Mathematical libraries

The ROOT Mathematical libraries consist of the following components:

MathCore library

The MathCore library provides a collection of functions, C++ classes and ROOT classes for HEP numerical computing.
The MathCore is a self-consistent minimal set of tools required for the basic numerical computing. More advanced mathematical functionalities is provided by the MathMore library. The following is included in the MathCore library:

In addition, the MathCore library contains the following ROOT classes that were originally part of libCore:


The TMath namespace provides a collection of free functions:

Elementary functions

Some of elementary mathematical functions refer to basic mathematical functions like the square root, the power to a number of the calculus of a logarithm, while others are used for number treatment, like rounding.

Although there are some functions that are not in the standard C math library (like Factorial), most of the functionality offered here is just a wrapper of the first ones. Nevertheless, some of the them also offer some security checks or a better precision, like the trigonometrical functions ASin(x), ACos(x) or ATan(x).


// Generate a vector with 10 random numbers.
   vector<double> v(10);
   std::generate(v.begin(), v.end(), rand);

// Find the minimum value of the vector (iterator version).
   vector<double>::iterator it;
   it = TMath::LocMin(v.begin(), v.end());
   std::cout << *it << std::endl;

// The same with the old-style version.
   int i;
   i = TMath::LocMin(10, &v[0]);
   std::cout << v[i] << std::endl;

Statistic functions operating on arrays

TMath provides functions that process arrays for calculation:

  • mean
  • median
  • geometrical mean
  • sample standard deviation (RMS)
  • the kth smallest element


// Size of the array.
   const int n = 100;

// Vector v with random values.
   vector<double> v(n);
   std::generate(v.begin(), v.end(), rand);

// Weight vector w.
   vector<double> w(n);
   std::fill(w.begin(), w.end, 1);
   double mean;

// Calculate the mean of the vector with iterators.
   mean = TMath::Mean(v.begin(), v.end());

Special and statistical functions

TMath provides special functions like Bessel, error functions, Gamma or similar plus statistical mathematical functions, including probability density functions, cumulative distribution and their inverse.

The majority of the special functions and the statistical distributions are provided also as free functions in the ROOT::Math namespace.

Functions not present in the ROOT::Math name that are provided only by TMath are:

  • Special functions:
    • DiLogarithm
    • Struve
  • Statistical functions:
    • KolmogorovProb
    • Voigt function
    • LaplaceDist
    • Vavilov


The ROOT::Math namespace provides a set of function interfaces to define the basic behaviour of a mathematical function:

In addition, helper classes, wrapping the user interfaces in the ROOT::Math function interfaces are provided. With wrapper functions you can insert your own type of function in the needed function interface.

To use the self-defined functions, they must have inherited from one of the following classes:

Figure: ROOT::Math function interface structure.

One-dimensional function interfaces

This interface is used for numerical algorithms operating only on one-dimensional functions. It cannot applied to multi-dimensional functions.

This interface provides a method to evaluate the function given a value (simple double) by implementing double operator() (const double). The user class defined only needs to reimplement the pure abstract method double DoEval(double x) that will do the work of evaluating the function at point x.


Example for the implementation of a class that represents a mathematical function.

   #include "Math/IFunction.h"
   class MyFunction: public ROOT::Math::IBaseFunctionOneDim
      double DoEval(double x) const
      return x*x;
   ROOT::Math::IBaseFunctionOneDim* Clone() const
      return new MyFunction();

This interface is needed by some numerical algorithms to calculate the derivatives of the function. It introduces the method double Derivative(double x) that will return the derivative of the function at the point x. The class inherit by the user will have to implement the abstract method double DoDerivative(double x), leaving the rest of the class untouched.


Example for the implementation of a gradient one-dimensional function.

   #include "Math/IFunction.h"
   class MyGradientFunction: public ROOT::Math::IGradientFunctionOneDim
   double DoEval(double x) const
      return sin(x);
   ROOT::Math::IBaseFunctionOneDim* Clone() const
      return new MyGradientFunction();
   double DoDerivative(double x) const
      return -cos(x);

Multi-dimensional function interfaces

This interface is used for numerical algorithms operating on multi-dimensional functions.

This interface provides the double operator() (const double*) that takes an array of doubles with all the values for the different dimensions. In this case, the user has to provide the functionality for two different functions: double DoEval(const double*) and unsigned int NDim(). The first ones evaluates the function given the array that represents the multiple variables. The second returns the number of dimensions of the function.


Example for the implementation of a basic multi-dimensional function.

   #include "Math/IFunction.h"
   class MyFunction: public ROOT::Math::IBaseFunctionMultiDim
   double DoEval(const double* x) const
      return x[0] + sin(x[1]);
   unsigned int NDim() const
      return 2;
   ROOT::Math::IBaseFunctionMultiDim* Clone() const
      return new MyFunction();

This interface offers the same functionality as the base function and additionally the calculation of the derivative. It only adds the double Derivative(double* x, uint ivar) method for the user to implement. This method must implement the derivative of the function with respect to the variable indicated with the second parameter.


Example for the implementation of a multi-dimensional gradient function.

   #include "Math/IFunction.h"
   class MyGradientFunction: public ROOT::Math::IGradientFunctionMultiDim
   double DoEval(const double* x) const
      return x[0] + sin(x[1]);
   unsigned int NDim() const
      return 2;
   ROOT::Math::IGradientFunctionMultiDim* Clone() const
      return new MyGradientFunction();
   double DoDerivative(const double* x, unsigned int ipar) const
      if ( ipar == 0 )
      return sin(x[1]);
      return x[0] + x[1] * cos(x[1]);

Parametric function interfaces

This interface is used for fitting after evaluating multi-dimensional functions.

This interface describes a multi-dimensional parametric function. Similarly to the one dimensional version, the user needs to provide the void SetParameters(double* p) method as well as the getter methods const double * Parameters() and uint NPar().


Example for the implementation of a parametric function.

   #include "Math/IFunction.h"
   #include "Math/IParamFunction.h"
   class MyParametricFunction: public ROOT::Math::IParametricFunctionMultiDim
   const double* pars;
   double DoEvalPar(const double* x, const double* p) const
      return p[0] * x[0] + sin(x[1]) + p[1];
   unsigned int NDim() const
      return 2;
   ROOT::Math::IParametricFunctionMultiDim* Clone() const
     return new MyParametricFunction();
   const double* Parameters() const
      return pars;
   void SetParameters(const double* p)
      pars = p;
   unsigned int NPar() const
      return 2;

This interface provides an interface for parametric gradient multi-dimensional functions. In addition to function evaluation, it provides the gradient with respect to the parameters, via the ParameterGradient() method. This interface is only used in case of some dedicated fitting algorithms, when is required or more efficient to provide derivatives with respect to the parameters.


Example for the implementation of a parametric gradient function.

   #include "Math/IFunction.h"
   #include "Math/IParamFunction.h"
   class MyParametricGradFunction:
   public ROOT::Math::IParametricGradFunctionMultiDim
   const double* pars;
   double DoEvalPar(const double* x, const double* p) const
      return p[0] * x[0] + sin(x[1]) + p[1];
   unsigned int NDim() const
      return 2;
   ROOT::Math::IParametricGradFunctionMultiDim* Clone() const
      return new MyParametricGradFunction();
   const double* Parameters() const
      return pars;
   void SetParameters(const double* p)
      pars = p;
   unsigned int NPar() const
      return 2;
   double DoParameterDerivative(const double* x, const double* p,
   unsigned int ipar) const
      if ( ipar == 0 )
      return sin(x[1]) + p[1];
      return p[0] * x[0] + x[1] * cos(x[1]) + p[1];

Wrapper functions

To insert your own type of function in the needed function interface, helper classes, wrapping the user interface in the ROOT::Math function interfaces are provided.

There is one possible wrapper for every interface.

Interface Wrapper Description
ROOT::Math::IBaseFunctionOneDim ROOT::Math::Functor1D See → Wrapping one-dimensional functions
ROOT::Math::IGradientFunctionOneDim ROOT::Math::GradFunctor1D See → Wrapping one-dimensional gradient functions
ROOT::Math::IBaseFunctionMultiDim ROOT::Math::Functor See → Wrapping multi-dimensional functions
ROOT::Math::IGradientFunctionMultiDim ROOT::Math::GradFunctor See → Wrapping multi-dimensional gradient functions

Note the special case when wrapping TF1 objects in parametric function interfaces.

Wrapping one-dimensional functions

Use ROOT::Math::Functor1D to wrap one-dimensional functions.

ROOT::Math::Functor1D can wrap the following types:

  • A free C function of type double ()(double ).
  • Any C++ callable object implementation double operator()( double).
  • A class member function with the correct signature like double Foo::Eval(double ). In this case one pass the object pointer and a pointer to the member function (&Foo::Eval).


   #include "Math/Functor.h"

   class MyFunction1D {
   double operator()(double x) const {
   return x*x;
   double Eval(double x) const { return x+x; }
   double freeFunction1D(double x ) {
   return 2*x;
     int main()

// Wrapping a free function.
      ROOT::Math::Functor1D f1(&freeFunction1D);
      MyFunction1D myf1;
// Wrapping a function object implementing operator().
      ROOT::Math::Functor1D f2(myf1);

// Wrapping a class member function.
      ROOT::Math::Functor1D f3(&myf1,&MyFunction1D::Eval);
      cout << f1(2) << endl;
      cout << f2(2) << endl;
      cout << f3(2) << endl;
      return 0;

Wrapping one-dimensional gradient functions

Use ROOT::Math::GradFunctor1D to wrap one-dimensional gradient functions.

It can be constructed in three different ways:

  • Any object implementing both double operator()( double) for the function evaluation and double Derivative(double) for the function derivative.
  • Any object implementing any member function like Foo::XXX(double ) for the function evaluation and any other member function like Foo::YYY(double) for the derivative.
  • Any two function objects implementing double operator()( double). One object provides the function evaluation, the other the derivative. One or both function object can be a free C function of type double ()(double).

Wrapping multi-dimensional functions

Use the ROOT::Math::Functor to wrap multi-dimensional function objects.

It can wrap all the following types:

  • Any C++ callable object implementing double operator()( const double * ).
  • A free C function of type double ()(const double *).
  • A member function with the correct signature like Foo::Eval(const double *). In this case one pass the object pointer and a pointer to the member function (&Foo::Eval).


   #include "Math/Functor.h"
   class MyFunction {
   double operator()(const double *x) const {
   return x[0]+x[1];
   double Eval(const double * x) const { return x[0]+x[1]; }
   double freeFunction(const double * x )
     return x[0]+x[1];
   int main()
// Test directly calling the function object.
    MyFunction myf;

// Test from a free function pointer.
    ROOT::Math::Functor f1(&freeFunction,2);

// Test from function object.
    ROOT::Math::Functor f2(myf,2);

// Test from a member function.
    ROOT::Math::Functor f3(&myf,&MyFunction::Eval,2);
    double x[] = {1,2};
    cout << f1(x) << endl;
    cout << f2(x) << endl;
    cout << f3(x) << endl;
    return 0;

Wrapping multi-dimensional gradient functions

Use ROOT::Math::GradFunctor to wrap C++ callable objects to make gradient functions.

It can be constructed in three different way:

  • From an object implementing both double operator()( const double*) for the function evaluation and double Derivative(const double *, int icoord) for the partial derivatives.
  • From an object implementing any member function like Foo::XXX(const double *) for the function evaluation and any member function like Foo::XXX(const double *, int icoord) for the partial derivatives.
  • From an function object implementing double operator()( const double *) for the function evaluation and another function object implementing double operator() (const double *, int icoord) for the partial derivatives.

The function dimension is required when constructing the functor.

Wrapping TF1 objects in parametric function interfaces

Often the TF1 class is used.
Use the ROOT::Math::WrappedTF1 class, if the interface to be wrapped is one-dimensional.

The default constructor takes a TF1 reference as argument, that will be wrapped with the interfaces of a ROOT::Math::IParametricGradFunctionOneDim.


   #include "TF1.h"
   #include "Math/WrappedTF1.h"
   int main()
      TF1 f("Sin Function", "sin(x)+y",0,3);
      ROOT::Math::WrappedTF1 wf1(f);
      cout << f(1) << endl;
      cout << wf1(1) << endl;
      return 0;

Use the ROOT::Math::WrappedMultiTF1 class, if the interface to be wrapped is multi-dimensional.

Following the usual procedure, setting the TF1 though the constructor, will wrap it into a ROOT::Math::IParametricGradFunctionMultiDim.


   #include "TF1.h"
   #include "Math/WrappedMultiTF1.h"
   int main()
      TF2 f("Sin Function", "sin(x) + y",0,3,0,2);
      ROOT::Math::WrappedMultiTF1 wf1(f);
      double x[] = {1,2};
      cout << f(x) << endl;
      cout << wf1(x) << endl;
      return 0;

Random numbers

The MathCore library provides the following classes for generating pseudo-random numbers:

  • TRandom Using a linear congruential random generator.
  • TRandom1 Random number generator based on the Ranlux engine.
  • TRandom2 Based on the maximally equi-distributed combined Tausworthe generator by L’Ecuyer.
  • TRandom3 Based on the Mersenne and Twister pseudo-random number generator.


For generating non-uniform random numbers, the UNU.RAN package (see → UNU.RAN) is available.

You can work with the random number generators as follows:

Seeding the random number generators

When no value is given, the generator default seed is used. In this case an identical sequence will be generated every time the application is run.
When the 0 value is used as seed, then a unique seed is generated using a TUUID, for TRandom , TRandom1 and TRandom3 .
For TRandom the seed is generated using only the machine clock, which has a resolution of about 1 s. Therefore, identical sequences will be generated if the elapsed time is less than a second.

Using the random number generators

  • Use the Rndm() method for generating a pseudo-random number distributed between 0 and 1.


// Use the default seed (same random numbers will be generated each time).
// Generate a number in the interval ]0,1] (0 is excluded).
   TRandom3 r;
   double x[100];

// Generate an array of random numbers in ]0,1].

// Construct with a user-defined seed.
   TRandom3 rdm(111);

// Use 0: a unique seed will be automatically generated using TUUID.
   TRandom1 r1(0);
   TRandom2 r2(0);
   TRandom3 r3(0);

// Seed generated using machine clock (different every second).
   TRandom r0(0);

Random number distributions

The TRandom class provides functions that can be used by all other derived classes to generate random variables according to predefined distributions. In the simplest cases, as in the exponential distribution, the non-uniform random number is obtained by suitable transformations. In the more complicated cases, the random variables are obtained by acceptance-rejection methods that require several random numbers.


   TRandom3 r;
// Generate a gaussian distributed number with:
// mu=0, sigma=1 (default values)
   double x1 = r.Gaus();
   double x2 = r.Gaus(10,3);
// Use mu = 10, sigma = 3;

The following table shows the various distributions that can be generated using methods of the TRandom classes.
In addition, you can use TF1::GetRandom() or TH1::GetRandom() to generate random numbers distributed according to a user defined function, in a limited interval, or to a user defined histogram.

Distributions Description
Double_t Uniform(Double_t x1,Double_t x2) Uniform random numbers between x1,x2.
Double_t Gaus(Double_t mu,Double_t sigma) Gaussian random numbers. Default values: mu=0, sigma=1.
Double_t Exp(Double_t tau) Exponential random numbers with mean tau.
Double_t Landau(Double_t mean,Double_t sigma) Landau distributed random numbers. Default values: mean=0, sigma=1.
Double_t BreitWigner(Double_t mean,Double_t gamma) Breit-Wigner distributed random numbers. Default values mean=0, gamma=1.
Int_t Poisson(Double_t mean) Poisson random numbers.
Double_t PoissonD(Double_t mean) Poisson random numbers.
Int_t Binomial(Int_t ntot,Double_t prob) Binomial Random numbers
Circle(Double_t &x,Double_t &y,Double_t r) Generate a random 2D point (x,y) in a circle of radius r.
Sphere(Double_t &x,Double_t &y,Double_t &z,Double_t r) Generate a random 3D point (x,y,z) in a sphere of radius r.
Rannor(Double_t &a,Double_t &b) Generate a pair of Gaussian random numbers with mu=0 and sigma=1.

Complex numbers

The MathCore library provides with TComplex a class for complex numbers.

MathMore library

The MathMore library provides an advanced collection of functions and C++ classes for numerical computing. This is an extension of the functionality provided by the MathCore library. The MathMore library is implemented wrapping in C++ the GNU Scientific Library (GSL). The mathematical functions are implemented as a set of free functions in the namespace ROOT::Math.

The MathMore library includes classes and functions for:

Linear algebra packages

The linear algebra packages provide a complete environment in ROOT to perform calculations like equation solving and eigenvalue decompositions.

There are the following linear algebra packages available:

Matrix package

Overview of the matrix package

  • matrix classes
  • matrix properties
  • creating and filling a matrix
  • inverting a matrix
  • matrix operators and methods
  • matrix views
  • matrix decompositions
  • matrix Eigen analysis

Matrix classes

ROOT provides the following matrix classes, among others:

  • TMatrixDBase: Base class for matrices.

  • TMatrixF: Matrix with single precision (float).

  • TMatrixFSym: Symmetrical matrix with single precision (float).

  • TVectorF: Vector with single precision (float).

  • TMatrixD: Matrix with double precision (double).

  • TMatrixDSym: Symmetrical matrix with double precision (double).

  • TMatrixDSparse: Sparse matrix with double precision (double).

  • TDecompBase
    Decomposition base class.
  • TDecompChol
    Cholesky decomposition class.

Matrix properties

A matrix has five properties, which are all set in the constructor:

  • precision
    If the precision is float (this is single precision), use the TMatrixF class family. If the precision is double, use the TMatrixD class family.

  • type
    Possible values are: general (TMatrixD), symmetric (TMatrixDSym) or sparse (TMatrixDSparse).

  • size
    Number of rows and columns.

  • index
    Range start of row and column index. By default these start at 0.

  • sparse map
    Only relevant for a sparse matrix. It indicates where elements are unequal 0.

You can:

Accessing matrix properties

Use one of the following methods to access the information about the relevant matrix property:

GetRowIndexArray()]({:target=”_blank”} and [GetColIndexArray() are specific to the sparse matrix, which is implemented according to the Harwell- Boeing format. Here, besides the usual shape/size descriptors of the matrix like fNrows, fRowLwb, fNcols and fColLwb, also a row index fRowIndex and a column index fColIndex are stored:

  • fRowIndex[0,..,fNrows]: Stores for each row the index range of the elements in the data and column array.
  • fColIndex[0,..,fNelems-1]: Stores the column number for each data element != 0.

Setting matrix properties

Use one of the following methods to set a matrix property:

  • SetTol (Double_t tol)
    Sets the tolerance number.

  • ResizeTo (Int_t nrows,Int_t ncols, Int_t nr_nonzeros=-1)
    Changes the matrix shape to nrows x ncols. Index will start at 0.

  • ResizeTo(Int_t row_lwb,Int_t row_upb, Int_t col_lwb,Int_t col_upb, Int_t nr_nonzeros=-1)
    Changes the matrix shape to row_lwb:row_upb x col_lwb:col_upb.

  • SetRowIndexArray (Int_t *data)
    For sparse matrices, it sets the row index. The array data should contain at least fNrows+1 entries column lower-bound index.

  • SetColIndexArray (Int_t *data)
    For sparse matrices, it sets the column index. The array data should contain at least fNelems entries.

  • SetSparseIndex (Int_t nelems new)
    Allocates memory for a sparse map of nelems_new elements and copies (if exists) at most nelems_new matrix elements over to the new structure.

  • SetSparseIndex (const TMatrixDBase &a)
    Copies the sparse map from matrix a.

  • SetSparseIndexAB (const TMatrixDSparse &a, const TMatrixDSparse &b)
    Sets the sparse map to the same map of matrix a and b.

Creating and filling a matrix

Use one of the following constructors to create a matrix:

  • TMatrixD(Int_t nrows,Int_t ncols)
  • TMatrixD(Int_t row_lwb,Int_t row_upb,Int_t col_lwb,Int_t col_upb)
  • TMatrixD(Int_t nrows,Int_t ncols,const Double_t *data, Option_t option= "")
  • TMatrixD(Int_t row_lwb,Int_t row_upb,Int_t col_lwb,Int_t col_upb, const Double_t *data,Option_t *option="")
  • TMatrixDSym(Int_t nrows)
  • TMatrixDSym(Int_t row_lwb,Int_t row_upb)
  • TMatrixDSym(Int_t nrows,const Double_t *data,Option_t *option="")
  • TMatrixDSym(Int_t row_lwb,Int_t row_upb, const Double_t *data, Option_t *opt="")
  • TMatrixDSparse(Int_t nrows,Int_t ncols)
  • TMatrixDSparse(Int_t row_lwb,Int_t row_upb,Int_t col_lwb, Int_t col_upb)
  • TMatrixDSparse(Int_t row_lwb,Int_t row_upb,Int_t col_lwb,Int_t col_upb, Int_t nr_nonzeros,Int_t *row,Int_t *col,Double_t *data)

Use one of the following methods to fill a matrix:

  • SetMatrixArray(const Double_t*data,Option_t*option="")
    Copies array data. If option="F", the array fills the matrix column-wise else row-wise. This option is implemented for TMatrixD and TMatrixDSym. It is expected that the array data contains at least fNelems entries.

  • SetMatrixArray(Int_t nr,Int_t *irow,Int_t *icol,Double_t *data)
    Only available for sparse matrices. The three arrays should each contain nr entries with row index, column index and data entry. Only the entries with non-zero data value are inserted.

  • operator(), operator[]
    These operators provide the easiest way to fill a matrix but are in particular for a sparse matrix expensive. If no entry for slot (i,j) is found in the sparse index table it will be entered, which involves some memory management. Therefore, before invoking this method in a loop set the index table first through a call to the SetSparseIndex() method.

  • SetSub(Int_t row_lwb,Int_t col_lwb,const TMatrixDBase &source)
    The matrix to be inserted at position (row_lwb,col_lwb) can be both, dense or sparse.

  • Use()
    Allows inserting another matrix or data array without actually copying the data.
    The following list shows the application of the Use() method:

    • Use(TMatrixD &a)
    • Use(Int_t row_lwb,Int_t row_upb,Int_t col_lwb,Int_t col_upb,Double_t *data)
    • Use(Int_t nrows,Int_t ncols,Double_t *data)
    • Use(TMatrixDSym &a)
    • Use(Int_t nrows,Double_t *data)
    • Use(Int_t row_lwb,Int_t row_upb,Double_t *data)
    • Use(TMatrixDSparse &a)
    • Use(Int_t row_lwb,Int_t row_upb,Int_t col_lwb,Int_t col_upb,Int_t nr_no nzeros, Int_t *pRowIndex,Int_t *pColIndex,Double_t *pData)
    • Use(Int_t nrows,Int_t ncols,Int_t nr_nonzeros,Int_t *pRowIndex,Int_t *pColIndex,Double_t *pData)


A Hilbert matrix is created by copying an array.

   TMatrixD h(5,5);
   TArrayD data(25);
   for (Int_t = 0; i < 25; i++) {
      const Int_t ir = i/5;
      const Int_t ic = i%5;
      data[i] = 1./(ir+ic);

You can also assign the data array to the matrix without actually copying it.

   TMatrixD h; h.Use(5,5,data.GetArray());

The array data now contains the inverted matrix.

Now a unit matrix in sparse format is created.

   TMatrixDSparse unit1(5,5);
   TArrayI row(5),col(5);
   for (Int_t i = 0; i < 5; i++) row[i] = col[i] = i;
   TArrayD data(5); data.Reset(1.);

   TMatrixDSparse unit2(5,5);

Inverting a matrix

  • Use the Invert(Double_t &det=0) function to invert a matrix:
   TMatrixD a(...);

– or –

  • Use the appropriate constructor to invert a matrix:
   TMatrixD b(kInvert,a);

Both methods are available for general and symmetric matrices.

For matrices whose size is less than or equal to 6x6, the InvertFast(Double_t &det=0) function is available. Here the Cramer algorithm will be applied, which is faster but less accurate.

Using decomposition classes for inverting

You can also use the following decomposition classes (see → Matrix decompositions) for inverting a matrix:

Name Matrix type Comment
TDecompLU General
TDecompQRH General
TDecompSVD General Can manipulate singular matrix.
TDecompBK symmetric
TDecompChol Symmetric Matrix should also be positive definite.
TDecompSparse Sparse

If the required matrix type is general, you also can handle symmetric matrices.


This example shows how to check whether the matrix is singular before attempting to invert it.

   TDecompLU lu(a);
   TMatrixD b;
   if (!lu.Decompose()) {
      cout << "Decomposition failed, matrix singular ?" << endl;
      cout << "condition number = " << = a.GetCondition() << endl;
   } else {

Matrix operators and methods

The matrix/vector operations are classified according to BLAS (basic linear algebra subroutines) levels.

The following operations and methods are available:

Arithmetic operations between matrices

Description Format Comment
Element C=A+B overwrites A
Wise sum A+=B
Add (A,alpha,B)
A = A + α B constructor
Element wise substraction C=A-B A-=B
overwrites A
Matrix multiplication C=A*B
TMatrixD(A, TMatrixD(A, TMatrixD::kTransposeMult,B)
TMatrixD(A, TMatrixD::kMultTranspose,B)
overwrites A
constructor of A.B
constructor of AT .B
constructor of A.BT
Element wise multiplication ElementMult(A,B) A(i,j)*= B(i,j)
Element wise division ElementDiv(A,B) A(i,j)/= B(i,j)

Arithmetic operations between matrices and real numbers

Description Format Comment
Element wise sum C=r+A C=A+r A+=r overwrites A
Element wise subtraction C=r-A C=A-r A-=r overwrites A
Matrix multiplication C=r*A C=A*r A*=r overwrites A

Comparison between two matrices

Description Output Descriptions
A == B Bool_t equal to
A != B matrix not equal
A > B matrix greater than
A >= B matrix greater than or equal to
A < B matrix smaller than
A <= B matrix smaller than or equal to
AreCompatible(A,B) Bool_t compare matrix properties
Compare(A,B) Bool_t return summary of comparison
VerifyMatrixIdentity(A,B,verb, maxDev)   check matrix identity within maxDev tolerance

Comparison between matrix and real number

Format Output Description
A == r Bool_t equal to
A != r Bool_t not equal
A > r Bool_t greater than
A >= r Bool_t greater than or equal to
A < r Bool_t smaller than
A <= r Bool_t smaller than or equal to
VerifyMatrixValue(A,r,verb, maxDev) Bool_t compare matrix value with r within maxDev tolerance
A.RowNorm() Double_t norm induced by the infinity vector norm
A.NormInf() Double_t  
A.ColNorm() Double_t norm induced by the 1 vector norm
A.Norm1() Double_t  
A.E2Norm() Double_t square of the Euclidean norm
A.NonZeros() Int_t  
A.Sum() Double_t number of elements unequal zero
A.Min() Double_t  
A.Max() Double_t  
A.NormByColumn (v,"D") TMatrixD  
A.NormByRow (v,"D") TMatrixD  

Matrix views

With the following matrix view classes, you can access the matrix elements:

  • TMatrixDRow
  • TMatrixDColumn
  • TMatrixDDiag
  • TMatrixDSub

Matrix view operators

For the matrix view classes TMatrixDRow, TMatrixDColumn and TMatrixDDiag, the necessary assignment operators are available to interact with the vector class TVectorD.
The sub matrix view classes TMatrixDSub has links to the matrix classes TMatrixD and TMatrixDSym.

The next table summarizes how to access the individual matrix elements in the matrix view classes.

Format Comment
TMatrixDRow(A,i)(j) TMatrixDRow(A,i)[j] element Aij
TMatrixDColumn(A,j)(i) TMatrixDColumn(A,j)[i] element Aij
TMatrixDDiag(A(i) TMatrixDDiag(A[i] element Aij
TMatrixDSub(A(i) TMatrixDSub(A,rl,rh,cl,ch)(i,j) element Aij
element Arl+i,cl+j

Matrix decompositions

There are the following classes available for matrix decompositions:

  • TDecompLU
    Decomposes a general n x n matrix A into P A = L U.
  • TDecompBK
    The Bunch-Kaufman diagonal pivoting method decomposes a real symmetric matrix A.
  • TDecompChol
    The Cholesky decomposition class, which decomposes a symmetric, positive definite matrix A = U^T * U where U is a upper triangular matrix.
  • TDecompQRH
    QR decomposition class.
  • TDecompSVD
    Single value decomposition class.
  • TDecompSparse
    Sparse symmetric decomposition class.

Matrix Eigen analysis

With the TMatrixDEigen and TMatrixDSymEigen classes, you can compute eigenvalues and eigenvectors for general dense and symmetric real matrices.


SMatrix is a C++ package for high performance vector and matrix computations. It can be used only in problems when the size of the matrices is known at compile time, like in the tracking reconstruction of HEP experiments. It is based on a C++ technique, called expression templates, to achieve an high level optimization. The C++ templates can be used to implement vector and matrix expressions such that these expressions can be transformed at compile time to code which is equivalent to hand optimized code in a low-level language like FORTRAN or C.

The SMatrix has been developed initially by T. Glebe of the Max-Planck-Institut, Heidelberg, as part of the HeraB analysis framework. A subset of the original package has been now incorporated in the ROOT distribution, with the aim to provide to the LHC experiments a stand-alone and high performance matrix package for reconstruction. The API of the current package differs from the original one, in order to be compliant to the ROOT coding conventions.

SMatrix contains the following generic classes for describing matrices and vectors of arbitrary dimensions and of arbitrary type:


The template class ROOT::Math::SVector represents n-dimensional vectors for objects of arbitrary type. This class has 2 template parameters, which define at compile time, its properties:

  1. type of the contained elements (for example float or double)
  2. size of the vector


The template class ROOT::Math::SMatrix represents a matrix of arbitrary type with nrows x ncoldimension. The class has 4 template parameters, which define at compile time, its properties:

  • type of the contained elements, T, for example float or double
  • number of rows
  • number of columns
  • representation type

Minimization libraries and classes

ROOT provides several minimization libraries and classes:


The Minuit minimization package was originally written in Fortran by Fred James and part of PACKLIB (patch D506). It has been converted to the C++ class TMinuit , by R.Brun.

Topical manual

For TMinuit, a topical manual it available at Topical Manual - TMinuit.
It contains in-depth information about TMinuit.

Minuit2 library

The Minuit2 library is a new object-oriented implementation, written in C++, of the popular MINUIT minimization package. These new version provides basically all the functionality present in the old Fortran version, with almost equivalent numerical accuracy and computational performances. Furthermore, it contains new functionality, like the possibility to set single side parameter limits or the FUMILI algorithm, which is an optimized method for least square and log likelihood minimizations. The package has been originally developed by M. Winkler and F. James.

Topical manuals

For Minuit2, topical manuals are available at Topical Manuals - Minuit2.
They contain in-depth information about Minuit2.


UNU.RAN (Universal Non Uniform RAndom Number generator for generating non-uniform pseudo-random numbers) contains universal (also called automatic or black-box) algorithms that can generate random numbers from large classes of continuous (in one or multi-dimensions), discrete distributions, empirical distributions (like histograms) and also from practically all standard distributions.

UNU.RAN is an ANSI C library licensed under GPL.

The TUnuran class is used to interface the UNURAN package.


UNU:RAN tutorials are available at →

Initializing TUnuran with string API

You can initialize UNU.RAN with the string API via TUnuran::Init(), passing the distribution type and the method.


TUnuran unr;
// Initialize UNU.RAN to generate normal random numbers using an "arou" method.

// Sample distributions N times (generate N random numbers).
   for (int i = 0; i<N; ++i)
   double x = unr.Sample();

Using TUnuranContDist for a one-dimensional distribution

  • Use TUnuranContDist for creating a continuous 1-D distribution object (for example from a TF1 object providing the PDF (probability density function).
    You can provide additional information via TUnuranContDist::SetDomain(min,max) like the domain() for generating numbers in a restricted region.


// 1D case: create a distribution from two TF1 object, pointers pdfFunc.
   TUnuranContDist dist(pdfFunc);

// Initialize UNU.RAN, passing the distribution and a string.
// Define the method.
   unr.Init(dist, "method=hinv");

// Sample distribution N times (generate N random numbers).
   for (int i = 0; i < N; ++i)
   double x = unr.Sample();

Using TUnuranMultiContDist for a multi-dimensional distribution

  • Use TUnuranMultiContDist to create a multi-dimensional distribution, which can be created from a multi-dimensional PDF (probability density function).


// Multi- dimensional case from TF1 (TF2 or TF3) objects.
   TUnuranMultiContDist dist(pdfFuncMulti);

// The recommended method for a multi-dimensional function is "hitro".

// Sample distribution N times (generate N random numbers).
   double x[NDIM];
   for (int i = 0; i<N; ++i)

Using TUnuranDiscrDist for a discrete one-dimensional distribution

  • Use TUnuranDiscrDist to create a discrete one-dimensional distribution, which can be initialized from a TF1 object or from a vector of probabilities.


// Create a distribution from a vector of probabilities.
   double pv[NSize] = {0.1,0.2,...};
   TUnuranDiscrDist dist(pv,pv+NSize);

// The recommended method for a discrete distribution is "dgt".
   unr.Init(dist, "method=dgt");

// Sample N times (generate N random numbers).
   for (int i = 0; i < N; ++i)
   int k = unr.SampleDiscr();

Using TUnuranEmpDist an empirical distribution

Use TUnuranEmpDist for creating an empirical distribution, which can be initialized from a TH1 object (using the bins or from its buffer for un-binned data) or from a vector of data.


// Create a distribution from a set of data.
// vdata is an std::vector containing the data.
   TUnuranEmpDist dist(vdata.begin(),vdata.end());

// Sample N times (generate N random numbers).
   for (int i = 0; i<N; ++i)
   double x = unr.Sample();