wiso · June 21, 2014 17:51
diff --git a/Poisson2 b/Poisson2
 {
 "metadata": {
  "name": "poisson"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "import numpy as np\nimport ROOT\nimport rootnotes",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 1
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Definitions\n============"
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "NEVENTS_PER_CAT = [1, 2, 4] # first category 1 event on average, second category 2 events on average, ...\nNCAT = len(NEVENTS_PER_CAT) # numbers of categories\nNTOYS = 1000000\nweights = np.random.random(NCAT) / 100. # weights for every categories\nprint \"weights: \", weights",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "weights:  [ 0.00633232  0.00998537  0.00152294]\n"
      }
     ],
     "prompt_number": 2
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Toys\n====\nI generate `NTOYS`. I fill the histogram with the sum of the weights."
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "nw = (NEVENTS_PER_CAT * weights).sum()\nnw2 = (NEVENTS_PER_CAT * weights ** 2).sum()\nlambda_tilde = nw ** 2 / nw2\nscale = nw2 / nw\nprint \"lambda_tilde: \", lambda_tilde\nprint \"scale: \", scale\nscaled_poisson = ROOT.TF1(\"scaled_poisson\", \"[0] * TMath::Poisson(x / [2], [1])\", 0, 100)\nscaled_poisson.SetParameter(0, NTOYS*100.)\nscaled_poisson.FixParameter(1, lambda_tilde)\nscaled_poisson.FixParameter(2, scale)\nscaled_poisson.SetLineColor(ROOT.kGreen)\n\nx = ROOT.RooRealVar(\"x\", \"x\", 0, 0.2)\nlambda_tilde_var = ROOT.RooRealVar(\"lambda_tilde\", \"lambda_tilde\", lambda_tilde)\nscale_var = ROOT.RooRealVar(\"scale\", \"scale\", scale)\nx_scaled = ROOT.RooFormulaVar(\"x_scaled\", \"x_scaled\", \"@0/@1\", ROOT.RooArgList(x, scale_var))\n#scaled_poisson_pdf = ROOT.RooGenericPdf(\"scaled_poisson_pdf\", \"scaled poisson\", \"TMath::Poisson(x / @0, @1)\",\n#                                        ROOT.RooArgList(scale_var, lambda_tilde_var))\nscaled_poisson_pdf = ROOT.RooPoisson(\"scaled_poisson_pdf\", \"scaled_poisson_pdf\", x_scaled, lambda_tilde_var)\n",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "lambda_tilde:  4.21810232008\nscale:  0.00767995660764\n"
      },
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": "TClass::TClass:0: RuntimeWarning: no dictionary for class stack<RooAbsArg*,deque<RooAbsArg*> > is available\n"
      }
     ],
     "prompt_number": 3
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "histo = ROOT.TH1F(\"histo\", \"histo\", 20, 0, 0.2)\n\nvars_set = ROOT.RooArgSet(x)\ndataset = ROOT.RooDataSet(\"dataset\", \"dataset\", vars_set)\nfor itoy in xrange(NTOYS):\n    n = [np.random.poisson(n) for n in NEVENTS_PER_CAT]\n    b = (n * weights).sum()\n    histo.Fill(b)\n    x.setVal(b)\n    set_to_add = ROOT.RooArgSet(x)\n    dataset.add(set_to_add)    \n\nhisto.Fit(scaled_poisson, \"\", \"goff\")\n\nprint \"expected sum of weights\", (NEVENTS_PER_CAT * weights).sum()\nprint \"average sum of weights\", histo.GetMean()",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": " expected sum of weights 0.0323948427848\naverage sum of weights 0.032373085747\n"
      },
      {
       "output_type": "stream",
       "stream": "stderr",
       "text": "TROOT::Append:0: RuntimeWarning: Replacing existing TH1: histo (Potential memory leak).\n"
      }
     ],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "canvas = rootnotes.default_canvas()\nhisto.Draw()\nscaled_poisson.Draw(\"same\")\ncanvas",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAxwAAAI8CAIAAAD0vjrdAAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE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\n1g2xPh17AAAAAElFTkSuQmCC\n",
       "prompt_number": 6,
       "text": "<ROOT.TCanvas object (\"icanvas\") at 0x4d661d0>"
      }
     ],
     "prompt_number": 6
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "canvas = rootnotes.default_canvas()\nframe = x.frame(20)\ndataset.plotOn(frame)\nscaled_poisson_pdf.plotOn(frame)\nframe.Draw()\ncanvas",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAxwAAAI8CAIAAAD0vjrdAAAABmJLR0QAAAAAAAD5Q7t/AAAgAElE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\nABH8/0WlCvihyITgAAAAAElFTkSuQmCC\n",
       "prompt_number": 7,
       "text": "<ROOT.TCanvas object (\"icanvas\") at 0x4d661d0>"
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "",
     "language": "python",
     "metadata": {},
     "outputs": []
    }
   ],
   "metadata": {}
  }
 ]
 }