deeplearning_intro.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "<center>\n",
    "<h1>Introduction to Deep Learning with Keras</h1>\n",
    "<h3>Michela Paganini - Yale University</h3>\n",
    "<h4>High Energy Phenomenology, Experiment and Cosmology Seminar Series</h4>\n",
    "<img src='http://www.edustart.org/wp-content/uploads/2014/03/MIST-Vert-Aggie-Maroon-page-001-300x200.jpg'>\n",
    "</center>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is the second of two notebooks for this tutorial. If you are unfamiliar with data handling techniques in a Python environment, see the <a href=\"data_handling.ipynb\">previous notebook</a>."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## What is Keras? Why Using Keras?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " *  Modular, powerful and intuitive Deep Learning Python library built on \n",
    " \n",
    " <center>\n",
    " <a href='http://deeplearning.net/software/theano/'><img src='./images/theano-logo.png' style=\"height:50px; display:inline;\"></a> and\n",
    " <a href='https://www.tensorflow.org/'><img src='./images/tf-logo.png' style=\"height:60px; display:inline\"></a> and\n",
    " <a href='https://docs.microsoft.com/en-us/cognitive-toolkit/'><img src='https://developer.nvidia.com/sites/default/files/akamai/cuda/images/deeplearning/cntk.png' style=\"height:80px; display:inline\"></a>\n",
    " </center>\n",
    "     * Let you call C++ efficiently while using syntax of higher level programming language\n",
    "     * Offer arbitrary tensor function definition, symbolic graph building, autodifferentiation, optimization, parallelism\n",
    "\n",
    " * Keras doesn’t itself handle any tensor ops; it relies on these tensor manipulation libraries\n",
    " * Connects with both of them via the abstract <a href='http://keras.io/backend/'>Keras backend</a>\n",
    " * Provides you with high level API, simple neural network building blocks\n",
    " * Now integral part of TensorFlow via <a href=\"https://www.tensorflow.org/api_docs/python/tf/keras\">`tf.keras`</a> module\n",
    " * Model and data parallelism across different devices"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Other Properties of Keras \n",
    "> <i>Developed with a focus on enabling fast experimentation. Being able to go from idea to result with the least possible delay is key to doing good research. </i>\n",
    "<div align=\"right\">\n",
    "  https://keras.io\n",
    "</div>\n",
    "\n",
    "*  Minimalist, user-friendly interface\n",
    "*  Extremely well documented, lots of <a href='https://github.com/fchollet/keras/tree/master/examples'>working examples</a>\n",
    "*  Very shallow learning curve $\\rightarrow$ by far one of the best tools for both beginners and experts\n",
    "*  Open-source, developed and maintained by a community of contributors, and publicly hosted on <a href='https://github.com/fchollet/keras'>GitHub</a>\n",
    "*  Extensible: possibility to customize layers\n",
    " \n",
    "From the Keras website: \n",
    "<img src='./images/keras_principles.jpg' style=\"width:800px;\">"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load and Plot Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's load the utility functions we defined in the <a href=\"root_numpy_pandas.ipynb\">previous notebook</a>."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "\n",
       "require(['notebook'],\n",
       "  function() {\n",
       "    IPython.CodeCell.config_defaults.highlight_modes['magic_text/x-c++src'] = {'reg':[/^%%cpp/]};\n",
       "    console.log(\"JupyROOT - %%cpp magic configured\");\n",
       "  }\n",
       ");\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Welcome to JupyROOT 6.08/02\n"
     ]
    }
   ],
   "source": [
    "import glob\n",
    "from root_numpy import root2array\n",
    "from numpy.lib.recfunctions import stack_arrays\n",
    "\n",
    "def root2pandas(files_path, tree_name, **kwargs):\n",
    "    '''\n",
    "    Args:\n",
    "    -----\n",
    "        files_path: a string like './data/*.root', for example\n",
    "        tree_name: a string like 'Collection_Tree' corresponding to the name of the folder inside the root \n",
    "                   file that we want to open\n",
    "        kwargs: arguments taken by root2array, such as branches to consider, start, stop, step, etc\n",
    "    Returns:\n",
    "    --------    \n",
    "        output_panda: a pandas dataframe like allbkg_df in which all the info from the root file will be stored\n",
    "    \n",
    "    Note:\n",
    "    -----\n",
    "        if you are working with .root files that contain different branches, you might have to mask your data\n",
    "        in that case, return pd.DataFrame(ss.data)\n",
    "    '''\n",
    "    # -- create list of .root files to process\n",
    "    files = glob.glob(files_path)\n",
    "    \n",
    "    # -- process ntuples into rec arrays\n",
    "    ss = stack_arrays([root2array(fpath, tree_name, **kwargs).view(np.recarray) for fpath in files])\n",
    "\n",
    "    try:\n",
    "        return pd.DataFrame(ss)\n",
    "    except Exception:\n",
    "        return pd.DataFrame(ss.data)\n",
    "    \n",
    "    \n",
    "def flatten(column):\n",
    "    '''\n",
    "    Args:\n",
    "    -----\n",
    "        column: a column of a pandas df whose entries are lists (or regular entries -- in which case nothing is done)\n",
    "                e.g.: my_df['some_variable'] \n",
    "\n",
    "    Returns:\n",
    "    --------    \n",
    "        flattened out version of the column. \n",
    "\n",
    "        For example, it will turn:\n",
    "        [1791, 2719, 1891]\n",
    "        [1717, 1, 0, 171, 9181, 537, 12]\n",
    "        [82, 11]\n",
    "        ...\n",
    "        into:\n",
    "        1791, 2719, 1891, 1717, 1, 0, 171, 9181, 537, 12, 82, 11, ...\n",
    "    '''\n",
    "    try:\n",
    "        return np.array([v for e in column for v in e])\n",
    "    except (TypeError, ValueError):\n",
    "        return column"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Load all data in memory into dataframes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# MC signal:\n",
    "ttbar = root2pandas('files/ttbar.root', 'events') \n",
    "# MC backgrounds:\n",
    "dy = root2pandas('files/dy.root', 'events')\n",
    "wj = root2pandas('files/wjets.root', 'events')\n",
    "ww = root2pandas('files/ww.root', 'events')\n",
    "wz = root2pandas('files/wz.root', 'events')\n",
    "zz = root2pandas('files/zz.root', 'events')\n",
    "singletop = root2pandas('files/single_top.root', 'events')\n",
    "qcd = root2pandas('files/qcd.root', 'events')\n",
    "# data:\n",
    "data = root2pandas('files/data.root', 'events')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "All samples except from the $t\\bar{t}$ one are produced using muon trigger. To enforce that on the $t\\bar{t}$ sample as well, we use the branch called `triggerIsoMu24` which contains a boolean to indicate if any event would have passed the specified trigger. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0        False\n",
       "1        False\n",
       "2        False\n",
       "3        False\n",
       "4        False\n",
       "5        False\n",
       "6        False\n",
       "7         True\n",
       "8        False\n",
       "9        False\n",
       "10       False\n",
       "11       False\n",
       "12       False\n",
       "13       False\n",
       "14       False\n",
       "15       False\n",
       "16        True\n",
       "17       False\n",
       "18       False\n",
       "19       False\n",
       "20       False\n",
       "21       False\n",
       "22       False\n",
       "23       False\n",
       "24        True\n",
       "25       False\n",
       "26       False\n",
       "27       False\n",
       "28       False\n",
       "29       False\n",
       "         ...  \n",
       "36911    False\n",
       "36912    False\n",
       "36913    False\n",
       "36914    False\n",
       "36915    False\n",
       "36916    False\n",
       "36917    False\n",
       "36918    False\n",
       "36919    False\n",
       "36920    False\n",
       "36921    False\n",
       "36922    False\n",
       "36923    False\n",
       "36924     True\n",
       "36925    False\n",
       "36926    False\n",
       "36927    False\n",
       "36928    False\n",
       "36929    False\n",
       "36930    False\n",
       "36931    False\n",
       "36932     True\n",
       "36933    False\n",
       "36934    False\n",
       "36935    False\n",
       "36936     True\n",
       "36937    False\n",
       "36938    False\n",
       "36939    False\n",
       "36940    False\n",
       "Name: triggerIsoMu24, dtype: bool"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# -- step by step:\n",
    "ttbar['triggerIsoMu24']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NJet</th>\n",
       "      <th>Jet_Px</th>\n",
       "      <th>Jet_Py</th>\n",
       "      <th>Jet_Pz</th>\n",
       "      <th>Jet_E</th>\n",
       "      <th>Jet_btag</th>\n",
       "      <th>Jet_ID</th>\n",
       "      <th>NMuon</th>\n",
       "      <th>Muon_Px</th>\n",
       "      <th>Muon_Py</th>\n",
       "      <th>...</th>\n",
       "      <th>MClepton_px</th>\n",
       "      <th>MClepton_py</th>\n",
       "      <th>MClepton_pz</th>\n",
       "      <th>MCleptonPDGid</th>\n",
       "      <th>MCneutrino_px</th>\n",
       "      <th>MCneutrino_py</th>\n",
       "      <th>MCneutrino_pz</th>\n",
       "      <th>NPrimaryVertices</th>\n",
       "      <th>triggerIsoMu24</th>\n",
       "      <th>EventWeight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>3</td>\n",
       "      <td>[-39.2862, 49.3453, -7.69329]</td>\n",
       "      <td>[-32.3978, 5.15671, 41.4265]</td>\n",
       "      <td>[-65.3803, -95.7182, -70.6679]</td>\n",
       "      <td>[83.5105, 108.047, 82.8404]</td>\n",
       "      <td>[2.42398, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-34.0692]</td>\n",
       "      <td>[5.64295]</td>\n",
       "      <td>...</td>\n",
       "      <td>-34.540741</td>\n",
       "      <td>5.717676</td>\n",
       "      <td>-10.324351</td>\n",
       "      <td>-13</td>\n",
       "      <td>43.943020</td>\n",
       "      <td>4.711743</td>\n",
       "      <td>-16.245531</td>\n",
       "      <td>4</td>\n",
       "      <td>True</td>\n",
       "      <td>0.279545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>5</td>\n",
       "      <td>[184.591, 10.8253, -72.3689, -43.3258, 2.57552]</td>\n",
       "      <td>[117.452, -166.958, 48.44, -5.91005, 42.4344]</td>\n",
       "      <td>[379.501, -51.8267, 452.426, 33.3998, -77.0341]</td>\n",
       "      <td>[438.532, 176.045, 461.069, 56.1953, 88.4542]</td>\n",
       "      <td>[-1.0, 3.20327, -1.0, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[42.3929]</td>\n",
       "      <td>[-35.9437]</td>\n",
       "      <td>...</td>\n",
       "      <td>43.464474</td>\n",
       "      <td>-37.058472</td>\n",
       "      <td>-30.771671</td>\n",
       "      <td>13</td>\n",
       "      <td>-21.166794</td>\n",
       "      <td>19.246643</td>\n",
       "      <td>-13.815475</td>\n",
       "      <td>4</td>\n",
       "      <td>True</td>\n",
       "      <td>0.279545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>5</td>\n",
       "      <td>[-92.04, 29.9482, 24.2262, 42.4436, -18.0844]</td>\n",
       "      <td>[51.4753, -84.1564, 48.5667, -8.55752, 24.7327]</td>\n",
       "      <td>[-52.9665, -181.678, -10.7411, -31.6566, -43.3...</td>\n",
       "      <td>[121.717, 203.098, 57.0088, 53.9551, 53.2984]</td>\n",
       "      <td>[-1.0, -1.0, -1.0, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[26.0595]</td>\n",
       "      <td>[0.139715]</td>\n",
       "      <td>...</td>\n",
       "      <td>27.888340</td>\n",
       "      <td>-2.613286</td>\n",
       "      <td>-114.308418</td>\n",
       "      <td>13</td>\n",
       "      <td>-43.849094</td>\n",
       "      <td>-32.715321</td>\n",
       "      <td>-46.295307</td>\n",
       "      <td>8</td>\n",
       "      <td>True</td>\n",
       "      <td>0.299037</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>3</td>\n",
       "      <td>[92.6861, -21.7365, 21.6693]</td>\n",
       "      <td>[-126.292, 52.302, 31.9107]</td>\n",
       "      <td>[88.5404, -12.5433, -18.5118]</td>\n",
       "      <td>[180.864, 58.5981, 43.3167]</td>\n",
       "      <td>[2.59038, 2.73967, -1.0]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-13.8492]</td>\n",
       "      <td>[38.0815]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "      <td>0.270251</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>3</td>\n",
       "      <td>[-97.5345, -19.9276, 2.89748]</td>\n",
       "      <td>[-13.185, -80.3989, 30.9127]</td>\n",
       "      <td>[-11.907, 51.528, -91.2712]</td>\n",
       "      <td>[101.636, 98.2916, 96.7967]</td>\n",
       "      <td>[2.89632, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[72.6027]</td>\n",
       "      <td>[-9.59597]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>16</td>\n",
       "      <td>True</td>\n",
       "      <td>0.010469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>4</td>\n",
       "      <td>[13.2568, 84.261, -4.9345, -33.4614]</td>\n",
       "      <td>[188.902, -129.347, -121.279, -62.2332]</td>\n",
       "      <td>[-67.8184, 185.424, 164.392, 53.9148]</td>\n",
       "      <td>[202.167, 242.673, 205.04, 89.4054]</td>\n",
       "      <td>[2.31161, -1.0, -1.0, 3.40863]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>2</td>\n",
       "      <td>[-41.887, 8.97538]</td>\n",
       "      <td>[108.296, 21.3534]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6</td>\n",
       "      <td>True</td>\n",
       "      <td>0.293529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57</th>\n",
       "      <td>4</td>\n",
       "      <td>[86.7032, -84.3351, -60.7989, 38.2517]</td>\n",
       "      <td>[0.745445, -0.972272, 49.1331, -22.5668]</td>\n",
       "      <td>[-77.6728, -19.5303, 3.48934, 48.336]</td>\n",
       "      <td>[117.088, 87.3901, 78.8732, 66.4349]</td>\n",
       "      <td>[1.94137, -1.0, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>0</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>...</td>\n",
       "      <td>-51.339859</td>\n",
       "      <td>47.421997</td>\n",
       "      <td>9.806538</td>\n",
       "      <td>-13</td>\n",
       "      <td>7.199954</td>\n",
       "      <td>33.037640</td>\n",
       "      <td>64.709595</td>\n",
       "      <td>13</td>\n",
       "      <td>True</td>\n",
       "      <td>0.026791</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58</th>\n",
       "      <td>0</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-73.5007]</td>\n",
       "      <td>[47.405]</td>\n",
       "      <td>...</td>\n",
       "      <td>-77.774635</td>\n",
       "      <td>49.710968</td>\n",
       "      <td>-163.584732</td>\n",
       "      <td>13</td>\n",
       "      <td>-14.959000</td>\n",
       "      <td>-17.021935</td>\n",
       "      <td>-129.785049</td>\n",
       "      <td>20</td>\n",
       "      <td>True</td>\n",
       "      <td>0.010469</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>68</th>\n",
       "      <td>3</td>\n",
       "      <td>[-72.5721, -37.3589, 54.3353]</td>\n",
       "      <td>[-21.6138, 44.3861, -11.165]</td>\n",
       "      <td>[-203.388, -141.191, -148.702]</td>\n",
       "      <td>[217.305, 152.979, 159.159]</td>\n",
       "      <td>[1.64367, 2.30578, -1.0]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-1.18855]</td>\n",
       "      <td>[-36.1914]</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.134651</td>\n",
       "      <td>-35.692688</td>\n",
       "      <td>-109.194305</td>\n",
       "      <td>13</td>\n",
       "      <td>67.564972</td>\n",
       "      <td>-8.594560</td>\n",
       "      <td>-78.256340</td>\n",
       "      <td>6</td>\n",
       "      <td>True</td>\n",
       "      <td>0.286201</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>80</th>\n",
       "      <td>3</td>\n",
       "      <td>[394.21, -66.0473, 8.33537]</td>\n",
       "      <td>[-174.289, 96.3359, -45.8379]</td>\n",
       "      <td>[-363.242, 1.3422, -32.5724]</td>\n",
       "      <td>[574.341, 117.572, 57.576]</td>\n",
       "      <td>[2.80352, 1.69599, -1.0]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-66.4606]</td>\n",
       "      <td>[-1.64239]</td>\n",
       "      <td>...</td>\n",
       "      <td>-100.356377</td>\n",
       "      <td>-2.219039</td>\n",
       "      <td>-10.412981</td>\n",
       "      <td>-13</td>\n",
       "      <td>-246.632401</td>\n",
       "      <td>112.364296</td>\n",
       "      <td>-81.893837</td>\n",
       "      <td>3</td>\n",
       "      <td>True</td>\n",
       "      <td>0.271386</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86</th>\n",
       "      <td>4</td>\n",
       "      <td>[65.0804, 47.7328, -41.3639, -15.2447]</td>\n",
       "      <td>[-84.0002, 73.0816, -21.3092, 30.1185]</td>\n",
       "      <td>[-19.6535, -286.044, -80.5942, -45.1493]</td>\n",
       "      <td>[109.014, 299.209, 93.831, 56.4601]</td>\n",
       "      <td>[3.79264, -1.0, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[9.05585]</td>\n",
       "      <td>[-24.4455]</td>\n",
       "      <td>...</td>\n",
       "      <td>8.528934</td>\n",
       "      <td>-24.427439</td>\n",
       "      <td>27.513771</td>\n",
       "      <td>13</td>\n",
       "      <td>-51.809479</td>\n",
       "      <td>8.691632</td>\n",
       "      <td>-20.814299</td>\n",
       "      <td>11</td>\n",
       "      <td>True</td>\n",
       "      <td>0.105165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>2</td>\n",
       "      <td>[-40.122, 27.6997]</td>\n",
       "      <td>[-44.2169, -41.4156]</td>\n",
       "      <td>[10.1575, 29.0914]</td>\n",
       "      <td>[62.1224, 58.2074]</td>\n",
       "      <td>[-1.0, 1.66511]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[42.9513]</td>\n",
       "      <td>[45.6999]</td>\n",
       "      <td>...</td>\n",
       "      <td>44.989742</td>\n",
       "      <td>46.907940</td>\n",
       "      <td>50.970924</td>\n",
       "      <td>-13</td>\n",
       "      <td>13.318195</td>\n",
       "      <td>-31.906128</td>\n",
       "      <td>19.553530</td>\n",
       "      <td>10</td>\n",
       "      <td>True</td>\n",
       "      <td>0.299981</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>107</th>\n",
       "      <td>5</td>\n",
       "      <td>[-100.62, 32.0011, -35.6917, 28.3667, 3.8716]</td>\n",
       "      <td>[-4.6241, -39.0739, 26.2937, -25.9119, 34.3113]</td>\n",
       "      <td>[-135.39, 38.0203, -23.802, -58.4347, 29.4247]</td>\n",
       "      <td>[171.725, 64.9216, 51.6265, 70.5236, 45.8437]</td>\n",
       "      <td>[-1.0, -1.0, -1.0, 3.47004, -1.0]</td>\n",
       "      <td>[True, True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[32.2751]</td>\n",
       "      <td>[-59.3023]</td>\n",
       "      <td>...</td>\n",
       "      <td>33.404510</td>\n",
       "      <td>-60.193768</td>\n",
       "      <td>-54.171921</td>\n",
       "      <td>13</td>\n",
       "      <td>44.509659</td>\n",
       "      <td>19.568098</td>\n",
       "      <td>-44.442898</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "      <td>0.270251</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>122</th>\n",
       "      <td>3</td>\n",
       "      <td>[10.4771, -43.1103, 23.1228]</td>\n",
       "      <td>[65.1101, -9.28895, 33.4949]</td>\n",
       "      <td>[-64.5278, 20.2662, 20.8267]</td>\n",
       "      <td>[92.8245, 48.7273, 46.165]</td>\n",
       "      <td>[4.1136, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-0.298645]</td>\n",
       "      <td>[-52.4379]</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.104305</td>\n",
       "      <td>-95.161293</td>\n",
       "      <td>-309.347382</td>\n",
       "      <td>-15</td>\n",
       "      <td>-8.769296</td>\n",
       "      <td>-14.520910</td>\n",
       "      <td>4.469729</td>\n",
       "      <td>6</td>\n",
       "      <td>True</td>\n",
       "      <td>0.286201</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>133</th>\n",
       "      <td>3</td>\n",
       "      <td>[-104.634, 48.1258, 12.1684]</td>\n",
       "      <td>[29.6702, -66.15, 34.2625]</td>\n",
       "      <td>[-72.2929, -229.216, 13.3559]</td>\n",
       "      <td>[131.668, 243.703, 38.9503]</td>\n",
       "      <td>[3.45501, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>0</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "      <td>0.270251</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>144</th>\n",
       "      <td>3</td>\n",
       "      <td>[-75.713, 58.5136, 31.0808]</td>\n",
       "      <td>[44.9549, -38.7253, -41.9073]</td>\n",
       "      <td>[-195.356, -125.271, -35.4589]</td>\n",
       "      <td>[214.638, 144.113, 64.1693]</td>\n",
       "      <td>[2.9908, 3.62441, -1.0]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-45.2115]</td>\n",
       "      <td>[5.31683]</td>\n",
       "      <td>...</td>\n",
       "      <td>-44.851936</td>\n",
       "      <td>5.279506</td>\n",
       "      <td>-0.475743</td>\n",
       "      <td>13</td>\n",
       "      <td>20.658106</td>\n",
       "      <td>36.965542</td>\n",
       "      <td>-45.506985</td>\n",
       "      <td>6</td>\n",
       "      <td>True</td>\n",
       "      <td>0.286201</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>148</th>\n",
       "      <td>4</td>\n",
       "      <td>[-113.872, -32.332, 34.4798, 10.6671]</td>\n",
       "      <td>[-23.9376, 35.5097, -6.87441, 30.7233]</td>\n",
       "      <td>[87.0612, 45.5352, 72.4027, -44.7416]</td>\n",
       "      <td>[145.824, 66.5828, 80.8721, 55.7227]</td>\n",
       "      <td>[2.06658, -1.0, -1.0, 3.2202]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[74.4426]</td>\n",
       "      <td>[-47.3361]</td>\n",
       "      <td>...</td>\n",
       "      <td>74.646675</td>\n",
       "      <td>-48.276752</td>\n",
       "      <td>-21.981726</td>\n",
       "      <td>13</td>\n",
       "      <td>-14.983271</td>\n",
       "      <td>-19.674046</td>\n",
       "      <td>-33.766380</td>\n",
       "      <td>19</td>\n",
       "      <td>True</td>\n",
       "      <td>0.014366</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>157</th>\n",
       "      <td>2</td>\n",
       "      <td>[39.0372, 18.8909]</td>\n",
       "      <td>[56.7757, -41.463]</td>\n",
       "      <td>[-154.183, -153.292]</td>\n",
       "      <td>[169.354, 160.027]</td>\n",
       "      <td>[-1.0, 2.21366]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-81.4154]</td>\n",
       "      <td>[11.1185]</td>\n",
       "      <td>...</td>\n",
       "      <td>-88.916168</td>\n",
       "      <td>11.958924</td>\n",
       "      <td>-100.715332</td>\n",
       "      <td>-13</td>\n",
       "      <td>-8.342011</td>\n",
       "      <td>-37.894085</td>\n",
       "      <td>-57.389412</td>\n",
       "      <td>5</td>\n",
       "      <td>True</td>\n",
       "      <td>0.279545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>168</th>\n",
       "      <td>4</td>\n",
       "      <td>[-23.9239, -6.59758, -14.6238, 45.4713]</td>\n",
       "      <td>[60.4995, -51.068, 46.1813, -10.2181]</td>\n",
       "      <td>[57.153, -172.568, -82.0874, -18.6692]</td>\n",
       "      <td>[87.7235, 180.256, 95.6667, 50.63]</td>\n",
       "      <td>[3.08056, -1.0, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[42.5798]</td>\n",
       "      <td>[-37.9202]</td>\n",
       "      <td>...</td>\n",
       "      <td>42.677971</td>\n",
       "      <td>-38.684910</td>\n",
       "      <td>25.826906</td>\n",
       "      <td>13</td>\n",
       "      <td>-35.230225</td>\n",
       "      <td>-32.083885</td>\n",
       "      <td>29.348593</td>\n",
       "      <td>4</td>\n",
       "      <td>True</td>\n",
       "      <td>0.274432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>171</th>\n",
       "      <td>2</td>\n",
       "      <td>[-18.0533, 45.1819]</td>\n",
       "      <td>[75.6471, -25.322]</td>\n",
       "      <td>[-59.945, -22.2292]</td>\n",
       "      <td>[98.8365, 57.0231]</td>\n",
       "      <td>[1.58192, -1.0]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-2.2919]</td>\n",
       "      <td>[-55.9058]</td>\n",
       "      <td>...</td>\n",
       "      <td>-2.266898</td>\n",
       "      <td>-56.387188</td>\n",
       "      <td>84.049629</td>\n",
       "      <td>13</td>\n",
       "      <td>17.545727</td>\n",
       "      <td>17.956202</td>\n",
       "      <td>95.806236</td>\n",
       "      <td>3</td>\n",
       "      <td>True</td>\n",
       "      <td>0.279545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>175</th>\n",
       "      <td>2</td>\n",
       "      <td>[-62.7099, -16.6717]</td>\n",
       "      <td>[62.0686, -68.4154]</td>\n",
       "      <td>[-56.106, -164.618]</td>\n",
       "      <td>[105.655, 179.547]</td>\n",
       "      <td>[-1.0, -1.0]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[52.0206]</td>\n",
       "      <td>[100.328]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3</td>\n",
       "      <td>True</td>\n",
       "      <td>0.293529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>188</th>\n",
       "      <td>2</td>\n",
       "      <td>[10.1792, -26.8712]</td>\n",
       "      <td>[55.9482, -16.998]</td>\n",
       "      <td>[-89.4134, 72.4999]</td>\n",
       "      <td>[106.531, 79.6631]</td>\n",
       "      <td>[-1.0, 2.40431]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-65.1655]</td>\n",
       "      <td>[22.1073]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6</td>\n",
       "      <td>True</td>\n",
       "      <td>0.293529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>194</th>\n",
       "      <td>3</td>\n",
       "      <td>[69.7044, 28.6426, -30.5756]</td>\n",
       "      <td>[-21.1823, 21.4084, -5.88853]</td>\n",
       "      <td>[189.136, 125.371, 28.3306]</td>\n",
       "      <td>[203.069, 130.501, 42.8366]</td>\n",
       "      <td>[2.25507, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-84.0036]</td>\n",
       "      <td>[-38.9238]</td>\n",
       "      <td>...</td>\n",
       "      <td>-79.938377</td>\n",
       "      <td>-35.873989</td>\n",
       "      <td>-74.092720</td>\n",
       "      <td>13</td>\n",
       "      <td>-14.628411</td>\n",
       "      <td>30.902468</td>\n",
       "      <td>-51.598186</td>\n",
       "      <td>5</td>\n",
       "      <td>True</td>\n",
       "      <td>0.293529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>197</th>\n",
       "      <td>4</td>\n",
       "      <td>[-81.0861, 35.7697, 45.4883, -7.59584]</td>\n",
       "      <td>[-22.679, -41.042, 24.9598, 37.9333]</td>\n",
       "      <td>[22.1584, -50.1099, -306.429, -61.9009]</td>\n",
       "      <td>[87.7846, 74.7338, 311.244, 73.3563]</td>\n",
       "      <td>[4.60365, 1.62578, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-21.7847]</td>\n",
       "      <td>[10.4468]</td>\n",
       "      <td>...</td>\n",
       "      <td>-21.555775</td>\n",
       "      <td>10.569125</td>\n",
       "      <td>16.874752</td>\n",
       "      <td>13</td>\n",
       "      <td>66.121849</td>\n",
       "      <td>17.503078</td>\n",
       "      <td>1.514331</td>\n",
       "      <td>5</td>\n",
       "      <td>True</td>\n",
       "      <td>0.286201</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>203</th>\n",
       "      <td>4</td>\n",
       "      <td>[-92.6663, -29.3145, 43.617, 22.991]</td>\n",
       "      <td>[-9.88577, -39.3474, 0.0992902, 32.9985]</td>\n",
       "      <td>[-19.9208, 24.1042, -28.9692, 73.4268]</td>\n",
       "      <td>[96.1515, 55.6889, 53.1124, 84.3405]</td>\n",
       "      <td>[2.15808, -1.0, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[51.9452]</td>\n",
       "      <td>[-34.679]</td>\n",
       "      <td>...</td>\n",
       "      <td>51.943840</td>\n",
       "      <td>-35.272839</td>\n",
       "      <td>-15.022160</td>\n",
       "      <td>13</td>\n",
       "      <td>-14.474733</td>\n",
       "      <td>10.930140</td>\n",
       "      <td>20.586962</td>\n",
       "      <td>11</td>\n",
       "      <td>True</td>\n",
       "      <td>0.160826</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>204</th>\n",
       "      <td>1</td>\n",
       "      <td>[21.5621]</td>\n",
       "      <td>[-30.0244]</td>\n",
       "      <td>[160.531]</td>\n",
       "      <td>[164.844]</td>\n",
       "      <td>[-1.0]</td>\n",
       "      <td>[True]</td>\n",
       "      <td>2</td>\n",
       "      <td>[109.452, -9.8738]</td>\n",
       "      <td>[191.047, -12.5724]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4</td>\n",
       "      <td>True</td>\n",
       "      <td>0.274432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>219</th>\n",
       "      <td>2</td>\n",
       "      <td>[55.6175, 32.9164]</td>\n",
       "      <td>[39.0514, -11.1956]</td>\n",
       "      <td>[-44.6516, -31.1788]</td>\n",
       "      <td>[82.3264, 47.1459]</td>\n",
       "      <td>[2.18723, -1.0]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-17.0189]</td>\n",
       "      <td>[47.6267]</td>\n",
       "      <td>...</td>\n",
       "      <td>-17.056755</td>\n",
       "      <td>47.598736</td>\n",
       "      <td>0.132445</td>\n",
       "      <td>-13</td>\n",
       "      <td>-18.238882</td>\n",
       "      <td>4.040609</td>\n",
       "      <td>71.326790</td>\n",
       "      <td>4</td>\n",
       "      <td>True</td>\n",
       "      <td>0.279545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>223</th>\n",
       "      <td>2</td>\n",
       "      <td>[-63.6374, 21.9639]</td>\n",
       "      <td>[58.9319, -21.1155]</td>\n",
       "      <td>[131.927, 79.1002]</td>\n",
       "      <td>[159.269, 85.0185]</td>\n",
       "      <td>[-1.0, -1.0]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[39.4852]</td>\n",
       "      <td>[-28.8416]</td>\n",
       "      <td>...</td>\n",
       "      <td>40.519344</td>\n",
       "      <td>-29.519676</td>\n",
       "      <td>67.463295</td>\n",
       "      <td>-13</td>\n",
       "      <td>2.028334</td>\n",
       "      <td>13.116418</td>\n",
       "      <td>-16.692287</td>\n",
       "      <td>25</td>\n",
       "      <td>True</td>\n",
       "      <td>0.001839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>235</th>\n",
       "      <td>1</td>\n",
       "      <td>[-22.13]</td>\n",
       "      <td>[-44.9609]</td>\n",
       "      <td>[-56.7319]</td>\n",
       "      <td>[77.0054]</td>\n",
       "      <td>[-1.0]</td>\n",
       "      <td>[True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[18.7418]</td>\n",
       "      <td>[37.8549]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>15</td>\n",
       "      <td>True</td>\n",
       "      <td>0.005580</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>268</th>\n",
       "      <td>3</td>\n",
       "      <td>[40.9751, -26.4305, 8.81603]</td>\n",
       "      <td>[0.197465, 19.2683, -30.2287]</td>\n",
       "      <td>[-10.3833, 126.664, 35.724]</td>\n",
       "      <td>[42.9108, 130.983, 47.8053]</td>\n",
       "      <td>[-1.0, 1.85155, -1.0]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>0</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>...</td>\n",
       "      <td>-53.865688</td>\n",
       "      <td>5.485102</td>\n",
       "      <td>147.813019</td>\n",
       "      <td>13</td>\n",
       "      <td>6.192799</td>\n",
       "      <td>-14.861897</td>\n",
       "      <td>285.897095</td>\n",
       "      <td>7</td>\n",
       "      <td>True</td>\n",
       "      <td>0.293529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36671</th>\n",
       "      <td>4</td>\n",
       "      <td>[63.6192, 15.5392, -24.603, -22.424]</td>\n",
       "      <td>[46.3087, -54.6324, 33.719, 33.788]</td>\n",
       "      <td>[-201.403, 121.437, -185.224, -41.4796]</td>\n",
       "      <td>[216.771, 134.415, 189.95, 59.4113]</td>\n",
       "      <td>[-1.0, 2.30456, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[44.2237]</td>\n",
       "      <td>[1.19632]</td>\n",
       "      <td>...</td>\n",
       "      <td>44.393093</td>\n",
       "      <td>4.441749</td>\n",
       "      <td>17.649075</td>\n",
       "      <td>13</td>\n",
       "      <td>-37.079338</td>\n",
       "      <td>4.005792</td>\n",
       "      <td>22.262280</td>\n",
       "      <td>13</td>\n",
       "      <td>True</td>\n",
       "      <td>0.192474</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36674</th>\n",
       "      <td>3</td>\n",
       "      <td>[-63.3877, -42.8766, 42.7456]</td>\n",
       "      <td>[16.5414, 9.93516, 3.20814]</td>\n",
       "      <td>[57.6559, 93.1895, 27.8355]</td>\n",
       "      <td>[87.8452, 103.46, 51.6133]</td>\n",
       "      <td>[2.389, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-19.4274]</td>\n",
       "      <td>[-29.8691]</td>\n",
       "      <td>...</td>\n",
       "      <td>-19.507946</td>\n",
       "      <td>-30.081053</td>\n",
       "      <td>47.349102</td>\n",
       "      <td>-13</td>\n",
       "      <td>57.827797</td>\n",
       "      <td>-55.537346</td>\n",
       "      <td>26.413647</td>\n",
       "      <td>16</td>\n",
       "      <td>True</td>\n",
       "      <td>0.105165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36678</th>\n",
       "      <td>4</td>\n",
       "      <td>[-42.7893, 41.4222, -12.0665, 17.7019]</td>\n",
       "      <td>[-36.8382, -15.6533, 37.0771, 30.5058]</td>\n",
       "      <td>[-41.9049, -4.82216, 5.23354, -70.8694]</td>\n",
       "      <td>[70.8273, 45.4714, 39.5967, 79.481]</td>\n",
       "      <td>[3.14142, -1.0, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-1.03951]</td>\n",
       "      <td>[-37.0587]</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.284460</td>\n",
       "      <td>-37.510612</td>\n",
       "      <td>-173.577927</td>\n",
       "      <td>-13</td>\n",
       "      <td>-17.592842</td>\n",
       "      <td>-28.825563</td>\n",
       "      <td>-12.680113</td>\n",
       "      <td>11</td>\n",
       "      <td>True</td>\n",
       "      <td>0.192474</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36691</th>\n",
       "      <td>4</td>\n",
       "      <td>[-58.3385, -18.4484, 15.4408, 37.792]</td>\n",
       "      <td>[-4.25548, -43.2476, 42.527, 3.44773]</td>\n",
       "      <td>[145.612, 57.8691, 70.6434, -58.7726]</td>\n",
       "      <td>[157.184, 75.1443, 84.1751, 70.7867]</td>\n",
       "      <td>[-1.0, -1.0, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-45.0365]</td>\n",
       "      <td>[-3.44792]</td>\n",
       "      <td>...</td>\n",
       "      <td>-46.260303</td>\n",
       "      <td>-3.330531</td>\n",
       "      <td>-12.245513</td>\n",
       "      <td>13</td>\n",
       "      <td>34.181591</td>\n",
       "      <td>14.543386</td>\n",
       "      <td>6.494123</td>\n",
       "      <td>4</td>\n",
       "      <td>True</td>\n",
       "      <td>0.279545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36697</th>\n",
       "      <td>0</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>2</td>\n",
       "      <td>[72.8115, -32.1987]</td>\n",
       "      <td>[-41.4184, 46.7349]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>15</td>\n",
       "      <td>True</td>\n",
       "      <td>0.036131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36698</th>\n",
       "      <td>2</td>\n",
       "      <td>[49.845, -38.1155]</td>\n",
       "      <td>[-31.4604, -12.8079]</td>\n",
       "      <td>[-105.783, -222.021]</td>\n",
       "      <td>[121.478, 225.912]</td>\n",
       "      <td>[-1.0, -1.0]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[39.8958]</td>\n",
       "      <td>[-8.11494]</td>\n",
       "      <td>...</td>\n",
       "      <td>41.240803</td>\n",
       "      <td>-6.950862</td>\n",
       "      <td>-118.234520</td>\n",
       "      <td>-13</td>\n",
       "      <td>-24.029903</td>\n",
       "      <td>32.869717</td>\n",
       "      <td>-56.536091</td>\n",
       "      <td>5</td>\n",
       "      <td>True</td>\n",
       "      <td>0.274432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36704</th>\n",
       "      <td>3</td>\n",
       "      <td>[44.5244, -22.9694, -5.49336]</td>\n",
       "      <td>[52.0617, -22.2976, -29.943]</td>\n",
       "      <td>[-123.699, 45.3537, -114.566]</td>\n",
       "      <td>[141.7, 56.1422, 118.853]</td>\n",
       "      <td>[-1.0, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-18.3659]</td>\n",
       "      <td>[38.4539]</td>\n",
       "      <td>...</td>\n",
       "      <td>-19.019670</td>\n",
       "      <td>38.604202</td>\n",
       "      <td>-15.343031</td>\n",
       "      <td>-13</td>\n",
       "      <td>-85.043907</td>\n",
       "      <td>6.205180</td>\n",
       "      <td>-118.760750</td>\n",
       "      <td>8</td>\n",
       "      <td>True</td>\n",
       "      <td>0.253713</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36705</th>\n",
       "      <td>4</td>\n",
       "      <td>[95.7703, -91.2611, -22.8162, 11.3628]</td>\n",
       "      <td>[-61.2077, -23.8256, -90.0745, 70.7234]</td>\n",
       "      <td>[14.1595, 200.652, 21.1822, 282.059]</td>\n",
       "      <td>[116.274, 222.271, 96.4095, 291.297]</td>\n",
       "      <td>[-1.0, -1.0, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-68.7636]</td>\n",
       "      <td>[41.4596]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>10</td>\n",
       "      <td>True</td>\n",
       "      <td>0.293189</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36723</th>\n",
       "      <td>2</td>\n",
       "      <td>[-29.0196, 40.0986]</td>\n",
       "      <td>[-28.8848, -6.73161]</td>\n",
       "      <td>[-10.89, 10.5535]</td>\n",
       "      <td>[43.0001, 42.402]</td>\n",
       "      <td>[1.72804, -1.0]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[48.2104]</td>\n",
       "      <td>[10.7573]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3</td>\n",
       "      <td>True</td>\n",
       "      <td>0.271386</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36728</th>\n",
       "      <td>4</td>\n",
       "      <td>[-37.2348, -19.6443, 35.2389, 36.7954]</td>\n",
       "      <td>[-58.4847, 53.7423, 23.7087, 0.172772]</td>\n",
       "      <td>[-140.538, -103.277, -81.6349, -79.4282]</td>\n",
       "      <td>[157.469, 118.497, 92.416, 87.6949]</td>\n",
       "      <td>[-1.0, -1.0, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>0</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5</td>\n",
       "      <td>True</td>\n",
       "      <td>0.293529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36733</th>\n",
       "      <td>2</td>\n",
       "      <td>[90.0657, -63.0919]</td>\n",
       "      <td>[57.5107, -48.7488]</td>\n",
       "      <td>[-176.365, 13.9594]</td>\n",
       "      <td>[207.504, 81.852]</td>\n",
       "      <td>[2.16326, 3.64163]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[29.2887]</td>\n",
       "      <td>[-13.7026]</td>\n",
       "      <td>...</td>\n",
       "      <td>29.956720</td>\n",
       "      <td>-13.935602</td>\n",
       "      <td>5.496824</td>\n",
       "      <td>13</td>\n",
       "      <td>-3.975806</td>\n",
       "      <td>18.813091</td>\n",
       "      <td>100.464325</td>\n",
       "      <td>4</td>\n",
       "      <td>True</td>\n",
       "      <td>0.293529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36761</th>\n",
       "      <td>5</td>\n",
       "      <td>[-72.7327, 87.8201, -48.6379, 35.3672, -29.3188]</td>\n",
       "      <td>[-71.4104, 25.7807, -24.6549, 37.8814, 26.399]</td>\n",
       "      <td>[-25.5412, -27.7219, 69.2121, 166.603, -101.946]</td>\n",
       "      <td>[105.677, 96.5188, 88.2745, 174.643, 109.666]</td>\n",
       "      <td>[2.64415, -1.0, 2.74785, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-15.0587]</td>\n",
       "      <td>[-19.3267]</td>\n",
       "      <td>...</td>\n",
       "      <td>-14.784895</td>\n",
       "      <td>-19.563482</td>\n",
       "      <td>-41.428085</td>\n",
       "      <td>13</td>\n",
       "      <td>44.433407</td>\n",
       "      <td>46.769634</td>\n",
       "      <td>-128.488770</td>\n",
       "      <td>9</td>\n",
       "      <td>True</td>\n",
       "      <td>0.277205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36773</th>\n",
       "      <td>2</td>\n",
       "      <td>[42.3535, 7.84763]</td>\n",
       "      <td>[7.22512, -38.6475]</td>\n",
       "      <td>[24.7581, 90.8231]</td>\n",
       "      <td>[50.1142, 99.2029]</td>\n",
       "      <td>[2.67679, -1.0]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-72.876]</td>\n",
       "      <td>[53.7722]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5</td>\n",
       "      <td>True</td>\n",
       "      <td>0.286201</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36777</th>\n",
       "      <td>5</td>\n",
       "      <td>[-99.6067, -3.34446, 32.1012, 42.7883, -11.6452]</td>\n",
       "      <td>[-25.3014, 92.4234, -32.9203, 0.946451, 36.2387]</td>\n",
       "      <td>[155.932, 174.57, 38.0382, 25.5367, 18.2656]</td>\n",
       "      <td>[187.385, 198.147, 60.2857, 50.5626, 42.798]</td>\n",
       "      <td>[-1.0, 3.07208, -1.0, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[23.5196]</td>\n",
       "      <td>[-33.3833]</td>\n",
       "      <td>...</td>\n",
       "      <td>19.755816</td>\n",
       "      <td>-44.092232</td>\n",
       "      <td>172.988556</td>\n",
       "      <td>-13</td>\n",
       "      <td>28.066628</td>\n",
       "      <td>-17.915775</td>\n",
       "      <td>9.881391</td>\n",
       "      <td>6</td>\n",
       "      <td>True</td>\n",
       "      <td>0.293529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36779</th>\n",
       "      <td>3</td>\n",
       "      <td>[-163.892, -104.229, 43.5479]</td>\n",
       "      <td>[15.7904, -1.47729, -1.82685]</td>\n",
       "      <td>[-387.744, -61.9309, -63.0709]</td>\n",
       "      <td>[423.49, 123.1, 77.3157]</td>\n",
       "      <td>[-1.0, -1.0, 2.4724]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[204.758]</td>\n",
       "      <td>[3.03201]</td>\n",
       "      <td>...</td>\n",
       "      <td>194.055786</td>\n",
       "      <td>2.588275</td>\n",
       "      <td>31.950975</td>\n",
       "      <td>-13</td>\n",
       "      <td>36.918297</td>\n",
       "      <td>-33.213650</td>\n",
       "      <td>-13.392416</td>\n",
       "      <td>17</td>\n",
       "      <td>True</td>\n",
       "      <td>0.014366</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36845</th>\n",
       "      <td>2</td>\n",
       "      <td>[56.3493, 19.3332]</td>\n",
       "      <td>[-62.0922, 23.8897]</td>\n",
       "      <td>[98.1019, 8.61856]</td>\n",
       "      <td>[129.654, 32.2253]</td>\n",
       "      <td>[2.90254, -1.0]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-58.4214]</td>\n",
       "      <td>[49.5135]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>10</td>\n",
       "      <td>True</td>\n",
       "      <td>0.277205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36846</th>\n",
       "      <td>2</td>\n",
       "      <td>[63.5405, 61.2992]</td>\n",
       "      <td>[33.7838, 6.70778]</td>\n",
       "      <td>[-10.9178, 69.2736]</td>\n",
       "      <td>[73.4685, 93.2766]</td>\n",
       "      <td>[-1.0, -1.0]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-79.0116]</td>\n",
       "      <td>[-19.7364]</td>\n",
       "      <td>...</td>\n",
       "      <td>-77.586594</td>\n",
       "      <td>-19.407608</td>\n",
       "      <td>-66.658394</td>\n",
       "      <td>13</td>\n",
       "      <td>-72.814438</td>\n",
       "      <td>-36.851547</td>\n",
       "      <td>17.604248</td>\n",
       "      <td>6</td>\n",
       "      <td>True</td>\n",
       "      <td>0.293529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36854</th>\n",
       "      <td>6</td>\n",
       "      <td>[23.5022, -29.0738, 27.8149, 27.4214, 0.959874...</td>\n",
       "      <td>[-79.9664, -53.5415, 44.6099, 26.0418, 33.4664...</td>\n",
       "      <td>[7.103, 22.0331, -156.058, -22.6882, -58.1608,...</td>\n",
       "      <td>[85.9293, 65.432, 165.279, 44.604, 67.6031, 13...</td>\n",
       "      <td>[-1.0, -1.0, -1.0, -1.0, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-32.9304]</td>\n",
       "      <td>[34.6853]</td>\n",
       "      <td>...</td>\n",
       "      <td>-32.763885</td>\n",
       "      <td>34.414585</td>\n",
       "      <td>-20.146509</td>\n",
       "      <td>13</td>\n",
       "      <td>-34.493790</td>\n",
       "      <td>9.358992</td>\n",
       "      <td>-145.859955</td>\n",
       "      <td>19</td>\n",
       "      <td>True</td>\n",
       "      <td>0.026791</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36862</th>\n",
       "      <td>4</td>\n",
       "      <td>[65.2048, 26.1962, -27.163, 38.1059]</td>\n",
       "      <td>[-77.5005, 70.3165, -45.9601, 33.931]</td>\n",
       "      <td>[-55.8269, -3.97152, 95.8268, 287.091]</td>\n",
       "      <td>[117.233, 75.8283, 110.272, 291.82]</td>\n",
       "      <td>[2.6823, 2.13792, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>0</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9</td>\n",
       "      <td>True</td>\n",
       "      <td>0.277205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36864</th>\n",
       "      <td>4</td>\n",
       "      <td>[-105.143, 96.585, -95.1253, 80.7612]</td>\n",
       "      <td>[-153.554, 39.1784, 37.6713, -11.9996]</td>\n",
       "      <td>[-238.466, -25.2841, -9.86635, -101.42]</td>\n",
       "      <td>[304.335, 108.426, 103.746, 130.728]</td>\n",
       "      <td>[-1.0, 4.3339, 4.28816, -1.0]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>0</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>...</td>\n",
       "      <td>18.866938</td>\n",
       "      <td>-15.149240</td>\n",
       "      <td>7.708319</td>\n",
       "      <td>-13</td>\n",
       "      <td>1.578786</td>\n",
       "      <td>85.071571</td>\n",
       "      <td>34.158749</td>\n",
       "      <td>11</td>\n",
       "      <td>True</td>\n",
       "      <td>0.160826</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36873</th>\n",
       "      <td>4</td>\n",
       "      <td>[237.71, -193.887, 35.7849, 28.4387]</td>\n",
       "      <td>[-69.9921, 0.191406, 13.5153, -21.4827]</td>\n",
       "      <td>[-152.786, 436.524, -0.336736, 2.15682]</td>\n",
       "      <td>[293.461, 478.051, 39.3145, 36.544]</td>\n",
       "      <td>[-1.0, 3.04585, -1.0, 3.54336]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-45.9498]</td>\n",
       "      <td>[59.5273]</td>\n",
       "      <td>...</td>\n",
       "      <td>-48.707134</td>\n",
       "      <td>60.437092</td>\n",
       "      <td>219.109711</td>\n",
       "      <td>-13</td>\n",
       "      <td>-32.495464</td>\n",
       "      <td>14.442889</td>\n",
       "      <td>15.860429</td>\n",
       "      <td>9</td>\n",
       "      <td>True</td>\n",
       "      <td>0.293189</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36884</th>\n",
       "      <td>3</td>\n",
       "      <td>[68.6953, -68.5364, 46.7583]</td>\n",
       "      <td>[22.5236, 14.8628, -42.646]</td>\n",
       "      <td>[15.618, -29.9445, 35.7689]</td>\n",
       "      <td>[75.5679, 77.1019, 73.1977]</td>\n",
       "      <td>[2.38775, 2.44255, 1.78135]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-10.2928]</td>\n",
       "      <td>[25.1898]</td>\n",
       "      <td>...</td>\n",
       "      <td>-10.396090</td>\n",
       "      <td>25.239723</td>\n",
       "      <td>-11.126865</td>\n",
       "      <td>13</td>\n",
       "      <td>-42.048183</td>\n",
       "      <td>-10.150770</td>\n",
       "      <td>74.172813</td>\n",
       "      <td>3</td>\n",
       "      <td>True</td>\n",
       "      <td>0.270077</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36888</th>\n",
       "      <td>2</td>\n",
       "      <td>[-99.3072, -29.3348]</td>\n",
       "      <td>[49.2617, -29.6785]</td>\n",
       "      <td>[155.757, -6.78339]</td>\n",
       "      <td>[191.845, 42.8853]</td>\n",
       "      <td>[-1.0, 4.18899]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[114.711]</td>\n",
       "      <td>[-47.2675]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>8</td>\n",
       "      <td>True</td>\n",
       "      <td>0.299981</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36892</th>\n",
       "      <td>4</td>\n",
       "      <td>[-158.273, -67.7991, -33.143, -1.02856]</td>\n",
       "      <td>[201.345, -24.0705, -49.9765, -31.1794]</td>\n",
       "      <td>[-56.5885, 108.054, -133.396, -40.5998]</td>\n",
       "      <td>[270.567, 130.381, 146.749, 51.5353]</td>\n",
       "      <td>[-1.0, 4.01285, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-101.12]</td>\n",
       "      <td>[-152.441]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9</td>\n",
       "      <td>True</td>\n",
       "      <td>0.277205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36897</th>\n",
       "      <td>3</td>\n",
       "      <td>[-107.69, -46.7154, -15.9445]</td>\n",
       "      <td>[-24.167, 8.60087, -26.171]</td>\n",
       "      <td>[-6.4825, 41.4792, 96.2556]</td>\n",
       "      <td>[111.979, 63.7637, 101.178]</td>\n",
       "      <td>[2.38118, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[137.716]</td>\n",
       "      <td>[21.6976]</td>\n",
       "      <td>...</td>\n",
       "      <td>146.659790</td>\n",
       "      <td>22.631247</td>\n",
       "      <td>36.031197</td>\n",
       "      <td>13</td>\n",
       "      <td>51.945667</td>\n",
       "      <td>-31.295113</td>\n",
       "      <td>-22.553812</td>\n",
       "      <td>9</td>\n",
       "      <td>True</td>\n",
       "      <td>0.277205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36900</th>\n",
       "      <td>4</td>\n",
       "      <td>[36.9388, 27.8715, 35.8173, -23.6561]</td>\n",
       "      <td>[35.4823, 34.4244, -21.416, -19.9391]</td>\n",
       "      <td>[-53.2023, -18.4092, -60.2866, 21.9757]</td>\n",
       "      <td>[74.1006, 48.2303, 74.2057, 38.408]</td>\n",
       "      <td>[2.69591, 2.68174, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-31.3944]</td>\n",
       "      <td>[30.6773]</td>\n",
       "      <td>...</td>\n",
       "      <td>-31.754948</td>\n",
       "      <td>30.867546</td>\n",
       "      <td>0.585443</td>\n",
       "      <td>-13</td>\n",
       "      <td>-73.327507</td>\n",
       "      <td>-50.762230</td>\n",
       "      <td>-15.318438</td>\n",
       "      <td>5</td>\n",
       "      <td>True</td>\n",
       "      <td>0.271386</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36906</th>\n",
       "      <td>3</td>\n",
       "      <td>[60.7559, -6.67839, 10.8504]</td>\n",
       "      <td>[-23.811, -47.7918, -33.6265]</td>\n",
       "      <td>[78.5075, 165.349, 5.05383]</td>\n",
       "      <td>[103.611, 172.752, 36.5233]</td>\n",
       "      <td>[-1.0, 1.40379, 4.02371]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-78.0497]</td>\n",
       "      <td>[14.8932]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>19</td>\n",
       "      <td>True</td>\n",
       "      <td>0.036131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36924</th>\n",
       "      <td>1</td>\n",
       "      <td>[44.4301]</td>\n",
       "      <td>[22.9213]</td>\n",
       "      <td>[-53.3913]</td>\n",
       "      <td>[74.282]</td>\n",
       "      <td>[-1.0]</td>\n",
       "      <td>[True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[14.2536]</td>\n",
       "      <td>[-34.5261]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>15</td>\n",
       "      <td>True</td>\n",
       "      <td>0.105165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36932</th>\n",
       "      <td>2</td>\n",
       "      <td>[-189.224, 161.017]</td>\n",
       "      <td>[289.557, -116.198]</td>\n",
       "      <td>[-246.186, -161.373]</td>\n",
       "      <td>[425.525, 257.883]</td>\n",
       "      <td>[-1.0, -1.0]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-23.1623]</td>\n",
       "      <td>[-215.169]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>5</td>\n",
       "      <td>True</td>\n",
       "      <td>0.293529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36936</th>\n",
       "      <td>4</td>\n",
       "      <td>[130.752, -88.6566, 52.0555, -38.6537]</td>\n",
       "      <td>[-25.4033, 73.8251, -74.2651, -30.6417]</td>\n",
       "      <td>[-120.235, -287.902, -124.879, -128.15]</td>\n",
       "      <td>[180.098, 310.715, 154.604, 137.425]</td>\n",
       "      <td>[-1.0, 2.17192, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-44.8889]</td>\n",
       "      <td>[42.0714]</td>\n",
       "      <td>...</td>\n",
       "      <td>-42.990280</td>\n",
       "      <td>40.279636</td>\n",
       "      <td>-78.633194</td>\n",
       "      <td>-13</td>\n",
       "      <td>13.631536</td>\n",
       "      <td>-7.189492</td>\n",
       "      <td>9.089453</td>\n",
       "      <td>12</td>\n",
       "      <td>True</td>\n",
       "      <td>0.019638</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4515 rows × 51 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       NJet                                             Jet_Px  \\\n",
       "7         3                      [-39.2862, 49.3453, -7.69329]   \n",
       "16        5    [184.591, 10.8253, -72.3689, -43.3258, 2.57552]   \n",
       "24        5      [-92.04, 29.9482, 24.2262, 42.4436, -18.0844]   \n",
       "37        3                       [92.6861, -21.7365, 21.6693]   \n",
       "40        3                      [-97.5345, -19.9276, 2.89748]   \n",
       "51        4               [13.2568, 84.261, -4.9345, -33.4614]   \n",
       "57        4             [86.7032, -84.3351, -60.7989, 38.2517]   \n",
       "58        0                                                 []   \n",
       "68        3                      [-72.5721, -37.3589, 54.3353]   \n",
       "80        3                        [394.21, -66.0473, 8.33537]   \n",
       "86        4             [65.0804, 47.7328, -41.3639, -15.2447]   \n",
       "96        2                                 [-40.122, 27.6997]   \n",
       "107       5      [-100.62, 32.0011, -35.6917, 28.3667, 3.8716]   \n",
       "122       3                       [10.4771, -43.1103, 23.1228]   \n",
       "133       3                       [-104.634, 48.1258, 12.1684]   \n",
       "144       3                        [-75.713, 58.5136, 31.0808]   \n",
       "148       4              [-113.872, -32.332, 34.4798, 10.6671]   \n",
       "157       2                                 [39.0372, 18.8909]   \n",
       "168       4            [-23.9239, -6.59758, -14.6238, 45.4713]   \n",
       "171       2                                [-18.0533, 45.1819]   \n",
       "175       2                               [-62.7099, -16.6717]   \n",
       "188       2                                [10.1792, -26.8712]   \n",
       "194       3                       [69.7044, 28.6426, -30.5756]   \n",
       "197       4             [-81.0861, 35.7697, 45.4883, -7.59584]   \n",
       "203       4               [-92.6663, -29.3145, 43.617, 22.991]   \n",
       "204       1                                          [21.5621]   \n",
       "219       2                                 [55.6175, 32.9164]   \n",
       "223       2                                [-63.6374, 21.9639]   \n",
       "235       1                                           [-22.13]   \n",
       "268       3                       [40.9751, -26.4305, 8.81603]   \n",
       "...     ...                                                ...   \n",
       "36671     4               [63.6192, 15.5392, -24.603, -22.424]   \n",
       "36674     3                      [-63.3877, -42.8766, 42.7456]   \n",
       "36678     4             [-42.7893, 41.4222, -12.0665, 17.7019]   \n",
       "36691     4              [-58.3385, -18.4484, 15.4408, 37.792]   \n",
       "36697     0                                                 []   \n",
       "36698     2                                 [49.845, -38.1155]   \n",
       "36704     3                      [44.5244, -22.9694, -5.49336]   \n",
       "36705     4             [95.7703, -91.2611, -22.8162, 11.3628]   \n",
       "36723     2                                [-29.0196, 40.0986]   \n",
       "36728     4             [-37.2348, -19.6443, 35.2389, 36.7954]   \n",
       "36733     2                                [90.0657, -63.0919]   \n",
       "36761     5   [-72.7327, 87.8201, -48.6379, 35.3672, -29.3188]   \n",
       "36773     2                                 [42.3535, 7.84763]   \n",
       "36777     5   [-99.6067, -3.34446, 32.1012, 42.7883, -11.6452]   \n",
       "36779     3                      [-163.892, -104.229, 43.5479]   \n",
       "36845     2                                 [56.3493, 19.3332]   \n",
       "36846     2                                 [63.5405, 61.2992]   \n",
       "36854     6  [23.5022, -29.0738, 27.8149, 27.4214, 0.959874...   \n",
       "36862     4               [65.2048, 26.1962, -27.163, 38.1059]   \n",
       "36864     4              [-105.143, 96.585, -95.1253, 80.7612]   \n",
       "36873     4               [237.71, -193.887, 35.7849, 28.4387]   \n",
       "36884     3                       [68.6953, -68.5364, 46.7583]   \n",
       "36888     2                               [-99.3072, -29.3348]   \n",
       "36892     4            [-158.273, -67.7991, -33.143, -1.02856]   \n",
       "36897     3                      [-107.69, -46.7154, -15.9445]   \n",
       "36900     4              [36.9388, 27.8715, 35.8173, -23.6561]   \n",
       "36906     3                       [60.7559, -6.67839, 10.8504]   \n",
       "36924     1                                          [44.4301]   \n",
       "36932     2                                [-189.224, 161.017]   \n",
       "36936     4             [130.752, -88.6566, 52.0555, -38.6537]   \n",
       "\n",
       "                                                  Jet_Py  \\\n",
       "7                           [-32.3978, 5.15671, 41.4265]   \n",
       "16         [117.452, -166.958, 48.44, -5.91005, 42.4344]   \n",
       "24       [51.4753, -84.1564, 48.5667, -8.55752, 24.7327]   \n",
       "37                           [-126.292, 52.302, 31.9107]   \n",
       "40                          [-13.185, -80.3989, 30.9127]   \n",
       "51               [188.902, -129.347, -121.279, -62.2332]   \n",
       "57              [0.745445, -0.972272, 49.1331, -22.5668]   \n",
       "58                                                    []   \n",
       "68                          [-21.6138, 44.3861, -11.165]   \n",
       "80                         [-174.289, 96.3359, -45.8379]   \n",
       "86                [-84.0002, 73.0816, -21.3092, 30.1185]   \n",
       "96                                  [-44.2169, -41.4156]   \n",
       "107      [-4.6241, -39.0739, 26.2937, -25.9119, 34.3113]   \n",
       "122                         [65.1101, -9.28895, 33.4949]   \n",
       "133                           [29.6702, -66.15, 34.2625]   \n",
       "144                        [44.9549, -38.7253, -41.9073]   \n",
       "148               [-23.9376, 35.5097, -6.87441, 30.7233]   \n",
       "157                                   [56.7757, -41.463]   \n",
       "168                [60.4995, -51.068, 46.1813, -10.2181]   \n",
       "171                                   [75.6471, -25.322]   \n",
       "175                                  [62.0686, -68.4154]   \n",
       "188                                   [55.9482, -16.998]   \n",
       "194                        [-21.1823, 21.4084, -5.88853]   \n",
       "197                 [-22.679, -41.042, 24.9598, 37.9333]   \n",
       "203             [-9.88577, -39.3474, 0.0992902, 32.9985]   \n",
       "204                                           [-30.0244]   \n",
       "219                                  [39.0514, -11.1956]   \n",
       "223                                  [58.9319, -21.1155]   \n",
       "235                                           [-44.9609]   \n",
       "268                        [0.197465, 19.2683, -30.2287]   \n",
       "...                                                  ...   \n",
       "36671                [46.3087, -54.6324, 33.719, 33.788]   \n",
       "36674                        [16.5414, 9.93516, 3.20814]   \n",
       "36678             [-36.8382, -15.6533, 37.0771, 30.5058]   \n",
       "36691              [-4.25548, -43.2476, 42.527, 3.44773]   \n",
       "36697                                                 []   \n",
       "36698                               [-31.4604, -12.8079]   \n",
       "36704                       [52.0617, -22.2976, -29.943]   \n",
       "36705            [-61.2077, -23.8256, -90.0745, 70.7234]   \n",
       "36723                               [-28.8848, -6.73161]   \n",
       "36728             [-58.4847, 53.7423, 23.7087, 0.172772]   \n",
       "36733                                [57.5107, -48.7488]   \n",
       "36761     [-71.4104, 25.7807, -24.6549, 37.8814, 26.399]   \n",
       "36773                                [7.22512, -38.6475]   \n",
       "36777   [-25.3014, 92.4234, -32.9203, 0.946451, 36.2387]   \n",
       "36779                      [15.7904, -1.47729, -1.82685]   \n",
       "36845                                [-62.0922, 23.8897]   \n",
       "36846                                 [33.7838, 6.70778]   \n",
       "36854  [-79.9664, -53.5415, 44.6099, 26.0418, 33.4664...   \n",
       "36862              [-77.5005, 70.3165, -45.9601, 33.931]   \n",
       "36864             [-153.554, 39.1784, 37.6713, -11.9996]   \n",
       "36873            [-69.9921, 0.191406, 13.5153, -21.4827]   \n",
       "36884                        [22.5236, 14.8628, -42.646]   \n",
       "36888                                [49.2617, -29.6785]   \n",
       "36892            [201.345, -24.0705, -49.9765, -31.1794]   \n",
       "36897                        [-24.167, 8.60087, -26.171]   \n",
       "36900              [35.4823, 34.4244, -21.416, -19.9391]   \n",
       "36906                      [-23.811, -47.7918, -33.6265]   \n",
       "36924                                          [22.9213]   \n",
       "36932                                [289.557, -116.198]   \n",
       "36936            [-25.4033, 73.8251, -74.2651, -30.6417]   \n",
       "\n",
       "                                                  Jet_Pz  \\\n",
       "7                         [-65.3803, -95.7182, -70.6679]   \n",
       "16       [379.501, -51.8267, 452.426, 33.3998, -77.0341]   \n",
       "24     [-52.9665, -181.678, -10.7411, -31.6566, -43.3...   \n",
       "37                         [88.5404, -12.5433, -18.5118]   \n",
       "40                           [-11.907, 51.528, -91.2712]   \n",
       "51                 [-67.8184, 185.424, 164.392, 53.9148]   \n",
       "57                 [-77.6728, -19.5303, 3.48934, 48.336]   \n",
       "58                                                    []   \n",
       "68                        [-203.388, -141.191, -148.702]   \n",
       "80                          [-363.242, 1.3422, -32.5724]   \n",
       "86              [-19.6535, -286.044, -80.5942, -45.1493]   \n",
       "96                                    [10.1575, 29.0914]   \n",
       "107       [-135.39, 38.0203, -23.802, -58.4347, 29.4247]   \n",
       "122                         [-64.5278, 20.2662, 20.8267]   \n",
       "133                        [-72.2929, -229.216, 13.3559]   \n",
       "144                       [-195.356, -125.271, -35.4589]   \n",
       "148                [87.0612, 45.5352, 72.4027, -44.7416]   \n",
       "157                                 [-154.183, -153.292]   \n",
       "168               [57.153, -172.568, -82.0874, -18.6692]   \n",
       "171                                  [-59.945, -22.2292]   \n",
       "175                                  [-56.106, -164.618]   \n",
       "188                                  [-89.4134, 72.4999]   \n",
       "194                          [189.136, 125.371, 28.3306]   \n",
       "197              [22.1584, -50.1099, -306.429, -61.9009]   \n",
       "203               [-19.9208, 24.1042, -28.9692, 73.4268]   \n",
       "204                                            [160.531]   \n",
       "219                                 [-44.6516, -31.1788]   \n",
       "223                                   [131.927, 79.1002]   \n",
       "235                                           [-56.7319]   \n",
       "268                          [-10.3833, 126.664, 35.724]   \n",
       "...                                                  ...   \n",
       "36671            [-201.403, 121.437, -185.224, -41.4796]   \n",
       "36674                        [57.6559, 93.1895, 27.8355]   \n",
       "36678            [-41.9049, -4.82216, 5.23354, -70.8694]   \n",
       "36691              [145.612, 57.8691, 70.6434, -58.7726]   \n",
       "36697                                                 []   \n",
       "36698                               [-105.783, -222.021]   \n",
       "36704                      [-123.699, 45.3537, -114.566]   \n",
       "36705               [14.1595, 200.652, 21.1822, 282.059]   \n",
       "36723                                  [-10.89, 10.5535]   \n",
       "36728           [-140.538, -103.277, -81.6349, -79.4282]   \n",
       "36733                                [-176.365, 13.9594]   \n",
       "36761   [-25.5412, -27.7219, 69.2121, 166.603, -101.946]   \n",
       "36773                                 [24.7581, 90.8231]   \n",
       "36777       [155.932, 174.57, 38.0382, 25.5367, 18.2656]   \n",
       "36779                     [-387.744, -61.9309, -63.0709]   \n",
       "36845                                 [98.1019, 8.61856]   \n",
       "36846                                [-10.9178, 69.2736]   \n",
       "36854  [7.103, 22.0331, -156.058, -22.6882, -58.1608,...   \n",
       "36862             [-55.8269, -3.97152, 95.8268, 287.091]   \n",
       "36864            [-238.466, -25.2841, -9.86635, -101.42]   \n",
       "36873            [-152.786, 436.524, -0.336736, 2.15682]   \n",
       "36884                        [15.618, -29.9445, 35.7689]   \n",
       "36888                                [155.757, -6.78339]   \n",
       "36892            [-56.5885, 108.054, -133.396, -40.5998]   \n",
       "36897                        [-6.4825, 41.4792, 96.2556]   \n",
       "36900            [-53.2023, -18.4092, -60.2866, 21.9757]   \n",
       "36906                        [78.5075, 165.349, 5.05383]   \n",
       "36924                                         [-53.3913]   \n",
       "36932                               [-246.186, -161.373]   \n",
       "36936            [-120.235, -287.902, -124.879, -128.15]   \n",
       "\n",
       "                                                   Jet_E  \\\n",
       "7                            [83.5105, 108.047, 82.8404]   \n",
       "16         [438.532, 176.045, 461.069, 56.1953, 88.4542]   \n",
       "24         [121.717, 203.098, 57.0088, 53.9551, 53.2984]   \n",
       "37                           [180.864, 58.5981, 43.3167]   \n",
       "40                           [101.636, 98.2916, 96.7967]   \n",
       "51                   [202.167, 242.673, 205.04, 89.4054]   \n",
       "57                  [117.088, 87.3901, 78.8732, 66.4349]   \n",
       "58                                                    []   \n",
       "68                           [217.305, 152.979, 159.159]   \n",
       "80                            [574.341, 117.572, 57.576]   \n",
       "86                   [109.014, 299.209, 93.831, 56.4601]   \n",
       "96                                    [62.1224, 58.2074]   \n",
       "107        [171.725, 64.9216, 51.6265, 70.5236, 45.8437]   \n",
       "122                           [92.8245, 48.7273, 46.165]   \n",
       "133                          [131.668, 243.703, 38.9503]   \n",
       "144                          [214.638, 144.113, 64.1693]   \n",
       "148                 [145.824, 66.5828, 80.8721, 55.7227]   \n",
       "157                                   [169.354, 160.027]   \n",
       "168                   [87.7235, 180.256, 95.6667, 50.63]   \n",
       "171                                   [98.8365, 57.0231]   \n",
       "175                                   [105.655, 179.547]   \n",
       "188                                   [106.531, 79.6631]   \n",
       "194                          [203.069, 130.501, 42.8366]   \n",
       "197                 [87.7846, 74.7338, 311.244, 73.3563]   \n",
       "203                 [96.1515, 55.6889, 53.1124, 84.3405]   \n",
       "204                                            [164.844]   \n",
       "219                                   [82.3264, 47.1459]   \n",
       "223                                   [159.269, 85.0185]   \n",
       "235                                            [77.0054]   \n",
       "268                          [42.9108, 130.983, 47.8053]   \n",
       "...                                                  ...   \n",
       "36671                [216.771, 134.415, 189.95, 59.4113]   \n",
       "36674                         [87.8452, 103.46, 51.6133]   \n",
       "36678                [70.8273, 45.4714, 39.5967, 79.481]   \n",
       "36691               [157.184, 75.1443, 84.1751, 70.7867]   \n",
       "36697                                                 []   \n",
       "36698                                 [121.478, 225.912]   \n",
       "36704                          [141.7, 56.1422, 118.853]   \n",
       "36705               [116.274, 222.271, 96.4095, 291.297]   \n",
       "36723                                  [43.0001, 42.402]   \n",
       "36728                [157.469, 118.497, 92.416, 87.6949]   \n",
       "36733                                  [207.504, 81.852]   \n",
       "36761      [105.677, 96.5188, 88.2745, 174.643, 109.666]   \n",
       "36773                                 [50.1142, 99.2029]   \n",
       "36777       [187.385, 198.147, 60.2857, 50.5626, 42.798]   \n",
       "36779                           [423.49, 123.1, 77.3157]   \n",
       "36845                                 [129.654, 32.2253]   \n",
       "36846                                 [73.4685, 93.2766]   \n",
       "36854  [85.9293, 65.432, 165.279, 44.604, 67.6031, 13...   \n",
       "36862                [117.233, 75.8283, 110.272, 291.82]   \n",
       "36864               [304.335, 108.426, 103.746, 130.728]   \n",
       "36873                [293.461, 478.051, 39.3145, 36.544]   \n",
       "36884                        [75.5679, 77.1019, 73.1977]   \n",
       "36888                                 [191.845, 42.8853]   \n",
       "36892               [270.567, 130.381, 146.749, 51.5353]   \n",
       "36897                        [111.979, 63.7637, 101.178]   \n",
       "36900                [74.1006, 48.2303, 74.2057, 38.408]   \n",
       "36906                        [103.611, 172.752, 36.5233]   \n",
       "36924                                           [74.282]   \n",
       "36932                                 [425.525, 257.883]   \n",
       "36936               [180.098, 310.715, 154.604, 137.425]   \n",
       "\n",
       "                                   Jet_btag  \\\n",
       "7                     [2.42398, -1.0, -1.0]   \n",
       "16        [-1.0, 3.20327, -1.0, -1.0, -1.0]   \n",
       "24           [-1.0, -1.0, -1.0, -1.0, -1.0]   \n",
       "37                 [2.59038, 2.73967, -1.0]   \n",
       "40                    [2.89632, -1.0, -1.0]   \n",
       "51           [2.31161, -1.0, -1.0, 3.40863]   \n",
       "57              [1.94137, -1.0, -1.0, -1.0]   \n",
       "58                                       []   \n",
       "68                 [1.64367, 2.30578, -1.0]   \n",
       "80                 [2.80352, 1.69599, -1.0]   \n",
       "86              [3.79264, -1.0, -1.0, -1.0]   \n",
       "96                          [-1.0, 1.66511]   \n",
       "107       [-1.0, -1.0, -1.0, 3.47004, -1.0]   \n",
       "122                    [4.1136, -1.0, -1.0]   \n",
       "133                   [3.45501, -1.0, -1.0]   \n",
       "144                 [2.9908, 3.62441, -1.0]   \n",
       "148           [2.06658, -1.0, -1.0, 3.2202]   \n",
       "157                         [-1.0, 2.21366]   \n",
       "168             [3.08056, -1.0, -1.0, -1.0]   \n",
       "171                         [1.58192, -1.0]   \n",
       "175                            [-1.0, -1.0]   \n",
       "188                         [-1.0, 2.40431]   \n",
       "194                   [2.25507, -1.0, -1.0]   \n",
       "197          [4.60365, 1.62578, -1.0, -1.0]   \n",
       "203             [2.15808, -1.0, -1.0, -1.0]   \n",
       "204                                  [-1.0]   \n",
       "219                         [2.18723, -1.0]   \n",
       "223                            [-1.0, -1.0]   \n",
       "235                                  [-1.0]   \n",
       "268                   [-1.0, 1.85155, -1.0]   \n",
       "...                                     ...   \n",
       "36671           [-1.0, 2.30456, -1.0, -1.0]   \n",
       "36674                   [2.389, -1.0, -1.0]   \n",
       "36678           [3.14142, -1.0, -1.0, -1.0]   \n",
       "36691              [-1.0, -1.0, -1.0, -1.0]   \n",
       "36697                                    []   \n",
       "36698                          [-1.0, -1.0]   \n",
       "36704                    [-1.0, -1.0, -1.0]   \n",
       "36705              [-1.0, -1.0, -1.0, -1.0]   \n",
       "36723                       [1.72804, -1.0]   \n",
       "36728              [-1.0, -1.0, -1.0, -1.0]   \n",
       "36733                    [2.16326, 3.64163]   \n",
       "36761  [2.64415, -1.0, 2.74785, -1.0, -1.0]   \n",
       "36773                       [2.67679, -1.0]   \n",
       "36777     [-1.0, 3.07208, -1.0, -1.0, -1.0]   \n",
       "36779                  [-1.0, -1.0, 2.4724]   \n",
       "36845                       [2.90254, -1.0]   \n",
       "36846                          [-1.0, -1.0]   \n",
       "36854  [-1.0, -1.0, -1.0, -1.0, -1.0, -1.0]   \n",
       "36862         [2.6823, 2.13792, -1.0, -1.0]   \n",
       "36864         [-1.0, 4.3339, 4.28816, -1.0]   \n",
       "36873        [-1.0, 3.04585, -1.0, 3.54336]   \n",
       "36884           [2.38775, 2.44255, 1.78135]   \n",
       "36888                       [-1.0, 4.18899]   \n",
       "36892           [-1.0, 4.01285, -1.0, -1.0]   \n",
       "36897                 [2.38118, -1.0, -1.0]   \n",
       "36900        [2.69591, 2.68174, -1.0, -1.0]   \n",
       "36906              [-1.0, 1.40379, 4.02371]   \n",
       "36924                                [-1.0]   \n",
       "36932                          [-1.0, -1.0]   \n",
       "36936           [-1.0, 2.17192, -1.0, -1.0]   \n",
       "\n",
       "                                     Jet_ID  NMuon              Muon_Px  \\\n",
       "7                        [True, True, True]      1           [-34.0692]   \n",
       "16           [True, True, True, True, True]      1            [42.3929]   \n",
       "24           [True, True, True, True, True]      1            [26.0595]   \n",
       "37                       [True, True, True]      1           [-13.8492]   \n",
       "40                       [True, True, True]      1            [72.6027]   \n",
       "51                 [True, True, True, True]      2   [-41.887, 8.97538]   \n",
       "57                 [True, True, True, True]      0                   []   \n",
       "58                                       []      1           [-73.5007]   \n",
       "68                       [True, True, True]      1           [-1.18855]   \n",
       "80                       [True, True, True]      1           [-66.4606]   \n",
       "86                 [True, True, True, True]      1            [9.05585]   \n",
       "96                             [True, True]      1            [42.9513]   \n",
       "107          [True, True, True, True, True]      1            [32.2751]   \n",
       "122                      [True, True, True]      1          [-0.298645]   \n",
       "133                      [True, True, True]      0                   []   \n",
       "144                      [True, True, True]      1           [-45.2115]   \n",
       "148                [True, True, True, True]      1            [74.4426]   \n",
       "157                            [True, True]      1           [-81.4154]   \n",
       "168                [True, True, True, True]      1            [42.5798]   \n",
       "171                            [True, True]      1            [-2.2919]   \n",
       "175                            [True, True]      1            [52.0206]   \n",
       "188                            [True, True]      1           [-65.1655]   \n",
       "194                      [True, True, True]      1           [-84.0036]   \n",
       "197                [True, True, True, True]      1           [-21.7847]   \n",
       "203                [True, True, True, True]      1            [51.9452]   \n",
       "204                                  [True]      2   [109.452, -9.8738]   \n",
       "219                            [True, True]      1           [-17.0189]   \n",
       "223                            [True, True]      1            [39.4852]   \n",
       "235                                  [True]      1            [18.7418]   \n",
       "268                      [True, True, True]      0                   []   \n",
       "...                                     ...    ...                  ...   \n",
       "36671              [True, True, True, True]      1            [44.2237]   \n",
       "36674                    [True, True, True]      1           [-19.4274]   \n",
       "36678              [True, True, True, True]      1           [-1.03951]   \n",
       "36691              [True, True, True, True]      1           [-45.0365]   \n",
       "36697                                    []      2  [72.8115, -32.1987]   \n",
       "36698                          [True, True]      1            [39.8958]   \n",
       "36704                    [True, True, True]      1           [-18.3659]   \n",
       "36705              [True, True, True, True]      1           [-68.7636]   \n",
       "36723                          [True, True]      1            [48.2104]   \n",
       "36728              [True, True, True, True]      0                   []   \n",
       "36733                          [True, True]      1            [29.2887]   \n",
       "36761        [True, True, True, True, True]      1           [-15.0587]   \n",
       "36773                          [True, True]      1            [-72.876]   \n",
       "36777        [True, True, True, True, True]      1            [23.5196]   \n",
       "36779                    [True, True, True]      1            [204.758]   \n",
       "36845                          [True, True]      1           [-58.4214]   \n",
       "36846                          [True, True]      1           [-79.0116]   \n",
       "36854  [True, True, True, True, True, True]      1           [-32.9304]   \n",
       "36862              [True, True, True, True]      0                   []   \n",
       "36864              [True, True, True, True]      0                   []   \n",
       "36873              [True, True, True, True]      1           [-45.9498]   \n",
       "36884                    [True, True, True]      1           [-10.2928]   \n",
       "36888                          [True, True]      1            [114.711]   \n",
       "36892              [True, True, True, True]      1            [-101.12]   \n",
       "36897                    [True, True, True]      1            [137.716]   \n",
       "36900              [True, True, True, True]      1           [-31.3944]   \n",
       "36906                    [True, True, True]      1           [-78.0497]   \n",
       "36924                                [True]      1            [14.2536]   \n",
       "36932                          [True, True]      1           [-23.1623]   \n",
       "36936              [True, True, True, True]      1           [-44.8889]   \n",
       "\n",
       "                   Muon_Py     ...     MClepton_px MClepton_py MClepton_pz  \\\n",
       "7                [5.64295]     ...      -34.540741    5.717676  -10.324351   \n",
       "16              [-35.9437]     ...       43.464474  -37.058472  -30.771671   \n",
       "24              [0.139715]     ...       27.888340   -2.613286 -114.308418   \n",
       "37               [38.0815]     ...        0.000000    0.000000    0.000000   \n",
       "40              [-9.59597]     ...        0.000000    0.000000    0.000000   \n",
       "51      [108.296, 21.3534]     ...        0.000000    0.000000    0.000000   \n",
       "57                      []     ...      -51.339859   47.421997    9.806538   \n",
       "58                [47.405]     ...      -77.774635   49.710968 -163.584732   \n",
       "68              [-36.1914]     ...       -1.134651  -35.692688 -109.194305   \n",
       "80              [-1.64239]     ...     -100.356377   -2.219039  -10.412981   \n",
       "86              [-24.4455]     ...        8.528934  -24.427439   27.513771   \n",
       "96               [45.6999]     ...       44.989742   46.907940   50.970924   \n",
       "107             [-59.3023]     ...       33.404510  -60.193768  -54.171921   \n",
       "122             [-52.4379]     ...       -2.104305  -95.161293 -309.347382   \n",
       "133                     []     ...        0.000000    0.000000    0.000000   \n",
       "144              [5.31683]     ...      -44.851936    5.279506   -0.475743   \n",
       "148             [-47.3361]     ...       74.646675  -48.276752  -21.981726   \n",
       "157              [11.1185]     ...      -88.916168   11.958924 -100.715332   \n",
       "168             [-37.9202]     ...       42.677971  -38.684910   25.826906   \n",
       "171             [-55.9058]     ...       -2.266898  -56.387188   84.049629   \n",
       "175              [100.328]     ...        0.000000    0.000000    0.000000   \n",
       "188              [22.1073]     ...        0.000000    0.000000    0.000000   \n",
       "194             [-38.9238]     ...      -79.938377  -35.873989  -74.092720   \n",
       "197              [10.4468]     ...      -21.555775   10.569125   16.874752   \n",
       "203              [-34.679]     ...       51.943840  -35.272839  -15.022160   \n",
       "204    [191.047, -12.5724]     ...        0.000000    0.000000    0.000000   \n",
       "219              [47.6267]     ...      -17.056755   47.598736    0.132445   \n",
       "223             [-28.8416]     ...       40.519344  -29.519676   67.463295   \n",
       "235              [37.8549]     ...        0.000000    0.000000    0.000000   \n",
       "268                     []     ...      -53.865688    5.485102  147.813019   \n",
       "...                    ...     ...             ...         ...         ...   \n",
       "36671            [1.19632]     ...       44.393093    4.441749   17.649075   \n",
       "36674           [-29.8691]     ...      -19.507946  -30.081053   47.349102   \n",
       "36678           [-37.0587]     ...       -0.284460  -37.510612 -173.577927   \n",
       "36691           [-3.44792]     ...      -46.260303   -3.330531  -12.245513   \n",
       "36697  [-41.4184, 46.7349]     ...        0.000000    0.000000    0.000000   \n",
       "36698           [-8.11494]     ...       41.240803   -6.950862 -118.234520   \n",
       "36704            [38.4539]     ...      -19.019670   38.604202  -15.343031   \n",
       "36705            [41.4596]     ...        0.000000    0.000000    0.000000   \n",
       "36723            [10.7573]     ...        0.000000    0.000000    0.000000   \n",
       "36728                   []     ...        0.000000    0.000000    0.000000   \n",
       "36733           [-13.7026]     ...       29.956720  -13.935602    5.496824   \n",
       "36761           [-19.3267]     ...      -14.784895  -19.563482  -41.428085   \n",
       "36773            [53.7722]     ...        0.000000    0.000000    0.000000   \n",
       "36777           [-33.3833]     ...       19.755816  -44.092232  172.988556   \n",
       "36779            [3.03201]     ...      194.055786    2.588275   31.950975   \n",
       "36845            [49.5135]     ...        0.000000    0.000000    0.000000   \n",
       "36846           [-19.7364]     ...      -77.586594  -19.407608  -66.658394   \n",
       "36854            [34.6853]     ...      -32.763885   34.414585  -20.146509   \n",
       "36862                   []     ...        0.000000    0.000000    0.000000   \n",
       "36864                   []     ...       18.866938  -15.149240    7.708319   \n",
       "36873            [59.5273]     ...      -48.707134   60.437092  219.109711   \n",
       "36884            [25.1898]     ...      -10.396090   25.239723  -11.126865   \n",
       "36888           [-47.2675]     ...        0.000000    0.000000    0.000000   \n",
       "36892           [-152.441]     ...        0.000000    0.000000    0.000000   \n",
       "36897            [21.6976]     ...      146.659790   22.631247   36.031197   \n",
       "36900            [30.6773]     ...      -31.754948   30.867546    0.585443   \n",
       "36906            [14.8932]     ...        0.000000    0.000000    0.000000   \n",
       "36924           [-34.5261]     ...        0.000000    0.000000    0.000000   \n",
       "36932           [-215.169]     ...        0.000000    0.000000    0.000000   \n",
       "36936            [42.0714]     ...      -42.990280   40.279636  -78.633194   \n",
       "\n",
       "      MCleptonPDGid  MCneutrino_px MCneutrino_py MCneutrino_pz  \\\n",
       "7               -13      43.943020      4.711743    -16.245531   \n",
       "16               13     -21.166794     19.246643    -13.815475   \n",
       "24               13     -43.849094    -32.715321    -46.295307   \n",
       "37                0       0.000000      0.000000      0.000000   \n",
       "40                0       0.000000      0.000000      0.000000   \n",
       "51                0       0.000000      0.000000      0.000000   \n",
       "57              -13       7.199954     33.037640     64.709595   \n",
       "58               13     -14.959000    -17.021935   -129.785049   \n",
       "68               13      67.564972     -8.594560    -78.256340   \n",
       "80              -13    -246.632401    112.364296    -81.893837   \n",
       "86               13     -51.809479      8.691632    -20.814299   \n",
       "96              -13      13.318195    -31.906128     19.553530   \n",
       "107              13      44.509659     19.568098    -44.442898   \n",
       "122             -15      -8.769296    -14.520910      4.469729   \n",
       "133               0       0.000000      0.000000      0.000000   \n",
       "144              13      20.658106     36.965542    -45.506985   \n",
       "148              13     -14.983271    -19.674046    -33.766380   \n",
       "157             -13      -8.342011    -37.894085    -57.389412   \n",
       "168              13     -35.230225    -32.083885     29.348593   \n",
       "171              13      17.545727     17.956202     95.806236   \n",
       "175               0       0.000000      0.000000      0.000000   \n",
       "188               0       0.000000      0.000000      0.000000   \n",
       "194              13     -14.628411     30.902468    -51.598186   \n",
       "197              13      66.121849     17.503078      1.514331   \n",
       "203              13     -14.474733     10.930140     20.586962   \n",
       "204               0       0.000000      0.000000      0.000000   \n",
       "219             -13     -18.238882      4.040609     71.326790   \n",
       "223             -13       2.028334     13.116418    -16.692287   \n",
       "235               0       0.000000      0.000000      0.000000   \n",
       "268              13       6.192799    -14.861897    285.897095   \n",
       "...             ...            ...           ...           ...   \n",
       "36671            13     -37.079338      4.005792     22.262280   \n",
       "36674           -13      57.827797    -55.537346     26.413647   \n",
       "36678           -13     -17.592842    -28.825563    -12.680113   \n",
       "36691            13      34.181591     14.543386      6.494123   \n",
       "36697             0       0.000000      0.000000      0.000000   \n",
       "36698           -13     -24.029903     32.869717    -56.536091   \n",
       "36704           -13     -85.043907      6.205180   -118.760750   \n",
       "36705             0       0.000000      0.000000      0.000000   \n",
       "36723             0       0.000000      0.000000      0.000000   \n",
       "36728             0       0.000000      0.000000      0.000000   \n",
       "36733            13      -3.975806     18.813091    100.464325   \n",
       "36761            13      44.433407     46.769634   -128.488770   \n",
       "36773             0       0.000000      0.000000      0.000000   \n",
       "36777           -13      28.066628    -17.915775      9.881391   \n",
       "36779           -13      36.918297    -33.213650    -13.392416   \n",
       "36845             0       0.000000      0.000000      0.000000   \n",
       "36846            13     -72.814438    -36.851547     17.604248   \n",
       "36854            13     -34.493790      9.358992   -145.859955   \n",
       "36862             0       0.000000      0.000000      0.000000   \n",
       "36864           -13       1.578786     85.071571     34.158749   \n",
       "36873           -13     -32.495464     14.442889     15.860429   \n",
       "36884            13     -42.048183    -10.150770     74.172813   \n",
       "36888             0       0.000000      0.000000      0.000000   \n",
       "36892             0       0.000000      0.000000      0.000000   \n",
       "36897            13      51.945667    -31.295113    -22.553812   \n",
       "36900           -13     -73.327507    -50.762230    -15.318438   \n",
       "36906             0       0.000000      0.000000      0.000000   \n",
       "36924             0       0.000000      0.000000      0.000000   \n",
       "36932             0       0.000000      0.000000      0.000000   \n",
       "36936           -13      13.631536     -7.189492      9.089453   \n",
       "\n",
       "      NPrimaryVertices triggerIsoMu24 EventWeight  \n",
       "7                    4           True    0.279545  \n",
       "16                   4           True    0.279545  \n",
       "24                   8           True    0.299037  \n",
       "37                   1           True    0.270251  \n",
       "40                  16           True    0.010469  \n",
       "51                   6           True    0.293529  \n",
       "57                  13           True    0.026791  \n",
       "58                  20           True    0.010469  \n",
       "68                   6           True    0.286201  \n",
       "80                   3           True    0.271386  \n",
       "86                  11           True    0.105165  \n",
       "96                  10           True    0.299981  \n",
       "107                  1           True    0.270251  \n",
       "122                  6           True    0.286201  \n",
       "133                  1           True    0.270251  \n",
       "144                  6           True    0.286201  \n",
       "148                 19           True    0.014366  \n",
       "157                  5           True    0.279545  \n",
       "168                  4           True    0.274432  \n",
       "171                  3           True    0.279545  \n",
       "175                  3           True    0.293529  \n",
       "188                  6           True    0.293529  \n",
       "194                  5           True    0.293529  \n",
       "197                  5           True    0.286201  \n",
       "203                 11           True    0.160826  \n",
       "204                  4           True    0.274432  \n",
       "219                  4           True    0.279545  \n",
       "223                 25           True    0.001839  \n",
       "235                 15           True    0.005580  \n",
       "268                  7           True    0.293529  \n",
       "...                ...            ...         ...  \n",
       "36671               13           True    0.192474  \n",
       "36674               16           True    0.105165  \n",
       "36678               11           True    0.192474  \n",
       "36691                4           True    0.279545  \n",
       "36697               15           True    0.036131  \n",
       "36698                5           True    0.274432  \n",
       "36704                8           True    0.253713  \n",
       "36705               10           True    0.293189  \n",
       "36723                3           True    0.271386  \n",
       "36728                5           True    0.293529  \n",
       "36733                4           True    0.293529  \n",
       "36761                9           True    0.277205  \n",
       "36773                5           True    0.286201  \n",
       "36777                6           True    0.293529  \n",
       "36779               17           True    0.014366  \n",
       "36845               10           True    0.277205  \n",
       "36846                6           True    0.293529  \n",
       "36854               19           True    0.026791  \n",
       "36862                9           True    0.277205  \n",
       "36864               11           True    0.160826  \n",
       "36873                9           True    0.293189  \n",
       "36884                3           True    0.270077  \n",
       "36888                8           True    0.299981  \n",
       "36892                9           True    0.277205  \n",
       "36897                9           True    0.277205  \n",
       "36900                5           True    0.271386  \n",
       "36906               19           True    0.036131  \n",
       "36924               15           True    0.105165  \n",
       "36932                5           True    0.293529  \n",
       "36936               12           True    0.019638  \n",
       "\n",
       "[4515 rows x 51 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ttbar[ttbar['triggerIsoMu24']] # slicing doesn't automatically reset the indices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "ttbar = ttbar[ttbar['triggerIsoMu24']].reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NJet</th>\n",
       "      <th>Jet_Px</th>\n",
       "      <th>Jet_Py</th>\n",
       "      <th>Jet_Pz</th>\n",
       "      <th>Jet_E</th>\n",
       "      <th>Jet_btag</th>\n",
       "      <th>Jet_ID</th>\n",
       "      <th>NMuon</th>\n",
       "      <th>Muon_Px</th>\n",
       "      <th>Muon_Py</th>\n",
       "      <th>...</th>\n",
       "      <th>MClepton_px</th>\n",
       "      <th>MClepton_py</th>\n",
       "      <th>MClepton_pz</th>\n",
       "      <th>MCleptonPDGid</th>\n",
       "      <th>MCneutrino_px</th>\n",
       "      <th>MCneutrino_py</th>\n",
       "      <th>MCneutrino_pz</th>\n",
       "      <th>NPrimaryVertices</th>\n",
       "      <th>triggerIsoMu24</th>\n",
       "      <th>EventWeight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>[-39.2862, 49.3453, -7.69329]</td>\n",
       "      <td>[-32.3978, 5.15671, 41.4265]</td>\n",
       "      <td>[-65.3803, -95.7182, -70.6679]</td>\n",
       "      <td>[83.5105, 108.047, 82.8404]</td>\n",
       "      <td>[2.42398, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-34.0692]</td>\n",
       "      <td>[5.64295]</td>\n",
       "      <td>...</td>\n",
       "      <td>-34.540741</td>\n",
       "      <td>5.717676</td>\n",
       "      <td>-10.324351</td>\n",
       "      <td>-13</td>\n",
       "      <td>43.943020</td>\n",
       "      <td>4.711743</td>\n",
       "      <td>-16.245531</td>\n",
       "      <td>4</td>\n",
       "      <td>True</td>\n",
       "      <td>0.279545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5</td>\n",
       "      <td>[184.591, 10.8253, -72.3689, -43.3258, 2.57552]</td>\n",
       "      <td>[117.452, -166.958, 48.44, -5.91005, 42.4344]</td>\n",
       "      <td>[379.501, -51.8267, 452.426, 33.3998, -77.0341]</td>\n",
       "      <td>[438.532, 176.045, 461.069, 56.1953, 88.4542]</td>\n",
       "      <td>[-1.0, 3.20327, -1.0, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[42.3929]</td>\n",
       "      <td>[-35.9437]</td>\n",
       "      <td>...</td>\n",
       "      <td>43.464474</td>\n",
       "      <td>-37.058472</td>\n",
       "      <td>-30.771671</td>\n",
       "      <td>13</td>\n",
       "      <td>-21.166794</td>\n",
       "      <td>19.246643</td>\n",
       "      <td>-13.815475</td>\n",
       "      <td>4</td>\n",
       "      <td>True</td>\n",
       "      <td>0.279545</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5</td>\n",
       "      <td>[-92.04, 29.9482, 24.2262, 42.4436, -18.0844]</td>\n",
       "      <td>[51.4753, -84.1564, 48.5667, -8.55752, 24.7327]</td>\n",
       "      <td>[-52.9665, -181.678, -10.7411, -31.6566, -43.3...</td>\n",
       "      <td>[121.717, 203.098, 57.0088, 53.9551, 53.2984]</td>\n",
       "      <td>[-1.0, -1.0, -1.0, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[26.0595]</td>\n",
       "      <td>[0.139715]</td>\n",
       "      <td>...</td>\n",
       "      <td>27.888340</td>\n",
       "      <td>-2.613286</td>\n",
       "      <td>-114.308418</td>\n",
       "      <td>13</td>\n",
       "      <td>-43.849094</td>\n",
       "      <td>-32.715321</td>\n",
       "      <td>-46.295307</td>\n",
       "      <td>8</td>\n",
       "      <td>True</td>\n",
       "      <td>0.299037</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>[92.6861, -21.7365, 21.6693]</td>\n",
       "      <td>[-126.292, 52.302, 31.9107]</td>\n",
       "      <td>[88.5404, -12.5433, -18.5118]</td>\n",
       "      <td>[180.864, 58.5981, 43.3167]</td>\n",
       "      <td>[2.59038, 2.73967, -1.0]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[-13.8492]</td>\n",
       "      <td>[38.0815]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>True</td>\n",
       "      <td>0.270251</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>[-97.5345, -19.9276, 2.89748]</td>\n",
       "      <td>[-13.185, -80.3989, 30.9127]</td>\n",
       "      <td>[-11.907, 51.528, -91.2712]</td>\n",
       "      <td>[101.636, 98.2916, 96.7967]</td>\n",
       "      <td>[2.89632, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[72.6027]</td>\n",
       "      <td>[-9.59597]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>16</td>\n",
       "      <td>True</td>\n",
       "      <td>0.010469</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 51 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   NJet                                           Jet_Px  \\\n",
       "0     3                    [-39.2862, 49.3453, -7.69329]   \n",
       "1     5  [184.591, 10.8253, -72.3689, -43.3258, 2.57552]   \n",
       "2     5    [-92.04, 29.9482, 24.2262, 42.4436, -18.0844]   \n",
       "3     3                     [92.6861, -21.7365, 21.6693]   \n",
       "4     3                    [-97.5345, -19.9276, 2.89748]   \n",
       "\n",
       "                                            Jet_Py  \\\n",
       "0                     [-32.3978, 5.15671, 41.4265]   \n",
       "1    [117.452, -166.958, 48.44, -5.91005, 42.4344]   \n",
       "2  [51.4753, -84.1564, 48.5667, -8.55752, 24.7327]   \n",
       "3                      [-126.292, 52.302, 31.9107]   \n",
       "4                     [-13.185, -80.3989, 30.9127]   \n",
       "\n",
       "                                              Jet_Pz  \\\n",
       "0                     [-65.3803, -95.7182, -70.6679]   \n",
       "1    [379.501, -51.8267, 452.426, 33.3998, -77.0341]   \n",
       "2  [-52.9665, -181.678, -10.7411, -31.6566, -43.3...   \n",
       "3                      [88.5404, -12.5433, -18.5118]   \n",
       "4                        [-11.907, 51.528, -91.2712]   \n",
       "\n",
       "                                           Jet_E  \\\n",
       "0                    [83.5105, 108.047, 82.8404]   \n",
       "1  [438.532, 176.045, 461.069, 56.1953, 88.4542]   \n",
       "2  [121.717, 203.098, 57.0088, 53.9551, 53.2984]   \n",
       "3                    [180.864, 58.5981, 43.3167]   \n",
       "4                    [101.636, 98.2916, 96.7967]   \n",
       "\n",
       "                            Jet_btag                          Jet_ID  NMuon  \\\n",
       "0              [2.42398, -1.0, -1.0]              [True, True, True]      1   \n",
       "1  [-1.0, 3.20327, -1.0, -1.0, -1.0]  [True, True, True, True, True]      1   \n",
       "2     [-1.0, -1.0, -1.0, -1.0, -1.0]  [True, True, True, True, True]      1   \n",
       "3           [2.59038, 2.73967, -1.0]              [True, True, True]      1   \n",
       "4              [2.89632, -1.0, -1.0]              [True, True, True]      1   \n",
       "\n",
       "      Muon_Px     Muon_Py     ...     MClepton_px MClepton_py MClepton_pz  \\\n",
       "0  [-34.0692]   [5.64295]     ...      -34.540741    5.717676  -10.324351   \n",
       "1   [42.3929]  [-35.9437]     ...       43.464474  -37.058472  -30.771671   \n",
       "2   [26.0595]  [0.139715]     ...       27.888340   -2.613286 -114.308418   \n",
       "3  [-13.8492]   [38.0815]     ...        0.000000    0.000000    0.000000   \n",
       "4   [72.6027]  [-9.59597]     ...        0.000000    0.000000    0.000000   \n",
       "\n",
       "  MCleptonPDGid  MCneutrino_px MCneutrino_py MCneutrino_pz NPrimaryVertices  \\\n",
       "0           -13      43.943020      4.711743    -16.245531                4   \n",
       "1            13     -21.166794     19.246643    -13.815475                4   \n",
       "2            13     -43.849094    -32.715321    -46.295307                8   \n",
       "3             0       0.000000      0.000000      0.000000                1   \n",
       "4             0       0.000000      0.000000      0.000000               16   \n",
       "\n",
       "  triggerIsoMu24 EventWeight  \n",
       "0           True    0.279545  \n",
       "1           True    0.279545  \n",
       "2           True    0.299037  \n",
       "3           True    0.270251  \n",
       "4           True    0.010469  \n",
       "\n",
       "[5 rows x 51 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ttbar.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's say we want to start by training a simple model that only relies on event-level variables. The ones available in these samples are:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# names of event-level branches\n",
    "npart = ['NJet', 'NMuon', 'NElectron', 'NPhoton', 'MET_px', 'MET_py']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you are a physicist, the first thing you might want to do is to plot them. We do so very easily using <a href=\"https://matplotlib.org/\">`matplotlib`</a>:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsEAAAKqCAYAAADbgoPtAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\npNr5/wG9iZQN3MqxGgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11f4e8950>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArQAAAKqCAYAAADL4s0xAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x123f8ba50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArQAAAKqCAYAAADL4s0xAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/wE1l871UHbAgAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x125f1d990>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsEAAAKqCAYAAADbgoPtAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x12209a490>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsEAAAKrCAYAAAAQ3lBIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x122e675d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAsEAAAKrCAYAAAAQ3lBIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nM+s7/w89NKMrKftQPQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1257b1c90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for key in npart: # loop through the event-level branches and plot them on separate histograms\n",
    "    \n",
    "    # -- set font and canvas size (optional)\n",
    "    matplotlib.rcParams.update({'font.size': 16})\n",
    "    fig = plt.figure(figsize=(8,8), dpi=100)\n",
    "    \n",
    "    # -- declare common binning strategy (otherwise every histogram will have its own binning)\n",
    "    bins = np.linspace(min(ttbar[key]), max(ttbar[key]) + 1, 30)\n",
    "    \n",
    "    # plot!\n",
    "    _ = plt.hist(ttbar[key], histtype='step', normed=False, bins=bins, weights=ttbar['EventWeight'], label=r'$t\\overline{t}$', linewidth=2)\n",
    "    _ = plt.hist(dy[key], histtype='step', normed=False, bins=bins, weights=dy['EventWeight'], label='Drell Yan')\n",
    "    _ = plt.hist(wj[key], histtype='step', normed=False, bins=bins, weights=wj['EventWeight'], label=r'$W$ + jets')\n",
    "    _ = plt.hist(ww[key], histtype='step', normed=False, bins=bins, weights=ww['EventWeight'], label=r'$WW$')\n",
    "    _ = plt.hist(wz[key], histtype='step', normed=False, bins=bins, weights=wz['EventWeight'], label=r'$WZ$')\n",
    "    _ = plt.hist(zz[key], histtype='step', normed=False, bins=bins, weights=zz['EventWeight'], label=r'$ZZ$')\n",
    "    _ = plt.hist(singletop[key], histtype='step', normed=False, bins=bins, weights=singletop['EventWeight'], label=r'single $t$')\n",
    "    _ = plt.hist(qcd[key], histtype='step', normed=False, bins=bins, weights=qcd['EventWeight'], label='QCD', color='salmon')\n",
    "    \n",
    "    plt.xlabel(key)\n",
    "    plt.yscale('log')\n",
    "    plt.legend(loc='best')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Stack all the backgrounds (shown only for one branch here):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.cm as cm\n",
    "colors = cm.cool(np.linspace(0, 1, 7))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x122178c10>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAcUAAAG3CAYAAADFK05OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x143161210>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(7,7))\n",
    "bins = np.linspace(0, 10, 11)\n",
    "_ = plt.hist([\n",
    "                dy['NJet'], wj['NJet'], ww['NJet'], wz['NJet'], zz['NJet'], singletop['NJet'], qcd['NJet']\n",
    "             ],\n",
    "             histtype='stepfilled',\n",
    "             normed=False,\n",
    "             bins=bins,\n",
    "             weights=[\n",
    "                 dy['EventWeight'], wj['EventWeight'], ww['EventWeight'], wz['EventWeight'], zz['EventWeight'], singletop['EventWeight'], qcd['EventWeight']\n",
    "             ],\n",
    "             label=[ r'Drell Yan', r'$W$ + jets', r'$WW$', r'$WZ$', r'$ZZ$', r'single $t$', 'QCD'\n",
    "             ],\n",
    "             stacked=True,\n",
    "             color=colors)\n",
    "\n",
    "plt.hist(ttbar['NJet'],\n",
    "         histtype='step', normed=False, bins=bins, weights=ttbar['EventWeight'], label=r'$t\\overline{t}$',\n",
    "         linewidth=2, color='black', linestyle='dashed')\n",
    "\n",
    "plt.yscale('log')\n",
    "plt.xlabel('Number of Jets')\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plot 2D histogram of $t\\bar{t}$ MET_px and MET_py:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAFQCAYAAABTS665AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nsxpx0jczqxEnfTOzGvn/79Xn2X7BdbQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1232f7c50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib.colors import LogNorm\n",
    "\n",
    "plt.figure(figsize=(5,5)) # square canvas\n",
    "_ = plt.hist2d(ttbar['MET_px'], ttbar['MET_py'], bins=40, cmap='binary', norm=LogNorm())\n",
    "\n",
    "# add decorations\n",
    "plt.xlabel(r'MET$_{p_x}$') # LaTeX!\n",
    "plt.ylabel(r'MET$_{p_y}$', fontsize=20)\n",
    "_ = plt.colorbar()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Turn Data into ML-Compatible Inputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### What are ML-Compatible Inputs?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In the simplest scenarios of tabular data, Keras, just like scikit-learn, takes as inputs the following objects:\n",
    " *  <h3>$X$</h3>\n",
    " an `ndarray` of dimensions `[nb_examples, nb_features]` containing the distributions to be used as inputs to the model. Each row is an object to classify, each column corresponds to a specific variable.\n",
    " *  <h3>$y$</h3>\n",
    " an `array` of dimensions `[nb_examples]` containing the truth labels indicating the class each object belongs to (for classification), or the continuous target values (for regression).\n",
    " *  <h3>$w$</h3> \n",
    " (optional) an `array` of dimensions `[nb_examples]` containing the weights to be assigned to each example\n",
    " \n",
    "The indices of these objects must map to the same examples. In general, you will want to shuffle and split them into a training and test set."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Forward Pass"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"./images/forward_pass.jpg\">"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If we want to stack multiple dataframes into a single one, we can \"concatenate\" them. To simplify our classification problem, in this tutorial I will only focus on a three-class classification task, in which we aim to separate $t\\bar{t}$ events from two of the main background sources: Drell Yan and W+jets events."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "del ww, wz, zz, singletop, qcd, data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "# -- this will only contain TTbar, Drell Yan and W+jets events (all branches)\n",
    "df_full = pd.concat((ttbar, dy, wj), ignore_index=True)  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "However, we decided we were only going to train on **event-level variables**, so this is would be a more useful `df`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NJet</th>\n",
       "      <th>NMuon</th>\n",
       "      <th>NElectron</th>\n",
       "      <th>NPhoton</th>\n",
       "      <th>MET_px</th>\n",
       "      <th>MET_py</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>20.238705</td>\n",
       "      <td>1.876128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-29.892015</td>\n",
       "      <td>16.498444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-24.056999</td>\n",
       "      <td>-46.160618</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>-30.039965</td>\n",
       "      <td>5.851555</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4.518302</td>\n",
       "      <td>117.570068</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   NJet  NMuon  NElectron  NPhoton     MET_px      MET_py\n",
       "0     3      1          0        0  20.238705    1.876128\n",
       "1     5      1          0        0 -29.892015   16.498444\n",
       "2     5      1          0        0 -24.056999  -46.160618\n",
       "3     3      1          0        0 -30.039965    5.851555\n",
       "4     3      1          0        0   4.518302  117.570068"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# recall: npart was the list of branch names corresponding to event-level variables\n",
    "df =  pd.concat((ttbar[npart], dy[npart], wj[npart]), ignore_index=True)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, turn your new `df` the desired ndarray $X$ that can be directly used for ML applications using this handy `pandas` function:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df.as_matrix()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "numpy.ndarray"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(191981, 6)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The weight array $w$ can also easily be extracted using a similar procedure:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "w =  pd.concat((ttbar['EventWeight'], dy['EventWeight'], wj['EventWeight']), ignore_index=True).values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "numpy.ndarray"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "type(w)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally, generate an array of truth labels $y$ to distinguish among the different classes in the problem:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = []\n",
    "for _df, ID in [(ttbar, 0), (dy, 1), (wj, 2)]:\n",
    "    y.extend([ID] * _df.shape[0])\n",
    "y = np.array(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0, 0, 0, ..., 2, 2, 2])"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# -- check what we created\n",
    "y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Extra Pre-Processing Steps: Shuffling, Splitting into Train-Test, Scaling Inputs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The `sklearn` function <a href='http://scikit-learn.org/stable/modules/generated/sklearn.model_selection.train_test_split.html#sklearn.model_selection.train_test_split'>`train_test_split`</a> will randomly shuffle and automatically split all your objects into train and test subsets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/mp744/venvs/keras2tf1/lib/python2.7/site-packages/sklearn/model_selection/_split.py:2026: FutureWarning: From version 0.21, test_size will always complement train_size unless both are specified.\n",
      "  FutureWarning)\n"
     ]
    }
   ],
   "source": [
    "ix = range(X.shape[0]) # array of indices, just to keep track of them for safety reasons and future checks\n",
    "X_train, X_test, y_train, y_test, w_train, w_test, ix_train, ix_test = train_test_split(X, y, w, ix, train_size=0.6)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is common practice to scale the inputs to Neural Nets such that they have approximately similar ranges. Without this step, you might end up with variables whose values span very different orders of magnitude. This will create problems in the NN convergence due to very wild fluctuations in the magnitude of the internal weights. To take care of the scaling, we use the `sklearn` `StandardScaler`:\n",
    "<a href='http://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html'><img src='./images/standardscaler.jpg' style=\"width:800px\"></a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaler = StandardScaler()\n",
    "X_train = scaler.fit_transform(X_train)\n",
    "X_test = scaler.transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Training a NN using Keras"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Neural Netwroks"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " *  Stack of tensor operators\n",
    " *  Series of linear and non-linear transformations with the goal of finding optimal parameters to transform inputs and approximate targets\n",
    " *  Deep nets > shallow nets"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Multilayer Perceptron (MLP)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3>Dense</h3>\n",
    " * Densely connected layer, where all inputs are connected to all outputs\n",
    " *  Core layer of an MLP\n",
    " *  Linear transformation of the input vector $x \\in \\mathbb{R}^n$, which can be expressed using the $n \\times m$ matrix $W \\in \\mathbb{R}^{n \\times m}$ as:\n",
    "<center> $u = Wx + b$ </center>\n",
    "where $b \\in \\mathbb{R}^m$ is the bias unit\n",
    "\n",
    " *  All entries in both $W$ and $b$ are trainable\n",
    " *  In Keras:\n",
    " ```\n",
    "keras.layers.Dense(\n",
    "                    units,\n",
    "                    activation=None,\n",
    "                    use_bias=True,\n",
    "                    kernel_initializer='glorot_uniform',\n",
    "                    bias_initializer='zeros', kernel_regularizer=None,\n",
    "                    bias_regularizer=None,\n",
    "                    activity_regularizer=None,\n",
    "                    kernel_constraint=None,\n",
    "                    bias_constraint=None\n",
    ")\n",
    "```\n",
    "\n",
    " *  `input_dim` (or `input_shape`) are necessary arguments for the 1st layer of the net:\n",
    "\n",
    "```python\n",
    "# as first layer in a sequential model:\n",
    "model = Sequential()\n",
    "model.add(Dense(32, input_shape=(16,)))\n",
    "# now the model will take as input arrays of shape (*, 16)\n",
    "# and output arrays of shape (*, 32)\n",
    "\n",
    "# after the first layer, you don't need to specify\n",
    "# the size of the input anymore:\n",
    "model.add(Dense(32))\n",
    " ```\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3>Activation Functions</h3>\n",
    " *  Mathematical way of quantifying the activation state of a node $\\rightarrow$ whether it's firing or not\n",
    " *  Non-linear activation functions are the key to Deep Learning\n",
    " *  Allow NNs to learn complex, non-linear transformations of the inputs\n",
    " *  Some popular choices:\n",
    " <img src='./images/activations_table.jpg' style='width:700px'>\n",
    " <img src='./images/activation.jpg' style='width:800px'>\n",
    " Available activations: **softmax**, **elu**, **selu**, **softplus**, **softsign**, **relu**, **tanh**, **sigmoid**, **hard_sigmoid**, **linear**.\n",
    " Activations that are more complex than a simple TensorFlow/Theano/CNTK function (eg. learnable activations, which maintain a state) are available as *Advanced Activation* layers, and can be found in the module  `keras.layers.advanced_activations`. These include `PReLU` and `LeakyReLU`."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3>Regularization</h3>\n",
    " *  Series of methods to avoid overfitting\n",
    " *  Mathematical encouragement towards simpler models\n",
    " *  Explicitly penalize weights that get too large\n",
    " *  Two main categories: \n",
    " <img src='./images/regtypes.jpg' style='width:700px'>\n",
    " *  In Keras, Dropout is added in as a layer. It masks the outputs of the previous layer such that some of them will randomly become inactive and will not contribute to information propagation:\n",
    "```python\n",
    "keras.layers.Dropout(rate, noise_shape=None, seed=None)\n",
    "```\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Build a simple Keras classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n",
      "2017-11-14 23:55:27.843006: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.\n",
      "2017-11-14 23:55:27.843031: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.\n",
      "2017-11-14 23:55:27.843040: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations.\n",
      "2017-11-14 23:55:27.843046: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations.\n"
     ]
    }
   ],
   "source": [
    "from keras.models import Model\n",
    "from keras.layers import Dense, Dropout, Input"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "inputs = Input(shape=(X_train.shape[1], )) # placeholder\n",
    "\n",
    "hidden = Dense(10, activation='relu')(inputs)\n",
    "hidden = Dropout(0.2)(hidden)\n",
    "hidden = Dense(20, activation='relu')(hidden)\n",
    "hidden = Dropout(0.2)(hidden)\n",
    "hidden = Dense(30, activation='relu')(hidden)\n",
    "hidden = Dropout(0.2)(hidden)\n",
    "outputs = Dense(3, activation='softmax')(hidden)\n",
    "# last layer has to have the same dimensionality as the number of classes we want to predict, here 3\n",
    "\n",
    "model = Model(inputs, outputs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         (None, 6)                 0         \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 10)                70        \n",
      "_________________________________________________________________\n",
      "dropout_1 (Dropout)          (None, 10)                0         \n",
      "_________________________________________________________________\n",
      "dense_2 (Dense)              (None, 20)                220       \n",
      "_________________________________________________________________\n",
      "dropout_2 (Dropout)          (None, 20)                0         \n",
      "_________________________________________________________________\n",
      "dense_3 (Dense)              (None, 30)                630       \n",
      "_________________________________________________________________\n",
      "dropout_3 (Dropout)          (None, 30)                0         \n",
      "_________________________________________________________________\n",
      "dense_4 (Dense)              (None, 3)                 93        \n",
      "=================================================================\n",
      "Total params: 1,013\n",
      "Trainable params: 1,013\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.utils.vis_utils import plot_model\n",
    "plot_model(model, 'temp.png', show_shapes=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src='temp.png'>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Satisfied with your net?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Objective / Loss / Cost\n",
    "* Mathematical way of quantifying how much ŷ deviates from y\n",
    "* Dictates how strongly we penalize certain types of mistakes\n",
    "* Cost of inaccurately classifying an event (“cost function”)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"./images/loss.jpg\">"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"./images/training.jpg\">"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Optimizers\n",
    "<img src=\"./images/optimizers.jpg\">"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "(animations from A. Radford)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"http://2.bp.blogspot.com/-q6l20Vs4P_w/VPmIC7sEhnI/AAAAAAAACC4/g3UOUX2r_yA/s1600/s25RsOr%2B-%2BImgur.gif\">\n",
    "<img src=\"http://2.bp.blogspot.com/-L98w-SBmF58/VPmICIjKEKI/AAAAAAAACCs/rrFz3VetYmM/s1600/Beale%26amp%3B%23039%3Bs%2Bfunction%2B-%2BImgur.gif\">\n",
    "<img src=\"http://3.bp.blogspot.com/-nrtJPrdBWuE/VPmIB46F2aI/AAAAAAAACCw/vaE_B0SVy5k/s1600/Long%2BValley%2B-%2BImgur.gif\">\n",
    "<img src=\"http://1.bp.blogspot.com/-K_X-yud8nj8/VPmIBxwGlsI/AAAAAAAACC0/JS-h1fa09EQ/s1600/Saddle%2BPoint%2B-%2BImgur.gif\">"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now you need to declare what loss function and optimizer to use. We pass these as arguments to `compile`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile('adam', 'sparse_categorical_crossentropy') "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Training"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.callbacks import EarlyStopping, ModelCheckpoint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Counter({0: 1796, 1: 31146, 2: 43851})"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from collections import Counter\n",
    "Counter(y_test)\n",
    "# uneven classes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training:\n",
      "Train on 92150 samples, validate on 23038 samples\n",
      "Epoch 1/20\n",
      "92096/92150 [============================>.] - ETA: 0s - loss: 0.4397Epoch 00001: val_loss improved from inf to 0.42322, saving model to ./models/tutorial-progress.h5\n",
      "92150/92150 [==============================] - 8s 86us/step - loss: 0.4398 - val_loss: 0.4232\n",
      "Epoch 2/20\n",
      "92032/92150 [============================>.] - ETA: 0s - loss: 0.4388Epoch 00002: val_loss improved from 0.42322 to 0.41837, saving model to ./models/tutorial-progress.h5\n",
      "92150/92150 [==============================] - 8s 86us/step - loss: 0.4388 - val_loss: 0.4184\n",
      "Epoch 3/20\n",
      "91680/92150 [============================>.] - ETA: 0s - loss: 0.4400Epoch 00003: val_loss did not improve\n",
      "92150/92150 [==============================] - 8s 88us/step - loss: 0.4401 - val_loss: 0.4207\n",
      "Epoch 4/20\n",
      "92096/92150 [============================>.] - ETA: 0s - loss: 0.4416Epoch 00004: val_loss improved from 0.41837 to 0.41834, saving model to ./models/tutorial-progress.h5\n",
      "92150/92150 [==============================] - 8s 86us/step - loss: 0.4416 - val_loss: 0.4183\n",
      "Epoch 5/20\n",
      "91648/92150 [============================>.] - ETA: 0s - loss: 0.4420Epoch 00005: val_loss improved from 0.41834 to 0.41505, saving model to ./models/tutorial-progress.h5\n",
      "92150/92150 [==============================] - 8s 87us/step - loss: 0.4429 - val_loss: 0.4150\n",
      "Epoch 6/20\n",
      "91680/92150 [============================>.] - ETA: 0s - loss: 0.4401Epoch 00006: val_loss did not improve\n",
      "92150/92150 [==============================] - 8s 86us/step - loss: 0.4405 - val_loss: 0.4218\n",
      "Epoch 7/20\n",
      "91968/92150 [============================>.] - ETA: 0s - loss: 0.4383Epoch 00007: val_loss did not improve\n",
      "92150/92150 [==============================] - 8s 86us/step - loss: 0.4383 - val_loss: 0.4208\n",
      "Epoch 8/20\n",
      "91552/92150 [============================>.] - ETA: 0s - loss: 0.4384Epoch 00008: val_loss did not improve\n",
      "92150/92150 [==============================] - 8s 85us/step - loss: 0.4386 - val_loss: 0.4232\n",
      "Epoch 9/20\n",
      "91968/92150 [============================>.] - ETA: 0s - loss: 0.4391Epoch 00009: val_loss did not improve\n",
      "92150/92150 [==============================] - 8s 87us/step - loss: 0.4391 - val_loss: 0.4194\n",
      "Epoch 10/20\n",
      "92096/92150 [============================>.] - ETA: 0s - loss: 0.4384Epoch 00010: val_loss improved from 0.41505 to 0.41338, saving model to ./models/tutorial-progress.h5\n",
      "92150/92150 [==============================] - 8s 87us/step - loss: 0.4383 - val_loss: 0.4134\n",
      "Epoch 11/20\n",
      "91936/92150 [============================>.] - ETA: 0s - loss: 0.4409Epoch 00011: val_loss did not improve\n",
      "92150/92150 [==============================] - 8s 87us/step - loss: 0.4407 - val_loss: 0.4186\n",
      "Epoch 12/20\n",
      "92032/92150 [============================>.] - ETA: 0s - loss: 0.4404Epoch 00012: val_loss did not improve\n",
      "92150/92150 [==============================] - 8s 86us/step - loss: 0.4403 - val_loss: 0.4164\n",
      "Epoch 13/20\n",
      "92128/92150 [============================>.] - ETA: 0s - loss: 0.4405Epoch 00013: val_loss did not improve\n",
      "92150/92150 [==============================] - 8s 86us/step - loss: 0.4404 - val_loss: 0.4312\n",
      "Epoch 14/20\n",
      "91968/92150 [============================>.] - ETA: 0s - loss: 0.4389Epoch 00014: val_loss did not improve\n",
      "92150/92150 [==============================] - 8s 86us/step - loss: 0.4390 - val_loss: 0.4165\n",
      "Epoch 15/20\n",
      "91840/92150 [============================>.] - ETA: 0s - loss: 0.4416Epoch 00015: val_loss improved from 0.41338 to 0.41242, saving model to ./models/tutorial-progress.h5\n",
      "92150/92150 [==============================] - 8s 85us/step - loss: 0.4417 - val_loss: 0.4124\n",
      "Epoch 16/20\n",
      "91616/92150 [============================>.] - ETA: 0s - loss: 0.4365Epoch 00016: val_loss improved from 0.41242 to 0.41008, saving model to ./models/tutorial-progress.h5\n",
      "92150/92150 [==============================] - 8s 85us/step - loss: 0.4366 - val_loss: 0.4101\n",
      "Epoch 17/20\n",
      "91744/92150 [============================>.] - ETA: 0s - loss: 0.4374Epoch 00017: val_loss did not improve\n",
      "92150/92150 [==============================] - 8s 85us/step - loss: 0.4378 - val_loss: 0.4219\n",
      "Epoch 18/20\n",
      "91584/92150 [============================>.] - ETA: 0s - loss: 0.4395Epoch 00018: val_loss did not improve\n",
      "92150/92150 [==============================] - 8s 86us/step - loss: 0.4393 - val_loss: 0.4243\n",
      "Epoch 19/20\n",
      "91872/92150 [============================>.] - ETA: 0s - loss: 0.4410Epoch 00019: val_loss did not improve\n",
      "92150/92150 [==============================] - 8s 86us/step - loss: 0.4412 - val_loss: 0.4162\n",
      "Epoch 20/20\n",
      "91584/92150 [============================>.] - ETA: 0s - loss: 0.4349Epoch 00020: val_loss did not improve\n",
      "92150/92150 [==============================] - 8s 86us/step - loss: 0.4348 - val_loss: 0.4209\n"
     ]
    }
   ],
   "source": [
    "print 'Training:'\n",
    "try:\n",
    "    model.fit(\n",
    "        X_train, y_train, class_weight={ # rebalance class representation\n",
    "            0 : 0.20 * (float(len(y)) / (y == 0).sum()),\n",
    "            1 : 0.40 * (float(len(y)) / (y == 1).sum()),\n",
    "            2 : 0.40 * (float(len(y)) / (y == 2).sum())\n",
    "        },\n",
    "        callbacks = [\n",
    "            EarlyStopping(verbose=True, patience=10, monitor='val_loss'),\n",
    "            ModelCheckpoint('./models/tutorial-progress.h5', monitor='val_loss', verbose=True, save_best_only=True)\n",
    "        ],\n",
    "        epochs=20, \n",
    "        validation_split = 0.2,\n",
    "        verbose=True\n",
    ") \n",
    "except KeyboardInterrupt:\n",
    "    print 'Training ended early.'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving model...\n",
      "Done\n"
     ]
    }
   ],
   "source": [
    "# -- load in best network\n",
    "model.load_weights('./models/tutorial-progress.h5')\n",
    "\n",
    "# -- Save network weights and structure\n",
    "print 'Saving model...'\n",
    "model.save_weights('./models/tutorial.h5', overwrite=True)\n",
    "json_string = model.to_json()\n",
    "open('./models/tutorial.json', 'w').write(json_string)\n",
    "print 'Done'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Testing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Testing...\n",
      "76793/76793 [==============================] - 0s 3us/step\n"
     ]
    }
   ],
   "source": [
    "print 'Testing...'\n",
    "yhat = model.predict(X_test, verbose = True, batch_size = 512) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[  1.80225354e-04,   9.99819815e-01,   2.92606961e-09],\n",
       "       [  6.79788063e-04,   2.35158458e-01,   7.64161706e-01],\n",
       "       [  3.93254653e-04,   2.32965067e-01,   7.66641676e-01],\n",
       "       ..., \n",
       "       [  9.69352317e-04,   2.21174046e-01,   7.77856588e-01],\n",
       "       [  4.15727874e-04,   2.39787340e-01,   7.59796917e-01],\n",
       "       [  2.84072390e-04,   2.67122030e-01,   7.32593954e-01]], dtype=float32)"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# predictions\n",
    "yhat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# -- turn them into classes\n",
    "yhat_cls = np.argmax(yhat, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Visualize performance with confusion matrix:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "import itertools\n",
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "def plot_confusion_matrix(cm, classes,\n",
    "                          normalize=False,\n",
    "                          title='Confusion matrix',\n",
    "                          cmap=plt.cm.Blues):\n",
    "    \"\"\"\n",
    "    This function prints and plots the confusion matrix.\n",
    "    Normalization can be applied by setting `normalize=True`.\n",
    "    \"\"\"\n",
    "    if normalize:\n",
    "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "        print(\"Normalized confusion matrix\")\n",
    "    else:\n",
    "        print('Confusion matrix, without normalization')\n",
    "\n",
    "    print(cm)\n",
    "\n",
    "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "    tick_marks = np.arange(len(classes))\n",
    "    plt.xticks(tick_marks, classes, rotation=45)\n",
    "    plt.yticks(tick_marks, classes)\n",
    "\n",
    "    fmt = '.2f' if normalize else 'd'\n",
    "    thresh = cm.max() / 2.\n",
    "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
    "        plt.text(j, i, format(cm[i, j], fmt),\n",
    "                 horizontalalignment=\"center\",\n",
    "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.ylabel('True label')\n",
    "    plt.xlabel('Predicted label')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Compute confusion matrix\n",
    "cnf_matrix = confusion_matrix(y_test, yhat_cls, sample_weight=w_test)\n",
    "np.set_printoptions(precision=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Normalized confusion matrix\n",
      "[[ 0.94  0.03  0.03]\n",
      " [ 0.03  0.67  0.3 ]\n",
      " [ 0.03  0.05  0.92]]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAU8AAAEeCAYAAAAKH5egAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n1EAqzyRJkhpI5ZkkSVIDqTyTJElq4P8BOL0YUSS4I58AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fda01d72a90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_confusion_matrix(cnf_matrix, classes=[r'$t\\overline{t}$', 'Drell Yan', r'$W$ + jets'],\n",
    "                      normalize=True,\n",
    "                      title='Normalized confusion matrix')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Signal efficiency: 0.941405\n",
      "Background efficiency: 0.0267202\n",
      "Background rej: 37.4248125521\n"
     ]
    }
   ],
   "source": [
    "# signal eff = weighted tpr --> out of all signal events, what % for we classify as signal?\n",
    "print 'Signal efficiency:', w_test[(y_test == 0) & (yhat_cls == 0)].sum() / w_test[y_test == 0].sum()\n",
    "\n",
    "# bkg eff = weighted fpr --> out of all bkg events, what % do we classify as signal?\n",
    "b_eff = w_test[(y_test != 0) & (yhat_cls == 0)].sum() / w_test[y_test != 0].sum()\n",
    "print 'Background efficiency:', b_eff\n",
    "print 'Background rej:', 1 / b_eff"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you want to, you could further analyze the results by, for example, looking at all events that were assigned to class 0 ($t\\bar{t}$) and learn more about their physical characteristics. This could provide insight into what the network is focusing in its decision-making process."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "# -- events that got assigned to class 0\n",
    "predicted_ttbar = df_full.iloc[np.array(ix_test)[yhat_cls == 0]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ubuntu/venv/micky-work/lib/python2.7/site-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  \"\"\"Entry point for launching an IPython kernel.\n"
     ]
    }
   ],
   "source": [
    "predicted_ttbar['true flavor'] = y_test[yhat_cls == 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>NJet</th>\n",
       "      <th>Jet_Px</th>\n",
       "      <th>Jet_Py</th>\n",
       "      <th>Jet_Pz</th>\n",
       "      <th>Jet_E</th>\n",
       "      <th>Jet_btag</th>\n",
       "      <th>Jet_ID</th>\n",
       "      <th>NMuon</th>\n",
       "      <th>Muon_Px</th>\n",
       "      <th>Muon_Py</th>\n",
       "      <th>...</th>\n",
       "      <th>MClepton_py</th>\n",
       "      <th>MClepton_pz</th>\n",
       "      <th>MCleptonPDGid</th>\n",
       "      <th>MCneutrino_px</th>\n",
       "      <th>MCneutrino_py</th>\n",
       "      <th>MCneutrino_pz</th>\n",
       "      <th>NPrimaryVertices</th>\n",
       "      <th>triggerIsoMu24</th>\n",
       "      <th>EventWeight</th>\n",
       "      <th>true flavor</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>71068</th>\n",
       "      <td>2</td>\n",
       "      <td>[-39.515, -32.7724]</td>\n",
       "      <td>[-7.06055, 4.96038]</td>\n",
       "      <td>[-59.9577, 1.15056]</td>\n",
       "      <td>[72.4383, 33.8005]</td>\n",
       "      <td>[-1.0, -1.0]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[36.5116]</td>\n",
       "      <td>[-27.2592]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7</td>\n",
       "      <td>True</td>\n",
       "      <td>0.540515</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1249</th>\n",
       "      <td>5</td>\n",
       "      <td>[60.9035, -111.827, -27.5385, 52.0547, 52.3307]</td>\n",
       "      <td>[117.735, -9.45221, -51.6361, 13.4452, 7.99291]</td>\n",
       "      <td>[151.038, -72.7721, -35.0566, 143.468, -8.85212]</td>\n",
       "      <td>[202.914, 135.274, 70.2694, 153.809, 55.7241]</td>\n",
       "      <td>[-1.0, -1.0, -1.0, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True, True, True]</td>\n",
       "      <td>0</td>\n",
       "      <td>[]</td>\n",
       "      <td>[]</td>\n",
       "      <td>...</td>\n",
       "      <td>-27.402386</td>\n",
       "      <td>-25.099463</td>\n",
       "      <td>13</td>\n",
       "      <td>-129.215714</td>\n",
       "      <td>-97.106895</td>\n",
       "      <td>43.848606</td>\n",
       "      <td>14</td>\n",
       "      <td>True</td>\n",
       "      <td>0.192474</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>143500</th>\n",
       "      <td>2</td>\n",
       "      <td>[-53.283, -38.1472]</td>\n",
       "      <td>[-76.5362, -9.61194]</td>\n",
       "      <td>[-86.3619, 32.1003]</td>\n",
       "      <td>[128.014, 51.698]</td>\n",
       "      <td>[-1.0, -1.0]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>1</td>\n",
       "      <td>[39.2392]</td>\n",
       "      <td>[16.2889]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>10</td>\n",
       "      <td>True</td>\n",
       "      <td>2.393087</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1186</th>\n",
       "      <td>3</td>\n",
       "      <td>[-17.622, -42.9345, -26.1204]</td>\n",
       "      <td>[-59.8529, 38.5907, 16.7819]</td>\n",
       "      <td>[-13.7983, -1.9199, 128.346]</td>\n",
       "      <td>[64.3067, 58.404, 132.271]</td>\n",
       "      <td>[1.52388, -1.0, -1.0]</td>\n",
       "      <td>[True, True, True]</td>\n",
       "      <td>2</td>\n",
       "      <td>[63.2695, 34.3971]</td>\n",
       "      <td>[3.4336, -23.4732]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9</td>\n",
       "      <td>True</td>\n",
       "      <td>0.299981</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28224</th>\n",
       "      <td>2</td>\n",
       "      <td>[49.646, 34.655]</td>\n",
       "      <td>[19.6789, -29.2292]</td>\n",
       "      <td>[-194.703, -144.563]</td>\n",
       "      <td>[202.0, 151.752]</td>\n",
       "      <td>[-1.0, -1.0]</td>\n",
       "      <td>[True, True]</td>\n",
       "      <td>2</td>\n",
       "      <td>[-60.5106, -35.1742]</td>\n",
       "      <td>[37.2119, -34.9203]</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>20</td>\n",
       "      <td>True</td>\n",
       "      <td>0.005656</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 52 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        NJet                                           Jet_Px  \\\n",
       "71068      2                              [-39.515, -32.7724]   \n",
       "1249       5  [60.9035, -111.827, -27.5385, 52.0547, 52.3307]   \n",
       "143500     2                              [-53.283, -38.1472]   \n",
       "1186       3                    [-17.622, -42.9345, -26.1204]   \n",
       "28224      2                                 [49.646, 34.655]   \n",
       "\n",
       "                                                 Jet_Py  \\\n",
       "71068                               [-7.06055, 4.96038]   \n",
       "1249    [117.735, -9.45221, -51.6361, 13.4452, 7.99291]   \n",
       "143500                             [-76.5362, -9.61194]   \n",
       "1186                       [-59.8529, 38.5907, 16.7819]   \n",
       "28224                               [19.6789, -29.2292]   \n",
       "\n",
       "                                                  Jet_Pz  \\\n",
       "71068                                [-59.9577, 1.15056]   \n",
       "1249    [151.038, -72.7721, -35.0566, 143.468, -8.85212]   \n",
       "143500                               [-86.3619, 32.1003]   \n",
       "1186                        [-13.7983, -1.9199, 128.346]   \n",
       "28224                               [-194.703, -144.563]   \n",
       "\n",
       "                                                Jet_E  \\\n",
       "71068                              [72.4383, 33.8005]   \n",
       "1249    [202.914, 135.274, 70.2694, 153.809, 55.7241]   \n",
       "143500                              [128.014, 51.698]   \n",
       "1186                       [64.3067, 58.404, 132.271]   \n",
       "28224                                [202.0, 151.752]   \n",
       "\n",
       "                              Jet_btag                          Jet_ID  NMuon  \\\n",
       "71068                     [-1.0, -1.0]                    [True, True]      1   \n",
       "1249    [-1.0, -1.0, -1.0, -1.0, -1.0]  [True, True, True, True, True]      0   \n",
       "143500                    [-1.0, -1.0]                    [True, True]      1   \n",
       "1186             [1.52388, -1.0, -1.0]              [True, True, True]      2   \n",
       "28224                     [-1.0, -1.0]                    [True, True]      2   \n",
       "\n",
       "                     Muon_Px              Muon_Py     ...     MClepton_py  \\\n",
       "71068              [36.5116]           [-27.2592]     ...        0.000000   \n",
       "1249                      []                   []     ...      -27.402386   \n",
       "143500             [39.2392]            [16.2889]     ...        0.000000   \n",
       "1186      [63.2695, 34.3971]   [3.4336, -23.4732]     ...        0.000000   \n",
       "28224   [-60.5106, -35.1742]  [37.2119, -34.9203]     ...        0.000000   \n",
       "\n",
       "       MClepton_pz MCleptonPDGid MCneutrino_px  MCneutrino_py MCneutrino_pz  \\\n",
       "71068     0.000000             0      0.000000       0.000000      0.000000   \n",
       "1249    -25.099463            13   -129.215714     -97.106895     43.848606   \n",
       "143500    0.000000             0      0.000000       0.000000      0.000000   \n",
       "1186      0.000000             0      0.000000       0.000000      0.000000   \n",
       "28224     0.000000             0      0.000000       0.000000      0.000000   \n",
       "\n",
       "       NPrimaryVertices triggerIsoMu24 EventWeight true flavor  \n",
       "71068                 7           True    0.540515           1  \n",
       "1249                 14           True    0.192474           0  \n",
       "143500               10           True    2.393087           2  \n",
       "1186                  9           True    0.299981           0  \n",
       "28224                20           True    0.005656           1  \n",
       "\n",
       "[5 rows x 52 columns]"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predicted_ttbar.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In a 3-class classifier, the outputs lie on a 2D triangle with vertices at (1,0,0), (0,1,0), (0,0,1)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZwAAAGRCAYAAABR3wXnAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngQkOg8FgMOoFJjgMBoPBqBf+P3KPf1R4PbvcAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fda01ac6990>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "\n",
    "fig = plt.figure(figsize=(7,7))\n",
    "ax = fig.add_subplot(111, projection='3d')\n",
    "for label, color, marker in [(0, 'r', '^'), (1, 'b', 'o'), (2, 'yellow', 'x')]:\n",
    "    ax.scatter(yhat[y_test == label, 0], yhat[y_test == label, 1], yhat[y_test == label, 2], c=color, marker=marker, alpha=0.01)\n",
    "\n",
    "ax.set_xlabel(r'$P(t\\bar{t})$')\n",
    "ax.set_ylabel('P(Drell Yan)')\n",
    "ax.set_zlabel('P(W + jets)')\n",
    "ax.view_init(30, 30)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To do some model introspection, add intermediate hidden layers to the outputs: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# model = Model(inputs, [outputs, hidden])\n",
    "# model.compile('adam', 'sparse_categorical_crossentropy') \n",
    "# model.load_weights('./models/tutorial-progress.h5')\n",
    "# _, last_layer = model.predict(X_test, verbose = True, batch_size = 512)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## 2. Recurrent Neural Networks"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's look at a fancier way of solving the same classification problem. In this case we will use Recurrent Neural Netwroks. These allow you to process variable length sequences of data. For example, we can use them to describe an event in terms of the properties of its jets, whose number varies event by event. We could also describe the same event using the properties of its muons, or any other particle that appears in it. Because the number of particles of each type changes in each event, we need the flexibility of RNNs to process this type of data.\n",
    "<img src='./images/RNNdetail.png'>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will build an event-level classifier with 4 recurrent branches for the 4 different types of particles in the event."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Jet_Px', 'Jet_Py', 'Jet_Pz', 'Jet_E', 'Jet_btag']\n",
      "['Muon_Px', 'Muon_Py', 'Muon_Pz', 'Muon_E', 'Muon_Charge', 'Muon_Iso']\n",
      "['Photon_Px', 'Photon_Py', 'Photon_Pz', 'Photon_E', 'Photon_Iso']\n",
      "['Electron_Px', 'Electron_Py', 'Electron_Pz', 'Electron_E', 'Electron_Charge', 'Electron_Iso']\n"
     ]
    }
   ],
   "source": [
    "jetvars = [key for key in df_full.keys() if key.startswith('Jet')]\n",
    "jetvars.remove('Jet_ID')\n",
    "print jetvars\n",
    "\n",
    "muonvars = [key for key in df_full.keys() if key.startswith('Muon')]\n",
    "print muonvars\n",
    "\n",
    "photonvars = [key for key in df_full.keys() if key.startswith('Photon')]\n",
    "print photonvars\n",
    "\n",
    "electronvars = [key for key in df_full.keys() if key.startswith('Electron')]\n",
    "print electronvars"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_jets = df_full[jetvars].copy()\n",
    "df_electrons = df_full[electronvars].copy()\n",
    "df_muons = df_full[muonvars].copy()\n",
    "df_photons = df_full[photonvars].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "num_electrons = max([len(e) for e in df_electrons.Electron_E])\n",
    "num_electrons"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "num_muons = max([len(m) for m in df_muons.Muon_E])\n",
    "num_muons"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "num_photons = max([len(gam) for gam in df_photons.Photon_E])\n",
    "num_photons"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "9"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "num_jets = max([len(j) for j in df_jets.Jet_E])\n",
    "num_jets"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Just for the sake of variety, you can either have class labels like (0, 1, 2, 3, ...) and train using `spare_categorical_crossentropy` as you loss function -- like we did before -- or, equivalently, you can have class labels like ([1, 0, 0, 0, ...], [0, 1, 0, 0, ...], ...) and train using `categorical_crossentropy`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 0.,  0.,  1.],\n",
       "       [ 0.,  1.,  0.],\n",
       "       [ 0.,  0.,  1.],\n",
       "       ..., \n",
       "       [ 0.,  1.,  0.],\n",
       "       [ 0.,  0.,  1.],\n",
       "       [ 0.,  1.,  0.]])"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from keras.utils.np_utils import to_categorical\n",
    "y_train = to_categorical(y_train)\n",
    "y_test = to_categorical(y_test)\n",
    "y_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_stream(df, num_obj, sort_col):\n",
    "   \n",
    "    n_variables = df.shape[1]\n",
    "    var_names = df.keys()\n",
    "\n",
    "    data = np.zeros((df.shape[0], num_obj, n_variables), dtype='float32')\n",
    "\n",
    "    # -- call functions to build X (a.k.a. data)                                                                                                                                                                      \n",
    "    sort_objects(df, data, sort_col, num_obj)\n",
    "    \n",
    "    # -- ix_{train, test} from above or from previously stored ordering\n",
    "    Xobj_train = data[ix_train]\n",
    "    Xobj_test = data[ix_test]\n",
    "    \n",
    "    #print 'Scaling features ...'\n",
    "    scale(Xobj_train, var_names, savevars=True) # scale training sample and save scaling\n",
    "    scale(Xobj_test, var_names, savevars=False) # apply scaling to test set\n",
    "    return Xobj_train, Xobj_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "def sort_objects(df, data, SORT_COL, max_nobj):\n",
    "    ''' \n",
    "    sort objects using your preferred variable\n",
    "    \n",
    "    Args:\n",
    "    -----\n",
    "        df: a dataframe with event-level structure where each event is described by a sequence of jets, muons, etc.\n",
    "        data: an array of shape (nb_events, nb_particles, nb_features)\n",
    "        SORT_COL: a string representing the column to sort the objects by\n",
    "        max_nobj: number of particles to cut off at. if >, truncate, else, -999 pad\n",
    "    \n",
    "    Returns:\n",
    "    --------\n",
    "        modifies @a data in place. Pads with -999\n",
    "    \n",
    "    '''\n",
    "    import tqdm\n",
    "    # i = event number, event = all the variables for that event \n",
    "    for i, event in tqdm.tqdm(df.iterrows(), total=df.shape[0]): \n",
    "\n",
    "        # objs = [[pt's], [eta's], ...] of particles for each event \n",
    "        objs = np.array(\n",
    "                [v.tolist() for v in event.get_values()], \n",
    "                dtype='float32'\n",
    "            )[:, (np.argsort(event[SORT_COL]))[::-1]]\n",
    "\n",
    "        # total number of tracks per jet      \n",
    "        nobjs = objs.shape[1] \n",
    "\n",
    "        # take all tracks unless there are more than n_tracks \n",
    "        data[i, :(min(nobjs, max_nobj)), :] = objs.T[:(min(nobjs, max_nobj)), :] \n",
    "\n",
    "        # default value for missing tracks \n",
    "        data[i, (min(nobjs, max_nobj)):, :  ] = -999 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "def scale(data, var_names, savevars, VAR_FILE_NAME='scaling.json'):\n",
    "    ''' \n",
    "    Args:\n",
    "    -----\n",
    "        data: a numpy array of shape (nb_events, nb_particles, n_variables)\n",
    "        var_names: list of keys to be used for the model\n",
    "        savevars: bool -- True for training, False for testing\n",
    "                  it decides whether we want to fit on data to find mean and std \n",
    "                  or if we want to use those stored in the json file \n",
    "    \n",
    "    Returns:\n",
    "    --------\n",
    "        modifies data in place, writes out scaling dictionary\n",
    "    '''\n",
    "    import json\n",
    "    \n",
    "    scale = {}\n",
    "    if savevars: \n",
    "        for v, name in enumerate(var_names):\n",
    "            #print 'Scaling feature %s of %s (%s).' % (v + 1, len(var_names), name)\n",
    "            f = data[:, :, v]\n",
    "            slc = f[f != -999]\n",
    "            m, s = slc.mean(), slc.std()\n",
    "            slc -= m\n",
    "            slc /= s\n",
    "            data[:, :, v][f != -999] = slc.astype('float32')\n",
    "            scale[name] = {'mean' : float(m), 'sd' : float(s)}\n",
    "            \n",
    "        with open(VAR_FILE_NAME, 'wb') as varfile:\n",
    "            json.dump(scale, varfile)\n",
    "\n",
    "    else:\n",
    "        with open(VAR_FILE_NAME, 'rb') as varfile:\n",
    "            varinfo = json.load(varfile)\n",
    "\n",
    "        for v, name in enumerate(var_names):\n",
    "            #print 'Scaling feature %s of %s (%s).' % (v + 1, len(var_names), name)\n",
    "            f = data[:, :, v]\n",
    "            slc = f[f != -999]\n",
    "            m = varinfo[name]['mean']\n",
    "            s = varinfo[name]['sd']\n",
    "            slc -= m\n",
    "            slc /= s\n",
    "            data[:, :, v][f != -999] = slc.astype('float32')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 191981/191981 [00:22<00:00, 8468.88it/s]\n"
     ]
    }
   ],
   "source": [
    "Xjet_train, Xjet_test = create_stream(df_jets, num_jets, sort_col='Jet_btag')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 191981/191981 [00:24<00:00, 7774.78it/s]\n"
     ]
    }
   ],
   "source": [
    "Xphoton_train, Xphoton_test = create_stream(df_photons, num_photons, sort_col='Photon_E')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 191981/191981 [00:23<00:00, 8088.31it/s]\n"
     ]
    }
   ],
   "source": [
    "Xmuon_train, Xmuon_test = create_stream(df_muons, num_muons, sort_col='Muon_E')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 191981/191981 [00:23<00:00, 8074.72it/s]\n"
     ]
    }
   ],
   "source": [
    "Xelectron_train, Xelectron_test = create_stream(df_electrons, num_electrons, sort_col='Electron_E')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.layers import Masking, GRU, concatenate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [],
   "source": [
    "JET_SHAPE = Xjet_train.shape[1:]\n",
    "MUON_SHAPE = Xmuon_train.shape[1:]\n",
    "ELECTRON_SHAPE = Xelectron_train.shape[1:]\n",
    "PHOTON_SHAPE = Xphoton_train.shape[1:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "jet_input = Input(JET_SHAPE)\n",
    "jet_channel = Masking(mask_value=-999, name='jet_masking')(jet_input)\n",
    "jet_channel = GRU(25, name='jet_gru')(jet_channel)\n",
    "jet_channel = Dropout(0.3, name='jet_dropout')(jet_channel)\n",
    "\n",
    "muon_input = Input(MUON_SHAPE)\n",
    "muon_channel = Masking(mask_value=-999, name='muon_masking')(muon_input)\n",
    "muon_channel = GRU(10, name='muon_gru')(muon_channel)\n",
    "muon_channel = Dropout(0.3, name='muon_dropout')(muon_channel)\n",
    "\n",
    "electron_input = Input(ELECTRON_SHAPE)\n",
    "electron_channel = Masking(mask_value=-999, name='electron_masking')(electron_input)\n",
    "electron_channel = GRU(10, name='electron_gru')(electron_channel)\n",
    "electron_channel = Dropout(0.3, name='electron_dropout')(electron_channel)\n",
    "\n",
    "photon_input = Input(PHOTON_SHAPE)\n",
    "photon_channel = Masking(mask_value=-999, name='photon_masking')(photon_input)\n",
    "photon_channel = GRU(10, name='photon_gru')(photon_channel)\n",
    "photon_channel = Dropout(0.3, name='photon_dropout')(photon_channel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "combined = concatenate([jet_channel, muon_channel, electron_channel, photon_channel])\n",
    "combined = Dense(64, activation = 'relu')(combined)\n",
    "combined = Dropout(0.3)(combined)\n",
    "combined = Dense(64, activation = 'relu')(combined)\n",
    "combined = Dropout(0.3)(combined)\n",
    "combined = Dense(64, activation = 'relu')(combined)\n",
    "combined = Dropout(0.3)(combined)\n",
    "combined_outputs = Dense(3, activation='softmax')(combined)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_2 (InputLayer)            (None, 9, 5)         0                                            \n",
      "__________________________________________________________________________________________________\n",
      "input_3 (InputLayer)            (None, 3, 6)         0                                            \n",
      "__________________________________________________________________________________________________\n",
      "input_4 (InputLayer)            (None, 2, 6)         0                                            \n",
      "__________________________________________________________________________________________________\n",
      "input_5 (InputLayer)            (None, 2, 5)         0                                            \n",
      "__________________________________________________________________________________________________\n",
      "jet_masking (Masking)           (None, 9, 5)         0           input_2[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "muon_masking (Masking)          (None, 3, 6)         0           input_3[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "electron_masking (Masking)      (None, 2, 6)         0           input_4[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "photon_masking (Masking)        (None, 2, 5)         0           input_5[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "jet_gru (GRU)                   (None, 25)           2325        jet_masking[0][0]                \n",
      "__________________________________________________________________________________________________\n",
      "muon_gru (GRU)                  (None, 10)           510         muon_masking[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "electron_gru (GRU)              (None, 10)           510         electron_masking[0][0]           \n",
      "__________________________________________________________________________________________________\n",
      "photon_gru (GRU)                (None, 10)           480         photon_masking[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "jet_dropout (Dropout)           (None, 25)           0           jet_gru[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "muon_dropout (Dropout)          (None, 10)           0           muon_gru[0][0]                   \n",
      "__________________________________________________________________________________________________\n",
      "electron_dropout (Dropout)      (None, 10)           0           electron_gru[0][0]               \n",
      "__________________________________________________________________________________________________\n",
      "photon_dropout (Dropout)        (None, 10)           0           photon_gru[0][0]                 \n",
      "__________________________________________________________________________________________________\n",
      "concatenate_3 (Concatenate)     (None, 55)           0           jet_dropout[0][0]                \n",
      "                                                                 muon_dropout[0][0]               \n",
      "                                                                 electron_dropout[0][0]           \n",
      "                                                                 photon_dropout[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "dense_6 (Dense)                 (None, 64)           3584        concatenate_3[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "dropout_4 (Dropout)             (None, 64)           0           dense_6[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "dense_7 (Dense)                 (None, 64)           4160        dropout_4[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "dropout_5 (Dropout)             (None, 64)           0           dense_7[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "dense_8 (Dense)                 (None, 64)           4160        dropout_5[0][0]                  \n",
      "__________________________________________________________________________________________________\n",
      "dropout_6 (Dropout)             (None, 64)           0           dense_8[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "dense_9 (Dense)                 (None, 3)            195         dropout_6[0][0]                  \n",
      "==================================================================================================\n",
      "Total params: 15,924\n",
      "Trainable params: 15,924\n",
      "Non-trainable params: 0\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "combined_rnn = Model(inputs=[jet_input, muon_input, electron_input, photon_input], outputs=combined_outputs)\n",
    "combined_rnn.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [],
   "source": [
    "combined_rnn.compile('adam', 'categorical_crossentropy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training:\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ubuntu/venv/micky-work/lib/python2.7/site-packages/ipykernel_launcher.py:14: UserWarning: The `nb_epoch` argument in `fit` has been renamed `epochs`.\n",
      "  \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 92150 samples, validate on 23038 samples\n",
      "Epoch 1/30\n",
      "92128/92150 [============================>.] - ETA: 0s - loss: 0.5260Epoch 00001: val_loss improved from inf to 0.43021, saving model to ./models/combinedrnn_tutorial-progress\n",
      "92150/92150 [==============================] - 96s 1ms/step - loss: 0.5260 - val_loss: 0.4302\n",
      "Epoch 2/30\n",
      "92144/92150 [============================>.] - ETA: 0s - loss: 0.4370Epoch 00002: val_loss improved from 0.43021 to 0.42557, saving model to ./models/combinedrnn_tutorial-progress\n",
      "92150/92150 [==============================] - 94s 1ms/step - loss: 0.4370 - val_loss: 0.4256\n",
      "Epoch 3/30\n",
      "92128/92150 [============================>.] - ETA: 0s - loss: 0.4305Epoch 00003: val_loss did not improve\n",
      "92150/92150 [==============================] - 86s 932us/step - loss: 0.4306 - val_loss: 0.4304\n",
      "Epoch 4/30\n",
      "92128/92150 [============================>.] - ETA: 0s - loss: 0.4315Epoch 00004: val_loss did not improve\n",
      "92150/92150 [==============================] - 84s 910us/step - loss: 0.4315 - val_loss: 0.4307\n",
      "Epoch 5/30\n",
      "92096/92150 [============================>.] - ETA: 0s - loss: 0.4333Epoch 00005: val_loss did not improve\n",
      "92150/92150 [==============================] - 82s 895us/step - loss: 0.4332 - val_loss: 0.4455\n",
      "Epoch 6/30\n",
      "92128/92150 [============================>.] - ETA: 0s - loss: 0.4316Epoch 00006: val_loss improved from 0.42557 to 0.42022, saving model to ./models/combinedrnn_tutorial-progress\n",
      "92150/92150 [==============================] - 83s 897us/step - loss: 0.4316 - val_loss: 0.4202\n",
      "Epoch 7/30\n",
      "92096/92150 [============================>.] - ETA: 0s - loss: 0.4289Epoch 00007: val_loss did not improve\n",
      "92150/92150 [==============================] - 84s 915us/step - loss: 0.4289 - val_loss: 0.4412\n",
      "Epoch 8/30\n",
      "92112/92150 [============================>.] - ETA: 0s - loss: 0.4317Epoch 00008: val_loss did not improve\n",
      "92150/92150 [==============================] - 83s 898us/step - loss: 0.4317 - val_loss: 0.4434\n",
      "Epoch 9/30\n",
      "92144/92150 [============================>.] - ETA: 0s - loss: 0.4327Epoch 00009: val_loss did not improve\n",
      "92150/92150 [==============================] - 83s 904us/step - loss: 0.4327 - val_loss: 0.4300\n",
      "Epoch 10/30\n",
      "92128/92150 [============================>.] - ETA: 0s - loss: 0.4270Epoch 00010: val_loss improved from 0.42022 to 0.41622, saving model to ./models/combinedrnn_tutorial-progress\n",
      "92150/92150 [==============================] - 84s 915us/step - loss: 0.4270 - val_loss: 0.4162\n",
      "Epoch 11/30\n",
      "92128/92150 [============================>.] - ETA: 0s - loss: 0.4276Epoch 00011: val_loss did not improve\n",
      "92150/92150 [==============================] - 85s 925us/step - loss: 0.4276 - val_loss: 0.4386\n",
      "Epoch 12/30\n",
      "92144/92150 [============================>.] - ETA: 0s - loss: 0.4262Epoch 00012: val_loss did not improve\n",
      "92150/92150 [==============================] - 83s 897us/step - loss: 0.4262 - val_loss: 0.4508\n",
      "Epoch 13/30\n",
      "92128/92150 [============================>.] - ETA: 0s - loss: 0.4310Epoch 00013: val_loss did not improve\n",
      "92150/92150 [==============================] - 82s 895us/step - loss: 0.4310 - val_loss: 0.4336\n",
      "Epoch 14/30\n",
      "92096/92150 [============================>.] - ETA: 0s - loss: 0.4251Epoch 00014: val_loss did not improve\n",
      "92150/92150 [==============================] - 82s 895us/step - loss: 0.4251 - val_loss: 0.4474\n",
      "Epoch 15/30\n",
      "92112/92150 [============================>.] - ETA: 0s - loss: 0.4289Epoch 00015: val_loss did not improve\n",
      "92150/92150 [==============================] - 84s 907us/step - loss: 0.4289 - val_loss: 0.4300\n",
      "Epoch 16/30\n",
      "92112/92150 [============================>.] - ETA: 0s - loss: 0.4266Epoch 00016: val_loss did not improve\n",
      "92150/92150 [==============================] - 83s 898us/step - loss: 0.4265 - val_loss: 0.4411\n",
      "Epoch 17/30\n",
      "92112/92150 [============================>.] - ETA: 0s - loss: 0.4262Epoch 00017: val_loss did not improve\n",
      "92150/92150 [==============================] - 83s 895us/step - loss: 0.4262 - val_loss: 0.4372\n",
      "Epoch 18/30\n",
      "92096/92150 [============================>.] - ETA: 0s - loss: 0.4283Epoch 00018: val_loss did not improve\n",
      "92150/92150 [==============================] - 83s 896us/step - loss: 0.4283 - val_loss: 0.4370\n",
      "Epoch 19/30\n",
      "92112/92150 [============================>.] - ETA: 0s - loss: 0.4239Epoch 00019: val_loss did not improve\n",
      "92150/92150 [==============================] - 83s 905us/step - loss: 0.4239 - val_loss: 0.4401\n",
      "Epoch 20/30\n",
      "92144/92150 [============================>.] - ETA: 0s - loss: 0.4298Epoch 00020: val_loss did not improve\n",
      "92150/92150 [==============================] - 83s 902us/step - loss: 0.4298 - val_loss: 0.4304\n",
      "Epoch 00020: early stopping\n"
     ]
    }
   ],
   "source": [
    "print 'Training:'\n",
    "try:\n",
    "    combined_rnn.fit([Xjet_train, Xmuon_train, Xelectron_train, Xphoton_train], y_train, batch_size=16,\n",
    "            class_weight={\n",
    "                0 : 0.20 * (float(len(y)) / (y == 0).sum()),\n",
    "                1 : 0.40 * (float(len(y)) / (y == 1).sum()),\n",
    "                2 : 0.40 * (float(len(y)) / (y == 2).sum())\n",
    "        },\n",
    "        callbacks = [\n",
    "            EarlyStopping(verbose=True, patience=10, monitor='val_loss'),\n",
    "            ModelCheckpoint('./models/combinedrnn_tutorial-progress', monitor='val_loss', verbose=True, save_best_only=True)\n",
    "        ],\n",
    "    epochs=30, \n",
    "    validation_split = 0.2) \n",
    "\n",
    "except KeyboardInterrupt:\n",
    "    print 'Training ended early.'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving weights...\n"
     ]
    }
   ],
   "source": [
    "# -- load in best network\n",
    "combined_rnn.load_weights('./models/combinedrnn_tutorial-progress')\n",
    "print 'Saving weights...'\n",
    "combined_rnn.save_weights('./models/combinedrnn_tutorial.h5', overwrite=True)\n",
    "\n",
    "json_string = combined_rnn.to_json()\n",
    "open('./models/combinedrnn_tutorial.json', 'w').write(json_string)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "76793/76793 [==============================] - 2s 27us/step\n"
     ]
    }
   ],
   "source": [
    "yhat_rnn = combined_rnn.predict([Xjet_test, Xmuon_test, Xelectron_test, Xphoton_test], verbose = True, batch_size = 512) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Normalized confusion matrix\n",
      "[[ 0.93  0.02  0.05]\n",
      " [ 0.03  0.6   0.37]\n",
      " [ 0.02  0.03  0.95]]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAU8AAAEeCAYAAAAKH5egAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fda02493a10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Compute confusion matrix\n",
    "cnf_matrix_rnn = confusion_matrix(np.argmax(y_test, axis=1), np.argmax(yhat_rnn, axis=1), sample_weight=w_test)\n",
    "np.set_printoptions(precision=2)\n",
    "plot_confusion_matrix(cnf_matrix_rnn, classes=[r'$t\\overline{t}$', 'Drell Yan', r'$W$ + jets'],\n",
    "                      normalize=True,\n",
    "                      title='Normalized confusion matrix')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [],
   "source": [
    "# -- turn the predictions back into class labels\n",
    "yhat_rnn_cls = np.argmax(yhat_rnn, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "# -- do the same for the truth labels\n",
    "y_test_cls = np.argmax(y_test, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Signal efficiency: 0.934622\n"
     ]
    }
   ],
   "source": [
    "print 'Signal efficiency:', w_test[(y_test_cls == 0) & (yhat_rnn_cls == 0)].sum() / w_test[y_test_cls == 0].sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Background efficiency: 0.0179076\n",
      "Background rej: 55.8421979781\n"
     ]
    }
   ],
   "source": [
    "b_eff = w_test[(y_test_cls != 0) & (yhat_rnn_cls == 0)].sum() / w_test[y_test_cls != 0].sum()\n",
    "print 'Background efficiency:', b_eff\n",
    "print 'Background rej:', 1 / b_eff"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Further Readings"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1. <a href=\"https://iml.web.cern.ch/meetings\">List</a> of all IML meetings\n",
    "1. CERN Academic Training lecture series on Machine Learning (<a href=\"https://indico.cern.ch/event/619370/\">slides and videos</a>)\n",
    "1. \"Machine Learning Algorithms for b-Jet Tagging at the ATLAS Experiment\" (<a href=\"https://cds.cern.ch/record/2289214?ln=en\">ATL-PHYS-PROC-2017-211</a>)\n",
    "1. \"Deep Learning in Flavour Tagging at the ATLAS Experiment (<a href=\"https://cds.cern.ch/record/2287551?ln=en\">ATL-PHYS-PROC-2017-191</a>)\n",
    "1. \"Machine and Deep Learning Techniques in Heavy-Ion Collisions with ALICE\" ([arXiv](https://arxiv.org/pdf/1709.08497.pdf))\n",
    "1. \"Deep learning approach to the Higgs boson CP measurement in H → τ τ decay and associated systematics\" ([arXiv](https://arxiv.org/pdf/1706.07983.pdf))\n",
    "1. Luke de Oliveira's DataScience@LHC 2015 talk on \"A ground-up construction of deep learning\" ([video](https://cds.cern.ch/record/2093518?ln=en))\n",
    "1. Peter Sadowski's DataScience@LHC 2015 deep learning tutorial ([video](https://cds.cern.ch/record/2093535?ln=en))\n",
    "1. Pierre Baldi's DataScience@LHC 2015 talk on \"Deep learning and its apllications to the natural sciences\" ([video](https://cds.cern.ch/record/2093593?ln=en))\n",
    "1. Yann LeCun's CERN colloquium ([video](https://cds.cern.ch/record/2142287?ln=en))\n",
    "1. Fernanda Psihas's 2017 IML Workshop talk on \"Deep convolutional networks for event reconstruction and particle tagging on NOvA and DUNE\" ([video](https://cds.cern.ch/record/2256882?ln=en))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.10"
  },
  "latex_envs": {
   "bibliofile": "biblio.bib",
   "cite_by": "apalike",
   "current_citInitial": 1,
   "eqLabelWithNumbers": true,
   "eqNumInitial": 0
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}