ROOT Basics

Deb Mohapatra

This lecture is based on Fermilab ROOT tutorial and ROOT lecture in CERN summer school

WHAT IS ROOT ? ROOT is an object oriented framework It has a C/C++ interpreter (CINT) and C/C++ compiler (ACLIC) ROOT is used extensively in High Energy Physics for data analysis - Reading and writing data files - Calculations to produce plots, numbers and fits. WHY ROOT ? It can handle large files (in GB) containing N-tuples and Histograms Multiplatform software Its based on widely known programming language C++ Its free

Introduction

Outline of this lecture

Overview of ROOT framework GUI and command line basics CINT: Interpreter for C and C++ code Graphs, Histograms and Root Trees Functions and fitting

Learning ROOT
http://root.cern.ch/root/Tutorials.html http://root.cern.ch/root/HowTo.html http://www-root.fnal.gov/root/ http://www.slac.stanford.edu/BFROOT/www/doc/w orkbook/root{1,2,3}/root{1,2,3}.html

ROOT Mailing Lists

[email protected] http://root.cern.ch/root/roottalk/AboutRootTalk.html

Root Interactive Session

Set ROOTSYS to the directory where ROOT is installed Add ROOT libraries to the LD_LIBRARY_PATH Include the ROOT executable binary files to the binary path

BASH

export ROOTSYS=/cern/root export LD_LIBRARY_PATH=$ROOTSYS/lib:$LD_LIBRARY_PATH export PATH=$ROOTSYS/bin:$PATH setenv ROOTSYS /cern/root setenv LD_LIBRARY_PATH $ROOTSYS/lib:$LD_LIBRARY_PATH setenv PATH $ROOTSYS/bin:$PATH

TCSH

You may add the above lines to your ~/.cshrc or ~/.bashrc You may define your root settings in ~/.rootlogon.C History of all commands are stored in ~/.root_hist

Object Oriented Concepts

Class: the description of a thing in the system Object: instance of a class Methods: functions for a class

Members: a has a relationship to the class. Inheritance: an is a relationship to the class.

TObject
IS A

Event
HAS A HAS A HAS A

Segment
HAS A

Track
HAS A HAS A

Vertex

Momentum

MassSquare

InterceptAtVert

The Framework Organization

$ROOTSYS bin
cint makecint new proofd proofserv rmkdepend root root.exe rootcint root-config rootd * Optional Installation libCint.so libCore.so libEG.so *libEGPythia.so *libEGPythia6.so libEGVenus.so libGpad.so libGraf.so libGraf3d.so libGui.so libGX11.so *libGX11TTF.so libHist.so libHistPainter.so libHtml.so libMatrix.so libMinuit.so libNew.so libPhysics.so libPostscript.so libProof.so *libRFIO.so *libRGL.so libRint.so *libThread.so libTree.so libTreePlayer.so libTreeViewer.so *libttf.so libX3d.so libXpm.a

lib
EditorBar.C Ifit.C analyze.C archi.C arrow.C basic.C basic.dat basic3d.C benchmarks.C canvas.C classcat.C cleanup.C compile.C copytree.C copytree2.C demos.C demoshelp.C dialogs.C dirs.C ellipse.C eval.C event.C exec1.C exec2.C feynman.C fildir.C file.C fillrandom.C first.C fit1.C fit1_C.C

tutorials
fitslicesy.C formula1.C framework.C games.C gaxis.C geometry.C gerrors.C gerrors2.C graph.C h1draw.C hadd.C hclient.C hcons.C hprod.C hserv.C hserv2.C hsimple.C hsum.C hsumTimer.C htmlex.C io.C latex.C latex2.C latex3.C manyaxis.C multifit.C myfit.C na49.C na49geomfile.C na49view.C na49visible.C

test
ntuple1.C oldbenchmarks.C pdg.dat psexam.C pstable.C rootalias.C rootenv.C rootlogoff.C rootlogon.C rootmarks.C runcatalog.sql runzdemo.C second.C shapes.C shared.C splines.C sqlcreatedb.C sqlfilldb.C sqlselect.C staff.C staff.dat surfaces.C tcl.C testrandom.C tornado.C tree.C two.C xyslider.C xysliderAction.C zdemo.C

include
Aclock.cxx Aclock.h Event.cxx Event.h EventLinkDef.h Hello.cxx Hello.h MainEvent.cxx Makefile Makefile.in Makefile.win32 README TestVectors.cxx Tetris.cxx Tetris.h eventa.cxx eventb.cxx eventload.cxx guitest.cxx hsimple.cxx hworld.cxx minexam.cxx stress.cxx tcollbm.cxx tcollex.cxx test2html.cxx tstring.cxx vlazy.cxx vmatrix.cxx vvector.cxx
*.h ...

User Interfaces

.q .L macro.C .x macro.C .x macro.C++

Quit Load a macro file Load and execute a macro file Compile and execute

GUI Basics
Start root
> root

Quit root (just in case)

root[0]>.q

Display the browser

TBrowser b;

Displaying a Histogram
Open the root file Browse the file

Display a histogram The Canvas

Basic Navigation by Clicking

Left Click
select the object drag the object resize the object

Right Click
context menu class::name methods

Middle Click
activate canvas freezes event status bar

The Draw Panel

Fitting, Coloring, and Zooming

Adding a gaussian fit Coloring the histogram Zooming/unzooming

Adding Objects to the Canvas

The Editor Adding an Arrow Adding Text

Adding another Pad

Add a Pad Select the new Pad Draw a histogram Add a title for the axis

Modifying the Statistics

The Canvas in the Browser Setting the (7) statistics options
default = 0001111

Reading & Storing Data in Root

Data can be read from files/database/network Data is generally stored as a TTree/TNtuple ( similar to a table with rows and columns ) Each row represents an event Each column represents a quantity Trees can be created from ASCII files.

Read data from ASCII file to Root File

basic.dat
-1.102279 1.867178 -0.524181 -0.380611 0.552454 -0.184954 0.205643 1.079222 -0.274919 2.047779 -0.458677 0.304731 -0.712336 -0.271866 0.886202 -2.035552 -1.459047 1.230661 0.088787 -0.314154 -0.198253 -1.636217 1.221109 1.413135 -0.174494 -1.464173 -1.799389 4.452822 -0.596622 3.842313 1.868521 3.766139 0.969128 1.084074 -0.212309 0.350281 1.187305 1.443902 -0.770148 0.635417 -0.327389 1.271904 -1.721429 3.038899 -0.062677 4.197329 -1.443219 2.293266 -0.884636 0.875442 -0.222392 0.556881 1.181767 1.470484 -0.654106 1.213209 0.527648 4.421883 -0.463998 2.344113 -0.005650 1.514559 1.885329 3.562347 -0.329161 0.207040 0.646070 0.456712 1.049551 3.778762 0.814383 2.154327 1.549837 4.398942 -1.330937 1.801841 -0.912864 2.977124

tree.C
#include "Riostream.h" void tree() { ifstream in; in.open(Form("basic.dat")); Float_t x,y,z; Int_t nlines = 0;

TFile *f = new TFile("basic.root","RECREATE"); TH1F *h1 = new TH1F("h1","x distribution",100,-4,4); TNtuple *ntuple = new TNtuple("ntuple","data from ascii file","x:y:z");
while (1) { in >> x >> y >> z; if (!in.good()) break; if (nlines < 5) printf("x=%8f, y=%8f, z=%8f\n",x,y,z);

h1->Fill(x); ntuple->Fill(x,y,z);
nlines++; } printf(" found %d points\n",nlines);

in.close(); f->Write();
}

Read the Root file and the Tree

ntuple->Scain();

TBrowser t; nTuple->Print();

What do we learn from this Macro

How to Create a Root file ? Commands
TFile *f = new TFile("basic.root","RECREATE");
//option: NEW, CREATE, RECREATE, UPDATE, or READ

//Book and fill histograms and trees //-----f->Write(); //write the file f->Close(); //close the file

Book and fill a histogram ?

TH1F *h1 = new TH1F("h1","x distribution",100,-4,4); /*do some calculation and get the parameter that you want to fill*/ h1->Fill(x);

Book and fill a tree ?

TNtuple *ntuple = new TNtuple("ntuple","data from ascii file","x:y:z");

/*do some calculation and get the parameter that you want to fill*/ ntuple->Fill(x,y,z);

CINT Data types

Int_t and Float_t (see http://root.cern.ch/root/html/ListOfTypes.html )

Root Data Types

C++ type Size (bytes) ROOT types Size (bytes) FORTRAN analog

(unsigned)char
(unsigned)short (int) (unsigned)int (unsigned)long (int)

1
2 2 or 4 4 or 8

(U)Char_t
(U)Short_t (U)Int_t (U)Long_t

1
2 4 8

CHARACTER*1
INTEGER*2 INTEGER*4 INTEGER*8

float
double long double

4
8 (=4) 16 (= double)

Float_t
Double_t

4
8

REAL*4
REAL*8 REAL*16

How to Compile a Macro ?

Running the macro root [0] .x tree.C or, root [0] .L tree.C root [1] tree()
#include "Riostream.h" void tree(){ ifstream in; in.open(Form("basic.dat")); Float_t x,y,z; Int_t nlines = 0; TFile *f = new TFile("basic.root","RECREATE"); TH1F *h1 = new TH1F("h1","x distribution",100,-4,4); TNtuple *ntuple = new TNtuple("ntuple","data from ascii file","x:y:z"); while (1) { in >> x >> y >> z; if (!in.good()) break; if (nlines < 5) printf("x=%8f, y=%8f, z=%8f\n",x,y,z); h1->Fill(x); ntuple->Fill(x,y,z); nlines++; } printf(" found %d points\n",nlines); in.close(); f->Write(); f->Close(); }

Compiling the macro root [0] .L tree1.C++ root [1] main()

#include "Riostream.h" #include "TFile.h" #include "TH1.h" #include "TNtuple.h" int main() { ifstream in; in.open(Form("basic.dat")); Float_t x,y,z; Int_t nlines = 0; TFile *f = new TFile("basic.root","RECREATE"); TH1F *h1 = new TH1F("h1","x distribution",100,-4,4); TNtuple *ntuple = new TNtuple("ntuple","data from ascii file","x:y:z"); while (1) { in >> x >> y >> z; if (!in.good()) break; if (nlines < 5) printf("x=%8f, y=%8f, z=%8f\n",x,y,z); h1->Fill(x); ntuple->Fill(x,y,z); nlines++; } printf(" found %d points\n",nlines); in.close(); f->Write(); f->Close(); return 0; }

Histograms in Root: 1D, 2D and 3D

Floats: Max bin content 7 digits Double: Max bin content 14 digits

1D Histograms: TH1
TH1F *name = new TH1F(name",Title, Bins, lowest bin, highest bin); Example: TH1F *h1 = new TH1F("h1","x distribution",100,-4,4); h1->Fill(x); h1->Draw();

Histogram Properties
Command h1.GetMean() h1.GetRMS() Parameters Mean Root of Variance

h1.GetMaximum();
h1.GetMaximumBin(int bin_number); h1.GetBinCenter(int bin_number); h1.GetBinContent(int bin_number);

Maximum bin content

location of maximum Center of bin Content of bin

Histogram Cosmetics
h1.SetMarkerStyle();

h1.SetFillColor();

Histogram cosmetics: Lines

LineStyle h1->SetLineStyle(); LineColor h1.SetLineColor();

1D Histogram
Overlapping h1->Draw(); h2->Draw(same); Fit ( will be covered in detail later) h1->Fit(gaus);

Canvas: an area mapped to a window

Command c1 = new TCanvas(c1,Title, w, h) c1->Divide(2,2); c1->cd(3) c1->SetGridx(); c1->SetGridy(); c1->SetLogy(); Action Creates a new canvas with width equal to w number of pixels and height equal to h number of pixels. Divides the canvas to 4 pads. Select the 3rd Pad You can set grid along x and y axis. You can also set log scale plots.

Canvas: Demo
root [1] c1 = new TCanvas("c1","Title",800,600); root [2] c1->Divide(2,2); root [3] c1->cd(1); root [4] h1->Draw(); root [5] c1->cd(2); root [6] h2->Draw(); root [7] c1->cd(3); root [8] h1->SetLineColor(2) root [9] h2->SetLineColor(4) root [10] h1->Draw(); root [11] h2->Draw("same"); root [12] c1->cd(4); root [13] h1->Fit("gaus");

2D Histograms: TH2
TH2F *name = new TH2F(name",Title, xBins, low xbin, up xbin, yBins, low ybin, up y bin); Example: TH2F *h12 = new TH2F("h12","x vs y",100,4,4,100, -4, 4); h12->Fill(x,y); h12->Draw();

3D Histograms: TH3
TH3F *name = new TH3F(name",Title, xBins, low xbin, up xbin, yBins, low ybin, up ybin, zBins, low zbin, up zbin); Example: TH3F *h123 = new TH3F("h123","x vs y vs z",100,-4,4,100, -4, 4,100,0,20); h123->Fill(x,y,z); h123->Draw();

Histogram Drawing Options

" SAME": Superimpose on previous picture in the same pad. " CYL": Use cylindrical coordinates. " POL": Use polar coordinates. " SPH": Use spherical coordinates. " PSR": Use pseudo-rapidity/phi coordinates. " LEGO": Draw a lego plot with hidden line removal. " LEGO1": Draw a lego plot with hidden surface removal. " LEGO2": Draw a lego plot using colors to show the cell contents. " SURF": Draw a surface plot with hidden line removal. " SURF1": Draw a surface plot with hidden surface removal. " SURF2": Draw a surface plot using colors to show the cell contents. " SURF3": Same as SURF with a contour view on the top. " SURF4": Draw a surface plot using Gouraud shading. SURF5: Same as SURF3 but only the colored contour is drawn. Note: Please check chapter 3 in users guide to learn more about options.

Graphs
Graphics object made of two arrays X and Y, holding the x, y coordinates of n points

Graphs:
Int_t n = 20; Double_t x[n], y[n]; for (Int_t i=0; i<n; i++){ x[i] = i*0.1; y[i] = 10*sin(x[i]+0.2); } TGraph *gr1 = new TGraph (n, x, y); gr1->Draw("AC*");

Graph Drawing Options

AB AC*

AF

AL

Superimpose two Graphs

TGraph *gr1 = new TGraph(n,x,y); TGraph *gr2 = new TGraph(n,x1,y1); gr1->SetLineColor(4); gr1->Draw("AC*"); gr2->SetLineWidth(3); gr2->SetMarkerStyle(21); gr2->SetLineColor(2); gr2->Draw("CP");

Graph with Error bar

Float_t x[n] = {-.22,.05,.25,.35,.5,.61,.7,.85,.89,.95}; Float_t y[n] = {1,2.9,5.6,7.4,9,9.6,8.7,6.3,4.5,1}; Float_t ex[n] = {.05,.1,.07,.07,.04,.05,.06,.07,.08,.05}; Float_t ey[n] = {.8,.7,.6,.5,.4,.4,.5,.6,.7,.8}; gr = new TGraphErrors(n,x,y,ex,ey);

Polar Graphs

Generate or calculate r and theta TGraphPolar * grP1 = new TGraphPolar(1000,r,theta); grP1->Draw();

TTree
Saving data in a table with rows representing the event and columns representing quantities.

ROOT Tree
Store large quantities of same-class objects TTree class is optimized to reduce disk space and enhance access speed TTree can hold all kind of data TNtuple is a TTree that is limited to only hold floating-point numbers If we do not use TTree, we need to read each event in its entirety into memory extract the parameters from the event Compute quantities from the same fill a histogram

Tfile *F = new Tfile(test.root,RECREATE); TTree *T = new TTree("T","test"); T->Branch("x",&x,"x/F"); T->Branch("y",&y,"x/F"); T->Branch("z",&z,"x/F"); // Read/or calculate x,y and z T->Fill(); T->Close(); F->Close();

Create a Root TTree/TNtuple

Tfile *F = new Tfile(test.root,RECREATE); TNtuple *T = new TNtuple("ntuple","data from ascii file","x:y:z"); // Read/or calculate x,y and z T->Fill(x,y,z); T->Close(); F->Close();

Draw: T->Draw(x);

T->Print(); //Print the Root Content

T->Scan();//scan the Root rows and columns

Play with Root Tree

Command T->Print(); Action Prints the content of the tree

T->Scan();
T->Draw(x); How to apply cuts: T->Draw(x,x>0); T->Draw(x,x>0 && y>0); T->Draw(y, ,same);

Scans the rows and columns

Draw a branch of tree Draw x when x>0 Draw x when both x >0 and y >0 Superimpose y on x

T->Draw(y:x);
T->Draw(z:y:x); T->Draw(sqrt(x*x+y*y)); T->Draw(x>>h1);

Make y vs x 2d scatter plot

Make z:y:x 3d plot Plot calculated quantity Dump a root branch to a histogram

Play with Root Tree

Create Histogram from Root Tree

root [2] TH1F *h1 = new TH1F("h1","hist from tree",50, -4, 4); root [3] T->Draw("x>>h1");

How to deal with number of large Root files with same trees ?
TChain chain("T"); // name of the tree is the argument chain.Add("file1.root"); chain.Add("file2.root"); chain.Add("file3.root); You can draw x from all the files in the chain at the same time chain.Draw("x");

Fitting Histograms in Root

Histograms (1-D,2-D,3-D and Profiles) can be fitted with a user specified function via TH1::Fit. It uses MINUIT as the minimization routine for fitting,

Fitting Histogram with Fit Panel

Same as: h1->Fit(gaus);

Fitting Multiple Sub Ranges

Fitting Multiple Sub Ranges contd.

Double_t par[9]; TF1 *g1 = new TF1("g1","gaus",85,95); TF1 *g2 = new TF1("g2","gaus",98,108); TF1 *g3 = new TF1("g3","gaus",110,121); TF1 *total = new TF1("total","gaus(0)+gaus(3)+gaus(6)",85,125); total->SetLineColor(2); h->Fit(g1,"R"); h->Fit(g2,"R+"); h->Fit(g3,"R+"); g1->GetParameters(&par[0]); g2->GetParameters(&par[3]); g3->GetParameters(&par[6]); total->SetParameters(par); h->Fit(total,"R+");

Fitting with Combining Functions

Fitting with RooFit (http://roofit.sourceforge.net/)

RooFit packages provide a toolkit for modeling the expected distribution of events in a physics analysis Models can be used to perform likelihood fits, produce plots, and generate "toy Monte Carlo" samples for various studies

Next Lecture Analysis of Muon Calibration Simulation Data with ROOT