full_model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# (top)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    " * move Gaussian constraints out (multiply the simul pdf by them, not each component\n",
    " * get correct expectation values for all parameters\n",
    " * separate high mass and low mass categories, only apply the desired one\n",
    " * make sure bkg from Leo's file is the correct thing to fit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "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 ROOT\n",
    "import ROOT.RooFit as rf\n",
    "#import rootnotes\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from root_numpy import array2tree, hist2array\n",
    "from rootpy.io import root_open\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Import compile Double Sided Crystal Ball function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from ROOT import gInterpreter, gSystem\n",
    "gInterpreter.ProcessLine('#include \"../56_Hbbyy/inc/HggTwoSidedCBPdf.h\"')\n",
    "gSystem.Load('../56_Hbbyy/libDH.so')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#c1=rootnotes.default_canvas()\n",
    "#%jsroot on"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# Declare observable x: myy, the diphoton mass\n",
    "myy = ROOT.RooRealVar(\"myy\", \"m_yy\", 105, 160)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Get histograms from Leo's file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "from copy import deepcopy\n",
    "\n",
    "def _add(h1, h2):\n",
    "    res = deepcopy(h1)\n",
    "    res.add(h2)\n",
    "    return res"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "f = root_open('combined_backgrounds_20170201.root')\n",
    "histos_bkg_myy = {k.GetName(): f.Get(k.GetName()) for k in f.GetListOfKeys() if 'm_yy_' in k.GetName()}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:DataHandling -- RooDataHist::adjustBinning(m_yy_highMass_1tag_tightIsolated_bkg_non_Higgs): fit range of variable myy expanded to nearest bin boundaries: [105,160] --> [105,160]\n",
      "[#1] INFO:DataHandling -- RooDataHist::adjustBinning(m_yy_highMass_2tag_tightIsolated_bkg_single_Higgs): fit range of variable myy expanded to nearest bin boundaries: [105,160] --> [105,160]\n",
      "[#1] INFO:DataHandling -- RooDataHist::adjustBinning(m_yy_highMass_0tag_tightIsolated_bkg_single_Higgs): fit range of variable myy expanded to nearest bin boundaries: [105,160] --> [105,160]\n",
      "[#1] INFO:DataHandling -- RooDataHist::adjustBinning(m_yy_lowMass_2tag_tightIsolated_bkg_non_Higgs): fit range of variable myy expanded to nearest bin boundaries: [105,160] --> [105,160]\n",
      "[#1] INFO:DataHandling -- RooDataHist::adjustBinning(m_yy_lowMass_0tag_tightIsolated_bkg_single_Higgs): fit range of variable myy expanded to nearest bin boundaries: [105,160] --> [105,160]\n",
      "[#1] INFO:DataHandling -- RooDataHist::adjustBinning(m_yy_lowMass_1tag_tightIsolated_bkg_single_Higgs): fit range of variable myy expanded to nearest bin boundaries: [105,160] --> [105,160]\n",
      "[#1] INFO:DataHandling -- RooDataHist::adjustBinning(m_yy_lowMass_0tag_tightIsolated_bkg_non_Higgs): fit range of variable myy expanded to nearest bin boundaries: [105,160] --> [105,160]\n",
      "[#1] INFO:DataHandling -- RooDataHist::adjustBinning(m_yy_highMass_0tag_tightIsolated_bkg_non_Higgs): fit range of variable myy expanded to nearest bin boundaries: [105,160] --> [105,160]\n",
      "[#1] INFO:DataHandling -- RooDataHist::adjustBinning(m_yy_lowMass_2tag_tightIsolated_bkg_single_Higgs): fit range of variable myy expanded to nearest bin boundaries: [105,160] --> [105,160]\n",
      "[#1] INFO:DataHandling -- RooDataHist::adjustBinning(m_yy_highMass_1tag_tightIsolated_bkg_single_Higgs): fit range of variable myy expanded to nearest bin boundaries: [105,160] --> [105,160]\n",
      "[#1] INFO:DataHandling -- RooDataHist::adjustBinning(m_yy_lowMass_1tag_tightIsolated_bkg_non_Higgs): fit range of variable myy expanded to nearest bin boundaries: [105,160] --> [105,160]\n",
      "[#1] INFO:DataHandling -- RooDataHist::adjustBinning(m_yy_highMass_2tag_tightIsolated_bkg_non_Higgs): fit range of variable myy expanded to nearest bin boundaries: [105,160] --> [105,160]\n"
     ]
    }
   ],
   "source": [
    "myyarglist = ROOT.RooArgList(myy)\n",
    "for k, v in histos_bkg_myy.iteritems():\n",
    "    short_k = k.replace('m_yy_', '').replace('tightIsolated_bkg_', '')\n",
    "    exec (\"{} = {}\".format(\n",
    "        short_k,\n",
    "        'ROOT.RooDataHist(\"{k}\", \"{k}\", myyarglist, v)'.format(k=k)\n",
    "    )) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# Create a binned dataset that imports contents of TH1 and associates its contents to observable 'x'\n",
    "# 0 tag: nonHiggs bkg + single Higgs bkg\n",
    "# 1,2 tags: nonHiggs bkg + single Higgs bkg + diHiggs signal\n",
    "\n",
    "# dh_bkg_noH_0tag_highmass = ROOT.RooDataHist(\n",
    "#     \"0tag_highmass_bkg_nonHiggs\", \"0tag_highmass_bkg_nonHiggs\",\n",
    "#     ROOT.RooArgList(myy), histos_bkg['m_yy_highMass_0tag_tightIsolated_bkg_non_Higgs'])\n",
    "# dh_bkg_noH_1tag_highmass = ROOT.RooDataHist(\n",
    "#     \"1tag_highmass_bkg_nonHiggs\", \"1tag_highmass_bkg_nonHiggs\",\n",
    "#     ROOT.RooArgList(myy), histos_bkg['m_yy_highMass_1tag_tightIsolated_bkg_non_Higgs'])\n",
    "# dh_bkg_noH_2tag_highmass = ROOT.RooDataHist(\n",
    "#     \"2tag_highmass_bkg_nonHiggs\", \"2tag_highmass_bkg_nonHiggs\",\n",
    "#     ROOT.RooArgList(myy), histos_bkg['m_yy_highMass_2tag_tightIsolated_bkg_non_Higgs'])\n",
    "# dh_bkg_1H_0tag_highmass = ROOT.RooDataHist(\n",
    "#     \"0tag_highmass_bkg_1Higgs\", \"0tag_highmass_bkg_nonHiggs\",\n",
    "#     ROOT.RooArgList(myy), histos_bkg['m_yy_highMass_0tag_tightIsolated_bkg_single_Higgs'])\n",
    "# dh_bkg_1H_1tag_highmass = ROOT.RooDataHist(\n",
    "#     \"1tag_highmass_bkg_1Higgs\", \"1tag_highmass_bkg_nonHiggs\",\n",
    "#     ROOT.RooArgList(myy), histos_bkg['m_yy_highMass_1tag_tightIsolated_bkg_single_Higgs'])\n",
    "# dh_bkg_1H_2tag_highmass = ROOT.RooDataHist(\n",
    "#     \"2tag_highmass_bkg_1Higgs\", \"2tag_highmass_bkg_1Higgs\",\n",
    "#     ROOT.RooArgList(myy), histos_bkg['m_yy_highMass_2tag_tightIsolated_bkg_single_Higgs'])\n",
    "\n",
    "# dh_bkg_noH_0tag_lowmass = ROOT.RooDataHist(\n",
    "#     \"0tag_lowmass_bkg_nonHiggs\", \"0tag_lowmass_bkg_nonHiggs\",\n",
    "#     ROOT.RooArgList(myy), histos_bkg['m_yy_lowMass_0tag_tightIsolated_bkg_non_Higgs'])\n",
    "# dh_bkg_noH_1tag_lowmass = ROOT.RooDataHist(\n",
    "#     \"1tag_lowmass_bkg_nonHiggs\", \"1tag_lowmass_bkg_nonHiggs\",\n",
    "#     ROOT.RooArgList(myy), histos_bkg['m_yy_lowMass_1tag_tightIsolated_bkg_non_Higgs'])\n",
    "# dh_bkg_noH_2tag_lowmass = ROOT.RooDataHist(\n",
    "#     \"2tag_lowmass_bkg_nonHiggs\", \"2tag_lowmass_bkg_nonHiggs\",\n",
    "#     ROOT.RooArgList(myy), histos_bkg['m_yy_lowMass_2tag_tightIsolated_bkg_non_Higgs'])\n",
    "# dh_bkg_1H_0tag_lowmass = ROOT.RooDataHist(\n",
    "#     \"0tag_lowmass_bkg_1Higgs\", \"0tag_lowmass_bkg_nonHiggs\",\n",
    "#     ROOT.RooArgList(myy), histos_bkg['m_yy_lowMass_0tag_tightIsolated_bkg_single_Higgs'])\n",
    "# dh_bkg_1H_1tag_lowmass = ROOT.RooDataHist(\n",
    "#     \"1tag_lowmass_bkg_1Higgs\", \"1tag_lowmass_bkg_nonHiggs\",\n",
    "#     ROOT.RooArgList(myy), histos_bkg['m_yy_lowMass_1tag_tightIsolated_bkg_single_Higgs'])\n",
    "# dh_bkg_1H_2tag_lowmass = ROOT.RooDataHist(\n",
    "#     \"2tag_lowmass_bkg_1Higgs\", \"2tag_lowmass_bkg_1Higgs\",\n",
    "#     ROOT.RooArgList(myy), histos_bkg['m_yy_lowMass_2tag_tightIsolated_bkg_single_Higgs'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "^ are these the full backgrounds with templates? or only the sharpa ones?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Fit to the 0 tag category to get the parameters:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# -- create a workspace\n",
    "w = ROOT.RooWorkspace('w')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::myy\n"
     ]
    }
   ],
   "source": [
    "# -- import the variable myy created above\n",
    "getattr(w,'import')(myy)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# - build model with nuisance params, global observable, categories, ...\n",
    "# - do fit on MC, fix fit function params \n",
    "# - build two model configs, one for B, one for S+B\n",
    "# - make two asimov datasets for mu=0, mu=1 (?)\n",
    "# - use hypothesis test inverter with asymptotic calculator to get CL"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Build generic background PDF with $\\texttt{epoly2}$ function: $$PDF_\\text{bkg} = exp(a_0 + a_1x + a_2x^2)$$\n",
    "Set $a_0=1$ and include the normalization factor later by extending the PDF. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# -- epoly background (b = e^(a0+a1*x+a2*x^2))\n",
    "w.factory(\"PolyVar::bkg_exponent(myy,{a0[1], a1[-100, 100], a2[-100, 100]})\")\n",
    "w.factory(\"Exponential::bkg_pdf(bkg_exponent,1)\")\n",
    "w.var('a0').setConstant(True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# -- old exponential background (b = e^(cx))\n",
    "# c0_bkg = ROOT.RooRealVar('c0_bkg', 'coefficient c0 of exponential bkg fit', -100, 100)\n",
    "# bkg_pdf = ROOT.RooExponential('bkg_pdf', 'exponential bkg pdf', myy, c0_bkg)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Build generic signal PDF with Double Sided CB function:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# -- define single-Higgs bkg fit function (Double Sided Crystal Ball)\n",
    "m0_cb = ROOT.RooRealVar('m0_cb','m0_cb', 125, 105, 160)\n",
    "sigma_cb = ROOT.RooRealVar('sigma_cb','sigma_cb', 5, 0, 20)\n",
    "alphaLo_cb = ROOT.RooRealVar('alphaLo_cb','alphaLo_cb', 0, 5)\n",
    "nLo_cb = ROOT.RooRealVar('nLo_cb','nLo_cb', 0, 10)\n",
    "alphaHi_cb = ROOT.RooRealVar('alphaHi_cb','alphaHi_cb', 0, 5)\n",
    "nHi_cb = ROOT.RooRealVar('nHi_cb','nHi_cb', 0, 10)\n",
    "\n",
    "# -- double sided crystal ball\n",
    "dscb_pdf = ROOT.HggTwoSidedCBPdf(\n",
    "    'dscb_pdf', 'Double Sided Crystal Ball',\n",
    "    myy,\n",
    "    m0_cb,\n",
    "    sigma_cb,\n",
    "    alphaLo_cb,\n",
    "    nLo_cb,\n",
    "    alphaHi_cb,\n",
    "    nHi_cb\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing HggTwoSidedCBPdf::dscb_pdf\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::m0_cb\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::sigma_cb\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::alphaLo_cb\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::nLo_cb\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::alphaHi_cb\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::nHi_cb\n"
     ]
    }
   ],
   "source": [
    "# -- import the DSCB function into the workspace\n",
    "getattr(w,'import')(dscb_pdf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Fit the non-Higgs background function to the high mass and low mass binned data individually to find the values of $a_1$ and $a_2$ that will be applied to the corresponding 1 and 2 tag categories."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:NumericIntegration -- RooRealIntegral::init(bkg_pdf_Int[myy]) using numeric integrator RooIntegrator1D to calculate Int(myy)\n",
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: activating const optimization\n",
      " **********\n",
      " **    1 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **    2 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 a1           0.00000e+00  2.00000e+01   -1.00000e+02  1.00000e+02\n",
      "     2 a2           0.00000e+00  2.00000e+01   -1.00000e+02  1.00000e+02\n",
      " **********\n",
      " **    3 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **    4 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **    5 **SET STR           1\n",
      " **********\n",
      " NOW USING STRATEGY  1: TRY TO BALANCE SPEED AGAINST RELIABILITY\n",
      " **********\n",
      " **    6 **MIGRAD        1000           1\n",
      " **********\n",
      " FIRST CALL TO USER FUNCTION AT NEW START POINT, WITH IFLAG=4.\n",
      " START MIGRAD MINIMIZATION.  STRATEGY  1.  CONVERGENCE WHEN EDM .LT. 1.00e-03\n",
      "[#0] WARNING:Integration -- RooIntegrator1D::integral: integral of bkg_pdf over range (105,160) did not converge after 20 steps\n",
      "   [1] h = 1 , s = inf\n",
      "   [2] h = 0.25 , s = inf\n",
      "   [3] h = 0.0625 , s = inf\n",
      "   [4] h = 0.015625 , s = inf\n",
      "   [5] h = 0.00390625 , s = inf\n",
      "   [6] h = 0.000976562 , s = inf\n",
      "   [7] h = 0.000244141 , s = inf\n",
      "   [8] h = 6.10352e-05 , s = inf\n",
      "   [9] h = 1.52588e-05 , s = inf\n",
      "   [10] h = 3.8147e-06 , s = inf\n",
      "   [11] h = 9.53674e-07 , s = inf\n",
      "   [12] h = 2.38419e-07 , s = inf\n",
      "   [13] h = 5.96046e-08 , s = inf\n",
      "   [14] h = 1.49012e-08 , s = inf\n",
      "   [15] h = 3.72529e-09 , s = inf\n",
      "   [16] h = 9.31323e-10 , s = inf\n",
      "   [17] h = 2.32831e-10 , s = inf\n",
      "   [18] h = 5.82077e-11 , s = inf\n",
      "   [19] h = 1.45519e-11 , s = inf\n",
      "   [20] h = 3.63798e-12 , s = inf\n",
      "[#0] WARNING:Minization -- RooMinimizerFcn: Minimized function has error status.\n",
      "Returning maximum FCN so far (1.10364e+06) to force MIGRAD to back out of this region. Error log follows\n",
      "Parameter values: a1=0, a2=2.33523\n",
      "RooNLLVar::nll_bkg_pdf_m_yy_lowMass_0tag_tightIsolated_bkg_non_Higgs[ paramSet=(a0,a1,a2) ]\n",
      "     function value is NAN @ paramSet=(a0 = 1,a1 = 0,a2 = 2.33523)\n",
      "RooExponential::bkg_pdf[ x=bkg_exponent c=1 ]\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=26363.6, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=27618.8, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=28903.2, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=30216.7, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=31559.5, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=32931.4, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=34332.6, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=35762.9, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=37222.4, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=38711.1, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=40229, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=41776.1, c=1=1\n",
      "    ... (remaining 12 messages suppressed)\n",
      "\n",
      "[#0] WARNING:Minization -- RooMinimizerFcn: Minimized function has error status.\n",
      "Returning maximum FCN so far (1.10364e+06) to force MIGRAD to back out of this region. Error log follows\n",
      "Parameter values: a1=0, a2=-2.33523\n",
      "RooNLLVar::nll_bkg_pdf_m_yy_lowMass_0tag_tightIsolated_bkg_non_Higgs[ paramSet=(a0,a1,a2) ]\n",
      "     function value is NAN @ paramSet=(a0 = 1,a1 = 0,a2 = -2.33523)\n",
      "RooExponential::bkg_pdf[ x=bkg_exponent c=1 ]\n",
      "     p.d.f normalization integral is zero or negative @ x=bkg_exponent=-26361.6, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to zero @ x=bkg_exponent=-26361.6, c=1=1\n",
      "     p.d.f normalization integral is zero or negative @ x=bkg_exponent=-27616.8, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to zero @ x=bkg_exponent=-27616.8, c=1=1\n",
      "     p.d.f normalization integral is zero or negative @ x=bkg_exponent=-28901.2, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to zero @ x=bkg_exponent=-28901.2, c=1=1\n",
      "     p.d.f normalization integral is zero or negative @ x=bkg_exponent=-30214.7, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to zero @ x=bkg_exponent=-30214.7, c=1=1\n",
      "     p.d.f normalization integral is zero or negative @ x=bkg_exponent=-31557.5, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to zero @ x=bkg_exponent=-31557.5, c=1=1\n",
      "     p.d.f normalization integral is zero or negative @ x=bkg_exponent=-32929.4, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to zero @ x=bkg_exponent=-32929.4, c=1=1\n",
      "    ... (remaining 34 messages suppressed)\n",
      "\n",
      "[#0] WARNING:Integration -- RooIntegrator1D::integral: integral of bkg_pdf over range (105,160) did not converge after 20 steps\n",
      "   [1] h = 1 , s = inf\n",
      "   [2] h = 0.25 , s = inf\n",
      "   [3] h = 0.0625 , s = inf\n",
      "   [4] h = 0.015625 , s = inf\n",
      "   [5] h = 0.00390625 , s = inf\n",
      "   [6] h = 0.000976562 , s = inf\n",
      "   [7] h = 0.000244141 , s = inf\n",
      "   [8] h = 6.10352e-05 , s = inf\n",
      "   [9] h = 1.52588e-05 , s = inf\n",
      "   [10] h = 3.8147e-06 , s = inf\n",
      "   [11] h = 9.53674e-07 , s = inf\n",
      "   [12] h = 2.38419e-07 , s = inf\n",
      "   [13] h = 5.96046e-08 , s = inf\n",
      "   [14] h = 1.49012e-08 , s = inf\n",
      "   [15] h = 3.72529e-09 , s = inf\n",
      "   [16] h = 9.31323e-10 , s = inf\n",
      "   [17] h = 2.32831e-10 , s = inf\n",
      "   [18] h = 5.82077e-11 , s = inf\n",
      "   [19] h = 1.45519e-11 , s = inf\n",
      "   [20] h = 3.63798e-12 , s = inf\n",
      "[#0] WARNING:Minization -- RooMinimizerFcn: Minimized function has error status.\n",
      "Returning maximum FCN so far (1.10364e+06) to force MIGRAD to back out of this region. Error log follows\n",
      "Parameter values: a1=0, a2=0.233544\n",
      "RooNLLVar::nll_bkg_pdf_m_yy_lowMass_0tag_tightIsolated_bkg_non_Higgs[ paramSet=(a0,a1,a2) ]\n",
      "     function value is NAN @ paramSet=(a0 = 1,a1 = 0,a2 = 0.233544)\n",
      "RooExponential::bkg_pdf[ x=bkg_exponent c=1 ]\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=2637.5, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=2763.03, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=2891.48, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=3022.84, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=3157.13, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=3294.34, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=3434.47, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=3577.51, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=3723.48, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=3872.36, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=4024.17, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to NaN @ x=bkg_exponent=4178.89, c=1=1\n",
      "    ... (remaining 12 messages suppressed)\n",
      "\n",
      "[#0] WARNING:Minization -- RooMinimizerFcn: Minimized function has error status.\n",
      "Returning maximum FCN so far (1.10364e+06) to force MIGRAD to back out of this region. Error log follows\n",
      "Parameter values: a1=0, a2=-0.233544\n",
      "RooNLLVar::nll_bkg_pdf_m_yy_lowMass_0tag_tightIsolated_bkg_non_Higgs[ paramSet=(a0,a1,a2) ]\n",
      "     function value is NAN @ paramSet=(a0 = 1,a1 = 0,a2 = -0.233544)\n",
      "RooExponential::bkg_pdf[ x=bkg_exponent c=1 ]\n",
      "     p.d.f normalization integral is zero or negative @ x=bkg_exponent=-2635.5, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to zero @ x=bkg_exponent=-2635.5, c=1=1\n",
      "     p.d.f normalization integral is zero or negative @ x=bkg_exponent=-2761.03, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to zero @ x=bkg_exponent=-2761.03, c=1=1\n",
      "     p.d.f normalization integral is zero or negative @ x=bkg_exponent=-2889.48, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to zero @ x=bkg_exponent=-2889.48, c=1=1\n",
      "     p.d.f normalization integral is zero or negative @ x=bkg_exponent=-3020.84, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to zero @ x=bkg_exponent=-3020.84, c=1=1\n",
      "     p.d.f normalization integral is zero or negative @ x=bkg_exponent=-3155.13, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to zero @ x=bkg_exponent=-3155.13, c=1=1\n",
      "     p.d.f normalization integral is zero or negative @ x=bkg_exponent=-3292.34, c=1=1\n",
      "     getLogVal() top-level p.d.f evaluates to zero @ x=bkg_exponent=-3292.34, c=1=1\n",
      "    ... (remaining 34 messages suppressed)\n",
      "\n",
      " FCN=56423.3 FROM MIGRAD    STATUS=INITIATE        8 CALLS           9 TOTAL\n",
      "                     EDM= unknown      STRATEGY= 1      NO ERROR MATRIX       \n",
      "  EXT PARAMETER               CURRENT GUESS       STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  a1           0.00000e+00   2.00000e+01   2.01358e-01   9.05124e+06\n",
      "   2  a2           0.00000e+00   2.00000e+01   2.01358e-01   0.00000e+00\n",
      "                               ERR DEF= 0.5\n",
      " MIGRAD FAILS TO FIND IMPROVEMENT\n",
      " EIGENVALUES OF SECOND-DERIVATIVE MATRIX:\n",
      "        -1.2999e-01  2.1300e+00\n",
      " MINUIT WARNING IN HESSE   \n",
      " ============== MATRIX FORCED POS-DEF BY ADDING 0.132116 TO DIAGONAL.\n",
      " FCN=55249.7 FROM HESSE     STATUS=NOT POSDEF     12 CALLS          46 TOTAL\n",
      "                     EDM=1.58833    STRATEGY= 1      ERR MATRIX NOT POS-DEF\n",
      "  EXT PARAMETER                APPROXIMATE        STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  a1          -2.67655e-02   8.57498e-03   2.33545e-06  -1.21882e+05\n",
      "   2  a2           0.00000e+00   3.37477e-05   2.33545e-06  -3.58390e+07\n",
      "                               ERR DEF= 0.5\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=55248.1 FROM MIGRAD    STATUS=CONVERGED      67 CALLS          68 TOTAL\n",
      "                     EDM=4.86617e-05    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  a1          -4.29004e-02   1.46481e-02   9.01243e-07  -7.31379e+02\n",
      "   2  a2           6.35874e-05   5.63103e-05   3.46417e-09  -2.00193e+05\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  2.146e-04 -8.242e-07 \n",
      " -8.242e-07  3.171e-09 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.99928   1.000 -0.999\n",
      "        2  0.99928  -0.999  1.000\n",
      " **********\n",
      " **    7 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **    8 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **    9 **HESSE        1000\n",
      " **********\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=55248.1 FROM HESSE     STATUS=OK             10 CALLS          78 TOTAL\n",
      "                     EDM=3.40587e-05    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  a1          -4.29004e-02   1.06865e-02   1.80249e-07  -4.29004e-04\n",
      "   2  a2           6.35874e-05   4.10811e-05   6.92834e-10   6.35874e-07\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  1.142e-04 -4.384e-07 \n",
      " -4.384e-07  1.688e-09 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.99865   1.000 -0.999\n",
      "        2  0.99865  -0.999  1.000\n",
      "[#1] INFO:Fitting -- RooAbsPdf::fitTo(bkg_pdf) Calculating sum-of-weights-squared correction matrix for covariance matrix\n",
      " **********\n",
      " **   10 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   11 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   12 **HESSE        1000\n",
      " **********\n",
      " EIGENVALUES OF SECOND-DERIVATIVE MATRIX:\n",
      "        -1.7038e-03  2.0017e+00\n",
      " MINUIT WARNING IN HESSE   \n",
      " ============== MATRIX FORCED POS-DEF BY ADDING 0.003705 TO DIAGONAL.\n",
      " FCN=29389.1 FROM HESSE     STATUS=NOT POSDEF     14 CALLS          92 TOTAL\n",
      "                     EDM=12.0872    STRATEGY= 1      ERR MATRIX NOT POS-DEF\n",
      "  EXT PARAMETER                APPROXIMATE     INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  a1          -4.29004e-02   1.20083e-02   3.67707e-06  -4.29004e-04\n",
      "   2  a2           6.35874e-05   4.61625e-05   1.41338e-08   6.35874e-07\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  1.442e-04 -5.532e-07 \n",
      " -5.532e-07  2.131e-09 \n",
      "ERR MATRIX NOT POS-DEF\n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.99801   1.000 -0.998\n",
      "        2  0.99801  -0.998  1.000\n",
      " ERR MATRIX NOT POS-DEF\n",
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: deactivating const optimization\n"
     ]
    }
   ],
   "source": [
    "# low mass fit\n",
    "r = w.pdf(\"bkg_pdf\").fitTo(lowMass_0tag_non_Higgs, rf.SumW2Error(ROOT.kTRUE), rf.Save())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:InputArguments -- RooAbsData::plotOn(m_yy_lowMass_0tag_tightIsolated_bkg_non_Higgs) INFO: dataset has non-integer weights, auto-selecting SumW2 errors instead of Poisson errors\n",
      "[#1] INFO:NumericIntegration -- RooRealIntegral::init(bkg_pdf_Int[myy]) using numeric integrator RooIntegrator1D to calculate Int(myy)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArgAAAHYCAIAAAApvgy/AAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE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\nCgAAwItAAQAAeBEoAAAAr/8H6TdpNle7aOAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "c = ROOT.TCanvas()\n",
    "frame = myy.frame()\n",
    "lowMass_0tag_non_Higgs.plotOn(frame)\n",
    "w.pdf(\"bkg_pdf\").plotOn(frame)\n",
    "w.pdf(\"bkg_pdf\").paramOn(frame)#,ROOT.RooFit.Layout(0.65,0.85,0.85),ROOT.RooFit.Parameters(ROOT.RooArgSet(w.var(\"a1\"),w.var(\"nbkg1\"))))\n",
    "frame.Draw()\n",
    "frame.SetTitle(\"Low Mass Non-Higgs Background - 0 Tag\")\n",
    "c.Draw()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Save a snapshot of the values of $a_1$ and $a_2$ that were found by fitting the low mass non-Higgs background in the 0 tag category. These values will be frozen for the 1 and 2 tag categories."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "w.defineSet(\"params_epoly2\", \"a1,a2\")\n",
    "w.saveSnapshot(\"lowmass_epoly2\",w.set(\"params_epoly2\"), ROOT.kTRUE)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Redo the fit for the high mass category:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:NumericIntegration -- RooRealIntegral::init(bkg_pdf_Int[myy]) using numeric integrator RooIntegrator1D to calculate Int(myy)\n",
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: activating const optimization\n",
      " **********\n",
      " **   13 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   14 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 a1          -4.29004e-02  9.51067e-03   -1.00000e+02  1.00000e+02\n",
      "     2 a2           6.35874e-05  3.65609e-05   -1.00000e+02  1.00000e+02\n",
      " **********\n",
      " **   15 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   16 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   17 **SET STR           1\n",
      " **********\n",
      " NOW USING STRATEGY  1: TRY TO BALANCE SPEED AGAINST RELIABILITY\n",
      " **********\n",
      " **   18 **MIGRAD        1000           1\n",
      " **********\n",
      " FIRST CALL TO USER FUNCTION AT NEW START POINT, WITH IFLAG=4.\n",
      " START MIGRAD MINIMIZATION.  STRATEGY  1.  CONVERGENCE WHEN EDM .LT. 1.00e-03\n",
      " FCN=4391.26 FROM MIGRAD    STATUS=INITIATE        8 CALLS           9 TOTAL\n",
      "                     EDM= unknown      STRATEGY= 1      NO ERROR MATRIX       \n",
      "  EXT PARAMETER               CURRENT GUESS       STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  a1          -4.29004e-02   9.51067e-03   9.51067e-05  -2.99255e+05\n",
      "   2  a2           6.35874e-05   3.65609e-05   3.65609e-07  -7.77438e+07\n",
      "                               ERR DEF= 0.5\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=4373.95 FROM MIGRAD    STATUS=CONVERGED      46 CALLS          47 TOTAL\n",
      "                     EDM=4.20831e-08    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  a1          -1.54718e-02   4.35775e-02   8.82011e-07  -1.03219e+01\n",
      "   2  a2           2.31129e-06   1.66082e-04   3.36156e-09  -2.73483e+03\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  1.899e-03 -7.230e-06 \n",
      " -7.230e-06  2.758e-08 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.99902   1.000 -0.999\n",
      "        2  0.99902  -0.999  1.000\n",
      " **********\n",
      " **   19 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   20 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   21 **HESSE        1000\n",
      " **********\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=4373.95 FROM HESSE     STATUS=OK             10 CALLS          57 TOTAL\n",
      "                     EDM=4.16816e-08    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  a1          -1.54718e-02   3.68526e-02   1.76402e-07  -1.54718e-04\n",
      "   2  a2           2.31129e-06   1.40452e-04   6.72313e-10   2.31129e-08\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  1.358e-03 -5.169e-06 \n",
      " -5.169e-06  1.973e-08 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.99863   1.000 -0.999\n",
      "        2  0.99863  -0.999  1.000\n",
      "[#1] INFO:Fitting -- RooAbsPdf::fitTo(bkg_pdf) Calculating sum-of-weights-squared correction matrix for covariance matrix\n",
      " **********\n",
      " **   22 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   23 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   24 **HESSE        1000\n",
      " **********\n",
      " EIGENVALUES OF SECOND-DERIVATIVE MATRIX:\n",
      "        -5.1369e-04  2.0005e+00\n",
      " MINUIT WARNING IN HESSE   \n",
      " ============== MATRIX FORCED POS-DEF BY ADDING 0.002514 TO DIAGONAL.\n",
      " FCN=1248.91 FROM HESSE     STATUS=NOT POSDEF     14 CALLS          71 TOTAL\n",
      "                     EDM=4.07958    STRATEGY= 1      ERR MATRIX NOT POS-DEF\n",
      "  EXT PARAMETER                APPROXIMATE     INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  a1          -1.54718e-02   5.69749e-02   3.59860e-06  -1.54718e-04\n",
      "   2  a2           2.31129e-06   2.17142e-04   1.37152e-08   2.31129e-08\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  3.246e-03 -1.235e-05 \n",
      " -1.235e-05  4.715e-08 \n",
      "ERR MATRIX NOT POS-DEF\n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.99800   1.000 -0.998\n",
      "        2  0.99800  -0.998  1.000\n",
      " ERR MATRIX NOT POS-DEF\n",
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: deactivating const optimization\n"
     ]
    }
   ],
   "source": [
    "# high mass fit\n",
    "r = w.pdf(\"bkg_pdf\").fitTo(highMass_0tag_non_Higgs, rf.SumW2Error(ROOT.kTRUE), rf.Save())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:InputArguments -- RooAbsData::plotOn(m_yy_highMass_0tag_tightIsolated_bkg_non_Higgs) INFO: dataset has non-integer weights, auto-selecting SumW2 errors instead of Poisson errors\n",
      "[#1] INFO:NumericIntegration -- RooRealIntegral::init(bkg_pdf_Int[myy]) using numeric integrator RooIntegrator1D to calculate Int(myy)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArgAAAHYCAIAAAApvgy/AAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE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\nLwIFAADgRaAAAAC8/j8vLmY2Lg6kaAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "c = ROOT.TCanvas()\n",
    "frame = myy.frame()\n",
    "highMass_0tag_non_Higgs.plotOn(frame)\n",
    "w.pdf(\"bkg_pdf\").plotOn(frame)\n",
    "w.pdf(\"bkg_pdf\").paramOn(frame)#,ROOT.RooFit.Layout(0.65,0.85,0.85),ROOT.RooFit.Parameters(ROOT.RooArgSet(w.var(\"a1\"),w.var(\"nbkg1\"))))\n",
    "frame.Draw()\n",
    "frame.SetTitle(\"High Mass Non-Higgs Background - 0 Tag\")\n",
    "c.Draw()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "w.saveSnapshot(\"highmass_epoly2\",w.set(\"params_epoly2\"), ROOT.kTRUE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "RooWorkspace(w) w contents\n",
      "\n",
      "variables\n",
      "---------\n",
      "(a0,a1,a2,alphaHi_cb,alphaLo_cb,m0_cb,myy,nHi_cb,nLo_cb,sigma_cb)\n",
      "\n",
      "p.d.f.s\n",
      "-------\n",
      "RooExponential::bkg_pdf[ x=bkg_exponent c=1 ] = 0.364423\n",
      "HggTwoSidedCBPdf::dscb_pdf[ m=myy m0=m0_cb sigma=sigma_cb alphaLo=alphaLo_cb nLo=nLo_cb alphaHi=alphaHi_cb nHi=nHi_cb ] = 0.324652\n",
      "\n",
      "functions\n",
      "--------\n",
      "RooPolyVar::bkg_exponent[ x=myy coefList=(a0,a1,a2) ] = -1.00944\n",
      "\n",
      "parameter snapshots\n",
      "-------------------\n",
      "lowmass_epoly2 = (a1=-0.0429004 +/- 0.00951067,a2=6.35874e-05 +/- 3.65609e-05)\n",
      "highmass_epoly2 = (a1=-0.0154718 +/- 0.0238383,a2=2.31129e-06 +/- 9.08521e-05)\n",
      "\n",
      "named sets\n",
      "----------\n",
      "params_epoly2:(a1,a2)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "w.Print()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "The cells below have been commented out. They tested the idea of doing a simultaneous fit in the high and low mass categories, which is of course a bad idea given that the two have different shapes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# -- individual fit to single higgs bkg\n",
    "#w.pdf(\"dscb_pdf\").fitTo(dh_bkg_1H_0tag_lowmass, rf.SumW2Error(ROOT.kTRUE))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#w.factory(\"tag0cat[tag0low,tag0high]\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# from copy import deepcopy\n",
    "# noH_0tag_combo = ROOT.RooDataHist(\"combData\", \"combined data\", ROOT.RooArgList(myy),\n",
    "#                             rf.Index(w.cat(\"tag0cat\")),\n",
    "#                             rf.Import(\"tag0low\", dh_bkg_noH_0tag_lowmass),\n",
    "#                             rf.Import(\"tag0high\", dh_bkg_noH_0tag_highmass))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# #w.factory(\"SIMUL:noH_0tag_jointmodel(tag0cat,tag0low=bkg_pdf,tag0high=bkg_pdf)\")\n",
    "# w.factory(\"SIMUL:noH_0tag_jointmodel(\\\n",
    "#     tag0cat,\\\n",
    "#     tag0low = ExtendPdf:pdf_low(bkg_pdf, n_low[100, 0, 100000]),\\\n",
    "#     tag0high = ExtendPdf:pdf_high(bkg_pdf, n_high[100, 0, 100000]))\"\n",
    "# )\n",
    "# #ExtendPdf:pdf(gaus, nevt[100,0,100000])\n",
    "# # Simultaneous p.d.f.s are constructed with SIMUL( index, state1=pdf1, state2=pdf2,...)\n",
    "# # w->factory(\"SIMUL::smodel( c[A=0,B=1], A=Gaussian::gs(x,m,s[1]), B=Landau::ls(x,0,1) )\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# c1 = ROOT.TCanvas()\n",
    "# plot1 = myy.frame()\n",
    "# plot2 = myy.frame()\n",
    "\n",
    "# r = w.pdf(\"noH_0tag_jointmodel\").fitTo(noH_0tag_combo, rf.Save(), rf.SumW2Error(ROOT.kTRUE))\n",
    "# r.Print()\n",
    "\n",
    "# #dh_bkg_noH_0tag_lowmass.plotOn(plot1, rf.MarkerColor(ROOT.kRed))\n",
    "# noH_0tag_combo.plotOn(plot1, ROOT.RooFit.Cut(\"tag0cat==tag0cat::tag0low\"))\n",
    "# w.pdf(\"noH_0tag_jointmodel\").plotOn(plot1, rf.ProjWData(noH_0tag_combo), rf.Slice(w.cat(\"tag0cat\"),\"tag0low\"))\n",
    "\n",
    "# #dh_bkg_noH_0tag_highmass.plotOn(plot2)\n",
    "# noH_0tag_combo.plotOn(plot2, ROOT.RooFit.Cut(\"tag0cat==tag0cat::tag0high\"))\n",
    "# w.pdf(\"noH_0tag_jointmodel\").plotOn(plot2, rf.ProjWData(noH_0tag_combo), rf.Slice(w.cat(\"tag0cat\"),\"tag0high\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# plot1.Draw()\n",
    "# c1.Draw()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# plot2.Draw()\n",
    "# c1.Draw()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# --> conclusion: fit high and low mass separately, not simultaneously"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# # -- generate fake data and re-fit it\n",
    "# data = w.pdf(\"noH_0tag_jointmodel\").generate(ROOT.RooArgSet(myy,w.cat(\"tag0cat\")))\n",
    "# data.SetName(\"data\")\n",
    "# getattr(w,'import')(data)\n",
    "# data.Print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "# c = ROOT.TCanvas()\n",
    "# plot1 = myy.frame(ROOT.RooFit.Title(\"Channel 1\"))\n",
    "# plot2 = myy.frame(ROOT.RooFit.Title(\"Channel 2\"))\n",
    "# data.plotOn(plot1,ROOT.RooFit.Cut(\"tag0cat==tag0cat::tag0low\"))\n",
    "# data.plotOn(plot2,ROOT.RooFit.Cut(\"tag0cat==tag0cat::tag0high\"))\n",
    "# r = w.pdf(\"noH_0tag_jointmodel\").fitTo(data, rf.SumW2Error(ROOT.kTRUE), rf.Save())\n",
    "# w.pdf(\"noH_0tag_jointmodel\").plotOn(plot1,ROOT.RooFit.ProjWData(data),ROOT.RooFit.Slice(w.cat(\"tag0cat\"),\"tag0low\"))\n",
    "# w.pdf(\"noH_0tag_jointmodel\").plotOn(plot2,ROOT.RooFit.ProjWData(data),ROOT.RooFit.Slice(w.cat(\"tag0cat\"),\"tag0high\"))\n",
    "# c.Divide(1,2)\n",
    "# c.cd(1)\n",
    "# plot1.Draw()\n",
    "# c.cd(2)\n",
    "# plot2.Draw()\n",
    "# c.Draw()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Check how well these fit the 1 and 2 tag categories:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:InputArguments -- RooAbsData::plotOn(m_yy_highMass_1tag_tightIsolated_bkg_non_Higgs) INFO: dataset has non-integer weights, auto-selecting SumW2 errors instead of Poisson errors\n",
      "[#1] INFO:NumericIntegration -- RooRealIntegral::init(bkg_pdf_Int[myy]) using numeric integrator RooIntegrator1D to calculate Int(myy)\n",
      "[#1] INFO:InputArguments -- RooAbsData::plotOn(m_yy_highMass_2tag_tightIsolated_bkg_non_Higgs) INFO: dataset has non-integer weights, auto-selecting SumW2 errors instead of Poisson errors\n",
      "[#1] INFO:NumericIntegration -- RooRealIntegral::init(bkg_pdf_Int[myy]) using numeric integrator RooIntegrator1D to calculate Int(myy)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArgAAAHYCAIAAAApvgy/AAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE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\n/wGvyS2MMKhmDQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArgAAAHYCAIAAAApvgy/AAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE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\nAABY/R8e3TqTRuFWyAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# -- high mass (keep a1 and a2 values we just found)\n",
    "c = ROOT.TCanvas()\n",
    "frame = myy.frame()\n",
    "highMass_1tag_non_Higgs.plotOn(frame)\n",
    "w.pdf(\"bkg_pdf\").plotOn(frame)\n",
    "w.pdf(\"bkg_pdf\").paramOn(frame)#,ROOT.RooFit.Layout(0.65,0.85,0.85),ROOT.RooFit.Parameters(ROOT.RooArgSet(w.var(\"a1\"),w.var(\"nbkg1\"))))\n",
    "frame.Draw()\n",
    "frame.SetTitle(\"High Mass Non-Higgs Background - 1 Tag\")\n",
    "c.Draw()\n",
    "c1 = ROOT.TCanvas()\n",
    "frame = myy.frame()\n",
    "highMass_2tag_non_Higgs.plotOn(frame)\n",
    "w.pdf(\"bkg_pdf\").plotOn(frame)\n",
    "w.pdf(\"bkg_pdf\").paramOn(frame)#,ROOT.RooFit.Layout(0.65,0.85,0.85),ROOT.RooFit.Parameters(ROOT.RooArgSet(w.var(\"a1\"),w.var(\"nbkg1\"))))\n",
    "frame.Draw()\n",
    "frame.SetTitle(\"High Mass Non-Higgs Background - 2 Tag\")\n",
    "c1.Draw()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:InputArguments -- RooAbsData::plotOn(m_yy_lowMass_1tag_tightIsolated_bkg_non_Higgs) INFO: dataset has non-integer weights, auto-selecting SumW2 errors instead of Poisson errors\n",
      "[#1] INFO:NumericIntegration -- RooRealIntegral::init(bkg_pdf_Int[myy]) using numeric integrator RooIntegrator1D to calculate Int(myy)\n",
      "[#1] INFO:InputArguments -- RooAbsData::plotOn(m_yy_lowMass_2tag_tightIsolated_bkg_non_Higgs) INFO: dataset has non-integer weights, auto-selecting SumW2 errors instead of Poisson errors\n",
      "[#1] INFO:NumericIntegration -- RooRealIntegral::init(bkg_pdf_Int[myy]) using numeric integrator RooIntegrator1D to calculate Int(myy)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArgAAAHYCAIAAAApvgy/AAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE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\n4PX/Ad3D8tdf7tCEAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArgAAAHYCAIAAAApvgy/AAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE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\ncCJQAAAATv8fc0YwPBV9TvAAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# -- low mass (restore previous snapshot)\n",
    "w.loadSnapshot(\"lowmass_epoly2\")\n",
    "c = ROOT.TCanvas()\n",
    "frame = myy.frame()\n",
    "lowMass_1tag_non_Higgs.plotOn(frame)\n",
    "w.pdf(\"bkg_pdf\").plotOn(frame)\n",
    "w.pdf(\"bkg_pdf\").paramOn(frame)#,ROOT.RooFit.Layout(0.65,0.85,0.85),ROOT.RooFit.Parameters(ROOT.RooArgSet(w.var(\"a1\"),w.var(\"nbkg1\"))))\n",
    "frame.Draw()\n",
    "frame.SetTitle(\"Low Mass Non-Higgs Background - 1 Tag\")\n",
    "c.Draw()\n",
    "c1 = ROOT.TCanvas()\n",
    "frame = myy.frame()\n",
    "lowMass_2tag_non_Higgs.plotOn(frame)\n",
    "w.pdf(\"bkg_pdf\").plotOn(frame)\n",
    "w.pdf(\"bkg_pdf\").paramOn(frame)#,ROOT.RooFit.Layout(0.65,0.85,0.85),ROOT.RooFit.Parameters(ROOT.RooArgSet(w.var(\"a1\"),w.var(\"nbkg1\"))))\n",
    "frame.Draw()\n",
    "frame.SetTitle(\"Low Mass Non-Higgs Background - 2 Tag\")\n",
    "c1.Draw()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "These look less than ideal. How can we quantify this?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Do the same for the DSCB function that represents the single-Higgs background. Fit it in the 0 tag category (high mass and low mass separately) and then check how that fits the 1 and 2 tag categories. The difference here is that we will only fix the $\\alpha$ and $n$ parameters, while we we let $m_0$ and $\\sigma$ float with every fit."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: activating const optimization\n",
      " **********\n",
      " **    1 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **    2 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 alphaHi_cb   2.50000e+00  5.00000e-01    0.00000e+00  5.00000e+00\n",
      "     2 alphaLo_cb   2.50000e+00  5.00000e-01    0.00000e+00  5.00000e+00\n",
      "     3 m0_cb        1.25000e+02  5.50000e+00    1.05000e+02  1.60000e+02\n",
      "     4 nHi_cb       5.00000e+00  1.00000e+00    0.00000e+00  1.00000e+01\n",
      "     5 nLo_cb       5.00000e+00  1.00000e+00    0.00000e+00  1.00000e+01\n",
      "     6 sigma_cb     5.00000e+00  2.00000e+00    0.00000e+00  2.00000e+01\n",
      " **********\n",
      " **    3 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **    4 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **    5 **SET STR           1\n",
      " **********\n",
      " NOW USING STRATEGY  1: TRY TO BALANCE SPEED AGAINST RELIABILITY\n",
      " **********\n",
      " **    6 **MIGRAD        3000           1\n",
      " **********\n",
      " FIRST CALL TO USER FUNCTION AT NEW START POINT, WITH IFLAG=4.\n",
      " START MIGRAD MINIMIZATION.  STRATEGY  1.  CONVERGENCE WHEN EDM .LT. 1.00e-03\n",
      " FCN=150.956 FROM MIGRAD    STATUS=INITIATE       74 CALLS          75 TOTAL\n",
      "                     EDM= unknown      STRATEGY= 1      NO ERROR MATRIX       \n",
      "  EXT PARAMETER               CURRENT GUESS       STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  alphaHi_cb   2.50000e+00   5.00000e-01   0.00000e+00   6.66069e-01\n",
      "   2  alphaLo_cb   2.50000e+00   5.00000e-01   0.00000e+00   1.30876e+00\n",
      "   3  m0_cb        1.25000e+02   5.50000e+00   0.00000e+00  -5.32499e+00\n",
      "   4  nHi_cb       5.00000e+00   1.00000e+00   0.00000e+00   7.50284e-02\n",
      "   5  nLo_cb       2.53328e+00   1.00000e+00  -5.15931e-01   7.53276e-02\n",
      "   6  sigma_cb     2.05687e+00   2.00000e+00   0.00000e+00   1.19287e+01\n",
      "                               ERR DEF= 0.5\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=150.732 FROM MIGRAD    STATUS=CONVERGED     222 CALLS         223 TOTAL\n",
      "                     EDM=1.30474e-05    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  alphaHi_cb   2.32694e+00   8.79964e-01   2.20187e-03   5.73673e-03\n",
      "   2  alphaLo_cb   2.16478e+00   7.51938e-01   1.69566e-03  -1.03532e-02\n",
      "   3  m0_cb        1.25013e+02   2.41709e-01   7.71822e-05  -1.22117e-01\n",
      "   4  nHi_cb       4.16611e+00   6.80447e+00   1.13927e-02  -1.57077e-05\n",
      "   5  nLo_cb       3.59579e+00   6.74759e+00   7.62115e-03  -2.44724e-03\n",
      "   6  sigma_cb     1.96760e+00   1.84900e-01   2.54505e-04  -3.43312e-02\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  6    ERR DEF=0.5\n",
      "  8.088e-01  1.915e-03  1.117e-03 -5.768e+00  6.001e-03  6.670e-04 \n",
      "  1.915e-03  5.837e-01  3.637e-03 -3.832e-02 -4.126e+00 -7.653e-03 \n",
      "  1.117e-03  3.637e-03  5.842e-02  9.295e-02 -1.174e-01 -6.117e-04 \n",
      " -5.768e+00 -3.832e-02  9.295e-02  8.589e+01  1.224e+00  2.104e-01 \n",
      "  6.001e-03 -4.126e+00 -1.174e-01  1.224e+00  5.092e+01  2.285e-01 \n",
      "  6.670e-04 -7.653e-03 -6.117e-04  2.104e-01  2.285e-01  3.420e-02 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2      3      4      5      6\n",
      "        1  0.69872   1.000  0.003  0.005 -0.692  0.001  0.004\n",
      "        2  0.76147   0.003  1.000  0.020 -0.005 -0.757 -0.054\n",
      "        3  0.10658   0.005  0.020  1.000  0.041 -0.068 -0.014\n",
      "        4  0.70487  -0.692 -0.005  0.041  1.000  0.019  0.123\n",
      "        5  0.77000   0.001 -0.757 -0.068  0.019  1.000  0.173\n",
      "        6  0.26766   0.004 -0.054 -0.014  0.123  0.173  1.000\n",
      " **********\n",
      " **    7 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **    8 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **    9 **HESSE        3000\n",
      " **********\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=150.732 FROM HESSE     STATUS=OK             40 CALLS         263 TOTAL\n",
      "                     EDM=1.32847e-05    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  alphaHi_cb   2.32694e+00   9.00735e-01   4.40375e-04  -6.92778e-02\n",
      "   2  alphaLo_cb   2.16478e+00   7.66845e-01   3.39133e-04  -1.34491e-01\n",
      "   3  m0_cb        1.25013e+02   2.41734e-01   1.54364e-05  -2.75741e-01\n",
      "   4  nHi_cb       4.16611e+00   6.74763e+00   4.55710e-04  -1.67562e-01\n",
      "   5  nLo_cb       3.59579e+00   6.73100e+00   1.52423e-03  -2.84671e-01\n",
      "   6  sigma_cb     1.96760e+00   1.85033e-01   5.09009e-05  -9.32715e-01\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  6    ERR DEF=0.5\n",
      "  8.493e-01  2.180e-03  6.777e-04 -6.215e+00 -1.786e-04 -3.497e-04 \n",
      "  2.180e-03  6.079e-01  4.147e-03 -4.598e-02 -4.361e+00 -8.663e-03 \n",
      "  6.777e-04  4.147e-03  5.844e-02  9.650e-02 -1.219e-01 -6.513e-04 \n",
      " -6.215e+00 -4.598e-02  9.650e-02  9.027e+01  1.324e+00  2.196e-01 \n",
      " -1.786e-04 -4.361e+00 -1.219e-01  1.324e+00  5.306e+01  2.373e-01 \n",
      " -3.497e-04 -8.663e-03 -6.513e-04  2.196e-01  2.373e-01  3.425e-02 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2      3      4      5      6\n",
      "        1  0.71596   1.000  0.003  0.003 -0.710 -0.000 -0.002\n",
      "        2  0.77236   0.003  1.000  0.022 -0.006 -0.768 -0.060\n",
      "        3  0.10752   0.003  0.022  1.000  0.042 -0.069 -0.015\n",
      "        4  0.72189  -0.710 -0.006  0.042  1.000  0.019  0.125\n",
      "        5  0.78054  -0.000 -0.768 -0.069  0.019  1.000  0.176\n",
      "        6  0.27013  -0.002 -0.060 -0.015  0.125  0.176  1.000\n",
      "[#1] INFO:Fitting -- RooAbsPdf::fitTo(dscb_pdf) Calculating sum-of-weights-squared correction matrix for covariance matrix\n",
      " **********\n",
      " **   10 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   11 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   12 **HESSE        3000\n",
      " **********\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=0.424829 FROM HESSE     STATUS=OK             54 CALLS         317 TOTAL\n",
      "                     EDM=0.000368542    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  alphaHi_cb   2.32694e+00   2.92201e+00   8.98364e-03  -6.92778e-02\n",
      "   2  alphaLo_cb   2.16478e+00   3.44422e+00   6.91830e-03  -1.34491e-01\n",
      "   3  m0_cb        1.25013e+02   4.57436e+00   3.14903e-04  -2.75741e-01\n",
      "   4  nHi_cb       4.16611e+00   7.21702e+00   9.29648e-03  -1.67562e-01\n",
      "   5  nLo_cb       3.59579e+00   6.79302e+00   3.10943e-02  -2.84671e-01\n",
      "   6  sigma_cb     1.96760e+00   3.22497e+00   1.03838e-03  -9.32715e-01\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  6    ERR DEF=0.5\n",
      "  2.737e+02  5.160e-01  8.496e-01 -1.872e+03  7.736e+00  1.276e+00 \n",
      "  5.160e-01  1.842e+02  6.858e-01 -6.530e+00 -1.231e+03 -1.437e+00 \n",
      "  8.496e-01  6.858e-01  2.114e+01  4.043e+01 -3.830e+01 -1.012e-01 \n",
      " -1.872e+03 -6.530e+00  4.043e+01  3.010e+04  4.111e+02  7.898e+01 \n",
      "  7.736e+00 -1.231e+03 -3.830e+01  4.111e+02  1.615e+04  7.056e+01 \n",
      "  1.276e+00 -1.437e+00 -1.012e-01  7.898e+01  7.056e+01  1.161e+01 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2      3      4      5      6\n",
      "        1  0.66313   1.000  0.002  0.011 -0.652  0.004  0.023\n",
      "        2  0.71982   0.002  1.000  0.011 -0.003 -0.714 -0.031\n",
      "        3  0.11584   0.011  0.011  1.000  0.051 -0.066 -0.006\n",
      "        4  0.67147  -0.652 -0.003  0.051  1.000  0.019  0.134\n",
      "        5  0.72968   0.004 -0.714 -0.066  0.019  1.000  0.163\n",
      "        6  0.27789   0.023 -0.031 -0.006  0.134  0.163  1.000\n",
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: deactivating const optimization\n"
     ]
    }
   ],
   "source": [
    "# low mass fit\n",
    "r = w.pdf(\"dscb_pdf\").fitTo(lowMass_0tag_single_Higgs, rf.SumW2Error(ROOT.kTRUE), rf.Save())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:InputArguments -- RooAbsData::plotOn(m_yy_lowMass_0tag_tightIsolated_bkg_single_Higgs) INFO: dataset has non-integer weights, auto-selecting SumW2 errors instead of Poisson errors\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArgAAAHYCAIAAAApvgy/AAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE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\nEPT/AWs+FusZV0fnAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "c = ROOT.TCanvas()\n",
    "frame = myy.frame()\n",
    "lowMass_0tag_single_Higgs.plotOn(frame)\n",
    "w.pdf(\"dscb_pdf\").plotOn(frame)\n",
    "w.pdf(\"dscb_pdf\").paramOn(frame)\n",
    "frame.Draw()\n",
    "frame.SetTitle(\"Low Mass Single-Higgs Background - 0 Tag\")\n",
    "c.Draw()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "w.defineSet(\"params_dscb\", \"alphaHi_cb,alphaLo_cb,nHi_cb,nLo_cb\")\n",
    "w.saveSnapshot(\"lowmass_cb\", w.set(\"params_dscb\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<ROOT.RooFitResult object (\"fitresult_dscb_pdf_m_yy_highMass_0tag_tightIsolated_bkg_single_Higgs\") at 0x7fc4a069e950>"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: activating const optimization\n",
      " **********\n",
      " **   13 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   14 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 alphaHi_cb   2.32694e+00  5.14366e-02    0.00000e+00  5.00000e+00\n",
      "     2 alphaLo_cb   2.16478e+00  4.49161e-02    0.00000e+00  5.00000e+00\n",
      "     3 m0_cb        1.25013e+02  1.27139e-02    1.05000e+02  1.60000e+02\n",
      "     4 nHi_cb       4.16611e+00  5.22503e-01    0.00000e+00  1.00000e+01\n",
      "     5 nLo_cb       3.59579e+00  4.18940e-01    0.00000e+00  1.00000e+01\n",
      "     6 sigma_cb     1.96760e+00  1.00634e-02    0.00000e+00  2.00000e+01\n",
      " **********\n",
      " **   15 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   16 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   17 **SET STR           1\n",
      " **********\n",
      " NOW USING STRATEGY  1: TRY TO BALANCE SPEED AGAINST RELIABILITY\n",
      " **********\n",
      " **   18 **MIGRAD        3000           1\n",
      " **********\n",
      " FIRST CALL TO USER FUNCTION AT NEW START POINT, WITH IFLAG=4.\n",
      " START MIGRAD MINIMIZATION.  STRATEGY  1.  CONVERGENCE WHEN EDM .LT. 1.00e-03\n",
      " FCN=23.2219 FROM MIGRAD    STATUS=INITIATE       24 CALLS          25 TOTAL\n",
      "                     EDM= unknown      STRATEGY= 1      NO ERROR MATRIX       \n",
      "  EXT PARAMETER               CURRENT GUESS       STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  alphaHi_cb   2.32694e+00   5.14366e-02   2.06256e-02  -1.07820e-01\n",
      "   2  alphaLo_cb   2.16478e+00   4.49161e-02   1.81312e-02  -7.54843e-02\n",
      "   3  m0_cb        1.25013e+02   1.27139e-02   4.80475e-04   2.45283e+00\n",
      "   4  nHi_cb       4.16611e+00   5.22503e-01   1.06202e-01  -1.40019e-02\n",
      "   5  nLo_cb       3.59579e+00   4.18940e-01   8.74417e-02   5.56692e-03\n",
      "   6  sigma_cb     1.96760e+00   1.00634e-02   1.68947e-03   3.94279e+00\n",
      "                               ERR DEF= 0.5\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=23.1655 FROM MIGRAD    STATUS=CONVERGED     130 CALLS         131 TOTAL\n",
      "                     EDM=0.000131858    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  alphaHi_cb   2.45231e+00   2.06599e+00   2.07257e-03  -3.14678e-03\n",
      "   2  alphaLo_cb   2.35701e+00   1.77596e+00   1.55378e-03  -9.35495e-03\n",
      "   3  m0_cb        1.24984e+02   5.59714e-01   7.07202e-05  -2.15003e-02\n",
      "   4  nHi_cb       3.43126e+00   8.04801e+00   8.44008e-03   1.78828e-03\n",
      "   5  nLo_cb       2.15038e+00   5.58402e+00   4.85974e-03  -3.34505e-03\n",
      "   6  sigma_cb     1.82905e+00   4.19250e-01   2.37968e-04   5.70113e-02\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  6    ERR DEF=0.5\n",
      "  5.914e+00  1.655e-02 -7.524e-03 -3.300e+01 -4.699e-02 -2.266e-02 \n",
      "  1.655e-02  3.905e+00  2.511e-02 -2.172e-01 -1.426e+01 -5.184e-02 \n",
      " -7.524e-03  2.511e-02  3.133e-01  3.963e-01 -2.527e-01 -2.008e-03 \n",
      " -3.300e+01 -2.172e-01  3.963e-01  3.525e+02  2.145e+00  8.878e-01 \n",
      " -4.699e-02 -1.426e+01 -2.527e-01  2.145e+00  8.112e+01  4.869e-01 \n",
      " -2.266e-02 -5.184e-02 -2.008e-03  8.878e-01  4.869e-01  1.761e-01 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2      3      4      5      6\n",
      "        1  0.72560   1.000  0.003 -0.006 -0.723 -0.002 -0.022\n",
      "        2  0.80240   0.003  1.000  0.023 -0.006 -0.801 -0.063\n",
      "        3  0.07710  -0.006  0.023  1.000  0.038 -0.050 -0.009\n",
      "        4  0.73004  -0.723 -0.006  0.038  1.000  0.013  0.113\n",
      "        5  0.80564  -0.002 -0.801 -0.050  0.013  1.000  0.129\n",
      "        6  0.20122  -0.022 -0.063 -0.009  0.113  0.129  1.000\n",
      " **********\n",
      " **   19 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   20 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   21 **HESSE        3000\n",
      " **********\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=23.1655 FROM HESSE     STATUS=OK             40 CALLS         171 TOTAL\n",
      "                     EDM=0.00013692    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  alphaHi_cb   2.45231e+00   2.09605e+00   4.14513e-04  -1.90775e-02\n",
      "   2  alphaLo_cb   2.35701e+00   1.80714e+00   3.10757e-04  -5.72257e-02\n",
      "   3  m0_cb        1.24984e+02   5.59750e-01   1.41440e-05  -2.76828e-01\n",
      "   4  nHi_cb       3.43126e+00   8.13347e+00   1.68802e-03  -3.19138e-01\n",
      "   5  nLo_cb       2.15038e+00   5.70223e+00   9.71948e-04  -6.06413e-01\n",
      "   6  sigma_cb     1.82905e+00   4.19431e-01   4.75935e-05  -9.56354e-01\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  6    ERR DEF=0.5\n",
      "  6.181e+00  1.879e-02 -9.720e-03 -3.518e+01 -6.224e-02 -2.772e-02 \n",
      "  1.879e-02  4.084e+00  2.748e-02 -2.477e-01 -1.509e+01 -5.637e-02 \n",
      " -9.720e-03  2.748e-02  3.134e-01  4.103e-01 -2.630e-01 -2.173e-03 \n",
      " -3.518e+01 -2.477e-01  4.103e-01  3.686e+02  2.312e+00  9.225e-01 \n",
      " -6.224e-02 -1.509e+01 -2.630e-01  2.312e+00  8.484e+01  5.059e-01 \n",
      " -2.772e-02 -5.637e-02 -2.173e-03  9.225e-01  5.059e-01  1.762e-01 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2      3      4      5      6\n",
      "        1  0.73955   1.000  0.004 -0.007 -0.737 -0.003 -0.027\n",
      "        2  0.81207   0.004  1.000  0.024 -0.006 -0.811 -0.066\n",
      "        3  0.07792  -0.007  0.024  1.000  0.038 -0.051 -0.009\n",
      "        4  0.74379  -0.737 -0.006  0.038  1.000  0.013  0.114\n",
      "        5  0.81514  -0.003 -0.811 -0.051  0.013  1.000  0.131\n",
      "        6  0.20328  -0.027 -0.066 -0.009  0.114  0.131  1.000\n",
      "[#1] INFO:Fitting -- RooAbsPdf::fitTo(dscb_pdf) Calculating sum-of-weights-squared correction matrix for covariance matrix\n",
      " **********\n",
      " **   22 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   23 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   24 **HESSE        3000\n",
      " **********\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=0.0538759 FROM HESSE     STATUS=OK             54 CALLS         225 TOTAL\n",
      "                     EDM=0.000109412    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  alphaHi_cb   2.45231e+00   2.72939e+00   8.45607e-03  -1.90775e-02\n",
      "   2  alphaLo_cb   2.35701e+00   3.21881e+00   6.33944e-03  -5.72257e-02\n",
      "   3  m0_cb        1.24984e+02   1.13341e+01   2.88539e-04  -2.76828e-01\n",
      "   4  nHi_cb       3.43126e+00   6.47144e+00   3.44355e-02  -3.19138e-01\n",
      "   5  nLo_cb       2.15038e+00   8.64180e+00   1.98277e-02  -6.06413e-01\n",
      "   6  sigma_cb     1.82905e+00   1.09346e+01   9.70908e-04  -9.56354e-01\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  6    ERR DEF=0.5\n",
      "  2.177e+03  3.327e+00  1.286e+00 -1.097e+04  9.316e+00  1.178e+00 \n",
      "  3.327e+00  2.074e+03  2.101e+01 -2.145e+02 -8.527e+03 -5.270e+01 \n",
      "  1.286e+00  2.101e+01  1.372e+02  1.707e+02 -1.607e+02 -1.092e+00 \n",
      " -1.097e+04 -2.145e+02  1.707e+02  1.309e+05  1.566e+03  3.537e+02 \n",
      "  9.316e+00 -8.527e+03 -1.607e+02  1.566e+03  4.902e+04  3.729e+02 \n",
      "  1.178e+00 -5.270e+01 -1.092e+00  3.537e+02  3.729e+02  7.512e+01 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2      3      4      5      6\n",
      "        1  0.65496   1.000  0.002  0.002 -0.650  0.001  0.003\n",
      "        2  0.84629   0.002  1.000  0.039 -0.013 -0.846 -0.133\n",
      "        3  0.08787   0.002  0.039  1.000  0.040 -0.062 -0.011\n",
      "        4  0.66121  -0.650 -0.013  0.040  1.000  0.020  0.113\n",
      "        5  0.85005   0.001 -0.846 -0.062  0.020  1.000  0.194\n",
      "        6  0.24934   0.003 -0.133 -0.011  0.113  0.194  1.000\n",
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: deactivating const optimization\n"
     ]
    }
   ],
   "source": [
    "w.pdf(\"dscb_pdf\").fitTo(highMass_0tag_single_Higgs, rf.SumW2Error(ROOT.kTRUE), rf.Save())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:InputArguments -- RooAbsData::plotOn(m_yy_highMass_0tag_tightIsolated_bkg_single_Higgs) INFO: dataset has non-integer weights, auto-selecting SumW2 errors instead of Poisson errors\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArgAAAHYCAIAAAApvgy/AAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE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\nwItAAQAAeP0/N5Gm8jSfuvQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "c = ROOT.TCanvas()\n",
    "frame = myy.frame()\n",
    "highMass_0tag_single_Higgs.plotOn(frame)\n",
    "w.pdf(\"dscb_pdf\").plotOn(frame)\n",
    "w.pdf(\"dscb_pdf\").paramOn(frame)\n",
    "frame.Draw()\n",
    "frame.SetTitle(\"High Mass Single-Higgs Background - 0 Tag\")\n",
    "c.Draw()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "w.saveSnapshot(\"highmass_cb\", w.set(\"params_dscb\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "w.loadSnapshot(\"highmass_cb\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "def wrap_iter(it):\n",
    "    elem = it.Next()\n",
    "    while elem:\n",
    "        yield elem\n",
    "        elem = it.Next()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('alphaHi_cb', None),\n",
       " ('alphaLo_cb', None),\n",
       " ('nHi_cb', None),\n",
       " ('nLo_cb', None)]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# -- setting parameters to constant\n",
    "[(a.GetName(), a.setConstant(True)) for a in wrap_iter(w.set(\"params_dscb\").iterator())]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:InputArguments -- RooAbsData::plotOn(m_yy_highMass_1tag_tightIsolated_bkg_single_Higgs) INFO: dataset has non-integer weights, auto-selecting SumW2 errors instead of Poisson errors\n",
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: activating const optimization\n",
      " **********\n",
      " **    1 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **    2 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 m0_cb        1.24984e+02  2.67597e-02    1.05000e+02  1.60000e+02\n",
      "     2 sigma_cb     1.82905e+00  2.04158e-02    0.00000e+00  2.00000e+01\n",
      " **********\n",
      " **    3 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **    4 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **    5 **SET STR           1\n",
      " **********\n",
      " NOW USING STRATEGY  1: TRY TO BALANCE SPEED AGAINST RELIABILITY\n",
      " **********\n",
      " **    6 **MIGRAD        1000           1\n",
      " **********\n",
      " FIRST CALL TO USER FUNCTION AT NEW START POINT, WITH IFLAG=4.\n",
      " START MIGRAD MINIMIZATION.  STRATEGY  1.  CONVERGENCE WHEN EDM .LT. 1.00e-03\n",
      " FCN=3.45121 FROM MIGRAD    STATUS=INITIATE        8 CALLS           9 TOTAL\n",
      "                     EDM= unknown      STRATEGY= 1      NO ERROR MATRIX       \n",
      "  EXT PARAMETER               CURRENT GUESS       STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  m0_cb        1.24984e+02   2.67597e-02   1.01159e-03   6.87008e-01\n",
      "   2  sigma_cb     1.82905e+00   2.04158e-02   3.54137e-03  -5.21108e-02\n",
      "                               ERR DEF= 0.5\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=3.45044 FROM MIGRAD    STATUS=CONVERGED      23 CALLS          24 TOTAL\n",
      "                     EDM=5.77597e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  m0_cb        1.24928e+02   1.47288e+00   7.59894e-05   6.13982e-04\n",
      "   2  sigma_cb     1.83961e+00   1.08912e+00   2.57177e-04   4.00501e-03\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  2.172e+00  1.342e-03 \n",
      "  1.342e-03  1.200e+00 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.00083   1.000  0.001\n",
      "        2  0.00083   0.001  1.000\n",
      " **********\n",
      " **    7 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **    8 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **    9 **HESSE        1000\n",
      " **********\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=3.45044 FROM HESSE     STATUS=OK             10 CALLS          34 TOTAL\n",
      "                     EDM=5.77603e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  m0_cb        1.24928e+02   1.47288e+00   1.51979e-05  -2.78934e-01\n",
      "   2  sigma_cb     1.83961e+00   1.08912e+00   5.14355e-05  -9.54524e-01\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  2.172e+00  2.825e-04 \n",
      "  2.825e-04  1.200e+00 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.00017   1.000  0.000\n",
      "        2  0.00017   0.000  1.000\n",
      "[#1] INFO:Fitting -- RooAbsPdf::fitTo(dscb_pdf) Calculating sum-of-weights-squared correction matrix for covariance matrix\n",
      " **********\n",
      " **   10 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   11 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   12 **HESSE        1000\n",
      " **********\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=0.00525898 FROM HESSE     STATUS=OK             14 CALLS          48 TOTAL\n",
      "                     EDM=9.95812e-06    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  m0_cb        1.24928e+02   3.73641e+01   3.10037e-04  -2.78934e-01\n",
      "   2  sigma_cb     1.83961e+00   1.57276e+01   1.04928e-03  -9.54524e-01\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  1.395e+03  6.987e+01 \n",
      "  6.987e+01  8.328e+02 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.06483   1.000  0.065\n",
      "        2  0.06483   0.065  1.000\n",
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: deactivating const optimization\n",
      "[#1] INFO:InputArguments -- RooAbsData::plotOn(m_yy_highMass_2tag_tightIsolated_bkg_single_Higgs) INFO: dataset has non-integer weights, auto-selecting SumW2 errors instead of Poisson errors\n",
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: activating const optimization\n",
      " **********\n",
      " **   13 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   14 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 m0_cb        1.24928e+02  5.82736e-02    1.05000e+02  1.60000e+02\n",
      "     2 sigma_cb     1.83961e+00  4.16856e-02    0.00000e+00  2.00000e+01\n",
      " **********\n",
      " **   15 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   16 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   17 **SET STR           1\n",
      " **********\n",
      " NOW USING STRATEGY  1: TRY TO BALANCE SPEED AGAINST RELIABILITY\n",
      " **********\n",
      " **   18 **MIGRAD        1000           1\n",
      " **********\n",
      " FIRST CALL TO USER FUNCTION AT NEW START POINT, WITH IFLAG=4.\n",
      " START MIGRAD MINIMIZATION.  STRATEGY  1.  CONVERGENCE WHEN EDM .LT. 1.00e-03\n",
      " FCN=1.02901 FROM MIGRAD    STATUS=INITIATE        8 CALLS           9 TOTAL\n",
      "                     EDM= unknown      STRATEGY= 1      NO ERROR MATRIX       \n",
      "  EXT PARAMETER               CURRENT GUESS       STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  m0_cb        1.24928e+02   5.82736e-02   2.20424e-03   2.46933e-02\n",
      "   2  sigma_cb     1.83961e+00   4.16856e-02   7.21252e-03   1.44722e-01\n",
      "                               ERR DEF= 0.5\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=1.02782 FROM MIGRAD    STATUS=CONVERGED      24 CALLS          25 TOTAL\n",
      "                     EDM=1.68013e-06    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  m0_cb        1.24922e+02   2.47497e+00   8.42957e-05  -1.93847e-03\n",
      "   2  sigma_cb     1.74669e+00   1.75100e+00   2.96408e-04  -4.07376e-03\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  6.143e+00 -2.608e-02 \n",
      " -2.608e-02  3.170e+00 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.00591   1.000 -0.006\n",
      "        2  0.00591  -0.006  1.000\n",
      " **********\n",
      " **   19 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   20 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   21 **HESSE        1000\n",
      " **********\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=1.02782 FROM HESSE     STATUS=OK             10 CALLS          35 TOTAL\n",
      "                     EDM=1.67974e-06    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  m0_cb        1.24922e+02   2.47498e+00   1.68591e-05  -2.79158e-01\n",
      "   2  sigma_cb     1.74669e+00   1.75101e+00   5.92817e-05  -9.70788e-01\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  6.144e+00 -2.934e-02 \n",
      " -2.934e-02  3.170e+00 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.00665   1.000 -0.007\n",
      "        2  0.00665  -0.007  1.000\n",
      "[#1] INFO:Fitting -- RooAbsPdf::fitTo(dscb_pdf) Calculating sum-of-weights-squared correction matrix for covariance matrix\n",
      " **********\n",
      " **   22 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   23 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   24 **HESSE        1000\n",
      " **********\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=0.00111068 FROM HESSE     STATUS=OK             14 CALLS          49 TOTAL\n",
      "                     EDM=6.96603e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  m0_cb        1.24922e+02   3.07090e+01   3.43927e-04  -2.79158e-01\n",
      "   2  sigma_cb     1.74669e+00   1.89473e+01   1.20935e-03  -9.70788e-01\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  5.558e+03 -1.775e+01 \n",
      " -1.775e+01  3.035e+03 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.00432   1.000 -0.004\n",
      "        2  0.00432  -0.004  1.000\n",
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: deactivating const optimization\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArgAAAHYCAIAAAApvgy/AAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE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\ngQIAAAgiUAAAAEEECgAAIIhAAQAABP0/W+6ZwLQf3I4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArgAAAHYCAIAAAApvgy/AAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# -- high mass (keep a1 and a2 values we just found)\n",
    "c = ROOT.TCanvas()\n",
    "frame = myy.frame()\n",
    "highMass_1tag_single_Higgs.plotOn(frame)\n",
    "r = w.pdf(\"dscb_pdf\").fitTo(highMass_1tag_single_Higgs, rf.SumW2Error(ROOT.kTRUE), rf.Save())\n",
    "w.pdf(\"dscb_pdf\").plotOn(frame)\n",
    "w.pdf(\"dscb_pdf\").paramOn(frame)#,ROOT.RooFit.Layout(0.65,0.85,0.85),ROOT.RooFit.Parameters(ROOT.RooArgSet(w.var(\"a1\"),w.var(\"nbkg1\"))))\n",
    "frame.Draw()\n",
    "frame.SetTitle(\"High Mass Single-Higgs Background - 1 Tag\")\n",
    "c.Draw()\n",
    "c1 = ROOT.TCanvas()\n",
    "frame = myy.frame()\n",
    "highMass_2tag_single_Higgs.plotOn(frame)\n",
    "r = w.pdf(\"dscb_pdf\").fitTo(highMass_2tag_single_Higgs, rf.SumW2Error(ROOT.kTRUE), rf.Save())\n",
    "w.pdf(\"dscb_pdf\").plotOn(frame)\n",
    "w.pdf(\"dscb_pdf\").paramOn(frame)#,ROOT.RooFit.Layout(0.65,0.85,0.85),ROOT.RooFit.Parameters(ROOT.RooArgSet(w.var(\"a1\"),w.var(\"nbkg1\"))))\n",
    "frame.Draw()\n",
    "frame.SetTitle(\"High Mass Single-Higgs Background - 2 Tag\")\n",
    "c1.Draw()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:InputArguments -- RooAbsData::plotOn(m_yy_lowMass_1tag_tightIsolated_bkg_single_Higgs) INFO: dataset has non-integer weights, auto-selecting SumW2 errors instead of Poisson errors\n",
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: activating const optimization\n",
      " **********\n",
      " **   25 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   26 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 m0_cb        1.24922e+02  8.24031e-02    1.05000e+02  1.60000e+02\n",
      "     2 sigma_cb     1.74669e+00  5.75389e-02    0.00000e+00  2.00000e+01\n",
      " **********\n",
      " **   27 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   28 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   29 **SET STR           1\n",
      " **********\n",
      " NOW USING STRATEGY  1: TRY TO BALANCE SPEED AGAINST RELIABILITY\n",
      " **********\n",
      " **   30 **MIGRAD        1000           1\n",
      " **********\n",
      " FIRST CALL TO USER FUNCTION AT NEW START POINT, WITH IFLAG=4.\n",
      " START MIGRAD MINIMIZATION.  STRATEGY  1.  CONVERGENCE WHEN EDM .LT. 1.00e-03\n",
      " FCN=11.5041 FROM MIGRAD    STATUS=INITIATE        8 CALLS           9 TOTAL\n",
      "                     EDM= unknown      STRATEGY= 1      NO ERROR MATRIX       \n",
      "  EXT PARAMETER               CURRENT GUESS       STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  m0_cb        1.24922e+02   8.24031e-02   3.11715e-03  -2.91032e+00\n",
      "   2  sigma_cb     1.74669e+00   5.75389e-02   1.01915e-02  -3.84705e+00\n",
      "                               ERR DEF= 0.5\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=11.4366 FROM MIGRAD    STATUS=CONVERGED      33 CALLS          34 TOTAL\n",
      "                     EDM=1.1305e-08    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  m0_cb        1.24986e+02   8.66085e-01   7.78301e-05   3.24329e-03\n",
      "   2  sigma_cb     1.95499e+00   6.38948e-01   2.55505e-04   5.60369e-05\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  7.504e-01 -4.644e-03 \n",
      " -4.644e-03  4.098e-01 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.00837   1.000 -0.008\n",
      "        2  0.00837  -0.008  1.000\n",
      " **********\n",
      " **   31 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   32 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   33 **HESSE        1000\n",
      " **********\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=11.4366 FROM HESSE     STATUS=OK             10 CALLS          44 TOTAL\n",
      "                     EDM=1.13041e-08    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  m0_cb        1.24986e+02   8.66089e-01   1.55660e-05  -2.76759e-01\n",
      "   2  sigma_cb     1.95499e+00   6.38951e-01   1.02202e-05  -9.34836e-01\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  7.504e-01 -4.997e-03 \n",
      " -4.997e-03  4.098e-01 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.00901   1.000 -0.009\n",
      "        2  0.00901  -0.009  1.000\n",
      "[#1] INFO:Fitting -- RooAbsPdf::fitTo(dscb_pdf) Calculating sum-of-weights-squared correction matrix for covariance matrix\n",
      " **********\n",
      " **   34 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   35 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   36 **HESSE        1000\n",
      " **********\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=0.0221447 FROM HESSE     STATUS=OK             14 CALLS          58 TOTAL\n",
      "                     EDM=1.15288e-06    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  m0_cb        1.24986e+02   1.79264e+01   3.17547e-04  -2.76759e-01\n",
      "   2  sigma_cb     1.95499e+00   1.52599e+01   2.08492e-04  -9.34836e-01\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  3.880e+02 -4.198e-01 \n",
      " -4.198e-01  2.129e+02 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.00146   1.000 -0.001\n",
      "        2  0.00146  -0.001  1.000\n",
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: deactivating const optimization\n",
      "[#1] INFO:InputArguments -- RooAbsData::plotOn(m_yy_lowMass_2tag_tightIsolated_bkg_single_Higgs) INFO: dataset has non-integer weights, auto-selecting SumW2 errors instead of Poisson errors\n",
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: activating const optimization\n",
      " **********\n",
      " **   37 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   38 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 m0_cb        1.24986e+02  3.80981e-02    1.05000e+02  1.60000e+02\n",
      "     2 sigma_cb     1.95499e+00  2.80870e-02    0.00000e+00  2.00000e+01\n",
      " **********\n",
      " **   39 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   40 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   41 **SET STR           1\n",
      " **********\n",
      " NOW USING STRATEGY  1: TRY TO BALANCE SPEED AGAINST RELIABILITY\n",
      " **********\n",
      " **   42 **MIGRAD        1000           1\n",
      " **********\n",
      " FIRST CALL TO USER FUNCTION AT NEW START POINT, WITH IFLAG=4.\n",
      " START MIGRAD MINIMIZATION.  STRATEGY  1.  CONVERGENCE WHEN EDM .LT. 1.00e-03\n",
      " FCN=4.36477 FROM MIGRAD    STATUS=INITIATE        8 CALLS           9 TOTAL\n",
      "                     EDM= unknown      STRATEGY= 1      NO ERROR MATRIX       \n",
      "  EXT PARAMETER               CURRENT GUESS       STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  m0_cb        1.24986e+02   3.80981e-02   1.44019e-03   8.90843e-01\n",
      "   2  sigma_cb     1.95499e+00   2.80870e-02   4.72895e-03   5.52371e-01\n",
      "                               ERR DEF= 0.5\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=4.35903 FROM MIGRAD    STATUS=CONVERGED      29 CALLS          30 TOTAL\n",
      "                     EDM=7.98467e-10    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  m0_cb        1.24924e+02   1.30446e+00   7.47802e-05  -4.16509e-04\n",
      "   2  sigma_cb     1.85925e+00   9.45952e-01   2.46868e-04  -1.19990e-04\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  1.703e+00 -1.487e-02 \n",
      " -1.487e-02  9.029e-01 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.01199   1.000 -0.012\n",
      "        2  0.01199  -0.012  1.000\n",
      " **********\n",
      " **   43 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   44 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   45 **HESSE        1000\n",
      " **********\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=4.35903 FROM HESSE     STATUS=OK             10 CALLS          40 TOTAL\n",
      "                     EDM=7.96935e-10    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  m0_cb        1.24924e+02   1.30447e+00   1.49560e-05  -2.79093e-01\n",
      "   2  sigma_cb     1.85925e+00   9.45959e-01   4.93736e-05  -9.51135e-01\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  1.703e+00 -1.565e-02 \n",
      " -1.565e-02  9.029e-01 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.01262   1.000 -0.013\n",
      "        2  0.01262  -0.013  1.000\n",
      "[#1] INFO:Fitting -- RooAbsPdf::fitTo(dscb_pdf) Calculating sum-of-weights-squared correction matrix for covariance matrix\n",
      " **********\n",
      " **   46 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   47 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   48 **HESSE        1000\n",
      " **********\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=0.0065094 FROM HESSE     STATUS=OK             14 CALLS          54 TOTAL\n",
      "                     EDM=7.51893e-05    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  m0_cb        1.24924e+02   3.74515e+01   3.05103e-04  -2.79093e-01\n",
      "   2  sigma_cb     1.85925e+00   1.74954e+01   1.00722e-03  -9.51135e-01\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  2    ERR DEF=0.5\n",
      "  1.096e+03  1.090e+02 \n",
      "  1.090e+02  7.067e+02 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2\n",
      "        1  0.12388   1.000  0.124\n",
      "        2  0.12388   0.124  1.000\n",
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: deactivating const optimization\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArgAAAHYCAIAAAApvgy/AAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE\nQVR4nO3dT5Pbxpn48UYqV9dId8mll7B3ywYo78H+7Zb3ktcgj0Y5S6NJxQCdSiz7tBdbKr+CPa5U\n8soXmYCt3PMKnJQ9d41eAX6HJ9Pp6X8ESQAEmt9PzWEIgkB3A0Q/7G40srZtFQAAgM9v9p0AAAAw\nXQQKAAAgiEABAAAEESgAAIAgAgUAABBEoLAfdV1nWVYUxb4T8i9FUWRZlmVZXdeR1bJLY6Wrq6qq\nsquqqrLWGbTYJQHx0tt0a24WNDleeneStcj645MkWYqiKIpiuHROsByG8/7772dZ9t133w2x8S5n\nsvcQew/6ECnEaAgU9qlpmn0nwSNykZ3s9TfLsuVyaS1cLpfemnugYpcd9RUorN2alYt+994Lb2Ka\npmmaJnRo+trppMphOK9fvx5oy0VRLBaLtat1LOdpXujQHYECbJFv9TSvv7p5Y7VatYY8z5VS5vWu\nKIo8z8uy3E9ChyRZm+ZPN/OgrFYrXf5dqiKMr67rjlV7VVXtVbK8LEvvcszUb/edAExRVVXexoMJ\n/jLQsYt7MZJ2UflH16DTjHV2VxTFLLKmux7cQ4O9q+u6rmu3cW5THNPE0KIwG/IdHnov8mvPuyMJ\nHSI/x+u6rqpqbTr1apEVtLUJjq8Tyc4WCVPbHoUuWx5Ox71///3333//fXyd58+fP3z48Pnz5z2m\nzd1Fx73Imjsm4LvvvuvSx+897s8v7ZgGcy/xI/Xdd989evRooEEJ0t2we5TQRcdD3PHoYHAt9mG1\nWnUvf2lCN5lt7O52dF1urqYXxvcSSVhkBW/0YHUEeDNiNVG628nzPF44a/NlchOvX8YT5qZNd2Ho\nFErurA/qPXbPkbm1yMpWIcuOrPXXplm8//775mpnZ2ey5OzsTK/z4MEDKyPvvfdePAvxQ+OeJM+e\nPVMO714i63gz6D2mt2/fNjdyenoqS168eGF+ytxdJKlWOmUFN/HvvfeeUurBgwdmas3uGHdf4sWL\nF9YKp6en8o9O8I4kGSJy4OLcw+pdYe0hDh2d09PTLVKFHREo7Ef3QEF/VeQLrOta/VV0L4J6HXeh\nW/9ZK+hrVqiXsXVCE50XqYe8ieySEXc7+mWkfMyaOHJ5slZ2UyUJW61W3sTrhZI2M+KJBApmFsxP\ndYkVdg8UuqS5NaKEs7MzHSLoJbKOjhIeXNIvI1mI1Dfet/R+ZRdSp7p7sVbTL80tmxnUB9e7kdPT\nU13jCitQcI+ajhLee+89bxraDQMFfaytiEF/yowSIgnuUejAdfxg6JtoHTtddFZBRTJLoLAXBAr7\n0TFQ8F5PrYWhK6N3YaQq1YFC5LepXG2tJHlDEGuhN79WXejdzkalpHMdiocigcKmiXfjGDf9bpl3\nDxDNaM/L2rh71Ny9u2l++fKlmx4zdDA39ezZM72OriwjWdC7M5Ntxive5gRzC+5CqWi99Yokz/o6\n6ASY60vdc/v2bXcjygkUlPOtkYXe8EUv3ChQcFPoPUvXJrhHHc/S0Ae915mOh1hHBuZquoGBQGEv\nCBT2o2OFEfrWmcutTekKw3utiexLBwrelSPvSgVgbc1qlgjlV+KSSGbNFSLMlgBTGegI0Eu8Oy07\n9Cnoz4ZWC30q1GBj8WbHFQoUQg0SVprlpdnF4F3uPXbPnj0zQwdX6TSnm6y0PXv27MGDB+4GvcfL\nWkd+mErtax477zmgl1v1q/7VbgUK1sf1j2BvGvTyTQMFbwrl4OqEWeu4Ce6Rd4/dPxgKFLofYjdT\nspxAYS8YzDgD7hBiqQZk0JM1nl8vNNeRcYgd6x51OQbQvPFB7nfwDmZ2b5FwB07rD1qD82UMvJmp\nxWJhbs1cIUI22169+05dTqXQ5eORl6G8x8sz9ClZoguhvsrdxSogniPZu3vrijfNf/7zn+NbE1mW\nPXz4UL/85JNPPvnkky4ftFIu7QpN05hTKXzyySdffvmltUGro1rGvukuCe3LL79s2/bLL7+0Uiv/\nWIdAD477j//4D3O59VLznn7uiA1r7xuJn0gyWYJVFCqcYFP87BqZ9xCbZ5SpS+4wGgKF6Yp8sa3K\nxowJdGRgrmOFFGvJmrqyX3u/g96jTMTmvT9ePt40zWKxkMna3PBCMiK1u6yz6dVNNitRsE5wLzdr\nhar8OMmsSQpH32i6uMqt2ouA4dKslLKGN+r24a+++kpysdG9D1bKZWC/jgvNNZ8/f3779m1dVn/9\n61/Nd62XEeZ9vDvO1mCVWPc0bL0LSyhQ6MI6u6Zw16J1iL/66ivz3cg9DtuVAHrBPAopqKpqsVhI\nnaovkUVRLJdLc2H3eRWtC0o8zqiqSocUcvWXmsBqVKgu1XXdXFoul6vVymwUkTvEJMESVeR5Hr+X\nMpQ2iV0Wi8Uep39Y24pjhV9jXsrX3gypffLJJ23bPn/+/KuvvpKaUq7vDx482PqXtJy0+uXz58//\n67/+S79875K5cCNlWerZGtxGr+FqnefPn3dsaBnBHs8uV++HGOPZY7fHIdtojIK7PDJuTjk90B33\nZY5CaKOjF82XeuNWV+7annhzVEFondBQL5POY5cVQmMU3LS5xehuOT4Yc23C4na86yG0d2uzoazJ\ncnfsgnj27Jlu/48MU4jc9eDuXf53O/XNdUId/+5OdQbdNIS6/Fund9y7mjkeYoukescohEZRyOHz\nDr30JrhHXa4YkQ96T3s9ENU6Z7p8JVvGKOwVXQ8z4P6edvu/9U95c7m5MN5x4JJPLZfL+MclbXme\nu/0I1kt3dILZT698wxGqqirXzZhkdrtE7PJbyixGU7yhIpQwKYqhn5oR2rs3zWubFm7fvm2NTnj9\n+rVc97duivceMuvhBd7eDXePDx8+zLLM20igy1mfAKHO740m9rEazENJdW1RXJIv97EOs5uJSPL+\n4MGDLi0us8td2ggUJs0dVKh81z79v9XFYC7cuqaUHoT4x93qx31kkfQjeD+u09k0jRthxPeuAxpv\nxSMf7D6KM74Ld+Gmn1JKSVfI0I3A3r27RXR2dqaU+vjjj82Ff/jDH6zV/vrXv7pV4y7qupaTIX5o\nrJ3qCsYaASeruYMchbS1mKeWVL3WOIzHjx93SbnuarEiA2k/twY5WmFBaOBenI5sHj16ZC7vmOCJ\nc8M7aUGxcmcdLIxt300aB2rtqHXrpkHlTAcUaq40j6m5l7VJsroe2qu3t7k7sl5K8qybDtzb+mWS\nA3M1q+ne3JReEm/ANysb7/36boGE8mKu5p2awjoKKno/pJUjN8trM7V110Nk76GsnZ2dvXz50jvh\nkl4it7eZXQ+RLHRpxNIrm30ZwlxNN1abkx15U2J1PVgpsXJ9+/ZtmdLHrK7iXQ/t1emn5JY/Nzut\n0Zki6TRX26jroTW6S3SCzcKZV9eDeYjN2M7sj9ALT09PX7x44T06GBOBwn50DxRa36Sn3u+hvGVd\nIr0LvdxAoTWCAHehfunWB1Iz6ZeRLFsJ8/6+7NLNH6qTrO1vHSi4acsvp0OOT5zgJqzLsWj7CBS6\npNnclKbnZ3z58mVoHbHLPAqRSFczF7qxgrWmudPQBBJ6p+6MyKF5FLxZ6zKhtZtOPRfhpoGCN8Hu\nnNM9iuS9ywdD31n3wJnhgjnyY8zMYq2sDVwCMCn6Nujut8aNTA+PsG5hUFdb6c37ub2t9+YKm2bW\nulm893EA1lEoiqJpGhldv9EH+03VRruWW1TcNH///fc//fSTUurPf/5zlmXvv//+Tz/99PLlS7NX\n4tmzZ7otfaDB6noX5valcrW6IbxrburFixfS979dM36XNOiQope+mx0TPAXdD7E3sy9evPjP//zP\n8ZJ7kNyogEABWENu6dRTU2hy6900v0GS4FCa9S2pf/jDH3766aezs7OPPvrIWkdNNWu7k/YSCYxM\nUz6gh+PRo0evX78+PT21hp1ydMaRZZ6ogMGMwHruSMyhb1vYXSTNZvTw008//eUvfzFXk3o07eFj\nr1+/tjIoL63uf4zv9evXr1+/9g5mZM6lfaFFAVhPehnU5Sh9fU+HOVvU1HRMs24/sH5nv3z50mxm\nSIzOtXXzIdfDKeDo7JG3RYHBjEAn1tA8eQrDvhO1Rsc0y02S2vvvvz96SvfA+nnKTD6TYjXteOeb\nwhC8UQEtCgCu8P+kAHAAvF9/nvUAICnmXTP7TAeQCgYzAtgPeWiT1tdzkEPP4RyaZMfNRXH5VFV5\nIGrks5EV3E15HzQa2dFApb3d9uMrb/Tu9IcVp2Dk/g8AEzfOZcE7HVMvwz5kUx0ntuqLnsvLO2XZ\n2quufmvtFFsm7wyb3h15S7tj1mSq07XrdD+a8ZXjs655P7s2eejOf36Onw4AU9a9CtlxL7q60rVs\nL7V7j5vqwqrYvJOgS00Wyqa5hVCy9aZk+9ZLK0xxn+lqbjz0xNeQLoHCRkczvrI3qd53W6PoumQE\nXRAoAFhvu8uunsE6/hQPcy/KmQB7o117n7jRGnWJmZKBblFxJyY3d+Rmyl2i596OV67ulN7m+u7E\n1eb6HR80H9I9UOh4NCMruw0z1qa8pUeg0CMCBQDrbXfZVd1Yv62tZ1V037W7ZetHp2u4WEG4e5GH\nk3mfE2ZlpF3XEGJVrm2gfvWub64ptf5GRbE2UNjoaEZW1s0h7pNurIOr0+M+7gs78hYm5Qvgiu0u\nu2YoYP3Otp5d2fqu76E+fi/rwZjWU0bju3aVAZsGFl3S763n5GU8UAi1KOhMRR4/5u31j3fNSIgj\n9PrCDRo2OprdVzaT7a5sHeVIXrAR79eE2yMB9CbPcxmFvlqtZALpsizleVR1XevZIXck29GPtqrr\nWubyq+vaHO0vVz1Zslwu3RWEvOWynjG2o7qu9YzaeqC+7LrLuH2ZZ7NpGl2Y7va7JCPPcym9pmnk\naSCh3YWWjHbTaV9nC3ZHoABgQMPVK25MYDJ/jxZFEYoGrDVD29+RPLRT/tdJle2vfeK8tQUJF9x3\nzeBJghKJKswAQu9a4qrlchkJFHT2ZZ3x70JcrVZ1XUuu9SxAEmzleS5hnGR8sVi4JwB6xDwKAMa2\nSx28xa9889Hn3g169VU16jBFWvv1cqnaF4uFzAegF+r/Le3V2yh0Z4S6bJnQzTlFUVgP+LB4bzLc\n2kZHM34srDWrqtJJNY+IfnJ6/Jn16AuBAoC90Rf67tMpemsaaUU3l5h1ZHzjoUChax7WpVb/1t99\nm1VVSZ+xbkIIFZe5fLTOgi2OprWy9CKZoRLTa07CKMMjAMzGdpcFuZ7oIXId73ML3R/fZV/KGcdn\nDWa0ZhEI5St0bdx0JgYrv+3VkXfuMMnVVXqn3qkRrNKzsmwN57SyHJploWMGu4zrjB9NayRmZGV3\njgfvzR3uu4xn7Iv3a0KgAOCKUIW69lNqk0DB2/rd8XIfajk3U+IK3eMXWn/3QCE0+sF/LQ4UoHVD\nqTeFoYEObuHH07C1+NG09hhfOX6Phjebmx4pRHhPDAIFAFdsV4WEqq5QoNA6VULHiQKFW9m4Kdl6\n49ux8ttGh9eFPh4KFFrfQ8PNj7uxgpVlq0C63MbpFb+B07t3N9fxo2O9a+0x/i525D05eZ4sgCvG\nfMy0jAYwx9hv+nH5J/RxPeptm8RNUrzE1r4r/wxUIBsdzbUDRyLvqhSP7ER4v/4ECgCuGDNQADAp\n3q8/dz0AmBDrIcIuboQDRsaESwAmxDvtoLXCSEkBoJQaouvBugV5697H7YTmKgHQHV0PwGEafIyC\nOU2pRc/KPjS6V4Ed8SUCDtaAYxSkW3G5XLpTc8jT25bLJZ2LAADMTg+BgvQ1SHxQVZXV0SDzdUvE\n0OPEqAAAYASptTHSagrsiC8RcLC4PRIAAGyGQAEAAAQRKAAAgCACBQAAENRPoGDdziDPBcmybO0k\nawAAYMr6Gd6cZVme5xITuNMujTbbkmLANrAzvkTAwRpwZkYzUDD/V0oVRdE0zWjXncgUzlz7gC6G\nCBSGfsDxcOabcmALY9weqRsV9BL5f8wOiDZgtAQAsFRVtVgsFovFvhOyMUn2+LPKynS3a6+c0snr\nfa6m9zmcep34uz1mQYvnJb6y9W48p0wB3LueAwWCbgDYUV3X0oEbr1yzLGuaRr9cLpfmFTj+2V1+\nvFVVtbYydvugF4tFaKfxld13l8ulGfFE3kUveut6kH/yPJcTVzZb17X8hhiz64HGA2AXfIlMcnEz\nu1MHJX21+mVkgJeuIFerlcyUb75Ulykvy9Lafpd34yRJ8cpYVwpt2+qKIFSM8ZWtQ2Cu7L6rC5Bz\neDv+r3+ooX4jZVnmeZ7nubVZ+d96TNSg+soRcLC2/hJZtY75xZeLQ57n5vr6iiGPjjNXkLfKstTb\ntN6SdyN7t95dy7x8mZ/VezdTMtA1bbVaWdfnSC7cIrU+Er/C73L9l+MSX8dKjC7eTVfWZaLL3Fpi\nbVa/u13W4C26YUtz0+/q7jg/MBCl/vWXtu2+RFY9vbaecFc2Vwi9G9p+fO9dsmzRFXBo18PFCmJt\nFtwVrNBBryC/4qxNxd+NWxsoRGr3jVaWopAUhtYPnQndswOTt+hSK03ODwzBjBKSjxW2+xLp+lVf\n3M3rdeilPJg+1BipV7BqaDMssNY3a9mOGQklxqyHIru2lGFblOcWLQrq6g9xk145/m4oX/klvX4o\nyHBrazca2G7l1jmXvHkZsxk7Md4T+7duEfdLOpBauoswW+FbbmGT7mGZaS3Shy2r5Xmu74oK3dgs\nK6xWK+m3LstS+tHrurb68pXRxa4HS3VPs5sYmThOryYXMVmiRxq6PfrWwDrNDIZ6IZ3xTdPoAjHf\n1S/zPNfF1TSNjEOMvxvaXWjJmGPY9RAEbxsSBjJ4oLDRNxaYiyxTRL9eeiSajLCLrGm+671QuJVr\nqE6qLm19tTET4/6wMVNSFEUoGlDhgKD32lSPXpQ63t2d7FHvV0c/a9/1Msc5dhnMOAQdTZpjPM2R\njxKeNk2zWCz4ddqncVs1BpdejrBfVnfDIfQ+bP0l8v7Ik7fiw9Nap/1Z/nebyvVHQn0Z6rJJvOP1\nLd7K7abEXDjoGKyOu9DlJl0DblK1eJl0L7F2kzEK7hK3nNcOV2yv9i9EThsxwtFJmPcc6LNFQceq\naxcCM9W2/+qJoFHBopu19TQAytc+r1+ab+1y86HuhtA/NLs3LXgTIxsxf0ObW4vP1RjKyEDXQLOz\nQH5wy47cLFifirzbI12qG01waa289u5KDK6vMCR0c5Es72sva425LxwCt/0g+UaFLb5Ea38UWj9Y\nreuPO/xQ/u/YouC+u9H1zVpZ/0i1BjNatx2GNh660saHCoa2Y15L40VqJTs04k+SEX93LRn12SX9\n1i0Y1j2uoXe9L9Xl7bLm4FD39hBrLCo25T2xe5tHQV0dM2x9n3vZSxdj7gvJC8UEBAreT+nru1lB\nyruhWs1lbq1710N8a3HeYfNrtxxq2Q6t32OgYNWR3r3EMxV/txfeQxyK5CIrh5KqfLeluOWATXlP\ng37ODOvAmMHB+IFCyGhpQDLWBgpJnlbbfVnid6m5XeBWPOGt+LsHCqUz25L+p0vi3YrKLA11ddxD\n983uwt2R++vZSrZVO7qZ8vbue98Npcdr7agINy9uOYdWjsSU7skQTw+6UL6vf/9Pj9RLpL9w/KdH\njrYvJE8PR7DOKfNWvvROt12+ROZFoHtvtHSW7zjz7u6PeVy7hWmOuJJBIaEBBx1HVAyUqXja+k3M\nNI/O7Az4mGk3GpBbd1arlYwqIlDAHIUChfhbczfCl8iarp/RasBEDBgo6O+5Mn4QmDOoEChgduLN\nBgQKu3Cf+Cd63695afIqw09dAg6Q9+vfz+2RRVG0bWtO+KWUkiWRmUkAHCbplDRvYhyuwo5PiUhL\nNbBWar+/aVFAX+JtBgkPU+BLBBws79f/N3tJCgAAmAUCBcBj7YOg+MkN4EBMK1CoqirLsizL3Ieh\nadlVdDECADCcwZ8e2Z3cYykjj+TxX6vVKhQH6AFKBAoYVJeWg/Qe+hB66DOAA9TDqCV5smeXu59l\nnHOoardmbXIncVKXNztFAgjGYaEXXe5+THg8I4CtpVcN9dD1IHV2pL9AZuYyn2zmXUc5Dzh3n/+m\n595iVhbsXVqXAgDw66frQebprKoqNLdJnueRZgB1ORODuUFlTNhu2ctUTgAAHKD+W0jqS+pyCvdN\nhxHoaZrctMk4hrIspVHhf//3f9++ffv3v/9dr7Bd3yrRBkzd+xQSnp8RwHbS63qYVn42nfJdP1FC\nxyLpHSGMj0ABwNbSq4YmdHukjhJWq1XHIQjc8oBBdf+yc5cAgFRNKFBYLBbyHPFI9W9NnMBzRQEA\nGNRUAgV9v0NhkIV1XWdZJv/ned40jcQHdV0vl8v4E1+AQaXVvggAHr3d9bDjz3qp+937IfVb5o70\nvRU8vR4AgEH1M+ZC7jXI8zwyn1K/QqFJeqNIML6NhigynhGAKb1qqJ+uh7Zt5daDxWIRf1JDXxiX\nAADACAYJfGRGZ3lwgx5qMI70QjmMbNOJmWlRAGBKrxoaZDCjBApy/4IMRWQkAQAAc5Ra4JNeKIeR\nbdpCwKOhAJjSq4amcnskAACYIAIFYCdp/XIAAFs/8yhMSui5UIm1BWEIzMQMAJZhWxTKsvzhhx8G\n3YWrDRg5GQAAJKDnQKEsyyzLJDjIsuzzzz//8MMPy7Lsdy/A0AgsAUD0HCh8/vnn8o8EB69evcrz\nXC8E0kbPBYD09BkoSEPCq1ev7ty5I09tuHPnjsygMH4HBAAA2N1QYxSapvnss88UIQIOAP0UABLW\n87wQ5h0Hr169Ukp9+OGHasQ7DtKb6QJj2no+ZiZyBiDSq4Z6blGQ4EAp9dlnn925c0dGJ+iFAABg\nXnqeR+E3v/mNGUnJAAUe9AAAwEz13/XgbjDLMnkIdY872igBQBe7PLWBJz4AEOlVQ/10PcgjImWA\nQuboZRcAAGB8vXU95HmulGqaRv4xVVU1TnMC0Istfgy0LZMoAEhTzy0kRVHsd0RCem0+GM2Ody5w\n4wMAlWI11H9+vIHCaC0KkZ6OxI4cekegAGB36QUKPd/1UBSFzMloGbPUEjtCAADsUZ+BQl3XMkah\nqqoeNwvMS5bRqAAgHT23KChmTQAAICF9zswoAxEIFHCYaEUAkKSeWxTKslwsFmVZWqMXuT0SAIA5\nGvChUCYeCoWJ62VqRW58AJBeNdRzi8Jqtep3g8DQLqPbpL7YANCXoQIfiRgWi8UQG49IL5TDOHpp\nDKBFAUB61VDPj5lWShVFkWXZnTt3mqaR/3vfBTBxPOUEQDJ6DhSyLGuapixLeSn/cMXEoWnbNrGf\nFAAOVp+BgnQ3/PDDD3rCpcVi8cMPPyjGLgAAME/9dz1Y4xLGH6YAdHd+fn52dnb9+nVzye6bpREN\nQDL6DBQkJrDmb5aXY4YLWcBoCcBcnJ+ff/rpp2/evLm4eKMX3r17d7tYwfrUo0ePeok5AGC/em5R\nKMtyuVxKrSz/LJdLPWRhHG3AmGnAdERixG+++ebdd999+vSpufDWrVtff/31pluTmOP4+J5e8vbt\n261jDgCYjv7v4litVnfu3NEvf/jhh5GbE4gJYIqcEtevX7+4uFBKGZMoZEqpo6Ojy+Vdt3Z2dvbm\nzZunT5+am7p3797R0dEXX3yxacIAzFd6X+2e81PX9X5na07vCGFH3XqdrgQKa1b1nWA9xhwAZi29\nr3bPXQ+LxSLLMh4zjUkJ9UZdu3bNu/7R0VGkA8s7AiYUDbx9+5ZBMwBmrf8xCnmey+iEoih4kiSm\n7Pj4+N69e9bCk5OT4+PjyKd6jDkAYPqGaiGpqmq5XMr/eZ6PFjGk1+aD4Zyfn9+9e/fWrVtPnnwj\nS05O7v/888/ffvvtzZs3vR8JnWCPHj16+/btkydPzK6Hk5OTd9555/HjxxttCsCspffVHjA/dV1X\nVdU0jeLpkZiq8/Pzmzdv6JcPH57ev38/FCXEt7NpzAEgSelVQ/1PuCTxQZZli8WiaZo8z5mWEZN1\n48YN8+Xjx4+3q9dv3Ljx7bffHh0d6SXvvPMOUQKABPQc+OghWnmeF0Ux/qjG9EI5DK3fRz6agxQ5\nE4EDlF419Nvet1iWJXc9AACQhp4DBTeMqutaOiP63RHQO2kP2/GnQNvyoAcASelhjEJd11mW6fsa\nZICC+a6+/QGYst3vWmSCBADp6b/rYe9CV+rEOo0wQZczMu07HQDQnwQDBQICdEelDgBx/d8eCQAA\nkkGgACjFrYwAEECgAAyFfg0ACehtjMJisTBfMvYbAIAE9BAoFEWR5/nu2wHSwFQKAFKS2kyT6c2d\niUH1O3/z0JsFMH3pVUOzHKMgczplWVYUxWgPsAYA4ADNL/ApikIeSqmUkmdYr1aroijk3fRCOQyK\nFgUA/UqvGppffrIsy/NcNyS4L2eXI+wRgQKAfqVXDc2s60ECAusRU9KuAAAAejezKZyLojAjNYkb\nyrLcW4IwZ9ybAABrzbiFpKoqeS6lmYXt5m+YbyFgF+bJQtcDgF6k1/UwsxYFUde1zO9kjk7QEjtC\nGAGnDACEzGyMgjKihNVqxb2RAAAMan4tJNZtDu67s8sR9mW4DoLhOjUATFx61dDMuh70/Q564gT5\n37oPAgAA9GJmgYI0JHA/JCaOxz0ASEZqLSTptflgOIPem8CND8BhSq8amt9gRgAAMBoCBQAAEESg\nAAAAgggUAABAEIECAAAIIlAAAABBM5tHoYvQc6ESu18FO2KeAwDoIsFAgYAAAIC+0PWAQzd0YEnT\nBYBZI1AAAABBBArAIOgBA5AGAgUAABBEoAAAAIIIFAAAQBCBAgAACCJQAAAAQQQKwOCYSgHAfBEo\n4BBRcwNARwQKwFCYSgFAAhJ81gMPhUJ3nBQAEJdgoEBAAABAX+h6AAAAQQQKAAAgiEABAAAEESgA\nY+CGTAAzRaAAAACCCBSAAXELDoC5I1AAAABBBAo4OAwXAIDuCBQAAEAQgTVR0L8AABSVSURBVAIO\nFwMIAGAtAgUAABCU4LMeeCgUpinLaMMAMD8JBgoEBAAA9IWuB2BYBK4AZo1AAQAABBEoAACAIAIF\nAAAQRKCAw8K0jACwEQIFAAAQRKCAA8XNCADQBYECAAAIIlAAxsMICQCzQ6AAAACCCBSAwTEeAsB8\nJfisBx4KBQBAXxIMFAgIEMIQAQDYFF0PAAAgiEABh4hWJwDoiEABAAAEESgAAIAgAgUAABBEoAAA\nAIKmGChkWVbXdeRdU1EU46UM2Bm3aAKYl8nNo1BVVZfV8jyXfwgU0BE1NABsYUKBQlVVy+Uyvo60\nNKxWK+IDzEvbEqkAmKUJdT0URVGWpW4q8JJAoSiKSN8EEMckCgDQ3YRaFIqikAhgsVjE1zSf5sCE\nzQAADGdCLQpdSENCWZar1aosS6VUlmW/+93vzHWyze0lLwAATN+EWhS6MHscZJjCcrn8/e9/b65D\nGwMAAH2ZWYuChSGNAAAMamaBgjVxgh7buKfkAACQuBkECnVdZ1km8yvked40jcQHdV0vl8v4XRKA\nYCAKAGxnBmMUzHEJdV0XRaFvi8jznPskAQAYTjbToX8SMbjLs2yuOcKgdIvCHs8Os1WDkxRIVXrV\nUHL5Se4IoRcECgDGkV41NIMxCkAa0rp0ADgUBAoAACCIQAEAAAQRKAAAgCACBQAAEDSDeRQ2FXrI\nU2LDUAEAGEGCgQIBASxMywgAW6PrAQAABBEo4IDQ2AQAmyJQAPaA3hAAc0GgAAAAgggUgPHQ9wFg\ndggUAABAEIECAAAIIlAAAABBBAoAACCIQAEAAAQRKCBxzFgAALtI8FkPPBQKAIC+JBgoEBDAi/MC\nALZA1wOwH/SJAJgFAgUAABBEoACMih4QAPNCoAAAAIIIFAAAQBCBAgAACCJQAAAAQQQKAAAgiEAB\nKWOuAgDYEYECAAAIIlDAQZjm7AU0eACYvgSf9cBDoQAA6EuCgQIBASaubWlLADAbdD0AAIAgAgUA\nABBEoAAAAIIIFAAAQBCBAgAACCJQAAAAQQQKAAAgiEABAAAEESggWUxqBAC7I1AAAABBBApIH5N6\nA8DWEnzWAw+FwoxkGXEMgElLMFAgIAAAoC90PQB7QDQLYC4IFAAAQBCBAgAACCJQAAAAQQQKAAAg\niEABAAAEzT5QyLKsrut9pwKTw/zNANCLeQcKVVXtOwkAAKRsrhMuVVW1XC73nQrMADMWAMAu5hoo\nFEWhlKrrummafacF2AmzOAOYshkHCkVR1HW9WCz2nRYAAJI17zEK4m9/+5v5MtvcvlKOQ0YrAoBZ\nmGuLgunf/u3fzJc8FAoAgL6k0KIAAAAGQqCABNGbBAB9IVAAAABBBApI2VzGq9AEAmCyssSG/mVZ\najnCFnS9O/FzwYwPJp5UAB2lVw3RogDsTVoXEwBpIlAAAABBBAoAACCIQAEAAAQRKAAAgCACBWAS\nuEMSwDSl8KwHS+ghT4ndr4IQalwA6FGCgQIBAWakbYlsAEwaXQ9IFhEjAOyOQAEAAAQRKAAAgCAC\nBQAAEESgAEwFoxoBTBCBAgAACCJQAPaMuzMATBmBAgAACCJQQFLo5geAfhEoAACAIAIFpImOfwDo\nRYLPeuChUAAA9CXBQIGAAACAvtD1AAAAgggUkI4EbnlIIAsAEkOgAAAAgggUkKDZDVOZXYIBHA4C\nBQAAEESgAAAAgggUAABAEIECEsH9AgAwBAIFAAAQRKCA1HAHAQD0iEABmBb6UABMSoLPeuChUAAA\n9CXBQIGAAHPUtrQlAJgiuh6QAqpYABgIgQIAAAgiUEBS6HcCgH4RKACTQ08KgOkgUAAAAEEECsBU\n0G8CYIIIFAAAQBCBAgAACCJQAAAAQQQKmD3uEQCA4RAoAACAoASf9cBDoQ5WSkc4y5LKDoD5SjBQ\nICAAAKAvdD0AE0KUC2BqCBQAAEAQgQIAAAgiUAAAAEEECpg3JlEAgEERKAAAgKBpBQpVVWVZlmVZ\nURShdbKrImvioKR3vwCNJQCmYELzKFRVtVwu8zxXSjVNUxRFXdehlWU1pRSBAgAAw8mmMz1RlmV5\nnktwIEGDm7a6rheLxWq1CsUHWTahHGEE+md3Soc9yUwBByK9amgqXQ86PpCX8o9+aa0Wb2zA4aBx\nHgCGNqGuB1coGjCf5pBY4AYAwKRMq0Vh7YADWa0sy9VqVZalUirLst/97nfmOtnmhskTxpNquMi5\nCWDvptWiUNd1PFYw2xhkzeVy+fvf/95chzYGAAD6MpUWBW98sLaBgVsekCRiXQDTMa1AQTcYhHoi\nrIkTOnZYAACA7UwlUFBK5Xm+XC6l7l8sFsqIHrIskzsg8jxvmkbWqetaz7sAAACGMKExChIQSIig\nlFqtVnq5uU5RFHodPe8CDtCBDPTLMnoiAOzT5OaF6H77g3ed9Ga6QIgZKKR3zNPOHZCw9Kqh5PKT\n3BFCSPLTFyafQSBJ6VVDExqjAAAApoZAAZi6AxmNAWCaCBQwP+fn5+Z8mlmWnZ+f7zE9Q7BylGQe\nAcwCgQJm5vz8/NNPP1XqX12A9+7du3v3bkr1qOTx+PieXpJeHgHMBYECpijyAI5vvvnm3XffNZc8\nefLk1q1bX3/99aab2q+1eXz69KleMtM8AkhAaoMzI1fMxHKatsiw4evXr19cXBgtCv884kdHRxcX\nFxttar865FFZ2ZxdHoEDlN73Mbn8JHeEDlOHn8h2oBBbdZKnRLdmgK7ZnGYegQOUXjVE1wMmqg24\ndu2aOUBBOzo68q4/fsq7i+bRXlfNM48A5m5CUzgDXRwfHz9+bC88OTl555139pGcQRwfH799+/bJ\nkyfmwsTyCGAuaFHAFEV+Jd+/f9949c8G+Z9//vnq8k6b2q94Hv/xj38opcwehznmEUACUutKSa9z\nCC5zDoWjo6Pj4+P79+/fvHlzj0nq3fn5+ddff/3kyZO3b/85evGXX35NLI9AktKrhpLLT3JHCBZz\nCOBq5X82WEp44gMwL+lVQ8nlJ7kjBMuhPVbx0PILzF161RBjFDBXaX0TAWCiCBSASSMeArBfBAqY\nE6YqBoCRESgAs0GcBGB8BAqYJRrkAWAcBArA1BEVAdijBKdwDj1rJ7H7VQ4QDe8AML4EAwUCAgAA\n+kLXA+bnkENBmlUAjIxAAQAABBEoADNwyI0oAPaLQAHzQJM7AOwFgQIwM8RMAMZEoICZoREeAMZE\noIAZ4De0IkICsCcECsD8EDkBGA2BAuaEX9UAMDICBUwdv5414iQA4yNQAAAAQQk+64GHQqWKA2jK\nMgoEwBgSDBQICFJCvwMA7BddD8CcEAYDGBmBAqbLbE6ggnTR3AJgBAQKAAAgiEABE0VzQohZGjQq\nABgagQIAAAgiUMAU0ZwQR6MCgNEQKAAAgCACBUwOzQld0KgAYBwECpgW6jwAmBQCBUyIFSXQnBBH\nowKAERAoYKKIEjZFrABgCAk+64GHQs0UQxO20LbEBwCGlWCLQhuw73RBKaWapnEXZhm13fbogAAw\nqAQDBUzQ+fn52dnZ9evXi6K4fv36o0ePzs/P5S23biOo2wVRF4B+EShgcOfn559++umbN28uLi6U\nUhcXF2/fvr179663SiNK2IJbaIQLAPqSJdYmn2Wp5WhGQoV/dnb25s2bp0+fGss8q11tQuc4biwU\nGeiCbJomz/PR0gMcpvQuX8nlJ7kjNCOhwr9+/bq0JSilvCGCcn4Tcxy3FgoXrl27fnFxce3atePj\n4/v379+4cWPcdAGHIr3LF10PWM87AtEr87mMEtpAlJC5n+sr5Qeobf3dNxcXb5Rqdb+PHiMCAHEE\nChPSvT4eZ2uREYghvptN2lCIcHR0rW39d6nsmPJD44ZWoXBBqfbJk29u3br19ddfd9+aq8dztd/T\nvkeTTRgwslkGClVVye/Ooii228KkLnNb1McjbC00AnHt1mQYnf4LrXVycv/4+HiLhKG7tlXXrl13\nlz958s3jx19s0XDT47na72mvTe372FeqBtraNDfV79amual5mV+gUFXVcrnM8zzP86ZpNooVJniZ\n27o+7mVrkd+O33zzzbvvvmuOQHzy5In+JWpFA5cxQduh7smUyk5OTn7++ef79++HVqJRYVPeTp/L\nfp9MKf+BcQ5irN+nx3O139NeTfj7OMHfAJPd1GQTNlBQOyPzCxQkSqjruq7rsiy7h3j7vcyFrr/x\n+nijTW23NWfj//z74ou/PH36xOg4aKXVertfor/+en56+ujo6JpS6ujo6J133vn2229v3ry58YYQ\nEJpn7Nq1a5erBMMFczPy5w0Eb9688X//9515Vmx9rvZ72o8WH2+0tf3+BpjdpiabsN6D2lkKXV+m\nabVaKaVWq5VeopQqy9J8Gfrso0eP3Lbue/funZ6eetfvtqmWP+9f5AhGji+2EzlXT09P7927Z532\nJycnez9DnD+llDo6Otoijz1+tY246ootErZpqkbLY48JI4+bJmymZnYXR13Xi8VitVrpHocsy6SB\nQb/UFx0MjXsap+/8/Pzu3bu3bt168uSJLJF+H6tFhxtNgB4ldiGcWdeDBATWuAR9j/7lxS7b/A9X\ntK06PX10796JVVAnJ/cfPjyVEfWJfRNSdePGjW+//fbo6Ojo6EiF+330MW1bpVRmvtR/165d5ysD\ndJFY5D2zQEHo9gOhWwsvq64tmz138M+r5+X9fp6/0MW3S33ccVNbbC20KaXU/fv3//GPf5it1oxA\nnKkbN2588cUXFxcXq9Xq4uLi8ePH240OOT4+Ns6Hnc7VqycqkJrEroUzCxS89zhYC4euRH2bUkqp\nk5OTLe7367c+3nRrkU11/CWKGel4i1DorOjxXL26qX9+Ez/66ONffvl1j/Fx26pffz3/+OP/Z25t\n64RN8zdAvwkjj6FNdfmizckex0dsRxmjF71jG0Mf/PXXXz/++GPrMvfRRx/98ssvm6Zhi03FE3Z6\neqrr44cPH26RpIG2JsYfgTjHM3OPRiuuHs+ufje10fcxXlx9JazHC06/W+vx8pVSHnvcVHqXr/nl\nR55qI1WXG+uMcwnYYlNdTp1+6+NZ31+Q3jdtUOMXV49nVy+b2uj72LG4dk/YZH8D9Hj5SiaPPW4q\nvcvXLPNjtohYX+bRLgGbbiq9U2dQFNdGKC6ty/dx1nFVv1vr8fI19zz2uKn0vo9zvZ/Ne/uDmvAd\nepNN2DRRXBuhuDZCcW2KEttIesWVXH6meoQmm7Bporg2QnFthOLaFCW2kfSKa2Z3PQAAgDEdVqDQ\n5fm5Q3x2R/tK9o5Z3leJzTTLcyyuHT8+x+Lacddz/OyOZprlPZbYBB1WoAAAADZCoAAAAIIIFAAA\nQNBv952A/oX6lhIbhgoAwAhSu4uDESgAgP1KrWJNLD8AAKBHjFEYw3//939bj8ZGxP/8z/9QXN29\nePGC4uqOs2tTnGAbSbK4aFEAAABBtCgAAIAgAoU+ZVnmbXSqqsp9flVVVVmWhT5yCNy813VdFEWW\nZUVRWG9RXG7e5byiuLwieS+Kwvo+UlzKKbG6rrOrzEKjxLx5l++j96s67+Ia9VmVSSvLUjmPvW7b\ndrVaKaXyPDcX5nkuC+Wffp/QOgtucUlBSbHIP2VZylsUl1tcsiTPc/nHfJfiCn0ZW6Pc9BKKqw1/\nH3MD30fNe4JZl6+Uvo8ECj3QV+pQzWddmGS59a0bNcV7FSouqxz0S4rLW1xWoK/PMYrLW1zCDEbN\nJQdbXG24xGS5u/6Bl1jH4tJfzzSKi66HHhRFUZalDiSt5eaJJaT1qaoqvZpeeAhCxdU0jblQt3NS\nXN7iUkq5p5aiuMLFpZRaLBbWWwdeXCpcYroQ3P5BdcAlFiqu5XJpLpT6VSVTXPuNU1IikaO3WUn5\nmjq7fDBha3Otz0+Kq43mWl+25F2Kqw3kWpeM+X2kuISbcasupMRMbq6V0bmQ57nV7xD54CzQorAH\nTdO4C+cXYw5GRlEppeRLRXFFVFW1XC6liOTHCsXlVVVV0zRmb6CguEKkZMqyXK1WZVk2TcMJFtc0\njTSLNk2zWCykFSGN4iJQ2ANvu6h7W8RhKopisVgopVarlZQJxRVRVVV7+TOF4opYLpdlWbrlQHGF\nyE9JubOmqiqp/xQlFtW2bV3Xbdvmeb5cLlUqxUWggAnJsqxpGhn4M7vv0sjMHyVFUejrOFzy2265\nXMotakqppmlmfK/aPvB9XMs7xCoNBAp7YJ1DcrVK7MTagpSA/Ihxl2sUl9BtmxaKy6WHFZuDi6WB\ngeLyku4/8wTTQRUl1kVqxbWvwRHp6T6YsXVuZrPePQTe0UDq6n3bulgorlBxyRJrbgCKKz5kzCoT\niqvtdoLpe/woMbe4zCKyiiWB4iJQ6M1GgYI1qGqcFE6KVVzuKDOzZCgu9+yyysSdqIPi6hgoUFxt\noMQ4wUIit9UkWVw8FGqf5toMtScUlytSJhTXRiiuEJlY3btcUWKOJIuLQAEAAAQxmBEAAAQRKAAA\ngCACBQAAEESgAAA4IHVdm4+8Yt6ttQgUAADpy7KsKIosyxaLxWKxMP+X+TrjnzXXyfNcZva05ve0\nZqlKBoECAOAgyIPB5F4/eYBTezlXUryC/+Mf/2i+/PHHHz/44AO511F/UP4hUAAAYK70g8EkOJDG\nAKna4x0Qn3/+uVLqs88+U0q9evVK/28+Y2W5XHofAZUAAgUAANb705/+pC6Dhg8//FAZQcas51Na\n67f7TgAAAFP3xz/+8U9/+tOrV69+/PFH3RPh7X1IDy0KAACsIQ0J//7v/67/F9L7ICMe9pa4gREo\nAACwpeSbExSBAgAAXUiPg3UHhB6XkOoABcVDoQAA6CLP8x9//NGtNLMsK8sy4RYFBjMCABDrO6iq\nSoYxfvDBB+byuq4Tnj5Bo0UBAIBY38EHH3wg90ZaNaZM15h2c4IiUAAAYK1Xr17J3AkHiEABAAAE\ncdcDAAAIIlAAAABBBAoAACCIQAEAAAQRKAAAgCACBQAAEESgAAAAgggUAABAEIECAAAIIlAAAABB\nBAoAACCIQAEAAAQRKAAAgKD/D78UzWpmINrrAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArgAAAHYCAIAAAApvgy/AAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE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\nRK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# -- low mass (load previous snapshot)\n",
    "w.loadSnapshot(\"lowmass_cb\")\n",
    "# -- setting parameters to constant\n",
    "[(a.GetName(), a.setConstant(True)) for a in wrap_iter(w.set(\"params_dscb\").iterator())]\n",
    "\n",
    "c = ROOT.TCanvas()\n",
    "frame = myy.frame()\n",
    "lowMass_1tag_single_Higgs.plotOn(frame)\n",
    "r = w.pdf(\"dscb_pdf\").fitTo(lowMass_1tag_single_Higgs, rf.SumW2Error(ROOT.kTRUE), rf.Save())\n",
    "w.pdf(\"dscb_pdf\").plotOn(frame)\n",
    "w.pdf(\"dscb_pdf\").paramOn(frame)#,ROOT.RooFit.Layout(0.65,0.85,0.85),ROOT.RooFit.Parameters(ROOT.RooArgSet(w.var(\"a1\"),w.var(\"nbkg1\"))))\n",
    "frame.Draw()\n",
    "frame.SetTitle(\"Low Mass Single-Higgs Background - 1 Tag\")\n",
    "c.Draw()\n",
    "c1 = ROOT.TCanvas()\n",
    "frame = myy.frame()\n",
    "lowMass_2tag_single_Higgs.plotOn(frame)\n",
    "r = w.pdf(\"dscb_pdf\").fitTo(lowMass_2tag_single_Higgs, rf.SumW2Error(ROOT.kTRUE), rf.Save())\n",
    "w.pdf(\"dscb_pdf\").plotOn(frame)\n",
    "w.pdf(\"dscb_pdf\").paramOn(frame)#,ROOT.RooFit.Layout(0.65,0.85,0.85),ROOT.RooFit.Parameters(ROOT.RooArgSet(w.var(\"a1\"),w.var(\"nbkg1\"))))\n",
    "frame.Draw()\n",
    "frame.SetTitle(\"Low Mass Single-Higgs Background - 2 Tag\")\n",
    "c1.Draw()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "# Build full model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<ROOT.RooCategory object (\"highmasscats\") at 0x7fc49e55f000>"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#w.factory(\"cat[tag0low, tag0high, tag1low, tag1high, tag2low, tag2high]\")\n",
    "w.factory(\"highmasscats[tag0, tag1, tag2]\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "RooWorkspace(w) w contents\n",
      "\n",
      "variables\n",
      "---------\n",
      "(a0,a1,a2,alphaHi_cb,alphaLo_cb,highmasscats,m0_cb,myy,nHi_cb,nLo_cb,sigma_cb)\n",
      "\n",
      "p.d.f.s\n",
      "-------\n",
      "RooExponential::bkg_pdf[ x=bkg_exponent c=1 ] = 0.0282139\n",
      "HggTwoSidedCBPdf::dscb_pdf[ m=myy m0=m0_cb sigma=sigma_cb alphaLo=alphaLo_cb nLo=nLo_cb alphaHi=alphaHi_cb nHi=nHi_cb ] = 0.00390644\n",
      "\n",
      "functions\n",
      "--------\n",
      "RooPolyVar::bkg_exponent[ x=myy coefList=(a0,a1,a2) ] = -3.56794\n",
      "\n",
      "parameter snapshots\n",
      "-------------------\n",
      "lowmass_epoly2 = (a1=-0.0429004 +/- 0.00951067,a2=6.35874e-05 +/- 3.65609e-05)\n",
      "highmass_epoly2 = (a1=-0.0154718 +/- 0.0238383,a2=2.31129e-06 +/- 9.08521e-05)\n",
      "lowmass_cb = (alphaLo_cb=2.16478 +/- 0.0449161,nLo_cb=3.59579 +/- 0.41894,alphaHi_cb=2.32694 +/- 0.0514366,nHi_cb=4.16611 +/- 0.522503)\n",
      "highmass_cb = (alphaLo_cb=2.35701 +/- 0.0908605,nLo_cb=2.15038 +/- 0.383769,alphaHi_cb=2.45231 +/- 0.133304,nHi_cb=3.43126 +/- 1.02606)\n",
      "\n",
      "named sets\n",
      "----------\n",
      "params_dscb:(alphaHi_cb,alphaLo_cb,nHi_cb,nLo_cb)\n",
      "params_epoly2:(a1,a2)\n",
      "\n"
     ]
    }
   ],
   "source": [
    "w.Print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#blah1 = _add(dh_bkg_noH_0tag_lowmass, dh_bkg_1H_0tag_lowmass)\n",
    "highMass_0tag_all = _add(highMass_0tag_non_Higgs, highMass_0tag_single_Higgs)\n",
    "#blah3 = _add(dh_bkg_noH_1tag_lowmass, dh_bkg_1H_1tag_lowmass)\n",
    "highMass_1tag_all = _add(highMass_1tag_non_Higgs, highMass_1tag_single_Higgs)\n",
    "#blah5 = _add(dh_bkg_noH_2tag_lowmass, dh_bkg_1H_2tag_lowmass)\n",
    "highMass_2tag_all = _add(highMass_2tag_non_Higgs, highMass_2tag_single_Higgs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "highmass_data = ROOT.RooDataHist(\n",
    "    \"highmass_data\", \"highmass_data\", ROOT.RooArgList(myy),\n",
    "    rf.Index(w.cat(\"highmasscats\")),\n",
    "    rf.Import(\"tag0\", highMass_0tag_all),\n",
    "    rf.Import(\"tag1\", highMass_1tag_all),\n",
    "    rf.Import(\"tag2\", highMass_2tag_all)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "# data_combo = ROOT.RooDataHist(\n",
    "#     \"data_combo\", \"combined data\", ROOT.RooArgList(myy),\n",
    "#     rf.Index(w.cat(\"cat\")),\n",
    "#     rf.Import(\"tag0low\", blah1),#_add(dh_bkg_noH_0tag_lowmass, dh_bkg_1H_0tag_lowmass)),\n",
    "#     rf.Import(\"tag0high\", blah2),#_add(dh_bkg_noH_0tag_highmass, dh_bkg_1H_0tag_highmass)),\n",
    "#     rf.Import(\"tag1low\", blah3),#_add(dh_bkg_noH_1tag_lowmass, dh_bkg_1H_1tag_lowmass)),\n",
    "#     rf.Import(\"tag1high\",blah4),# _add(dh_bkg_noH_1tag_highmass, dh_bkg_1H_1tag_highmass)),\n",
    "#     rf.Import(\"tag2low\", blah5),#_add(dh_bkg_noH_2tag_lowmass, dh_bkg_1H_2tag_lowmass)),\n",
    "#     rf.Import(\"tag2high\",blah6),# _add(dh_bkg_noH_2tag_highmass, dh_bkg_1H_2tag_highmass))\n",
    "# )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<ROOT.RooRealVar object (\"n_highMass_1tag_di_Higgs_bias\") at 0x7fc4a05f4e00>"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "w.factory(\"mu_1H[1]\")#, 0, 100]\")\n",
    "w.factory(\"mu_2H[1, 0, 20]\")\n",
    "w.factory(\"EG_SCALE[-10, 10]\")\n",
    "w.factory(\"EG_RES[-10, 10]\")\n",
    "w.factory(\"theta_yield[-10, 10]\")\n",
    "w.factory(\"n_highMass_2tag_di_Higgs_bias[-100, 100]\")\n",
    "w.factory(\"n_highMass_1tag_di_Higgs_bias[-100, 100]\")\n",
    "# w.factory(\"n_bias_tag2low_2H[-100, 100]\")\n",
    "# w.factory(\"n_bias_tag1low_2H[-100, 100]\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<ROOT.RooGaussian object (\"n_highMass_1tag_di_Higgs_bias_gaussian\") at 0x7fc49e685600>"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "w.factory(\"Gaussian::EG_SCALE_gaussian(EG_SCALE, EG_SCALE_global[0], 1)\")\n",
    "w.factory(\"Gaussian::EG_RES_gaussian(EG_RES, EG_RES_global[0], 1)\")\n",
    "w.factory(\"Gaussian::theta_yield_gaussian(theta_yield, theta_yield_global[0], 1)\")\n",
    "w.factory(\"Gaussian::n_highMass_2tag_di_Higgs_bias_gaussian(n_highMass_2tag_di_Higgs_bias, n_highMass_2tag_di_Higgs_bias_global[0], 0.8)\")\n",
    "w.factory(\"Gaussian::n_highMass_1tag_di_Higgs_bias_gaussian(n_highMass_1tag_di_Higgs_bias, n_highMass_1tag_di_Higgs_bias_global[0], 0.8)\")\n",
    "# w.factory(\"Gaussian::n_bias_tag2low_2H_gaussian(n_bias_tag2low_2H, n_bias_tag2low_2H_global[0], 0.8)\")\n",
    "# w.factory(\"Gaussian::n_bias_tag1low_2H_gaussian(n_bias_tag1low_2H, n_bias_tag1low_2H_global[0], 0.8)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<ROOT.RooFormulaVar object (\"coeff_2H\") at 0x7fc49c74c400>"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "w.factory(\"expr::coeff_1H('mu_1H * (1 + 0.04 * theta_yield)', mu_1H, theta_yield)\")\n",
    "w.factory(\"expr::coeff_2H('mu_2H * (1 + 0.1 * theta_yield)', mu_2H, theta_yield)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# -- check that integral and sum of weights actually give the same answer\n",
    "# this should be 0 if the two are the same for all histograms in histos_bkg\n",
    "sum([(v.Integral() != v.GetSumOfWeights()) for k,v in histos_bkg_myy.iteritems()])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# histos_bkg['m_yy_highMass_0tag_tightIsolated_bkg_non_Higgs'].Integral()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::n_highMass_1tag_non_Higgs_exp\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::n_highMass_2tag_single_Higgs_exp\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::n_highMass_0tag_single_Higgs_exp\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::n_highMass_0tag_non_Higgs_exp\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::n_highMass_1tag_single_Higgs_exp\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::n_highMass_2tag_non_Higgs_exp\n"
     ]
    }
   ],
   "source": [
    "highMass_yields_names = []\n",
    "for k, v in histos_bkg_myy.iteritems():\n",
    "    if 'high' in k:\n",
    "        short_k = k.replace('m_yy_', '').replace('tightIsolated_bkg_', '')\n",
    "        _name = 'n_' + short_k + '_exp'\n",
    "        highMass_yields_names.append(_name)\n",
    "        exec (\"{} = {}\".format(\n",
    "            _name,\n",
    "            'ROOT.RooRealVar(\"{k}\", \"{k}\", v.GetSumOfWeights())'.format(k=_name)\n",
    "        ))\n",
    "        exec(\"getattr(w, 'import')(ROOT.RooArgSet({}))\".format(_name))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::n_highMass_2tag_di_Higgs_exp\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::n_highMass_1tag_di_Higgs_exp\n"
     ]
    }
   ],
   "source": [
    "# n_tag2high_0H = ROOT.RooRealVar(\n",
    "#     \"n_tag2high_0H\", \n",
    "#     \"n_tag2high_0H\", \n",
    "#     histos_bkg_myy['m_yy_highMass_2tag_tightIsolated_bkg_non_Higgs'].GetSumOfWeights()\n",
    "# )\n",
    "# n_tag1high_0H = ROOT.RooRealVar(\n",
    "#     \"n_tag1high_0H\", \n",
    "#     \"n_tag1high_0H\", \n",
    "#     histos_bkg_myy['m_yy_highMass_1tag_tightIsolated_bkg_non_Higgs'].GetSumOfWeights()\n",
    "# )\n",
    "# n_tag0high_0H = ROOT.RooRealVar(\n",
    "#     \"n_tag0high_0H\",\n",
    "#     \"n_tag0high_0H\",\n",
    "#     histos_bkg_myy['m_yy_highMass_0tag_tightIsolated_bkg_non_Higgs'].GetSumOfWeights()\n",
    "# )\n",
    "\n",
    "# n_tag2high_1H_exp = ROOT.RooRealVar(\n",
    "#     \"n_tag2high_1H_exp\", \n",
    "#     \"n_tag2high_1H_exp\", \n",
    "#     histos_bkg['m_yy_highMass_2tag_tightIsolated_bkg_single_Higgs'].GetSumOfWeights()\n",
    "# )\n",
    "# n_tag1high_1H_exp = ROOT.RooRealVar(\n",
    "#     \"n_tag1high_1H_exp\", \n",
    "#     \"n_tag1high_1H_exp\", \n",
    "#     histos_bkg['m_yy_highMass_1tag_tightIsolated_bkg_single_Higgs'].GetSumOfWeights()\n",
    "# )\n",
    "# n_tag0high_1H_exp = ROOT.RooRealVar(\n",
    "#     \"n_tag0high_1H_exp\", \n",
    "#     \"n_tag0high_1H_exp\", \n",
    "#     histos_bkg['m_yy_highMass_0tag_tightIsolated_bkg_single_Higgs'].GetSumOfWeights()\n",
    "# )\n",
    "\n",
    "n_highMass_2tag_di_Higgs_exp = ROOT.RooRealVar(\n",
    "    \"n_highMass_2tag_di_Higgs_exp\", \n",
    "    \"n_highMass_2tag_di_Higgs_exp\", \n",
    "    0.19\n",
    ")#122.3) # correct up to a constant\n",
    "n_highMass_1tag_di_Higgs_exp = ROOT.RooRealVar(\n",
    "    \"n_highMass_1tag_di_Higgs_exp\", \n",
    "    \"n_highMass_1tag_di_Higgs_exp\", \n",
    "    0.13\n",
    ")#85.1) # correct up to a constant\n",
    "\n",
    "getattr(w, 'import')(ROOT.RooArgSet(n_highMass_2tag_di_Higgs_exp, n_highMass_1tag_di_Higgs_exp))\n",
    "highMass_yields_names.extend([\"n_highMass_2tag_di_Higgs_exp\", \"n_highMass_1tag_di_Higgs_exp\"])\n",
    "\n",
    "# n_tag2low_0H = ROOT.RooRealVar(\"n_tag2low_0H\", \"n_tag2low_0H\", histos_bkg['m_yy_lowMass_2tag_tightIsolated_bkg_non_Higgs'].GetSumOfWeights())\n",
    "# n_tag1low_0H = ROOT.RooRealVar(\"n_tag1low_0H\", \"n_tag1low_0H\", histos_bkg['m_yy_lowMass_1tag_tightIsolated_bkg_non_Higgs'].GetSumOfWeights())\n",
    "# n_tag0low_0H = ROOT.RooRealVar(\"n_tag0low_0H\", \"n_tag0low_0H\", histos_bkg['m_yy_lowMass_0tag_tightIsolated_bkg_non_Higgs'].GetSumOfWeights())\n",
    "\n",
    "# n_tag2low_1H_exp = ROOT.RooRealVar(\"n_tag2low_1H_exp\", \"n_tag2low_1H_exp\", histos_bkg['m_yy_lowMass_2tag_tightIsolated_bkg_single_Higgs'].GetSumOfWeights())\n",
    "# n_tag1low_1H_exp = ROOT.RooRealVar(\"n_tag1low_1H_exp\", \"n_tag1low_1H_exp\", histos_bkg['m_yy_lowMass_1tag_tightIsolated_bkg_single_Higgs'].GetSumOfWeights())\n",
    "# n_tag0low_1H_exp = ROOT.RooRealVar(\"n_tag0low_1H_exp\", \"n_tag0low_1H_exp\", histos_bkg['m_yy_lowMass_0tag_tightIsolated_bkg_single_Higgs'].GetSumOfWeights())\n",
    "\n",
    "# n_tag2low_2H_exp = ROOT.RooRealVar(\"n_tag2low_2H_exp\", \"n_tag2low_2H_exp\", 0.32)#122.3) # don't have low mass number, use high mass\n",
    "# n_tag1low_2H_exp = ROOT.RooRealVar(\"n_tag1low_2H_exp\", \"n_tag1low_2H_exp\", 0.23)#85.1) # don't have low mass number, use high mass\n",
    "\n",
    "\n",
    "\n",
    "# getattr(w, 'import')(\n",
    "#     ROOT.RooArgSet(\n",
    "#         n_tag2high_1H_exp,\n",
    "#         n_tag1high_1H_exp,\n",
    "#         n_tag0high_1H_exp,\n",
    "#         n_tag2high_2H_exp,\n",
    "#         n_tag1high_2H_exp,\n",
    "#         n_tag2high_0H,\n",
    "#         n_tag1high_0H,\n",
    "#         n_tag0high_0H\n",
    "# #         n_tag2low_1H_exp,\n",
    "# #         n_tag1low_1H_exp,\n",
    "# #         n_tag0low_1H_exp,\n",
    "# #         n_tag2low_2H_exp\n",
    "#     )\n",
    "# )\n",
    "\n",
    "# getattr(w, 'import')(\n",
    "#     ROOT.RooArgSet(\n",
    "#         n_tag1low_2H_exp,\n",
    "#         n_tag2high_0H,\n",
    "#         n_tag1high_0H,\n",
    "#         n_tag0high_0H,\n",
    "#         n_tag2lo_expw_0H,\n",
    "#         n_tag1low_0H,\n",
    "#         n_tag0low_0H\n",
    "#     )\n",
    "# ) # cannot import them all together because an ArgSet takes at most 10 elements :O\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['n_highMass_1tag_non_Higgs_exp',\n",
       " 'n_highMass_2tag_single_Higgs_exp',\n",
       " 'n_highMass_0tag_single_Higgs_exp',\n",
       " 'n_highMass_0tag_non_Higgs_exp',\n",
       " 'n_highMass_1tag_single_Higgs_exp',\n",
       " 'n_highMass_2tag_non_Higgs_exp',\n",
       " 'n_highMass_2tag_di_Higgs_exp',\n",
       " 'n_highMass_1tag_di_Higgs_exp']"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "highMass_yields_names"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "for name in highMass_yields_names:\n",
    "    if 'di_Higgs' in name:\n",
    "        w.factory(\"expr::{name_noexp}('{name} * {coeff} + {bias}', {name}, {coeff}, {bias})\".format(\n",
    "            name_noexp = name.replace('_exp', ''),\n",
    "            coeff = 'coeff_2H',\n",
    "            name = name,\n",
    "            bias = name.replace('_exp', '_bias')\n",
    "        )) \n",
    "    elif 'single_Higgs' in name:\n",
    "        w.factory(\"expr::{name_noexp}('{name} * {coeff}', {name}, {coeff})\".format(\n",
    "            name_noexp = name.replace('_exp', ''),\n",
    "            coeff = 'coeff_1H',\n",
    "            name = name  \n",
    "        ))\n",
    "    #elif 'non_Higgs' in name:\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# w.factory(\"expr::n_tag2high_1H('n_tag2high_1H_exp * coeff_1H', n_tag2high_1H_exp, coeff_1H)\")\n",
    "# w.factory(\"expr::n_tag1high_1H('n_tag1high_1H_exp * coeff_1H', n_tag1high_1H_exp, coeff_1H)\")\n",
    "# w.factory(\"expr::n_tag0high_1H('n_tag0high_1H_exp * coeff_1H', n_tag0high_1H_exp, coeff_1H)\")\n",
    "\n",
    "# # w.factory(\"expr::n_tag2low_1H('n_tag2low_1H_exp * coeff_1H', n_tag2low_1H_exp, coeff_1H)\")\n",
    "# # w.factory(\"expr::n_tag1low_1H('n_tag1low_1H_exp * coeff_1H', n_tag1low_1H_exp, coeff_1H)\")\n",
    "# # w.factory(\"expr::n_tag0low_1H('n_tag0low_1H_exp * coeff_1H', n_tag0low_1H_exp, coeff_1H)\")\n",
    "\n",
    "# w.factory(\"expr::n_tag2high_2H('n_tag2high_2H_exp * coeff_2H + n_bias_tag2high_2H', n_tag2high_2H_exp, coeff_2H, n_bias_tag2high_2H)\")\n",
    "# w.factory(\"expr::n_tag1high_2H('n_tag1high_2H_exp * coeff_2H + n_bias_tag1high_2H', n_tag1high_2H_exp, coeff_2H, n_bias_tag1high_2H)\")\n",
    "\n",
    "# # w.factory(\"expr::n_tag2low_2H('n_tag2low_2H_exp * coeff_2H + n_bias_tag2low_2H', n_tag2low_2H_exp, coeff_2H, n_bias_tag2low_2H)\")\n",
    "# # w.factory(\"expr::n_tag1low_2H('n_tag1low_2H_exp * coeff_2H + n_bias_tag1low_2H', n_tag1low_2H_exp, coeff_2H, n_bias_tag1low_2H)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "# w.loadSnapshot(\"lowmass_epoly2\")\n",
    "# [(a.GetName(), a.setConstant(True)) for a in wrap_iter(w.set(\"params_epoly2\").iterator())] # probably not needed\n",
    "# w.loadSnapshot(\"lowmass_cb\")\n",
    "# [(a.GetName(), a.setConstant(True)) for a in wrap_iter(w.set(\"params_dscb\").iterator())] # probably not needed\n",
    "\n",
    "# w.factory(\"PolyVar::bkg_exponent_lowmass(myy,{a0, a1_lowmass[-100, 100], a2_lowmass[-100, 100]})\")\n",
    "# w.factory(\"Exponential::bkg_pdf_lowmass(bkg_exponent_lowmass,1)\")\n",
    "\n",
    "# [w.var(a.GetName() + \"_lowmass\").setVal(a.getVal()) for a in wrap_iter(w.set(\"params_epoly2\").iterator())]\n",
    "# [w.var(a.GetName() + \"_lowmass\").setConstant(True) for a in wrap_iter(w.set(\"params_epoly2\").iterator())]\n",
    "\n",
    "# # w.factory(\"HggTwoSidedCBPdf::dscb_pdf_lowmass(\\\n",
    "# #           myy,\\\n",
    "# #           m0_cb,\\\n",
    "# #           sigma_cb,\\\n",
    "# #           alphaLo_cb_lowmass[0, 5],\\\n",
    "# #           nLo_cb_lowmass[0, 10],\\\n",
    "# #           alphaHi_cb_lowmass[0, 5],\\\n",
    "# #           nHi_cb_lowmass[0, 10],\\\n",
    "# #           )\"\n",
    "# # ) # no suitable constructor found (??)\n",
    "# # -- define single-Higgs bkg fit function (Double Sided Crystal Ball)\n",
    "# m0_cb_tag0low = ROOT.RooRealVar('m0_cb_tag0low','m0_cb_tag0low', 125)\n",
    "# # DON'T HAVE SIGMA FOR 0 TAG, USE 1 TAG\n",
    "# # ALSO, THE NUMBER WE HAVE IS FOR HIGH MASS NOT LOW MASS :(\n",
    "# sigma_cb_tag0low = ROOT.RooRealVar('sigma_cb_tag0low','sigma_cb_tag0low', 1.53) \n",
    "# m0_cb_tag1low = ROOT.RooRealVar('m0_cb_tag1low','m0_cb_tag1low', 125)\n",
    "# sigma_cb_tag1low = ROOT.RooRealVar('sigma_cb_tag1low','sigma_cb_tag1low', 1.53)\n",
    "# m0_cb_tag2low = ROOT.RooRealVar('m0_cb_tag2low','m0_cb_tag2low', 125)\n",
    "# sigma_cb_tag2low = ROOT.RooRealVar('sigma_cb_tag2low','sigma_cb_tag2low', 1.46)\n",
    "# # values for these variables will be set below\n",
    "# alphaLo_cb_lowmass = ROOT.RooRealVar('alphaLo_cb_lowmass','alphaLo_cb_lowmass', 0, 5)\n",
    "# nLo_cb_lowmass = ROOT.RooRealVar('nLo_cb_lowmass','nLo_cb_lowmass', 0, 10)\n",
    "# alphaHi_cb_lowmass = ROOT.RooRealVar('alphaHi_cb_lowmass','alphaHi_cb_lowmass', 0, 5)\n",
    "# nHi_cb_lowmass = ROOT.RooRealVar('nHi_cb_lowmass','nHi_cb_lowmass', 0, 10)\n",
    "# getattr(w, 'import')(ROOT.RooArgSet(m0_cb_tag0low,\n",
    "#     sigma_cb_tag0low,\n",
    "#     m0_cb_tag1low,\n",
    "#     sigma_cb_tag1low,\n",
    "#     m0_cb_tag2low,\n",
    "#     sigma_cb_tag2low,\n",
    "# ))\n",
    "# # NOTE: WE ARE SHARING THE FITTING PARAMETERS ACROSS CATEGORIES, BUT IN REALITY ALL CATEGORIES WILL HAVE \n",
    "# # THEIR OWN INDIVIDUAL m0, sigma0, alphaHi, alphaLo, nHi, nLo\n",
    "# # ALSO, FOR NOW WE ARE USING THE SAME CB FOR 1 AND 2 HIGGS, BUT THESE WILL IDEALLY BE COMPLETELY DIFFERENT\n",
    "\n",
    "# # -- double sided crystal ball\n",
    "# w.factory(\"expr::m0_tag0low('m0_cb_tag0low * (1 + 0.005 * EG_SCALE)', m0_cb_tag0low, EG_SCALE)\")\n",
    "# w.factory(\"expr::m0_tag1low('m0_cb_tag1low * (1 + 0.005 * EG_SCALE)', m0_cb_tag1low, EG_SCALE)\")\n",
    "# w.factory(\"expr::m0_tag2low('m0_cb_tag2low * (1 + 0.5 * EG_SCALE)', m0_cb_tag2low, EG_SCALE)\")\n",
    "\n",
    "# w.factory(\"expr::sigma_tag1low('sigma_cb_tag1low * (1 + 0.108 * EG_RES) * (1 + 0.03 * EG_SCALE)', sigma_cb_tag1low, EG_RES, EG_SCALE)\")\n",
    "# w.factory(\"expr::sigma_tag0low('sigma_cb_tag0low * (1 + 0.108 * EG_RES) * (1 + 0.03 * EG_SCALE)', sigma_cb_tag0low, EG_RES, EG_SCALE)\")\n",
    "# w.factory(\"expr::sigma_tag2low('sigma_cb_tag2low * (1 + 0.10 * EG_RES) * (1 + 0.02 * EG_SCALE)', sigma_cb_tag2low, EG_RES, EG_SCALE)\")\n",
    "\n",
    "    \n",
    "# dscb_pdf_tag0low = ROOT.HggTwoSidedCBPdf(\n",
    "#     'dscb_pdf_tag0low', 'Double Sided Crystal Ball',\n",
    "#     myy, w.function(\"m0_tag0low\"), w.function(\"sigma_tag0low\"),\n",
    "#     alphaLo_cb_lowmass,nLo_cb_lowmass,alphaHi_cb_lowmass,nHi_cb_lowmass\n",
    "# ) # AGAIN WE DON'T HAVE THE VARIATIONS FOR THE 0 TAG CATEGORY, SO I'M GOING TO USE THE ONE FROM 1 TAG FOR NOW\n",
    "# dscb_pdf_tag1low = ROOT.HggTwoSidedCBPdf(\n",
    "#     'dscb_pdf_tag1low', 'Double Sided Crystal Ball',\n",
    "#     myy, w.function(\"m0_tag1low\"), w.function(\"sigma_tag1low\"),\n",
    "#     alphaLo_cb_lowmass,nLo_cb_lowmass,alphaHi_cb_lowmass,nHi_cb_lowmass\n",
    "# )\n",
    "# dscb_pdf_tag2low = ROOT.HggTwoSidedCBPdf(\n",
    "#     'dscb_pdf_tag2low', 'Double Sided Crystal Ball',\n",
    "#     myy, w.function(\"m0_tag2low\"), w.function(\"sigma_tag2low\"),\n",
    "#     alphaLo_cb_lowmass,nLo_cb_lowmass,alphaHi_cb_lowmass,nHi_cb_lowmass\n",
    "# )\n",
    "# getattr(w, 'import')(ROOT.RooArgSet(dscb_pdf_tag0low, dscb_pdf_tag1low, dscb_pdf_tag2low))\n",
    "# [w.var(a.GetName() + \"_lowmass\").setVal(a.getVal()) for a in wrap_iter(w.set(\"params_dscb\").iterator())]\n",
    "# [w.var(a.GetName() + \"_lowmass\").setConstant(True) for a in wrap_iter(w.set(\"params_dscb\").iterator())]\n",
    "\n",
    "# w.factory(\"expr::n_tag1low_12H('n_tag1low_1H + n_tag1low_2H', n_tag1low_1H, n_tag1low_2H)\")\n",
    "# w.factory(\"expr::n_tag2low_12H('n_tag2low_1H + n_tag2low_2H', n_tag2low_1H, n_tag2low_2H)\")\n",
    "\n",
    "# w.factory(\"SUM:pdf_tag0low(\\\n",
    "#    n_tag0low_0H * bkg_pdf_lowmass,\\\n",
    "#    n_tag0low_1H * dscb_pdf_tag0low)\"\n",
    "# )\n",
    "# w.factory(\"SUM:pdf_tag1low(\\\n",
    "#    n_tag1low_0H * bkg_pdf_lowmass,\\\n",
    "#    n_tag1low_12H * dscb_pdf_tag1low)\"\n",
    "# )\n",
    "# w.factory(\"SUM:pdf_tag2low(\\\n",
    "#    n_tag2low_0H * bkg_pdf_lowmass,\\\n",
    "#    n_tag2low_12H * dscb_pdf_tag2low)\"\n",
    "# )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<ROOT.RooAddPdf object (\"pdf_highMass_2tag\") at 0x7fc49ced6200>"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing HggTwoSidedCBPdf::dscb_pdf_highmass_0tag\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing HggTwoSidedCBPdf::dscb_pdf_highmass_1tag\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing HggTwoSidedCBPdf::dscb_pdf_highmass_2tag\n"
     ]
    }
   ],
   "source": [
    "w.loadSnapshot(\"highmass_epoly2\")\n",
    "[(a.GetName(), a.setConstant(True)) for a in wrap_iter(w.set(\"params_epoly2\").iterator())] # probably not needed\n",
    "w.loadSnapshot(\"highmass_cb\")\n",
    "[(a.GetName(), a.setConstant(True)) for a in wrap_iter(w.set(\"params_dscb\").iterator())] # probably not needed\n",
    "\n",
    "w.factory(\"PolyVar::bkg_exponent_highmass(myy,{a0, a1_highmass[-100, 100], a2_highmass[-100, 100]})\")\n",
    "w.factory(\"Exponential::bkg_pdf_highmass(bkg_exponent_highmass,1)\")\n",
    "\n",
    "[w.var(a.GetName() + \"_highmass\").setVal(a.getVal()) for a in wrap_iter(w.set(\"params_epoly2\").iterator())]\n",
    "[w.var(a.GetName() + \"_highmass\").setConstant(True) for a in wrap_iter(w.set(\"params_epoly2\").iterator())]\n",
    "\n",
    "# w.factory(\"HggTwoSidedCBPdf::dscb_pdf_lowmass(\\\n",
    "#           myy,\\\n",
    "#           m0_cb,\\\n",
    "#           sigma_cb,\\\n",
    "#           alphaLo_cb_lowmass[0, 5],\\\n",
    "#           nLo_cb_lowmass[0, 10],\\\n",
    "#           alphaHi_cb_lowmass[0, 5],\\\n",
    "#           nHi_cb_lowmass[0, 10],\\\n",
    "#           )\"\n",
    "# ) # no suitable constructor found (??)\n",
    "# -- define single-Higgs bkg fit function (Double Sided Crystal Ball)\n",
    "for i in ['0', '1', '2']:\n",
    "    w.factory(\"m0_cb_highmass_\" + i + \"tag[125]\")\n",
    "    w.factory(\"sigma_cb_highmass_\" + i + \"tag[1.53]\")\n",
    "# w.factory(\"m0_cb_highMass_1tag\", 125)\n",
    "# w.factory(\"m0_cb_highMass_2tag\", 125)\n",
    "# w.factory(\"sigma_cb_highMass_0tag\", 1.53)\n",
    "# w.factory(\"sigma_cb_highMass_1tag\", 1.53)\n",
    "# w.factory(\"sigma_cb_highMass_2tag\", 1.53)\n",
    "w.factory(\"alphaLo_cb_highmass[0, 5]\")\n",
    "w.factory(\"alphaHi_cb_highmass[0, 5]\")\n",
    "w.factory(\"nLo_cb_highmass[0, 10]\")\n",
    "w.factory(\"nHi_cb_highmass[0, 10]\")\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "# m0_cb_tag0high = ROOT.RooRealVar('m0_cb_tag0high','m0_cb_tag0high', 125)\n",
    "# sigma_cb_tag0high = ROOT.RooRealVar('sigma_cb_tag0high','sigma_cb_tag0high', 1.53)\n",
    "# m0_cb_tag1high = ROOT.RooRealVar('m0_cb_tag1high','m0_cb_tag1high', 125)\n",
    "# sigma_cb_tag1high = ROOT.RooRealVar('sigma_cb_tag1high','sigma_cb_tag1high', 1.53)\n",
    "# m0_cb_tag2high = ROOT.RooRealVar('m0_cb_tag2high','m0_cb_tag2high', 125)\n",
    "# sigma_cb_tag2high = ROOT.RooRealVar('sigma_cb_tag2high','sigma_cb_tag2high', 1.46)\n",
    "# alphaLo_cb_highmass = ROOT.RooRealVar('alphaLo_cb_highmass','alphaLo_cb_highmass', 0, 5)\n",
    "# nLo_cb_highmass = ROOT.RooRealVar('nLo_cb_highmass','nLo_cb_highmass', 0, 10)\n",
    "# alphaHi_cb_highmass = ROOT.RooRealVar('alphaHi_cb_highmass','alphaHi_cb_highmass', 0, 5)\n",
    "# nHi_cb_highmass = ROOT.RooRealVar('nHi_cb_highmass','nHi_cb_highmass', 0, 10)\n",
    "\n",
    "# getattr(w, 'import')(ROOT.RooArgSet(m0_cb_tag0high,\n",
    "#     sigma_cb_tag0high,\n",
    "#     m0_cb_tag1high,\n",
    "#     sigma_cb_tag1high,\n",
    "#     m0_cb_tag2high,\n",
    "#     sigma_cb_tag2high,\n",
    "# ))\n",
    "# NOTE: WE ARE SHARING THE FITTING PARAMETERS ACROSS CATEGORIES, BUT IN REALITY ALL CATEGORIES WILL HAVE \n",
    "# THEIR OWN INDIVIDUAL m0, sigma0, alphaHi, alphaLo, nHi, nLo\n",
    "# ALSO, FOR NOW WE ARE USING THE SAME CB FOR 1 AND 2 HIGGS, BUT THESE WILL IDEALLY BE COMPLETELY DIFFERENT\n",
    "\n",
    "# -- double sided crystal ball\n",
    "w.factory(\"expr::m0_highmass_0tag('m0_cb_highmass_0tag * (1 + 0.005 * EG_SCALE)', m0_cb_highmass_0tag, EG_SCALE)\")\n",
    "w.factory(\"expr::m0_highmass_1tag('m0_cb_highmass_1tag * (1 + 0.005 * EG_SCALE)', m0_cb_highmass_1tag, EG_SCALE)\")\n",
    "w.factory(\"expr::m0_highmass_2tag('m0_cb_highmass_2tag * (1 + 0.5 * EG_SCALE)', m0_cb_highmass_2tag, EG_SCALE)\")\n",
    "\n",
    "w.factory(\"expr::sigma_highmass_0tag('sigma_cb_highmass_0tag * (1 + 0.108 * EG_RES) * (1 + 0.03 * EG_SCALE)', sigma_cb_highmass_0tag, EG_RES, EG_SCALE)\")\n",
    "w.factory(\"expr::sigma_highmass_1tag('sigma_cb_highmass_1tag * (1 + 0.108 * EG_RES) * (1 + 0.03 * EG_SCALE)', sigma_cb_highmass_1tag, EG_RES, EG_SCALE)\")\n",
    "w.factory(\"expr::sigma_highmass_2tag('sigma_cb_highmass_2tag * (1 + 0.10 * EG_RES) * (1 + 0.02 * EG_SCALE)', sigma_cb_highmass_2tag, EG_RES, EG_SCALE)\")\n",
    "\n",
    "    \n",
    "dscb_pdf_highmass_0tag = ROOT.HggTwoSidedCBPdf(\n",
    "    'dscb_pdf_highmass_0tag', 'Double Sided Crystal Ball',\n",
    "    myy, w.function(\"m0_highmass_0tag\"), w.function(\"sigma_highmass_0tag\"),\n",
    "    w.var(\"alphaLo_cb_highmass\"), w.var(\"nLo_cb_highmass\"), w.var(\"alphaHi_cb_highmass\"), w.var(\"nHi_cb_highmass\")\n",
    ") # AGAIN WE DON'T HAVE THE VARIATIONS FOR THE 0 TAG CATEGORY, SO I'M GOING TO USE THE ONE FROM 1 TAG FOR NOW\n",
    "dscb_pdf_highmass_1tag = ROOT.HggTwoSidedCBPdf(\n",
    "    'dscb_pdf_highmass_1tag', 'Double Sided Crystal Ball',\n",
    "    myy, w.function(\"m0_highmass_1tag\"), w.function(\"sigma_highmass_1tag\"),\n",
    "    w.var(\"alphaLo_cb_highmass\"),w.var(\"nLo_cb_highmass\"),w.var(\"alphaHi_cb_highmass\"),w.var(\"nHi_cb_highmass\")\n",
    ")\n",
    "dscb_pdf_highmass_2tag = ROOT.HggTwoSidedCBPdf(\n",
    "    'dscb_pdf_highmass_2tag', 'Double Sided Crystal Ball',\n",
    "    myy, w.function(\"m0_highmass_2tag\"), w.function(\"sigma_highmass_2tag\"),\n",
    "    w.var(\"alphaLo_cb_highmass\"),w.var(\"nLo_cb_highmass\"),w.var(\"alphaHi_cb_highmass\"),w.var(\"nHi_cb_highmass\")\n",
    ")\n",
    "getattr(w, 'import')(ROOT.RooArgSet(dscb_pdf_highmass_0tag, dscb_pdf_highmass_1tag, dscb_pdf_highmass_2tag))\n",
    "[w.var(a.GetName() + \"_highmass\").setVal(a.getVal()) for a in wrap_iter(w.set(\"params_dscb\").iterator())]\n",
    "[w.var(a.GetName() + \"_highmass\").setConstant(True) for a in wrap_iter(w.set(\"params_dscb\").iterator())]\n",
    "\n",
    "w.factory(\"expr::n_highMass_1tag_12_Higgs('n_highMass_1tag_single_Higgs + n_highMass_1tag_di_Higgs', n_highMass_1tag_single_Higgs, n_highMass_1tag_di_Higgs)\")\n",
    "w.factory(\"expr::n_highMass_2tag_12_Higgs('n_highMass_2tag_single_Higgs + n_highMass_2tag_di_Higgs', n_highMass_2tag_single_Higgs, n_highMass_2tag_di_Higgs)\")\n",
    "\n",
    "w.factory(\"SUM:pdf_highMass_0tag(\\\n",
    "   n_highMass_0tag_non_Higgs_exp * bkg_pdf_highmass,\\\n",
    "   n_highMass_0tag_single_Higgs * dscb_pdf_highmass_0tag)\"\n",
    ")\n",
    "w.factory(\"SUM:pdf_highMass_1tag(\\\n",
    "   n_highMass_1tag_non_Higgs_exp * bkg_pdf_highmass,\\\n",
    "   n_highMass_1tag_12_Higgs * dscb_pdf_highmass_1tag)\"\n",
    ")\n",
    "w.factory(\"SUM:pdf_highMass_2tag(\\\n",
    "   n_highMass_2tag_non_Higgs_exp * bkg_pdf_highmass,\\\n",
    "   n_highMass_2tag_12_Higgs * dscb_pdf_highmass_2tag)\"\n",
    ")\n",
    "\n",
    "# dscb_pdf_tag0high = ROOT.HggTwoSidedCBPdf(\n",
    "#     'dscb_pdf_tag0high', 'Double Sided Crystal Ball',\n",
    "#     myy,m0_cb_tag0high,sigma_cb_tag0high,alphaLo_cb_highmass,nLo_cb_highmass,alphaHi_cb_highmass,nHi_cb_highmass\n",
    "# )\n",
    "# dscb_pdf_tag1high = ROOT.HggTwoSidedCBPdf(\n",
    "#     'dscb_pdf_tag1high', 'Double Sided Crystal Ball',\n",
    "#     myy,m0_cb_tag1high,sigma_cb_tag1high,alphaLo_cb_highmass,nLo_cb_highmass,alphaHi_cb_highmass,nHi_cb_highmass\n",
    "# )\n",
    "# dscb_pdf_tag2high = ROOT.HggTwoSidedCBPdf(\n",
    "#     'dscb_pdf_tag2high', 'Double Sided Crystal Ball',\n",
    "#     myy,m0_cb_tag2high,sigma_cb_tag2high,alphaLo_cb_highmass,nLo_cb_highmass,alphaHi_cb_highmass,nHi_cb_highmass\n",
    "# )\n",
    "# getattr(w, 'import')(ROOT.RooArgSet(dscb_pdf_tag0high, dscb_pdf_tag1high, dscb_pdf_tag2high))\n",
    "# [w.var(a.GetName() + \"_highmass\").setVal(a.getVal()) for a in wrap_iter(w.set(\"params_dscb\").iterator())]\n",
    "# [w.var(a.GetName() + \"_highmass\").setConstant(True) for a in wrap_iter(w.set(\"params_dscb\").iterator())]\n",
    "\n",
    "# w.factory(\"expr::n_tag1high_12H('n_tag1high_1H + n_tag1high_2H', n_tag1high_1H, n_tag1high_2H)\")\n",
    "# w.factory(\"expr::n_tag2high_12H('n_tag2high_1H + n_tag2high_2H', n_tag2high_1H, n_tag2high_2H)\")\n",
    "\n",
    "# w.factory(\"SUM:pdf_tag0high(\\\n",
    "#    n_tag0high_0H[1000, 0, 200000] * bkg_pdf_highmass,\\\n",
    "#    n_tag0high_1H * dscb_pdf_tag0high)\"\n",
    "# )\n",
    "# w.factory(\"SUM:pdf_tag1high(\\\n",
    "#    n_tag1high_0H[1000, 0, 200000] * bkg_pdf_highmass,\\\n",
    "#    n_tag1high_12H * dscb_pdf_tag1high)\"\n",
    "# )\n",
    "# w.factory(\"SUM:pdf_tag2high(\\\n",
    "#    n_tag2high_0H[1000, 0, 200000] * bkg_pdf_highmass,\\\n",
    "#    n_tag2high_12H * dscb_pdf_tag2high)\"\n",
    "# )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# w.factory(\"SUM:pdf_tag0high(\\\n",
    "#    n_tag0high_0H[111901, 100000, 120000] * bkg_pdf_highmass,\\\n",
    "#    n_tag0high_1H[15980, 15000, 17000] * dscb_pdf_tag0high)\"\n",
    "# )\n",
    "# w.factory(\"SUM:pdf_tag1high(\\\n",
    "#    n_tag1high_0H[6568, 5500, 7500] * bkg_pdf_highmass,\\\n",
    "#    n_tag1high_1H[2260, 2000, 3000] * dscb_pdf_tag1high)\"\n",
    "# )\n",
    "# w.factory(\"SUM:pdf_tag2high(\\\n",
    "#    n_tag2high_0H[949, 800, 1000] * bkg_pdf_highmass,\\\n",
    "#    n_tag2high_1H[1135, 1000, 1200] * dscb_pdf_tag2high)\"\n",
    "# )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<ROOT.RooSimultaneous object (\"jointmodel\") at 0x7fc49fb4c800>"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# w.factory(\"SIMUL:jointmodel(\\\n",
    "#     cat,\\\n",
    "#         tag0low = PROD:constr_tag0low(pdf_tag0low, EG_SCALE_gaussian, EG_RES_gaussian, theta_yield_gaussian),\\\n",
    "#         tag0high = PROD:constr_tag0high(pdf_tag0high, EG_SCALE_gaussian, EG_RES_gaussian, theta_yield_gaussian),\\\n",
    "#         tag1low = PROD:constr_tag1low(pdf_tag1low, EG_SCALE_gaussian, EG_RES_gaussian, theta_yield_gaussian, n_bias_tag1low_2H_gaussian),\\\n",
    "#         tag1high = PROD:constr_tag1high(pdf_tag1high, EG_SCALE_gaussian, EG_RES_gaussian, theta_yield_gaussian, n_bias_tag1high_2H_gaussian),\\\n",
    "#         tag2low = PROD:constr_tag2low(pdf_tag2low, EG_SCALE_gaussian, EG_RES_gaussian, theta_yield_gaussian, n_bias_tag2low_2H_gaussian),\\\n",
    "#         tag2high = PROD:constr_tag2high(pdf_tag2high, EG_SCALE_gaussian, EG_RES_gaussian, theta_yield_gaussian, n_bias_tag2high_2H_gaussian))\"\n",
    "# )\n",
    "w.factory(\"SIMUL:jointmodel(\\\n",
    "    highmasscats,\\\n",
    "        tag0 = pdf_highMass_0tag,\\\n",
    "        tag1 = PROD:constr_highMass_1tag(pdf_highMass_1tag, n_highMass_1tag_di_Higgs_bias_gaussian),\\\n",
    "        tag2 = PROD:constr_highMass_2tag(pdf_highMass_2tag, n_highMass_2tag_di_Higgs_bias_gaussian))\"\n",
    ")\n",
    "# #ExtendPdf:pdf(gaus, nevt[100,0,100000])\n",
    "# # Simultaneous p.d.f.s are constructed with SIMUL( index, state1=pdf1, state2=pdf2,...)\n",
    "# # w->factory(\"SIMUL::smodel( c[A=0,B=1], A=Gaussian::gs(x,m,s[1]), B=Landau::ls(x,0,1) )\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<ROOT.RooProdPdf object (\"constrained_jointmodel\") at 0x7fc49fb78800>"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "w.factory(\"PROD::constrained_jointmodel(jointmodel,\\\n",
    "    EG_SCALE_gaussian, EG_RES_gaussian, theta_yield_gaussian)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "#data_combo.table(w.cat(\"cat\")).Print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:Minization -- p.d.f. provides expected number of events, including extended term in likelihood.\n",
      "[#1] INFO:NumericIntegration -- RooRealIntegral::init(bkg_pdf_highmass_Int[myy]) using numeric integrator RooIntegrator1D to calculate Int(myy)\n",
      "[#1] INFO:Minization --  Including the following contraint terms in minimization: (EG_SCALE_gaussian,EG_RES_gaussian,theta_yield_gaussian,n_highMass_1tag_di_Higgs_bias_gaussian,n_highMass_2tag_di_Higgs_bias_gaussian)\n",
      "[#1] INFO:Fitting -- RooAddition::defaultErrorLevel(nll_constrained_jointmodel_highmass_data_with_constr) Summation contains a RooNLLVar, using its error level\n",
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: activating const optimization\n",
      "[#1] INFO:NumericIntegration -- RooRealIntegral::init(bkg_pdf_highmass_Int[myy]) using numeric integrator RooIntegrator1D to calculate Int(myy)\n",
      "[#1] INFO:NumericIntegration -- RooRealIntegral::init(bkg_pdf_highmass_Int[myy]) using numeric integrator RooIntegrator1D to calculate Int(myy)\n",
      "[#1] INFO:Minization --  The following expressions have been identified as constant and will be precalculated and cached: (bkg_pdf_highmass)\n",
      "[#1] INFO:Minization --  The following expressions will be evaluated in cache-and-track mode: (jointmodel,dscb_pdf_highmass_0tag,dscb_pdf_highmass_1tag,dscb_pdf_highmass_2tag)\n",
      " **********\n",
      " **    1 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **    2 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       0.00000e+00  2.00000e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE     0.00000e+00  2.00000e+00   -1.00000e+01  1.00000e+01\n",
      "     3 mu_2H        1.00000e+00  5.00000e-01    0.00000e+00  2.00000e+01\n",
      "     4 n_highMass_1tag_di_Higgs_bias   0.00000e+00  2.00000e+01   -1.00000e+02  1.00000e+02\n",
      "     5 n_highMass_2tag_di_Higgs_bias   0.00000e+00  2.00000e+01   -1.00000e+02  1.00000e+02\n",
      "     6 theta_yield   0.00000e+00  2.00000e+00   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **    3 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **    4 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **    5 **SET STR           1\n",
      " **********\n",
      " NOW USING STRATEGY  1: TRY TO BALANCE SPEED AGAINST RELIABILITY\n",
      " **********\n",
      " **    6 **MIGRAD        3000           1\n",
      " **********\n",
      " FIRST CALL TO USER FUNCTION AT NEW START POINT, WITH IFLAG=4.\n",
      " START MIGRAD MINIMIZATION.  STRATEGY  1.  CONVERGENCE WHEN EDM .LT. 1.00e-03\n",
      " FCN=-2160.79 FROM MIGRAD    STATUS=INITIATE       26 CALLS          27 TOTAL\n",
      "                     EDM= unknown      STRATEGY= 1      NO ERROR MATRIX       \n",
      "  EXT PARAMETER               CURRENT GUESS       STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       0.00000e+00   2.00000e+00   2.01358e-01  -3.08352e-01\n",
      "   2  EG_SCALE     0.00000e+00   2.00000e+00   2.01358e-01   1.64307e+01\n",
      "   3  mu_2H        1.00000e+00   5.00000e-01   1.18625e-01   8.94602e-02\n",
      "   4  n_highMass_1tag_di_Higgs_bias   0.00000e+00   2.00000e+01   2.01358e-01  -5.96776e-01\n",
      "   5  n_highMass_2tag_di_Higgs_bias   0.00000e+00   2.00000e+01   2.01358e-01   1.11911e+01\n",
      "   6  theta_yield   0.00000e+00   2.00000e+00   2.01358e-01   1.28714e-01\n",
      "                               ERR DEF= 0.5\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=-2160.82 FROM MIGRAD    STATUS=CONVERGED     129 CALLS         130 TOTAL\n",
      "                     EDM=3.52105e-05    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       2.54608e-02   1.03731e+00   3.30548e-03  -2.81901e-02\n",
      "   2  EG_SCALE    -1.77933e-02   1.00852e-01   3.20544e-04  -2.32885e-01\n",
      "   3  mu_2H        2.11975e-03   1.64496e+01   1.28640e-01   9.62159e-04\n",
      "   4  n_highMass_1tag_di_Higgs_bias   9.50640e-03   7.78269e-01   2.49715e-04  -2.01306e-01\n",
      "   5  n_highMass_2tag_di_Higgs_bias  -2.25964e-02   7.44540e-01   2.36702e-04   8.40434e-02\n",
      "   6  theta_yield  -8.07127e-03   9.97463e-01   3.20729e-03   4.94341e-03\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  6    ERR DEF=0.5\n",
      "  1.080e+00 -7.195e-04  3.891e-03  1.907e-03  2.019e-03  2.945e-03 \n",
      " -7.195e-04  1.017e-02 -1.281e-03  2.813e-05 -1.093e-03 -4.446e-06 \n",
      "  3.891e-03 -1.281e-03  8.351e-01 -2.430e-02 -9.739e-02 -7.610e-03 \n",
      "  1.907e-03  2.813e-05 -2.430e-02  6.057e-01  2.850e-03 -3.296e-03 \n",
      "  2.019e-03 -1.093e-03 -9.739e-02  2.850e-03  5.544e-01 -2.137e-03 \n",
      "  2.945e-03 -4.446e-06 -7.610e-03 -3.296e-03 -2.137e-03  9.983e-01 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2      3      4      5      6\n",
      "        1  0.00935   1.000 -0.007  0.004  0.002  0.003  0.003\n",
      "        2  0.02277  -0.007  1.000 -0.014  0.000 -0.015 -0.000\n",
      "        3  0.14819   0.004 -0.014  1.000 -0.034 -0.143 -0.008\n",
      "        4  0.03456   0.002  0.000 -0.034  1.000  0.005 -0.004\n",
      "        5  0.14418   0.003 -0.015 -0.143  0.005  1.000 -0.003\n",
      "        6  0.01074   0.003 -0.000 -0.008 -0.004 -0.003  1.000\n",
      " **********\n",
      " **    7 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **    8 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **    9 **HESSE        3000\n",
      " **********\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=-2160.82 FROM HESSE     STATUS=OK             44 CALLS         174 TOTAL\n",
      "                     EDM=2.59604e-05    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  EG_RES       2.54608e-02   1.05706e+00   1.32219e-04   2.54608e-03\n",
      "   2  EG_SCALE    -1.77933e-02   1.00848e-01   6.41088e-05  -1.77933e-03\n",
      "   3  mu_2H        2.11975e-03   1.99942e+01   5.00000e-01  -1.55021e+00\n",
      "   4  n_highMass_1tag_di_Higgs_bias   9.50640e-03   7.79939e-01   4.99430e-05   9.50640e-05\n",
      "   5  n_highMass_2tag_di_Higgs_bias  -2.25964e-02   7.68433e-01   4.73405e-05  -2.25964e-04\n",
      "   6  theta_yield  -8.07127e-03   9.97825e-01   6.41458e-04  -8.07127e-04\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  6    ERR DEF=0.5\n",
      "  1.122e+00  4.891e-04  9.601e-03  1.175e-03 -9.313e-04  2.639e-03 \n",
      "  4.891e-04  1.017e-02 -2.404e-04  1.450e-05 -2.274e-04  2.818e-05 \n",
      "  9.601e-03 -2.404e-04  4.123e-01 -3.694e-02 -1.402e-01 -1.815e-02 \n",
      "  1.175e-03  1.450e-05 -3.694e-02  6.083e-01  1.257e-02 -1.898e-03 \n",
      " -9.313e-04 -2.274e-04 -1.402e-01  1.257e-02  5.905e-01  3.125e-03 \n",
      "  2.639e-03  2.818e-05 -1.815e-02 -1.898e-03  3.125e-03  9.990e-01 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2      3      4      5      6\n",
      "        1  0.01564   1.000  0.005  0.014  0.001 -0.001  0.002\n",
      "        2  0.00726   0.005  1.000 -0.004  0.000 -0.003  0.000\n",
      "        3  0.29368   0.014 -0.004  1.000 -0.074 -0.284 -0.028\n",
      "        4  0.07393   0.001  0.000 -0.074  1.000  0.021 -0.002\n",
      "        5  0.28413  -0.001 -0.003 -0.284  0.021  1.000  0.004\n",
      "        6  0.02909   0.002  0.000 -0.028 -0.002  0.004  1.000\n",
      "[#1] INFO:Fitting -- RooAbsPdf::fitTo(constrained_jointmodel) Calculating sum-of-weights-squared correction matrix for covariance matrix\n",
      " **********\n",
      " **   10 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   11 **SET PRINT           1\n",
      " **********\n",
      " **********\n",
      " **   12 **HESSE        3000\n",
      " **********\n",
      " MINUIT WARNING IN HESSE   \n",
      " ============== Negative diagonal element 3 in Error Matrix\n",
      " MINUIT WARNING IN HESSE   \n",
      " ============== 1.2157 added to diagonal of error matrix\n",
      " COVARIANCE MATRIX CALCULATED SUCCESSFULLY\n",
      " FCN=-615.851 FROM HESSE     STATUS=OK             58 CALLS         232 TOTAL\n",
      "                     EDM=0.412806    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                INTERNAL      INTERNAL  \n",
      "  NO.   NAME      VALUE            ERROR       STEP SIZE       VALUE   \n",
      "   1  EG_RES       2.54608e-02   1.00391e+00   2.69727e-03   2.54608e-03\n",
      "   2  EG_SCALE    -1.77933e-02   7.78868e-01   3.98320e-02  -1.77933e-03\n",
      "   3  mu_2H        2.11975e-03   1.22535e+01   5.00000e-01  -1.55021e+00\n",
      "   4  n_highMass_1tag_di_Higgs_bias   9.50640e-03   7.94499e-01   9.98861e-06   9.50640e-05\n",
      "   5  n_highMass_2tag_di_Higgs_bias  -2.25964e-02   7.88398e-01   9.65745e-04  -2.25964e-04\n",
      "   6  theta_yield  -8.07127e-03   9.92116e-01   1.28292e-04  -8.07127e-04\n",
      "                               ERR DEF= 0.5\n",
      " EXTERNAL ERROR MATRIX.    NDIM=  25    NPAR=  6    ERR DEF=0.5\n",
      "  1.011e+00 -1.047e-02  1.055e-03  1.874e-04  5.658e-03  1.239e-03 \n",
      " -1.047e-02  6.079e-01 -2.179e-02 -1.295e-03 -1.304e-01 -4.151e-03 \n",
      "  1.055e-03 -2.179e-02  4.325e-02 -9.059e-04 -2.178e-03  8.538e-04 \n",
      "  1.874e-04 -1.295e-03 -9.059e-04  6.312e-01  4.332e-04 -8.789e-04 \n",
      "  5.658e-03 -1.304e-01 -2.178e-03  4.332e-04  6.216e-01 -6.486e-04 \n",
      "  1.239e-03 -4.151e-03  8.538e-04 -8.789e-04 -6.486e-04  9.875e-01 \n",
      " PARAMETER  CORRELATION COEFFICIENTS  \n",
      "       NO.  GLOBAL      1      2      3      4      5      6\n",
      "        1  0.01453   1.000 -0.013  0.005  0.000  0.007  0.001\n",
      "        2  0.25289  -0.013  1.000 -0.134 -0.002 -0.212 -0.005\n",
      "        3  0.14122   0.005 -0.134  1.000 -0.005 -0.013  0.004\n",
      "        4  0.00628   0.000 -0.002 -0.005  1.000  0.001 -0.001\n",
      "        5  0.21627   0.007 -0.212 -0.013  0.001  1.000 -0.001\n",
      "        6  0.00683   0.001 -0.005  0.004 -0.001 -0.001  1.000\n",
      "[#1] INFO:Minization -- RooMinimizer::optimizeConst: deactivating const optimization\n",
      "\n",
      "  RooFitResult: minimized FCN value: -615.851, estimated distance to minimum: 0.412806\n",
      "                covariance matrix quality: Unknown, matrix was externally provided\n",
      "                Status : MINIMIZE=0 HESSE=0 HESSE=0 \n",
      "\n",
      "    Floating Parameter    FinalValue +/-  Error   \n",
      "  --------------------  --------------------------\n",
      "                EG_RES    2.5461e-02 +/-  1.12e+00\n",
      "              EG_SCALE   -1.7793e-02 +/-  1.33e-02\n",
      "                 mu_2H    2.1197e-03 +/-  2.00e+00\n",
      "  n_highMass_1tag_di_Higgs_bias    9.5064e-03 +/-  7.85e-01\n",
      "  n_highMass_2tag_di_Higgs_bias   -2.2596e-02 +/-  1.00e+00\n",
      "           theta_yield   -8.0713e-03 +/-  1.01e+00\n",
      "\n"
     ]
    }
   ],
   "source": [
    "r = w.pdf(\"constrained_jointmodel\").fitTo(highmass_data,\n",
    "                                          rf.Save(), rf.SumW2Error(ROOT.kTRUE))\n",
    "r.Print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# ^ not working because it probably doesn't like the input values for n I gave it. Plan to change them to realistic values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:InputArguments -- RooAbsData::plotOn(highmass_data) INFO: dataset has non-integer weights, auto-selecting SumW2 errors instead of Poisson errors\n",
      "[#1] INFO:Plotting -- RooTreeData::plotOn: plotting 1110.06 events out of 1222.14 total events\n",
      "[#1] INFO:Plotting -- RooAbsReal::plotOn(constrained_jointmodel) plot on myy represents a slice in (highmasscats)\n",
      "[#1] INFO:NumericIntegration -- RooRealIntegral::init(bkg_pdf_highmass_Int[myy]) using numeric integrator RooIntegrator1D to calculate Int(myy)\n",
      "[#1] INFO:InputArguments -- RooAbsData::plotOn(highmass_data) INFO: dataset has non-integer weights, auto-selecting SumW2 errors instead of Poisson errors\n",
      "[#1] INFO:Plotting -- RooTreeData::plotOn: plotting 84.5792 events out of 1222.14 total events\n",
      "[#1] INFO:Plotting -- RooAbsReal::plotOn(constrained_jointmodel) plot on myy represents a slice in (highmasscats)\n",
      "[#1] INFO:NumericIntegration -- RooRealIntegral::init(bkg_pdf_highmass_Int[myy]) using numeric integrator RooIntegrator1D to calculate Int(myy)\n",
      "[#1] INFO:InputArguments -- RooAbsData::plotOn(highmass_data) INFO: dataset has non-integer weights, auto-selecting SumW2 errors instead of Poisson errors\n",
      "[#1] INFO:Plotting -- RooTreeData::plotOn: plotting 27.4984 events out of 1222.14 total events\n",
      "[#1] INFO:Plotting -- RooAbsReal::plotOn(constrained_jointmodel) plot on myy represents a slice in (highmasscats)\n",
      "[#1] INFO:NumericIntegration -- RooRealIntegral::init(bkg_pdf_highmass_Int[myy]) using numeric integrator RooIntegrator1D to calculate Int(myy)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArgAAAHYCAIAAAApvgy/AAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE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\noAAAALQIFAAAgNb/B09VWSdnonNEAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArgAAAHYCAIAAAApvgy/AAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE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\n4AAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArgAAAHYCAIAAAApvgy/AAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE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\nAAAASUVORK5CYII=\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# c1 = ROOT.TCanvas()\n",
    "# plot1 = myy.frame(ROOT.RooFit.Title(\"0 Tag - Low Mass\"))\n",
    "# data_combo.plotOn(plot1,ROOT.RooFit.Cut(\"cat==cat::tag0low\"))\n",
    "# w.pdf(\"jointmodel\").plotOn(plot1, ROOT.RooFit.ProjWData(data_combo),ROOT.RooFit.Slice(w.cat(\"cat\"),\"tag0low\"))\n",
    "# plot1.Draw()\n",
    "# c1.Draw()\n",
    "\n",
    "c2 = ROOT.TCanvas()\n",
    "plot2 = myy.frame(ROOT.RooFit.Title(\"0 Tag - High Mass\"))\n",
    "highmass_data.plotOn(plot2,ROOT.RooFit.Cut(\"highmasscats==highmasscats::tag0\"))\n",
    "w.pdf(\"constrained_jointmodel\").plotOn(\n",
    "    plot2, ROOT.RooFit.ProjWData(highmass_data),\n",
    "    ROOT.RooFit.Slice(w.cat(\"highmasscats\"),\"tag0\"))\n",
    "plot2.Draw()\n",
    "c2.Draw()\n",
    "\n",
    "# c3 = ROOT.TCanvas()\n",
    "# plot3 = myy.frame(ROOT.RooFit.Title(\"1 Tag - Low Mass\"))\n",
    "# data_combo.plotOn(plot3,ROOT.RooFit.Cut(\"cat==cat::tag1low\"))\n",
    "# w.pdf(\"jointmodel\").plotOn(plot3, ROOT.RooFit.ProjWData(data_combo),ROOT.RooFit.Slice(w.cat(\"cat\"),\"tag1low\"))\n",
    "# plot3.Draw()\n",
    "# c3.Draw()\n",
    "\n",
    "c4 = ROOT.TCanvas()\n",
    "plot4 = myy.frame(ROOT.RooFit.Title(\"1 Tag - High Mass\"))\n",
    "highmass_data.plotOn(plot4,ROOT.RooFit.Cut(\"highmasscats==highmasscats::tag1\"))\n",
    "w.pdf(\"constrained_jointmodel\").plotOn(\n",
    "    plot4, ROOT.RooFit.ProjWData(highmass_data),\n",
    "    ROOT.RooFit.Slice(w.cat(\"highmasscats\"),\"tag1\"))\n",
    "plot4.Draw()\n",
    "c4.Draw()\n",
    "\n",
    "# c5 = ROOT.TCanvas()\n",
    "# plot5 = myy.frame(ROOT.RooFit.Title(\"2 Tag - Low Mass\"))\n",
    "# data_combo.plotOn(plot5,ROOT.RooFit.Cut(\"cat==cat::tag2low\"))\n",
    "# w.pdf(\"jointmodel\").plotOn(plot5, ROOT.RooFit.ProjWData(data_combo),ROOT.RooFit.Slice(w.cat(\"cat\"),\"tag2low\"))\n",
    "# plot5.Draw()\n",
    "# c5.Draw()\n",
    "\n",
    "c6 = ROOT.TCanvas()\n",
    "plot6 = myy.frame(ROOT.RooFit.Title(\"2 Tag - High Mass\"))\n",
    "highmass_data.plotOn(plot6,ROOT.RooFit.Cut(\"highmasscats==highmasscats::tag2\"))\n",
    "w.pdf(\"constrained_jointmodel\").plotOn(\n",
    "    plot6, ROOT.RooFit.ProjWData(highmass_data),\n",
    "    ROOT.RooFit.Slice(w.cat(\"highmasscats\"),\"tag2\"))\n",
    "plot6.Draw()\n",
    "c6.Draw()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "Okay whatever, we don't actually expect these fits to work without setting realistic expectations for the number of events, the boundaries on the resolution parameter etc."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "collapsed": true,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "import ROOT.RooStats as rs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "mconfig = rs.ModelConfig(w)\n",
    "mconfig.SetPdf(w.pdf(\"constrained_jointmodel\"))\n",
    "mconfig.SetParametersOfInterest(ROOT.RooArgSet(w.var(\"mu_2H\")))\n",
    "mconfig.SetObservables(ROOT.RooArgSet(w.var(\"myy\")))\n",
    "mconfig.SetNuisanceParameters(ROOT.RooArgSet(w.var(\"theta_yield\"), w.var(\"EG_RES\"), w.var(\"EG_SCALE\"),\n",
    "                                             w.var(\"n_highMass_2tag_di_Higgs_bias\"),  #w.var(\"n_bias_tag2low_2H\"),\n",
    "                                            w.var(\"n_highMass_1tag_di_Higgs_bias\") #, w.var(\"n_bias_tag1low_2H\")\n",
    "                                            ))\n",
    "mconfig.SetGlobalObservables(ROOT.RooArgSet(w.var(\"theta_yield_global\"), w.var(\"EG_SCALE_global\"), w.var(\"EG_RES_global\"),\n",
    "                                            w.var(\"n_highMass_2tag_di_Higgs_bias_global\"), # w.var(\"n_bias_tag2low_2H_global\"),\n",
    "                                            w.var(\"n_highMass_1tag_di_Higgs_bias_global\")#,  w.var(\"n_bias_tag1low_2H_global\")\n",
    "                                           ))\n",
    "# mconfig.SetNuisanceParameters(ROOT.RooArgSet(w.var(\"theta_yield\"), w.var(\"EG_RES\"), w.var(\"EG_SCALE\"),\n",
    "#                                              w.var(\"n_bias_tag2high_2H\"),\n",
    "#                                             w.var(\"n_bias_tag1high_2H\")))\n",
    "# mconfig.SetGlobalObservables(ROOT.RooArgSet(w.var(\"theta_yield_global\"), w.var(\"EG_SCALE_global\"), w.var(\"EG_RES_global\"),\n",
    "#                                             w.var(\"n_bias_tag2high_2H_global\"),\n",
    "#                                             w.var(\"n_bias_tag1high_2H_global\")\n",
    "#                                            ))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "=== Using the following for model ===\n",
      "Observables:             RooArgSet:: = (myy)\n",
      "Parameters of Interest:  RooArgSet:: = (mu_2H)\n",
      "Nuisance Parameters:     RooArgSet:: = (theta_yield,EG_RES,EG_SCALE,n_highMass_2tag_di_Higgs_bias,n_highMass_1tag_di_Higgs_bias)\n",
      "Global Observables:      RooArgSet:: = (theta_yield_global,EG_SCALE_global,EG_RES_global,n_highMass_2tag_di_Higgs_bias_global,n_highMass_1tag_di_Higgs_bias_global)\n",
      "PDF:                     RooProdPdf::constrained_jointmodel[ jointmodel * EG_SCALE_gaussian * EG_RES_gaussian * theta_yield_gaussian ] = 0.00804255\n",
      "\n"
     ]
    }
   ],
   "source": [
    "mconfig.SetName('model')\n",
    "mconfig.Print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "sb_model = mconfig\n",
    "poi = sb_model.GetParametersOfInterest().first()\n",
    "sb_model.SetSnapshot(ROOT.RooArgSet(poi))\n",
    "b_model = sb_model.Clone()\n",
    "b_model.SetName(sb_model.GetName()+\"_with_poi_0\")\n",
    "oldval = poi.getVal()\n",
    "poi.setVal(0)\n",
    "b_model.SetSnapshot(ROOT.RooArgSet(poi))\n",
    "poi.setVal(oldval)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::Initialize....\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::Initialize - Find  best unconditional NLL on observed data\n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **   13 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   14 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       2.54608e-02  1.11626e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.77933e-02  1.33245e-02   -1.00000e+01  1.00000e+01\n",
      "     3 mu_2H        2.11975e-03  2.00319e+00    0.00000e+00  2.00000e+01\n",
      " MINUIT WARNING IN PARAMETR\n",
      " ============== VARIABLE3 BROUGHT BACK INSIDE LIMITS.\n",
      "     4 n_highMass_1tag_di_Higgs_bias   9.50640e-03  7.85397e-01   -1.00000e+02  1.00000e+02\n",
      "     5 n_highMass_2tag_di_Higgs_bias  -2.25964e-02  1.00347e+00   -1.00000e+02  1.00000e+02\n",
      "     6 theta_yield  -8.07127e-03  1.00944e+00   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **   15 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   16 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   17 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **   18 **MIGRAD        3000           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=-2160.82 FROM MIGRAD    STATUS=CONVERGED      81 CALLS          82 TOTAL\n",
      "                     EDM=4.32489e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       2.78863e-02   1.03401e+00   3.33527e-03  -5.94709e-03\n",
      "   2  EG_SCALE    -1.75633e-02   1.00838e-01   3.23687e-04  -7.41201e-03\n",
      "   3  mu_2H        5.41440e-06   1.64788e+01   1.43247e-01  -4.82546e-05\n",
      "   4  n_highMass_1tag_di_Higgs_bias   1.07253e-02   7.78131e-01   2.49701e-04  -3.29300e-03\n",
      "   5  n_highMass_2tag_di_Higgs_bias  -2.30431e-02   7.42143e-01   2.36524e-04  -9.91745e-03\n",
      "   6  theta_yield  -8.56473e-03   9.97452e-01   3.20730e-03  -7.42777e-05\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = -2160.82\tfit time : Real time 0:00:00, CP time 0.010\n",
      "[#0] PROGRESS:Eval -- Best fitted POI value = 5.4144e-06 +/- 16.4788\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator: Building Asimov data Set\n",
      "[#0] WARNING:InputArguments -- AsymptoticCalculator: number of bins in myy are different than data bins  set the same data bins 66 in range  [ 105 , 160 ]\n",
      "[#1] INFO:InputArguments -- AsymptoticCalculator: Asimov data will be generated using fitted nuisance parameter values\n",
      "MakeAsimov: Setting poi mu_2H to a constant value = 0\n",
      "MakeAsimov: doing a conditional fit for finding best nuisance values \n",
      " **********\n",
      " **    1 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **    2 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       2.78863e-02  1.03401e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.75633e-02  1.00838e-01   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias   1.07253e-02  7.78131e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -2.30431e-02  7.42143e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -8.56473e-03  9.97452e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **    3 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **    4 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **    5 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **    6 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=-2160.82 FROM MIGRAD    STATUS=CONVERGED      60 CALLS          61 TOTAL\n",
      "                     EDM=2.37452e-08    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       2.84423e-02   1.03335e+00   3.32403e-03  -5.91937e-04\n",
      "   2  EG_SCALE    -1.75540e-02   1.00811e-01   3.23582e-04   2.27458e-03\n",
      "   3  n_highMass_1tag_di_Higgs_bias   1.07786e-02   7.77822e-01   2.49703e-04   5.33396e-03\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -2.28900e-02   7.36910e-01   2.36522e-04   1.81036e-02\n",
      "   5  theta_yield  -8.54550e-03   9.97429e-01   3.20730e-03   1.14490e-04\n",
      "                               ERR DEF= 0.5\n",
      "fit time Real time 0:00:00, CP time 0.020\n",
      "RooDataSet::AsimovData[myy,weight:binWeightAsimov] = 66 entries (0.617688 weighted)\n",
      "Generated Asimov data for observables RooArgSet:: = (myy)\n",
      "Generated Asimov data for global observables RooArgSet:: = (theta_yield_global,EG_SCALE_global,EG_RES_global,n_highMass_2tag_di_Higgs_bias_global,n_highMass_1tag_di_Higgs_bias_global)\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::Initialize Find  best conditional NLL on ASIMOV data set for given alt POI ( mu_2H ) = 0\n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **    7 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **    8 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       2.84423e-02  1.03335e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.75540e-02  1.00811e-01   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias   1.07786e-02  7.77822e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -2.28900e-02  7.36910e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -8.54550e-03  9.97429e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **    9 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   10 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   11 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **   12 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=34.0874 FROM MIGRAD    STATUS=CONVERGED      66 CALLS          67 TOTAL\n",
      "                     EDM=5.85173e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       2.84430e-02   9.98334e-01   4.06108e-04  -3.01758e-04\n",
      "   2  EG_SCALE    -1.78128e-02   1.28458e+00   5.38653e-04  -5.72970e-04\n",
      "   3  n_highMass_1tag_di_Higgs_bias   1.04551e-02   7.99992e-01   3.24888e-05   2.56162e-05\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -6.31669e-01   7.97300e-01   3.24153e-05  -9.45447e-02\n",
      "   5  theta_yield  -2.80467e-02   9.98332e-01   4.06109e-04  -1.05892e-03\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = 34.0874 for poi fixed at = 0\tfit time : Real time 0:00:00, CP time 0.010\n"
     ]
    }
   ],
   "source": [
    "ac = rs.AsymptoticCalculator(highmass_data, b_model, sb_model)\n",
    "ac.SetOneSided(True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "[#1] INFO:Eval -- AsymptoticCalculator::GetHypoTest: - perform  an hypothesis test for  POI ( mu_2H ) = 0.00211975\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -  Find  best conditional NLL on OBSERVED data set ..... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **   13 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   14 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       2.78863e-02  1.03401e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.75633e-02  1.00838e-01   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias   1.07253e-02  7.78131e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -2.30431e-02  7.42143e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -8.56473e-03  9.97452e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **   15 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   16 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   17 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **   18 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=-2160.82 FROM MIGRAD    STATUS=CONVERGED      61 CALLS          62 TOTAL\n",
      "                     EDM=2.86156e-08    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       2.84678e-02   1.03335e+00   3.32528e-03  -3.75087e-04\n",
      "   2  EG_SCALE    -1.75649e-02   1.00793e-01   3.23543e-04  -8.28482e-03\n",
      "   3  n_highMass_1tag_di_Higgs_bias   1.08219e-02   7.77823e-01   2.49703e-04   1.49875e-02\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -2.29967e-02   7.36919e-01   2.36526e-04   9.64711e-03\n",
      "   5  theta_yield  -8.52386e-03   9.97429e-01   3.20730e-03   3.75971e-04\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = -2160.82 for poi fixed at = 0.00211975\tfit time : Real time 0:00:00, CP time 0.010\n",
      "[#0] PROGRESS:Eval -- \t OBSERVED DATA :  qmu   = 1.93051e-05 condNLL = -2160.82 uncond -2160.82\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -- Find  best conditional NLL on ASIMOV data set .... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **   19 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   20 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       2.84678e-02  1.03335e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.75649e-02  1.00793e-01   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias   1.08219e-02  7.77823e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -2.29967e-02  7.36919e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -8.52386e-03  9.97429e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **   21 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   22 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   23 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **   24 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=34.0877 FROM MIGRAD    STATUS=CONVERGED      66 CALLS          67 TOTAL\n",
      "                     EDM=5.84604e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       2.84430e-02   9.98334e-01   4.06111e-04  -3.00245e-04\n",
      "   2  EG_SCALE    -1.78132e-02   1.28329e+00   5.38016e-04  -5.84426e-04\n",
      "   3  n_highMass_1tag_di_Higgs_bias   1.04551e-02   7.99992e-01   3.24890e-05   2.56123e-05\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -6.31672e-01   7.97300e-01   3.24155e-05  -9.44785e-02\n",
      "   5  theta_yield  -2.80853e-02   9.98332e-01   4.06112e-04  -1.06060e-03\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = 34.0877 for poi fixed at = 0.00211975\tfit time : Real time 0:00:00, CP time 0.010\n",
      "[#0] PROGRESS:Eval -- \t ASIMOV data qmu_A = 0.000763593 condNLL = 34.0877 uncond 34.0874\n",
      "[#1] INFO:InputArguments -- Minimum of POI is 0 corresponds to alt  snapshot   - using qtilde asymptotic formulae  \n",
      "[#0] PROGRESS:Eval -- poi = 0.00211975 qmu = 1.93051e-05 qmu_A = 0.000763593 sigma = 0.0767103  CLsplusb = 0.498247 CLb = 0.50927 CLs = 1.02212\n"
     ]
    }
   ],
   "source": [
    "results = ac.GetHypoTest()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.498247150061 0.50927035461\n"
     ]
    }
   ],
   "source": [
    "print results.NullPValue(), results.AlternatePValue()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:InputArguments -- HypoTestInverter ---- Input models: \n",
      "\t\t using as S+B (null) model     : model\n",
      "\t\t using as B (alternate) model  : model_with_poi_0\n",
      "\n"
     ]
    }
   ],
   "source": [
    "calc = rs.HypoTestInverter(ac)\n",
    "calc.SetConfidenceLevel(0.95)\n",
    "calc.UseCLs(True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "calc.SetFixedScan(8, poi.getMin(), poi.getMax())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true,
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[#1] INFO:Eval -- HypoTestInverter::GetInterval - run a fixed scan\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::saveSnaphot(w) replacing previous snapshot with name model__snapshot\n",
      "\n",
      "[#1] INFO:Eval -- AsymptoticCalculator::GetHypoTest: - perform  an hypothesis test for  POI ( mu_2H ) = 0\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -  Find  best conditional NLL on OBSERVED data set ..... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **   25 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   26 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       2.78863e-02  1.03401e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.75633e-02  1.00838e-01   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias   1.07253e-02  7.78131e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -2.30431e-02  7.42143e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -8.56473e-03  9.97452e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **   27 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   28 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   29 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **   30 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=-2160.82 FROM MIGRAD    STATUS=CONVERGED      60 CALLS          61 TOTAL\n",
      "                     EDM=3.09902e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       2.79833e-02   1.03389e+00   3.32528e-03  -5.02831e-03\n",
      "   2  EG_SCALE    -1.75630e-02   1.00815e-01   3.23630e-04  -6.90800e-03\n",
      "   3  n_highMass_1tag_di_Higgs_bias   1.07862e-02   7.77822e-01   2.49703e-04   6.73944e-03\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -2.28580e-02   7.36907e-01   2.36521e-04   2.41390e-02\n",
      "   5  theta_yield  -8.54335e-03   9.97429e-01   3.20730e-03   1.51083e-04\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = -2160.82 for poi fixed at = 0\tfit time : Real time 0:00:00, CP time 0.010\n",
      "[#0] PROGRESS:Eval -- \t OBSERVED DATA :  qmu   = -1.28925e-07 condNLL = -2160.82 uncond -2160.82\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -- Find  best conditional NLL on ASIMOV data set .... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **   31 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   32 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       2.79833e-02  1.03389e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.75630e-02  1.00815e-01   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias   1.07862e-02  7.77822e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -2.28580e-02  7.36907e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -8.54335e-03  9.97429e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **   33 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   34 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   35 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **   36 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=34.0874 FROM MIGRAD    STATUS=CONVERGED      66 CALLS          67 TOTAL\n",
      "                     EDM=5.85231e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       2.84429e-02   9.98334e-01   4.06108e-04  -3.01824e-04\n",
      "   2  EG_SCALE    -1.78131e-02   1.28459e+00   5.38661e-04  -5.75069e-04\n",
      "   3  n_highMass_1tag_di_Higgs_bias   1.04551e-02   7.99992e-01   3.24888e-05   2.56168e-05\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -6.31669e-01   7.97300e-01   3.24153e-05  -9.45462e-02\n",
      "   5  theta_yield  -2.80468e-02   9.98332e-01   4.06109e-04  -1.05897e-03\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = 34.0874 for poi fixed at = 0\tfit time : Real time 0:00:00, CP time 0.010\n",
      "[#0] PROGRESS:Eval -- \t ASIMOV data qmu_A = 6.38067e-11 condNLL = 34.0874 uncond 34.0874\n",
      "[#1] INFO:Eval -- Using one-sided qmu - setting qmu to zero  muHat = 5.4144e-06 muTest = 0\n",
      "[#0] PROGRESS:Eval -- poi = 0 qmu = 0 qmu_A = 6.38067e-11 sigma = 0  CLsplusb = 0.5 CLb = 0.500003 CLs = 1.00001\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::saveSnaphot(w) replacing previous snapshot with name model__snapshot\n",
      "\n",
      "[#1] INFO:Eval -- AsymptoticCalculator::GetHypoTest: - perform  an hypothesis test for  POI ( mu_2H ) = 2.85714\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -  Find  best conditional NLL on OBSERVED data set ..... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **   37 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   38 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       2.78863e-02  1.03401e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.75633e-02  1.00838e-01   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias   1.07253e-02  7.78131e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -2.30431e-02  7.42143e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -8.56473e-03  9.97452e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **   39 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   40 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   41 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **   42 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=-2160.77 FROM MIGRAD    STATUS=CONVERGED      56 CALLS          57 TOTAL\n",
      "                     EDM=2.67561e-06    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       3.50404e-02   1.02687e+00   3.32425e-03   7.99227e-03\n",
      "   2  EG_SCALE    -1.76416e-02   8.32968e-02   2.62301e-04  -4.37259e-02\n",
      "   3  n_highMass_1tag_di_Higgs_bias  -8.93188e-03   7.79185e-01   2.50156e-04  -1.56519e-02\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -9.55685e-02   7.51731e-01   2.41874e-04  -6.04579e-02\n",
      "   5  theta_yield  -2.26515e-02   9.96742e-01   3.20489e-03  -1.29505e-02\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = -2160.77 for poi fixed at = 2.85714\tfit time : Real time 0:00:00, CP time 0.010\n",
      "[#0] PROGRESS:Eval -- \t OBSERVED DATA :  qmu   = 0.101845 condNLL = -2160.77 uncond -2160.82\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -- Find  best conditional NLL on ASIMOV data set .... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **   43 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   44 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       3.50404e-02  1.02687e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.76416e-02  8.32968e-02   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias  -8.93188e-03  7.79185e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -9.55685e-02  7.51731e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -2.26515e-02  9.96742e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **   45 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   46 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   47 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **   48 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=34.602 FROM MIGRAD    STATUS=CONVERGED      62 CALLS          63 TOTAL\n",
      "                     EDM=5.89104e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       2.85518e-02   9.98328e-01   4.09113e-04   1.18696e-03\n",
      "   2  EG_SCALE    -1.81917e-02   6.68927e-01   2.17367e-04  -9.05350e-03\n",
      "   3  n_highMass_1tag_di_Higgs_bias   1.04553e-02   7.99992e-01   3.27296e-05   2.37952e-05\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -6.34395e-01   7.98105e-01   3.26716e-05  -4.75008e-02\n",
      "   5  theta_yield  -8.01528e-02   9.98314e-01   4.09112e-04  -2.55186e-03\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = 34.602 for poi fixed at = 2.85714\tfit time : Real time 0:00:00, CP time 0.020\n",
      "[#0] PROGRESS:Eval -- \t ASIMOV data qmu_A = 1.02925 condNLL = 34.602 uncond 34.0874\n",
      "[#0] PROGRESS:Eval -- poi = 2.85714 qmu = 0.101845 qmu_A = 1.02925 sigma = 2.81625  CLsplusb = 0.374813 CLb = 0.756594 CLs = 2.01859\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::saveSnaphot(w) replacing previous snapshot with name model__snapshot\n",
      "\n",
      "[#1] INFO:Eval -- AsymptoticCalculator::GetHypoTest: - perform  an hypothesis test for  POI ( mu_2H ) = 5.71429\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -  Find  best conditional NLL on OBSERVED data set ..... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **   49 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   50 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       2.78863e-02  1.03401e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.75633e-02  1.00838e-01   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias   1.07253e-02  7.78131e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -2.30431e-02  7.42143e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -8.56473e-03  9.97452e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **   51 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   52 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   53 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **   54 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=-2160.66 FROM MIGRAD    STATUS=CONVERGED      56 CALLS          57 TOTAL\n",
      "                     EDM=2.03785e-05    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       4.69958e-02   1.01856e+00   3.32170e-03   2.89016e-02\n",
      "   2  EG_SCALE    -1.77374e-02   7.54618e-02   2.36706e-04  -7.31467e-02\n",
      "   3  n_highMass_1tag_di_Higgs_bias  -2.74471e-02   7.80431e-01   2.50563e-04  -5.65541e-02\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -1.52357e-01   7.62083e-01   2.45297e-04  -1.78357e-01\n",
      "   5  theta_yield  -4.82245e-02   9.95751e-01   3.20155e-03  -3.11509e-02\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = -2160.66 for poi fixed at = 5.71429\tfit time : Real time 0:00:00, CP time 0.010\n",
      "[#0] PROGRESS:Eval -- \t OBSERVED DATA :  qmu   = 0.333196 condNLL = -2160.66 uncond -2160.82\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -- Find  best conditional NLL on ASIMOV data set .... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **   55 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   56 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       4.69958e-02  1.01856e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.77374e-02  7.54618e-02   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias  -2.74471e-02  7.80431e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -1.52357e-01  7.62083e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -4.82245e-02  9.95751e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **   57 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   58 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   59 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **   60 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=35.1161 FROM MIGRAD    STATUS=CONVERGED      62 CALLS          63 TOTAL\n",
      "                     EDM=5.57057e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       2.87232e-02   9.98311e-01   4.12094e-04   1.77166e-03\n",
      "   2  EG_SCALE    -1.85270e-02   5.26637e-01   1.93314e-04  -1.17145e-02\n",
      "   3  n_highMass_1tag_di_Higgs_bias   1.04555e-02   7.99992e-01   3.29684e-05   2.32268e-05\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -6.36449e-01   7.98568e-01   3.29211e-05  -2.86844e-02\n",
      "   5  theta_yield  -1.32502e-01   9.98288e-01   4.12086e-04  -3.05321e-03\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = 35.1161 for poi fixed at = 5.71429\tfit time : Real time 0:00:00, CP time 0.010\n",
      "[#0] PROGRESS:Eval -- \t ASIMOV data qmu_A = 2.0574 condNLL = 35.1161 uncond 34.0874\n",
      "[#0] PROGRESS:Eval -- poi = 5.71429 qmu = 0.333196 qmu_A = 2.0574 sigma = 3.98384  CLsplusb = 0.281892 CLb = 0.804315 CLs = 2.85328\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::saveSnaphot(w) replacing previous snapshot with name model__snapshot\n",
      "\n",
      "[#1] INFO:Eval -- AsymptoticCalculator::GetHypoTest: - perform  an hypothesis test for  POI ( mu_2H ) = 8.57143\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -  Find  best conditional NLL on OBSERVED data set ..... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **   61 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   62 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       2.78863e-02  1.03401e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.75633e-02  1.00838e-01   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias   1.07253e-02  7.78131e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -2.30431e-02  7.42143e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -8.56473e-03  9.97452e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **   63 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   64 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   65 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **   66 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=-2160.49 FROM MIGRAD    STATUS=CONVERGED      71 CALLS          72 TOTAL\n",
      "                     EDM=2.58613e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       5.62181e-02   1.01528e+00   3.26290e-03  -9.30360e-04\n",
      "   2  EG_SCALE    -1.79587e-02   6.97171e-02   2.23727e-04  -7.17140e-02\n",
      "   3  n_highMass_1tag_di_Higgs_bias  -4.41829e-02   7.81569e-01   2.50872e-04   3.07069e-03\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -1.95991e-01   7.69481e-01   2.46917e-04   2.25263e-03\n",
      "   5  theta_yield  -7.70027e-02   9.94634e-01   3.19745e-03  -4.07199e-04\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = -2160.49 for poi fixed at = 8.57143\tfit time : Real time 0:00:00, CP time 0.020\n",
      "[#0] PROGRESS:Eval -- \t OBSERVED DATA :  qmu   = 0.66867 condNLL = -2160.49 uncond -2160.82\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -- Find  best conditional NLL on ASIMOV data set .... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **   67 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   68 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       5.62181e-02  1.01528e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.79587e-02  6.97171e-02   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias  -4.41829e-02  7.81569e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -1.95991e-01  7.69481e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -7.70027e-02  9.94634e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **   69 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   70 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   71 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **   72 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=35.629 FROM MIGRAD    STATUS=CONVERGED      62 CALLS          63 TOTAL\n",
      "                     EDM=4.83008e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       2.89317e-02   9.98290e-01   4.15046e-04   1.99631e-03\n",
      "   2  EG_SCALE    -1.88219e-02   4.57770e-01   1.77568e-04  -1.23272e-02\n",
      "   3  n_highMass_1tag_di_Higgs_bias   1.04558e-02   7.99992e-01   3.32050e-05   2.44416e-05\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -6.38066e-01   7.98860e-01   3.31653e-05  -2.08116e-02\n",
      "   5  theta_yield  -1.85032e-01   9.98260e-01   4.15031e-04  -3.12568e-03\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = 35.629 for poi fixed at = 8.57143\tfit time : Real time 0:00:00, CP time 0.010\n",
      "[#0] PROGRESS:Eval -- \t ASIMOV data qmu_A = 3.08328 condNLL = 35.629 uncond 34.0874\n",
      "[#0] PROGRESS:Eval -- poi = 8.57143 qmu = 0.66867 qmu_A = 3.08328 sigma = 4.88143  CLsplusb = 0.206758 CLb = 0.825931 CLs = 3.99467\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::saveSnaphot(w) replacing previous snapshot with name model__snapshot\n",
      "\n",
      "[#1] INFO:Eval -- AsymptoticCalculator::GetHypoTest: - perform  an hypothesis test for  POI ( mu_2H ) = 11.4286\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -  Find  best conditional NLL on OBSERVED data set ..... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **   73 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   74 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       2.78863e-02  1.03401e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.75633e-02  1.00838e-01   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias   1.07253e-02  7.78131e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -2.30431e-02  7.42143e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -8.56473e-03  9.97452e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **   75 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   76 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   77 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **   78 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=-2160.28 FROM MIGRAD    STATUS=CONVERGED      71 CALLS          72 TOTAL\n",
      "                     EDM=2.86607e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       6.98909e-02   1.01503e+00   3.26317e-03  -1.09850e-03\n",
      "   2  EG_SCALE    -1.82955e-02   6.45198e-02   2.06950e-04  -8.08722e-02\n",
      "   3  n_highMass_1tag_di_Higgs_bias  -6.01098e-02   7.82609e-01   2.51188e-04   4.40294e-03\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -2.32177e-01   7.74870e-01   2.48617e-04   3.61214e-03\n",
      "   5  theta_yield  -1.14856e-01   9.93470e-01   3.19348e-03  -6.08967e-04\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = -2160.28 for poi fixed at = 11.4286\tfit time : Real time 0:00:00, CP time 0.020\n",
      "[#0] PROGRESS:Eval -- \t OBSERVED DATA :  qmu   = 1.08948 condNLL = -2160.28 uncond -2160.82\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -- Find  best conditional NLL on ASIMOV data set .... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **   79 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   80 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       6.98909e-02  1.01503e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.82955e-02  6.45198e-02   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias  -6.01098e-02  7.82609e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -2.32177e-01  7.74870e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -1.14856e-01  9.93470e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **   81 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   82 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   83 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **   84 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=36.1404 FROM MIGRAD    STATUS=CONVERGED      62 CALLS          63 TOTAL\n",
      "                     EDM=3.85767e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       2.91587e-02   9.98269e-01   4.17972e-04   2.04470e-03\n",
      "   2  EG_SCALE    -1.90737e-02   4.15714e-01   1.66424e-04  -1.17442e-02\n",
      "   3  n_highMass_1tag_di_Higgs_bias   1.04560e-02   7.99992e-01   3.34392e-05   2.73589e-05\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -6.39374e-01   7.99058e-01   3.34052e-05  -1.66312e-02\n",
      "   5  theta_yield  -2.37699e-01   9.98232e-01   4.17950e-04  -2.97488e-03\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = 36.1404 for poi fixed at = 11.4286\tfit time : Real time 0:00:00, CP time 0.010\n",
      "[#0] PROGRESS:Eval -- \t ASIMOV data qmu_A = 4.10615 condNLL = 36.1404 uncond 34.0874\n",
      "[#0] PROGRESS:Eval -- poi = 11.4286 qmu = 1.08948 qmu_A = 4.10615 sigma = 5.63994  CLsplusb = 0.148293 CLb = 0.837093 CLs = 5.64485\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::saveSnaphot(w) replacing previous snapshot with name model__snapshot\n",
      "\n",
      "[#1] INFO:Eval -- AsymptoticCalculator::GetHypoTest: - perform  an hypothesis test for  POI ( mu_2H ) = 14.2857\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -  Find  best conditional NLL on OBSERVED data set ..... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **   85 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   86 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       2.78863e-02  1.03401e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.75633e-02  1.00838e-01   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias   1.07253e-02  7.78131e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -2.30431e-02  7.42143e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -8.56473e-03  9.97452e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **   87 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   88 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   89 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **   90 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=-2160.03 FROM MIGRAD    STATUS=CONVERGED      71 CALLS          72 TOTAL\n",
      "                     EDM=2.70025e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       8.43137e-02   1.01513e+00   3.26388e-03  -1.23555e-03\n",
      "   2  EG_SCALE    -1.86597e-02   5.95916e-02   1.91033e-04  -8.38388e-02\n",
      "   3  n_highMass_1tag_di_Higgs_bias  -7.49932e-02   7.83560e-01   2.51472e-04   5.74775e-03\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -2.62084e-01   7.78896e-01   2.49887e-04   4.80143e-03\n",
      "   5  theta_yield  -1.57413e-01   9.92293e-01   3.18954e-03  -7.22204e-04\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = -2160.03 for poi fixed at = 14.2857\tfit time : Real time 0:00:00, CP time 0.030\n",
      "[#0] PROGRESS:Eval -- \t OBSERVED DATA :  qmu   = 1.58129 condNLL = -2160.03 uncond -2160.82\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -- Find  best conditional NLL on ASIMOV data set .... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **   91 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   92 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       8.43137e-02  1.01513e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.86597e-02  5.95916e-02   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias  -7.49932e-02  7.83560e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -2.62084e-01  7.78896e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -1.57413e-01  9.92293e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **   93 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **   94 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   95 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **   96 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=36.6501 FROM MIGRAD    STATUS=CONVERGED      62 CALLS          63 TOTAL\n",
      "                     EDM=2.97846e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       2.93944e-02   9.98249e-01   4.20870e-04   2.01020e-03\n",
      "   2  EG_SCALE    -1.92947e-02   3.86815e-01   1.58171e-04  -1.07050e-02\n",
      "   3  n_highMass_1tag_di_Higgs_bias   1.04563e-02   7.99992e-01   3.36710e-05   3.21307e-05\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -6.40464e-01   7.99200e-01   3.36412e-05  -1.42143e-02\n",
      "   5  theta_yield  -2.90477e-01   9.98203e-01   4.20845e-04  -2.71180e-03\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = 36.6501 for poi fixed at = 14.2857\tfit time : Real time 0:00:00, CP time 0.010\n",
      "[#0] PROGRESS:Eval -- \t ASIMOV data qmu_A = 5.12551 condNLL = 36.6501 uncond 34.0874\n",
      "[#0] PROGRESS:Eval -- poi = 14.2857 qmu = 1.58129 qmu_A = 5.12551 sigma = 6.31006  CLsplusb = 0.104288 CLb = 0.842905 CLs = 8.08248\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::saveSnaphot(w) replacing previous snapshot with name model__snapshot\n",
      "\n",
      "[#1] INFO:Eval -- AsymptoticCalculator::GetHypoTest: - perform  an hypothesis test for  POI ( mu_2H ) = 17.1429\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -  Find  best conditional NLL on OBSERVED data set ..... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **   97 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **   98 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       2.78863e-02  1.03401e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.75633e-02  1.00838e-01   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias   1.07253e-02  7.78131e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -2.30431e-02  7.42143e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -8.56473e-03  9.97452e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **   99 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **  100 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **  101 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **  102 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=-2159.76 FROM MIGRAD    STATUS=CONVERGED      69 CALLS          70 TOTAL\n",
      "                     EDM=2.38095e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       9.90557e-02   1.01531e+00   3.26513e-03  -1.33895e-03\n",
      "   2  EG_SCALE    -1.90001e-02   5.50402e-02   1.76370e-04  -8.39134e-02\n",
      "   3  n_highMass_1tag_di_Higgs_bias  -8.89182e-02   7.84430e-01   2.51729e-04   6.85823e-03\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -2.87217e-01   7.81970e-01   2.50857e-04   5.27252e-03\n",
      "   5  theta_yield  -2.03743e-01   9.91117e-01   3.18569e-03  -7.32316e-04\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = -2159.76 for poi fixed at = 17.1429\tfit time : Real time 0:00:00, CP time 0.020\n",
      "[#0] PROGRESS:Eval -- \t OBSERVED DATA :  qmu   = 2.13297 condNLL = -2159.76 uncond -2160.82\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -- Find  best conditional NLL on ASIMOV data set .... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **  103 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **  104 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       9.90557e-02  1.01531e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.90001e-02  5.50402e-02   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias  -8.89182e-02  7.84430e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -2.87217e-01  7.81970e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -2.03743e-01  9.91117e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **  105 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **  106 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **  107 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **  108 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=37.1579 FROM MIGRAD    STATUS=CONVERGED      62 CALLS          63 TOTAL\n",
      "                     EDM=2.26629e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       2.96324e-02   9.98229e-01   4.23740e-04   1.93144e-03\n",
      "   2  EG_SCALE    -1.94918e-02   3.65478e-01   1.51954e-04  -9.56626e-03\n",
      "   3  n_highMass_1tag_di_Higgs_bias   1.04565e-02   7.99992e-01   3.39003e-05   3.87175e-05\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -6.41391e-01   7.99305e-01   3.38739e-05  -1.25977e-02\n",
      "   5  theta_yield  -3.43346e-01   9.98173e-01   4.23719e-04  -2.39433e-03\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = 37.1579 for poi fixed at = 17.1429\tfit time : Real time 0:00:00, CP time 0.010\n",
      "[#0] PROGRESS:Eval -- \t ASIMOV data qmu_A = 6.14101 condNLL = 37.1579 uncond 34.0874\n",
      "[#0] PROGRESS:Eval -- poi = 17.1429 qmu = 2.13297 qmu_A = 6.14101 sigma = 6.91773  CLsplusb = 0.0720805 CLb = 0.845575 CLs = 11.731\n",
      "[#1] INFO:ObjectHandling -- RooWorkspace::saveSnaphot(w) replacing previous snapshot with name model__snapshot\n",
      "\n",
      "[#1] INFO:Eval -- AsymptoticCalculator::GetHypoTest: - perform  an hypothesis test for  POI ( mu_2H ) = 20\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -  Find  best conditional NLL on OBSERVED data set ..... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **  109 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **  110 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       2.78863e-02  1.03401e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.75633e-02  1.00838e-01   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias   1.07253e-02  7.78131e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -2.30431e-02  7.42143e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -8.56473e-03  9.97452e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **  111 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **  112 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **  113 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **  114 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=-2159.45 FROM MIGRAD    STATUS=CONVERGED      69 CALLS          70 TOTAL\n",
      "                     EDM=2.11108e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       1.13860e-01   1.01564e+00   3.26692e-03  -1.37261e-03\n",
      "   2  EG_SCALE    -1.92952e-02   5.09926e-02   1.63366e-04  -8.44987e-02\n",
      "   3  n_highMass_1tag_di_Higgs_bias  -1.01963e-01   7.85228e-01   2.51962e-04   7.56163e-03\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -3.08646e-01   7.84370e-01   2.51613e-04   4.86645e-03\n",
      "   5  theta_yield  -2.53172e-01   9.89950e-01   3.18195e-03  -6.47255e-04\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = -2159.45 for poi fixed at = 20\tfit time : Real time 0:00:00, CP time 0.010\n",
      "[#0] PROGRESS:Eval -- \t OBSERVED DATA :  qmu   = 2.73573 condNLL = -2159.45 uncond -2160.82\n",
      "[#0] PROGRESS:Eval -- AsymptoticCalculator::GetHypoTest -- Find  best conditional NLL on ASIMOV data set .... \n",
      "AsymptoticCalculator::EvaluateNLL  ........ using Minuit / Migrad with strategy  1 and tolerance 1\n",
      " **********\n",
      " **  115 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **  116 **SET NOGRAD\n",
      " **********\n",
      " PARAMETER DEFINITIONS:\n",
      "    NO.   NAME         VALUE      STEP SIZE      LIMITS\n",
      "     1 EG_RES       1.13860e-01  1.01564e+00   -1.00000e+01  1.00000e+01\n",
      "     2 EG_SCALE    -1.92952e-02  5.09926e-02   -1.00000e+01  1.00000e+01\n",
      "     3 n_highMass_1tag_di_Higgs_bias  -1.01963e-01  7.85228e-01   -1.00000e+02  1.00000e+02\n",
      "     4 n_highMass_2tag_di_Higgs_bias  -3.08646e-01  7.84370e-01   -1.00000e+02  1.00000e+02\n",
      "     5 theta_yield  -2.53172e-01  9.89950e-01   -1.00000e+01  1.00000e+01\n",
      " **********\n",
      " **  117 **SET ERR         0.5\n",
      " **********\n",
      " **********\n",
      " **  118 **SET PRINT           0\n",
      " **********\n",
      " **********\n",
      " **  119 **SET STR           1\n",
      " **********\n",
      " **********\n",
      " **  120 **MIGRAD        2500           1\n",
      " **********\n",
      " MIGRAD MINIMIZATION HAS CONVERGED.\n",
      " MIGRAD WILL VERIFY CONVERGENCE AND ERROR MATRIX.\n",
      " FCN=37.6635 FROM MIGRAD    STATUS=CONVERGED      62 CALLS          63 TOTAL\n",
      "                     EDM=1.7119e-07    STRATEGY= 1      ERROR MATRIX ACCURATE \n",
      "  EXT PARAMETER                                   STEP         FIRST   \n",
      "  NO.   NAME      VALUE            ERROR          SIZE      DERIVATIVE \n",
      "   1  EG_RES       2.98684e-02   9.98211e-01   4.26581e-04   1.82505e-03\n",
      "   2  EG_SCALE    -1.96695e-02   3.48947e-01   1.47209e-04  -8.48082e-03\n",
      "   3  n_highMass_1tag_di_Higgs_bias   1.04568e-02   7.99992e-01   3.41271e-05   4.70037e-05\n",
      "   4  n_highMass_2tag_di_Higgs_bias  -6.42194e-01   7.99386e-01   3.41034e-05  -1.13473e-02\n",
      "   5  theta_yield  -3.96293e-01   9.98144e-01   4.26574e-04  -2.05851e-03\n",
      "                               ERR DEF= 0.5\n",
      "AsymptoticCalculator::EvaluateNLL -  value = 37.6635 for poi fixed at = 20\tfit time : Real time 0:00:00, CP time 0.010\n",
      "[#0] PROGRESS:Eval -- \t ASIMOV data qmu_A = 7.15238 condNLL = 37.6635 uncond 34.0874\n",
      "[#0] PROGRESS:Eval -- poi = 20 qmu = 2.73573 qmu_A = 7.15238 sigma = 7.47833  CLsplusb = 0.0490633 CLb = 0.846228 CLs = 17.2477\n"
     ]
    }
   ],
   "source": [
    "r = calc.GetInterval()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "20.0\n",
      "0.0\n"
     ]
    }
   ],
   "source": [
    "print r.UpperLimit()\n",
    "print r.UpperLimitEstimatedError()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArgAAAHYCAIAAAApvgy/AAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE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\nAACAFQIFAAAAsPr/AcClREryL7G2AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "c = ROOT.TCanvas()\n",
    "plot = rs.HypoTestInverterPlot(r)\n",
    "plot.Draw(\"CLb 2CL\")\n",
    "c.Draw()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note to self: actually read this (https://root.cern.ch/root/html/tutorials/roostats/StandardHypoTestDemo.C.html) and try to understand what's going on!! Though I don't know if that's relevant. WHat I'm doing can be found here: https://twiki.cern.ch/twiki/bin/view/RooStats/RooStatsTutorialsAugust2012 and here: https://www.pp.rhul.ac.uk/~cowan/stat/roostats/SimpleCLs/SimpleCLs.C"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Andrew inverts this plot using this: https://github.com/rasumovsky/56_Hbbyy/blob/c83d975caf3f2cecb68cd470ad57b6292409573c/src/DHPlotCLvsMX.cxx"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {
    "collapsed": false,
    "deletable": true,
    "editable": true
   },
   "outputs": [],
   "source": [
    "# if you set the confidence interval above to 0.683, then the red line moves up"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Ignore Below"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "$$\\text{PDF}_{0tag,lowmass} = [\\color{red}{n_{noH,0tag,lowmass}}\\cdot f_{\\text{epoly2}}(m_{\\gamma\\gamma};a_{0,lowmass},a_{1,lowmass},a_{2,lowmass}) + \\mu_{1H}\\cdot \\color{red}{n_{1H,0tag,\n",
    "}}(1 + 0.3*\\theta_A)(1 + 0.05*\\theta_B)\\cdot f_{\\text{DSCB}}(m_{\\gamma\\gamma};\\color{red}{m_{0,0tag,lowmass}}(\\theta_A),\\color{red}{\\sigma_{0,0tag,lowmass}}(\\theta_B),\\alpha_\\text{low, lowmass},\\alpha_\\text{high, lowmass}, n_\\text{low, lowmass}, n_\\text{high, lowmass})] \\cdot \\text{Gauss}(\\theta_A; \\theta_{A0}, 1) \\cdot \\text{Gauss}(\\theta_B; \\theta_{B0}, 1)$$\n",
    "\n",
    "$$\\text{PDF}_{1tag,lowmass} = [\\color{red}{n_{noH,1tag,lowmass}}\\cdot f_{\\text{epoly2}}(m_{\\gamma\\gamma};a_{0,lowmass},a_{1,lowmass},a_{2,lowmass}) + (\\mu_{1H}\\cdot \\color{red}{n_{1H,1tag,lowmass}}+\\mu_{2H}\\cdot \\color{red}{n_{2H,1tag,lowmass}})(1 + 0.3*\\theta_A)(1 + 0.05*\\theta_B)\\cdot f_{\\text{DSCB}}(m_{\\gamma\\gamma};\\color{red}{m_{0,1tag,lowmass}}(\\theta_A),\\color{red}{\\sigma_{0,1tag,lowmass}}(\\theta_B),\\alpha_\\text{low, lowmass},\\alpha_\\text{high, lowmass}, n_\\text{low, lowmass}, n_\\text{high, lowmass})] \\cdot \\text{Gauss}(\\theta_A; \\theta_{A0}, 1) \\cdot \\text{Gauss}(\\theta_B; \\theta_{B0}, 1)$$\n",
    "\n",
    "$$\\text{PDF}_{2tag,lowmass} = [\\color{red}{n_{noH,2tag,lowmass}}\\cdot f_{\\text{epoly2}}(m_{\\gamma\\gamma};a_{0,lowmass},a_{1,lowmass},a_{2,lowmass}) + (\\mu_{1H}\\cdot \\color{red}{n_{1H,2tag,lowmass}}+\\mu_{2H}\\cdot \\color{red}{n_{2H,2tag,lowmass}})(1 + 0.3*\\theta_A)(1 + 0.05*\\theta_B)\\cdot f_{\\text{DSCB}}(m_{\\gamma\\gamma};\\color{red}{m_{0,2tag,lowmass}}(\\theta_A),\\color{red}{\\sigma_{0,2tag,lowmass}}(\\theta_B),\\alpha_\\text{low, lowmass},\\alpha_\\text{high, lowmass}, n_\\text{low, lowmass}, n_\\text{high, lowmass})] \\cdot \\text{Gauss}(\\theta_A; \\theta_{A0}, 1) \\cdot \\text{Gauss}(\\theta_B; \\theta_{B0}, 1)$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "deletable": true,
    "editable": true
   },
   "source": [
    "$$\\text{PDF}_{0tag,highmass} = [\\color{red}{n_{noH,0tag,highmass}}\\cdot f_{\\text{epoly2}}(m_{\\gamma\\gamma};a_{0,highmass},a_{1,highmass},a_{2,highmass}) + \\mu_{1H}\\cdot \\color{red}{n_{1H,0tag,highmass}}(1 + 0.3*\\theta_A)(1 + 0.05*\\theta_B)\\cdot f_{\\text{DSCB}}(m_{\\gamma\\gamma};m_0(\\theta_A),\\sigma_0(\\theta_B),\\alpha_\\text{low,highmass},\\alpha_\\text{high,highmass}, n_\\text{low,highmass}, n_\\text{high,highmass})] \\cdot \\text{Gauss}(\\theta_A; \\theta_{A0}, 1) \\cdot \\text{Gauss}(\\theta_B; \\theta_{B0}, 1)$$\n",
    "\n",
    "$$\\text{PDF}_{1tag,highmass} = [\\color{red}{n_{noH,1tag,highmass}}\\cdot f_{\\text{epoly2}}(m_{\\gamma\\gamma};a_{0,highmass},a_{1,highmass},a_{2,highmass}) + (\\mu_{1H}\\cdot \\color{red}{n_{1H,1tag,highmass}}+\\mu_{2H}\\cdot \\color{red}{n_{2H,1tag,highmass}})(1 + 0.3*\\theta_A)(1 + 0.05*\\theta_B)\\cdot f_{\\text{DSCB}}(m_{\\gamma\\gamma};m_0(\\theta_A),\\sigma_0(\\theta_B),\\alpha_\\text{low,highmass},\\alpha_\\text{high,highmass}, n_\\text{low,highmass}, n_\\text{high,highmass})] \\cdot \\text{Gauss}(\\theta_A; \\theta_{A0}, 1) \\cdot \\text{Gauss}(\\theta_B; \\theta_{B0}, 1)$$\n",
    "\n",
    "$$\\text{PDF}_{2tag,highmass} = [\\color{red}{n_{noH,2tag,highmass}}\\cdot f_{\\text{epoly2}}(m_{\\gamma\\gamma};a_{0,highmass},a_{1,highmass},a_{2,highmass}) + (\\mu_{1H}\\cdot \\color{red}{n_{1H,2tag,highmass}}+\\mu_{2H}\\cdot \\color{red}{n_{2H,2tag,highmass}})(1 + 0.3*\\theta_A)(1 + 0.05*\\theta_B)\\cdot f_{\\text{DSCB}}(m_{\\gamma\\gamma};m_0(\\theta_A),\\sigma_0(\\theta_B),\\alpha_\\text{low,highmass},\\alpha_\\text{high,highmass}, n_\\text{low,highmass}, n_\\text{high,highmass})] \\cdot \\text{Gauss}(\\theta_A; \\theta_{A0}, 1) \\cdot \\text{Gauss}(\\theta_B; \\theta_{B0}, 1)$$"
   ]
  }
 ],
 "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
}