{ "cells": [ { "cell_type": "markdown", "id": "728ed8a8", "metadata": {}, "source": [ "# rf510_wsnamedsets\n", "'ORGANIZATION AND SIMULTANEOUS FITS' RooFit tutorial macro #510\n", "\n", "Working with named parameter sets and parameter snapshots in\n", "workspaces\n", "\n", "\n", "\n", "\n", "**Author:** Clemens Lange, Wouter Verkerke (C version) \n", "This notebook tutorial was automatically generated with ROOTBOOK-izer from the macro found in the ROOT repository on Tuesday, May 19, 2026 at 08:32 PM." ] }, { "cell_type": "code", "execution_count": 1, "id": "4ca294f9", "metadata": { "collapsed": false, "execution": { "iopub.execute_input": "2026-05-19T20:32:32.125742Z", "iopub.status.busy": "2026-05-19T20:32:32.125614Z", "iopub.status.idle": "2026-05-19T20:32:33.099072Z", "shell.execute_reply": "2026-05-19T20:32:33.098650Z" } }, "outputs": [], "source": [ "import ROOT\n", "\n", "\n", "def fillWorkspace(w):\n", " # Create model\n", " # -----------------------\n", "\n", " # Declare observable x\n", " x = ROOT.RooRealVar(\"x\", \"x\", 0, 10)\n", "\n", " # Create two Gaussian PDFs g1(x,mean1,sigma) anf g2(x,mean2,sigma) and\n", " # their parameters\n", " mean = ROOT.RooRealVar(\"mean\", \"mean of gaussians\", 5, 0, 10)\n", " sigma1 = ROOT.RooRealVar(\"sigma1\", \"width of gaussians\", 0.5)\n", " sigma2 = ROOT.RooRealVar(\"sigma2\", \"width of gaussians\", 1)\n", "\n", " sig1 = ROOT.RooGaussian(\"sig1\", \"Signal component 1\", x, mean, sigma1)\n", " sig2 = ROOT.RooGaussian(\"sig2\", \"Signal component 2\", x, mean, sigma2)\n", "\n", " # Build Chebychev polynomial p.d.f.\n", " a0 = ROOT.RooRealVar(\"a0\", \"a0\", 0.5, 0.0, 1.0)\n", " a1 = ROOT.RooRealVar(\"a1\", \"a1\", -0.2, 0.0, 1.0)\n", " bkg = ROOT.RooChebychev(\"bkg\", \"Background\", x, [a0, a1])\n", "\n", " # Sum the signal components into a composite signal p.d.f.\n", " sig1frac = ROOT.RooRealVar(\"sig1frac\", \"fraction of component 1 in signal\", 0.8, 0.0, 1.0)\n", " sig = ROOT.RooAddPdf(\"sig\", \"Signal\", [sig1, sig2], [sig1frac])\n", "\n", " # Sum the composite signal and background\n", " bkgfrac = ROOT.RooRealVar(\"bkgfrac\", \"fraction of background\", 0.5, 0.0, 1.0)\n", " model = ROOT.RooAddPdf(\"model\", \"g1+g2+a\", [bkg, sig], [bkgfrac])\n", "\n", " # Import model into p.d.f.\n", " w.Import(model)\n", "\n", " # Encode definition of parameters in workspace\n", " # ---------------------------------------------------------------------------------------\n", "\n", " # Define named sets \"parameters\" and \"observables\", list which variables should be considered\n", " # parameters and observables by the users convention\n", " #\n", " # Variables appearing in sets _must_ live in the workspace already, the autoImport flag\n", " # of defineSet must be set to import them on the fly. Named sets contain only references\n", " # to the original variables, the value of observables in named sets already\n", " # reflect their 'current' value\n", " params = model.getParameters({x})\n", " w.defineSet(\"parameters\", params)\n", " w.defineSet(\"observables\", {x})\n", "\n", " # Encode reference value for parameters in workspace\n", " # ---------------------------------------------------------------------------------------------------\n", "\n", " # Define a parameter 'snapshot' in the p.d.f.\n", " # Unlike a named set, parameter snapshot stores an independent set of values for\n", " # a given set of variables in the workspace. The values can be stored and reloaded\n", " # into the workspace variable objects using the loadSnapshot() and saveSnapshot()\n", " # methods. A snapshot saves the value of each variable, errors that are stored\n", " # with it as well as the 'Constant' flag that is used in fits to determine if a\n", " # parameter is kept fixed or not.\n", "\n", " # Do a dummy fit to a (supposedly) reference dataset here and store the results\n", " # of that fit into a snapshot\n", " refData = model.generate({x}, 10000)\n", " model.fitTo(refData, PrintLevel=-1)\n", "\n", " # The kTRUE flag imports the values of the objects in (*params) into the workspace\n", " # If not set, present values of the workspace parameters objects are stored\n", " w.saveSnapshot(\"reference_fit\", params, True)\n", "\n", " # Make another fit with the signal componentforced to zero\n", " # and save those parameters too\n", "\n", " bkgfrac.setVal(1)\n", " bkgfrac.setConstant(True)\n", " bkgfrac.removeError()\n", " model.fitTo(refData, PrintLevel=-1)\n", "\n", " w.saveSnapshot(\"reference_fit_bkgonly\", params, True)" ] }, { "cell_type": "markdown", "id": "4f66cccf", "metadata": {}, "source": [ "Create model and dataset\n", "-----------------------------------------------" ] }, { "cell_type": "code", "execution_count": 2, "id": "103d036b", "metadata": { "collapsed": false, "execution": { "iopub.execute_input": "2026-05-19T20:32:33.105198Z", "iopub.status.busy": "2026-05-19T20:32:33.105060Z", "iopub.status.idle": "2026-05-19T20:32:33.608328Z", "shell.execute_reply": "2026-05-19T20:32:33.607639Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[#0] WARNING:InputArguments -- The parameter 'sigma1' with range [-inf, inf] of the RooGaussian 'sig1' exceeds the safe range of (0, inf). Advise to limit its range.\n", "[#0] WARNING:InputArguments -- The parameter 'sigma2' with range [-inf, inf] of the RooGaussian 'sig2' exceeds the safe range of (0, inf). Advise to limit its range.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooAddPdf::model\n", "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooChebychev::bkg\n", "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::x\n", "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::a0\n", "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::a1\n", "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::bkgfrac\n", "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooAddPdf::sig\n", "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooGaussian::sig1\n", "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::mean\n", "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::sigma1\n", "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::sig1frac\n", "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooGaussian::sig2\n", "[#1] INFO:ObjectHandling -- RooWorkspace::import(w) importing RooRealVar::sigma2\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "[#1] INFO:Fitting -- RooAbsPdf::fitTo(model) fixing normalization set for coefficient determination to observables in data\n", "[#1] INFO:Fitting -- using generic CPU library compiled with no vectorizations\n", "[#1] INFO:Fitting -- Creation of NLL object took 858.435 μs\n", "[#1] INFO:Fitting -- RooAddition::defaultErrorLevel(nll_model_modelData) Summation contains a RooNLLVar, using its error level\n", "[#1] INFO:Minimization -- [fitFCN] No discrete parameters, performing continuous minimization only\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "[#1] INFO:Fitting -- RooAbsPdf::fitTo(model) fixing normalization set for coefficient determination to observables in data\n", "[#1] INFO:Fitting -- Creation of NLL object took 325.222 μs\n", "[#1] INFO:Fitting -- RooAddition::defaultErrorLevel(nll_model_modelData) Summation contains a RooNLLVar, using its error level\n", "[#1] INFO:Minimization -- [fitFCN] No discrete parameters, performing continuous minimization only\n", "[#0] ERROR:Minimization -- RooMinimizer::calculateHessErrors() Error when calculating Hessian\n" ] } ], "source": [ "w = ROOT.RooWorkspace(\"w\")\n", "fillWorkspace(w)" ] }, { "cell_type": "markdown", "id": "2497daf4", "metadata": {}, "source": [ "Exploit convention encoded in named set \"parameters\" and \"observables\"\n", "to use workspace contents w/o need for introspected" ] }, { "cell_type": "code", "execution_count": 3, "id": "66b3b83e", "metadata": { "collapsed": false, "execution": { "iopub.execute_input": "2026-05-19T20:32:33.609942Z", "iopub.status.busy": "2026-05-19T20:32:33.609814Z", "iopub.status.idle": "2026-05-19T20:32:33.720243Z", "shell.execute_reply": "2026-05-19T20:32:33.719571Z" } }, "outputs": [], "source": [ "model = w[\"model\"]" ] }, { "cell_type": "markdown", "id": "673a0465", "metadata": {}, "source": [ "Generate data from p.d.f. in given observables" ] }, { "cell_type": "code", "execution_count": 4, "id": "54078328", "metadata": { "collapsed": false, "execution": { "iopub.execute_input": "2026-05-19T20:32:33.722343Z", "iopub.status.busy": "2026-05-19T20:32:33.722217Z", "iopub.status.idle": "2026-05-19T20:32:33.835175Z", "shell.execute_reply": "2026-05-19T20:32:33.834481Z" } }, "outputs": [], "source": [ "data = model.generate(w.set(\"observables\"), 1000)" ] }, { "cell_type": "markdown", "id": "1d7439e8", "metadata": {}, "source": [ "Fit model to data" ] }, { "cell_type": "code", "execution_count": 5, "id": "e85189c5", "metadata": { "collapsed": false, "execution": { "iopub.execute_input": "2026-05-19T20:32:33.837409Z", "iopub.status.busy": "2026-05-19T20:32:33.837278Z", "iopub.status.idle": "2026-05-19T20:32:33.946354Z", "shell.execute_reply": "2026-05-19T20:32:33.945745Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[#1] INFO:Fitting -- RooAbsPdf::fitTo(model) fixing normalization set for coefficient determination to observables in data\n", "[#1] INFO:Fitting -- Creation of NLL object took 237.872 μs\n", "[#1] INFO:Fitting -- RooAddition::defaultErrorLevel(nll_model_modelData) Summation contains a RooNLLVar, using its error level\n", "[#1] INFO:Minimization -- [fitFCN] No discrete parameters, performing continuous minimization only\n" ] }, { "data": { "text/plain": [ "" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "model.fitTo(data, PrintLevel=-1)" ] }, { "cell_type": "markdown", "id": "b94c2ebf", "metadata": {}, "source": [ "Plot fitted model and data on frame of first (only) observable" ] }, { "cell_type": "code", "execution_count": 6, "id": "5a9fe847", "metadata": { "collapsed": false, "execution": { "iopub.execute_input": "2026-05-19T20:32:33.948212Z", "iopub.status.busy": "2026-05-19T20:32:33.948084Z", "iopub.status.idle": "2026-05-19T20:32:34.113273Z", "shell.execute_reply": "2026-05-19T20:32:34.112564Z" } }, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "frame = (w.set(\"observables\").first()).frame()\n", "data.plotOn(frame)\n", "model.plotOn(frame)" ] }, { "cell_type": "markdown", "id": "8873e05f", "metadata": {}, "source": [ "Overlay plot with model with reference parameters as stored in snapshots" ] }, { "cell_type": "code", "execution_count": 7, "id": "2a537f6e", "metadata": { "collapsed": false, "execution": { "iopub.execute_input": "2026-05-19T20:32:34.115047Z", "iopub.status.busy": "2026-05-19T20:32:34.114920Z", "iopub.status.idle": "2026-05-19T20:32:34.233943Z", "shell.execute_reply": "2026-05-19T20:32:34.233399Z" } }, "outputs": [ { "data": { "text/plain": [ "" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "w.loadSnapshot(\"reference_fit\")\n", "model.plotOn(frame, LineColor=\"r\")\n", "w.loadSnapshot(\"reference_fit_bkgonly\")\n", "model.plotOn(frame, LineColor=\"r\", LineStyle=\"--\")" ] }, { "cell_type": "markdown", "id": "15ae409a", "metadata": {}, "source": [ "Draw the frame on the canvas" ] }, { "cell_type": "code", "execution_count": 8, "id": "5034b882", "metadata": { "collapsed": false, "execution": { "iopub.execute_input": "2026-05-19T20:32:34.236198Z", "iopub.status.busy": "2026-05-19T20:32:34.236042Z", "iopub.status.idle": "2026-05-19T20:32:34.450779Z", "shell.execute_reply": "2026-05-19T20:32:34.450166Z" } }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "Info in : png file rf510_wsnamedsets.png has been created\n" ] } ], "source": [ "c = ROOT.TCanvas(\"rf510_wsnamedsets\", \"rf503_wsnamedsets\", 600, 600)\n", "ROOT.gPad.SetLeftMargin(0.15)\n", "frame.GetYaxis().SetTitleOffset(1.4)\n", "frame.Draw()\n", "\n", "c.SaveAs(\"rf510_wsnamedsets.png\")" ] }, { "cell_type": "markdown", "id": "41241c26", "metadata": {}, "source": [ "Print workspace contents" ] }, { "cell_type": "code", "execution_count": 9, "id": "dbb125b8", "metadata": { "collapsed": false, "execution": { "iopub.execute_input": "2026-05-19T20:32:34.452986Z", "iopub.status.busy": "2026-05-19T20:32:34.452855Z", "iopub.status.idle": "2026-05-19T20:32:34.563940Z", "shell.execute_reply": "2026-05-19T20:32:34.563334Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "RooWorkspace(w) w contents\n", "\n", "variables\n", "---------\n", "(a0,a1,bkgfrac,mean,sig1frac,sigma1,sigma2,x)\n", "\n", "p.d.f.s\n", "-------\n", "RooChebychev::bkg[ x=x coefList=(a0,a1) ] = 1\n", "RooAddPdf::model[ bkgfrac * bkg + [%] * sig ] = 1/1\n", "RooAddPdf::sig[ sig1frac * sig1 + [%] * sig2 ] = 0.999807/1\n", "RooGaussian::sig1[ x=x mean=mean sigma=sigma1 ] = 0.999765\n", "RooGaussian::sig2[ x=x mean=mean sigma=sigma2 ] = 0.999941\n", "\n", "parameter snapshots\n", "-------------------\n", "reference_fit = (a0=0.488363 +/- 0.0241765,a1=0.0249649 +/- 0.0384439,bkgfrac=0.484424 +/- 0.0115194,mean=5.01085 +/- 0.0103629,sigma1=0.5[C],sig1frac=0.757184 +/- 0.0357497,sigma2=1[C])\n", "reference_fit_bkgonly = (a0=0.469143 +/- 0,a1=7.71653e-08 +/- 0,bkgfrac=1[C],mean=5.01085 +/- 0,sigma1=0.5[C],sig1frac=0.757184 +/- 0,sigma2=1[C])\n", "\n", "named sets\n", "----------\n", "observables:(x)\n", "parameters:(a0,a1,bkgfrac,mean,sig1frac,sigma1,sigma2)\n", "\n" ] } ], "source": [ "w.Print()" ] }, { "cell_type": "markdown", "id": "59075316", "metadata": {}, "source": [ "Workspace will remain in memory after macro finishes" ] }, { "cell_type": "code", "execution_count": 10, "id": "6555d497", "metadata": { "collapsed": false, "execution": { "iopub.execute_input": "2026-05-19T20:32:34.565860Z", "iopub.status.busy": "2026-05-19T20:32:34.565715Z", "iopub.status.idle": "2026-05-19T20:32:34.684249Z", "shell.execute_reply": "2026-05-19T20:32:34.683865Z" } }, "outputs": [], "source": [ "ROOT.gDirectory.Add(w)" ] }, { "cell_type": "markdown", "id": "4bca9fdc", "metadata": {}, "source": [ "Draw all canvases " ] }, { "cell_type": "code", "execution_count": 11, "id": "4bb01007", "metadata": { "collapsed": false, "execution": { "iopub.execute_input": "2026-05-19T20:32:34.686562Z", "iopub.status.busy": "2026-05-19T20:32:34.686436Z", "iopub.status.idle": "2026-05-19T20:32:34.870107Z", "shell.execute_reply": "2026-05-19T20:32:34.869343Z" } }, "outputs": [ { "data": { "text/html": [ "\n", "\n", "
\n", "
\n", "\n", "\n", "\n" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from ROOT import gROOT \n", "gROOT.GetListOfCanvases().Draw()" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.12.12" } }, "nbformat": 4, "nbformat_minor": 5 }