLgeWlfTlwXrm/EFhhe4ftu4ANwMndLDoiIkbW0Zi7pGmSbga2A9favgGYaXsrQLmdUVafDWxqefrm0jb0NS+RtFbS2oGBgQl0ISIihuoo3G0/bPtEYA5wsqQnjbC62r1Em9e80vYC2wumT5/eUbEREdGZMc2WsX0/cB3NWPo2SbMAyu32stpm4JiWp80Btky00IiI6Fwns2WmSzq83D8YeBZwG7AKWFxWWwxcXe6vAhZJOlDSscA8YE2X646IiBHs38E6s4DlZcbLfsBK2/8g6RvASkkXA3cD5wPYXidpJXArsBO41PbDk1N+neYuuabXJXRk47Jzel1CRAxj1HC3/R3gpDbt9wJnDPOcpcDSCVcXERHjkjNUIyIqlHCPiKhQwj0iokIJ94iICiXcIyIqlHCPiKhQwj0iokIJ94iICiXcIyIqlHCPiKhQwj0iokIJ94iICiXcIyIqlHCPiKhQwj0iokIJ94iICiXcIyIqlHCPiKhQwj0iokIJ94iICiXcIyIqlHCPiKhQwj0iokKjhrukYyR9RdJ6Seskva60HynpWkl3lNsjWp5zmaQNkm6XdOZkdiAiIvbUyZ77TuANtp8InAJcKmk+sARYbXsesLo8pixbBBwPnAVcIWnaZBQfERHtjRrutrfa/la5/yCwHpgNLASWl9WWA+eV+wuBFbZ32L4L2ACc3OW6IyJiBGMac5c0FzgJuAGYaXsrNG8AwIyy2mxgU8vTNpe2oa91iaS1ktYODAyMo/SIiBhOx+Eu6ZHAp4E/tP3TkVZt0+Y9GuwrbS+wvWD69OmdlhERER3oKNwlHUAT7B+z/ZnSvE3SrLJ8FrC9tG8Gjml5+hxgS3fKjYiITnQyW0bAB4H1tt/dsmgVsLjcXwxc3dK+SNKBko4F5gFruldyRESMZv8O1jkVeAnwXUk3l7Y3A8uAlZIuBu4GzgewvU7SSuBWmpk2l9p+uNuFR0TE8EYNd9tfo/04OsAZwzxnKbB0AnVFRMQE5AzViIgKJdwjIiqUcI+IqFDCPSKiQgn3iIgKJdwjIiqUcI+IqFDCPSKiQgn3iIgKJdwjIiqUcI+IqFDCPSKiQgn3iIgKJdwjIiqUcI+IqFDCPSKiQgn3iIgKJdwjIiqUcI+IqFDCPSKiQgn3iIgKJdwjIiqUcI+IqNCo4S7pQ5K2S/peS9uRkq6VdEe5PaJl2WWSNki6XdKZk1V4REQMr5M9978DzhrStgRYbXsesLo8RtJ8YBFwfHnOFZKmda3aiIjoyKjhbvtfgPuGNC8Elpf7y4HzWtpX2N5h+y5gA3Byd0qNiIhOjXfMfabtrQDldkZpnw1sallvc2nbg6RLJK2VtHZgYGCcZURERDvdPqCqNm1ut6LtK20vsL1g+vTpXS4jImLfNt5w3yZpFkC53V7aNwPHtKw3B9gy/vIiImI8xhvuq4DF5f5i4OqW9kWSDpR0LDAPWDOxEiMiYqz2H20FSZ8ATgeOkrQZeAuwDFgp6WLgbuB8ANvrJK0EbgV2ApfafniSao+IiGGMGu62Lxhm0RnDrL8UWDqRoiIiYmJyhmpERIUS7hERFUq4R0RUKOEeEVGhhHtERIUS7hERFUq4R0RUKOEeEVGhhHtERIUS7hERFUq4R0RUKOEeEVGhhHtERIUS7hERFUq4R0RUKOEeEVGhhHtERIUS7hERFUq4R0RUKOEeEVGhhHtERIUS7hERFUq4R0RUKOEeEVGhSQt3SWdJul3SBklLJms7ERGxp0kJd0nTgPcBZwPzgQskzZ+MbUVExJ4ma8/9ZGCD7Ttt/wpYASycpG1FRMQQst39F5VeAJxl+2Xl8UuAp9p+dcs6lwCXlIfHAbd3vZCJOQr4ca+L6KL0Z+qrrU+19QemXp8eY3t6uwX7T9IG1aZtt3cR21cCV07S9idM0lrbC3pdR7ekP1NfbX2qrT/QX32arGGZzcAxLY/nAFsmaVsRETHEZIX7jcA8ScdKegSwCFg1SduKiIghJmVYxvZOSa8GvghMAz5ke91kbGsSTdkho3FKf6a+2vpUW3+gj/o0KQdUIyKit3KGakREhRLuEREVSrhHRFQo4V5IOr+TtoiIfpADqoWkb9l+ymht/UDSE2zfJqlt7ba/tbdrmihJjwIuA84DBs/I2w5cDSyzfX9vKus+Sf9o++xe1zEekkRz+ZHZNCcubgHWuLKgGfwb63UdI5msM1T7hqSzgecCsyW9t2XRYcDO3lQ1Ya+nubTDu9osM/DMvVtOV6wEvgycbvtHAJJ+G1gMfBJ4dg9rG7Ph3nhpzu4+cS+W0jWSngNcAdwB3FOa5wCPl/Qq21/qWXHd9yXg0b0uYiT7/J67pBNo/pjeBvyflkUPAl+x/ZNe1BW7k3S77ePGumyqkvQw8FXaX6rjFNsH7+WSJkzSeuBs2xuHtB8LfMH2E3tS2DgN2dnbbRGw2PZhe7Oesdrn99xt3wLcIunjNN+PR9ueahcxGzdJTwfm0vKztv2RnhU0fj+U9CZgue1tAJJmAhcCm3pZ2DitB15h+46hCyT1Y3+g+R3b3Kb9HuCAvVxLN1wEvAHY0WbZBXu5ljHb58O9xVnAO4FHAMdKOhF4m+3n9bSqCZD0UeBxwM3Aw6XZQD+G+4uAJcBXJc0obdtoLmvxwp5VNX6XM/yEhtfsxTq66UPAjZJWsOsN9xiay498sGdVjd+NwPdsXz90gaTL9345Y7PPD8sMknQTzVj0dbZPKm3fsf3k3lY2fuVj8vzaDmaNRNJi28t7XUe39Ft/JD2R5n83zKYZvtgMrLJ9a08LGwdJRwIP2f5Fr2sZj+y577LT9gPNwf5qfA/4bWBrrwvZi14H9E0YdqCv+mN7Pc2Q07Akfdr2f99LJY2b7fs6WW+q9ifhvsv3JP0+ME3SPOC1wB4fx/qBpM/TDL8cCtwqaQ0t44b9PNTUgarenamvPwCP7XUBXTYl+5Nw3+U1wJ/QhODHaa5o+ec9rWj83tnrAnqotiGo2voD9fVpSvYn4V6UcbU/KV97kPRXtvviQJftrw7eL3PBT6b5BbxxcI54xWrb062tP7GX5PIDnTu11wWMlaSXAWuA5wMvAL4p6Q96W9XElDnTI7V9fS+WM2G19adDtb1hTcn+ZLZMh/rxUgSSbgeebvve8vi3gOv77YSfVsNcJuIm2/+5VzVNRG396YSk59R0tupU7U+GZeq2meZM20EP0p8n/CDpCcDxwKMkPb9l0WHAQb2pavxq6w+ApJW2Xyjpu+w+Di3Ag9OKp2IQttPv/Um4d25KfvQaxT3ADZKupvnlXAiskfR6ANvv7mVxY3Qc8LvA4cC5Le0PAi/vRUETVFt/oJm2CU2/atDX/cmwTCHpfNufHK5N0oW2/64nxY2TpLeMtNz2W/dWLd0i6Wm2v9HrOrqltv7E1JFwL2q65G+n+mkG0CBJjwX+EjiF5tPIN4A/sn1nTwsbp9r6A1CGmd4BzKD5xDs4jDGlL7Q1nH7tzz4/LFPpJX871XczgGjOQXgf8Hvl8SLgE8BTe1bRxNTWH4D/C5xbzlatQV/2J1Mhm38msBZ4CLip5WsVcGYP64r2ZPujtneWr6uYoieRdKi2/gBs67cgHEVf9ifDMoWkA6jwkr8j6cdhJ0nLgPuBFTQh+CLgQJq9346vBzJV1NSfllk/z6C5ptHn2P2yF5/pQVnj1u/9SbgXks6lXPLXdhWX/B2NpG8PXgGzX0i6a4TFtj0lr/MxnJr6I+nDIyy27b46ga7f+5NwLyq95G91M4BGI+nZtq/tdR3dUlt/Yu/JmPsuO20/0OsiuuyykdpqC/biHb0uoMv6rj+SHivp85IGJG2XdHW7yyz0i37tzz4/W6ZFTZf83ZdnAPXjyWYj6cf+tJsBtIL+nQHUl/3Jnvsur6E5HXzwkr8PsOsMtX6zL88Aqm2csR/7U9sMoL7sT8bcC0kLaC73O5ddn2jc52PumQHU5/qxPzXNAIL+7U/CvShXUHwjzb+m+/Vgu+0f9qyoCdpHZwB9xvbzR1+zP/Rjf2qaAQT925+EeyHpa7ZP63Ud3VTpDKCDgFcBp9HsRX0N+BvbD/W0sHGqrT+dqG0G0FTtT8K9kHQGcAGwmj46UWEkkm6w/dTW+ewVhPtKmisnXlWaLgCOsH1+76oav9r604l+HGoayVTtT2bL7HIR8ATgAHYNyxjo23CnohlALY6zfULL469IuqVn1Uxcbf3pRD/OABrJlOxPwn2XE2z/p14X0WXt/un3n/W0oon7tqRTbH8TQNJT6e9/RVdbfzpR23DBlOxPpkLu8k1J83tdRJfNL1/70/x3n4XAjT2taOKeClwvaaOkjTSXyH2GpO9K+k5vSxuX2voTU0T23Hc5DVhcjozvYMi/0upTH6PNDKA+d1avC+iy2vrTiY29LqDLNva6gHZyQLWQ9Jh27X0+FbK6GUAAkk4D5tn+sKSjgENtjzRdbUqrsD9VzQDq1/4k3CtW6QygtwALaA5E/kdJRwOftN2P/3ikuv5AfTOA+rU/GZapW40zgH4POAn4FoDtLZIO7W1JE1Jbf6C+GUB92Z+Ee91qnAH0K9uWZABJh/S6oAmqrT9Q3wygvuxPZsvUrcYZQCsl/S1wuKSXA/8MvL/HNU1Ebf2B+mYA9WV/MuZeMUnrgccBNc0AQtKzgefQ9OeLU/HU77GosD9tJycM6rdJCv3an4R7xWqcART9ocIZQH3Xn4R79AVJD9L+TMDBTyOH7eWSJqS2/rSqbQZQv/YnB1SjL9ju9xkku6mtP0PUNgOoL/uTA6oR0W2/cjMkUMsMoL7sT8I9IrqtthlAfdmfjLlHRNdVOAOo7/qTcI+IqFAOqEZEV9Q2A6jf+5M994iICuWAakREhRLuEREVSrhHRFQo4R4RUaF/B9EuciS6TJipAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.bar(data.columns.values, xgb.feature_importances_)\n",
    "plt.xticks(rotation=90)\n",
    "plt.title(\"Feature importances XGBoost Hist\")\n",
    "#plt.savefig(new_dir + \"/VarImp_BDT_XGBoost_Hist.pdf\",bbox_inches='tight')\n",
    "plt.show()\n",
    "plt.bar(data.columns.values, gbm.feature_importances_)\n",
    "plt.xticks(rotation=90)\n",
    "plt.title(\"Feature importances LightGBM\")\n",
    "#plt.savefig(new_dir + \"/VarImp_BDT_LightGBM.pdf\",bbox_inches='tight')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "colab_type": "text",
    "id": "i7WxH3bDrlIp"
   },
   "source": [
    "# Permutation importance"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A better way to show the importance of each feature is Permutation Importance, where each feature in turn is replaced by an instance of an other event (effectively switching it off by randomising).\n",
    "One possible implementation:\n",
    "https://github.com/aghoshpub/permutationImportancePhysics \n",
    "In particular it allows to : \n",
    "   * display directly the loss in whatever criteria (ROC auc, asimov significance) when the feature is switched off\n",
    "   * display the feature importance for a specific subset (for example the most signal like)\n",
    "   * it can even display which feature has the larges impact on systematics\n",
    "\n",
    "\n",
    "However, report can be misleading in case of highly correlated variables. \n",
    "   \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "if False:\n",
    "    !pip install PermutationImportancePhysics\n",
    "    from permutationimportancephysics.PermutationImportance import PermulationImportance # note the delibrate typo PermuLation\n",
    "    PI_xgb = PermulationImportance(model=xgb, X=X_test,y=y_test,weights=weights_test,\\\n",
    "                           n_iterations=1,usePredict_poba=True, scoreFunction=\"amsasimov\", colNames=list(data.columns.values))\n",
    "    #PI_xgb.dislayResults()\n",
    "    plott_xgb = PI_xgb.plotBars()\n",
    "    plott_xgb.show()\n",
    "    PI_gbm = PermulationImportance(model=gbm, X=X_test,y=y_test,weights=weights_test,\\\n",
    "                           n_iterations=1,usePredict_poba=True, scoreFunction=\"amsasimov\", colNames=list(data.columns.values))\n",
    "    #PI_gbm.dislayResults()\n",
    "    plott_gbm = PI_gbm.plotBars()\n",
    "    plott_gbm.show()\n"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [],
   "name": "HEPML_HandsOn_BDT.ipynb",
   "provenance": []
  },
  "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}