class THnSparse: public TNamed

 Destruct a THnSparse

 Add "v" to the content of bin with coordinates "coord"

Create a new chunk of bin content

 Create a new THnSparse object that is of the same type as *this,
 but with dimensions and bins given by axes.
 If keepTargetAxis is true, the axes will keep their original xmin / xmax,
 else they will be restricted to the range selected (first / last).

 Create an empty histogram with name and title with a given
 set of axes. Create a TH1D/TH2D/TH3D, depending on the number
 of elements in axes.

Arrays that will be needed later. Initialized to 0.

   Fit a THnSparse with function f

   since the data is sparse by default a likelihood fit is performed
   merging all the regions with empty bins for betetr performance efficiency

  Since the THnSparse is not drawn no graphics options are passed
  Here is the list of possible options

                = "I"  Use integral of function in bin instead of value at bin center
                = "X"  Use chi2 method (default is log-likelihood method)
                = "U"  Use a User specified fitting algorithm (via SetFCN)
                = "Q"  Quiet mode (minimum printing)
                = "V"  Verbose mode (default is between Q and V)
                = "E"  Perform better Errors estimation using Minos technique
                = "B"  Use this option when you want to fix one or more parameters
                       and the fitting function is like "gaus", "expo", "poln", "landau".
                = "M"  More. Improve fit results
                = "R"  Use the Range specified in the function range

 Get the bin index for the n dimensional tuple x,
 allocate one if it doesn't exist yet and "allocate" is true.

 Get the bin index for the n dimensional tuple addressed by "name",
 allocate one if it doesn't exist yet and "allocate" is true.

 Get the bin index for the n dimensional coordinates coord,
 allocate one if it doesn't exist yet and "allocate" is true.

 Get content of bin with coordinates "coord"

 Return the content of the filled bin number "idx".
 If coord is non-null, it will contain the bin's coordinates for each axis
 that correspond to the bin.

 Get error of bin with coordinates "coord" as


 If errors are not enabled (via Sumw2() or CalculateErrors())
 return sqrt(contents).

 Get error of bin addressed by linidx as


 If errors are not enabled (via Sumw2() or CalculateErrors())
 return sqrt(contents).

 Return the index for fCurrentBinIndex.
 If it doesn't exist then return -1, or allocate a new bin if allocate is set

 Return THnSparseCompactBinCoord object.

 Generate an n-dimensional random tuple based on the histogrammed
 distribution. If subBinRandom, the returned tuple will be additionally
 randomly distributed within the randomized bin, using a flat
 distribution.

 Return the amount of filled bins over all bins

 Return the amount of used memory over memory that would be used by a
 non-sparse n-dimensional histogram. The value is approximate.

 Check whether bin coord is in range, as defined by TAxis::SetRange().
 Currently, TAxis::SetRange() does not allow to select all but over- and
 underflow bins (it instead resets the axis to "no range selected").
 Instead, simply call
    TAxis* axis12 = hsparse.GetAxis(12);
    axis12->SetRange(1, axis12->GetNbins());
    axis12->SetBit(TAxis::kAxisRange);
 to deselect the under- and overflow bins in the 12th dimension.

 Project all bins into a 1-dimensional histogram,
 keeping only axis "xDim".
 If "option" contains "E" errors will be calculated.
                      "A" ranges of the taget axes will be ignored.
                      "O" original axis range of the taget axes will be
                          kept, but only bins inside the selected range
                          will be filled.

 Project all bins into a 2-dimensional histogram,
 keeping only axes "xDim" and "yDim".
 If "option" contains "E" errors will be calculated.
                      "A" ranges of the taget axes will be ignored.

 Project all bins into a 3-dimensional histogram,
 keeping only axes "xDim", "yDim", and "zDim".
 If "option" contains "E" errors will be calculated.
                      "A" ranges of the taget axes will be ignored.
                      "O" original axis range of the taget axes will be
                          kept, but only bins inside the selected range
                          will be filled.

 Project all bins into a ndim-dimensional THnSparse histogram,
 keeping only axes in dim (specifying ndim dimensions)
 If "option" contains "E" errors will be calculated.
                      "A" ranges of the taget axes will be ignored.
                      "O" original axis range of the taget axes will be
                          kept, but only bins inside the selected range
                          will be filled.

 Project all bins into a ndim-dimensional THnSparse histogram,
 keeping only axes in dim (specifying ndim dimensions)
 If "option" contains "E" errors will be calculated.
                      "A" ranges of the taget axes will be ignored.
                      "O" original axis range of the taget axes will be
                          kept, but only bins inside the selected range
                          will be filled.

 Scale contents and errors of this histogram by c:
 this = this * c
 It does not modify the histogram's number of entries.

 Add contents of h scaled by c to this histogram:
 this = this + c * h
 Note that if h has Sumw2 set, Sumw2 is automatically called for this
 if not already set.

 Add contents of h scaled by c to this histogram:
 this = this + c * h
 Note that if h has Sumw2 set, Sumw2 is automatically called for this
 if not already set.
 In contrast to Add(), RebinnedAdd() does not require consist binning of
 this and h; instead, each bin's center is used to determine the target bin.

 Merge this with a list of THnSparses. All THnSparses provided
 in the list must have the same bin layout!

 Multiply this histogram by histogram h
 this = this * h
 Note that if h has Sumw2 set, Sumw2 is automatically called for this
 if not already set.

 Performs the operation: this = this*c*f1
 if errors are defined, errors are also recalculated.

 Only bins inside the function range are recomputed.
 IMPORTANT NOTE: If you intend to use the errors of this histogram later
 you should call Sumw2 before making this operation.
 This is particularly important if you fit the histogram after THnSparse::Multiply

 Divide this histogram by h
 this = this/(h)
 Note that if h has Sumw2 set, Sumw2 is automatically called for
 this if not already set.
 The resulting errors are calculated assuming uncorrelated content.

 Replace contents of this histogram by multiplication of h1 by h2
 this = (c1*h1)/(c2*h2)
 Note that if h1 or h2 have Sumw2 set, Sumw2 is automatically called for
 this if not already set.
 The resulting errors are calculated assuming uncorrelated content.
 However, if option ="B" is specified, Binomial errors are computed.
 In this case c1 and c2 do not make real sense and they are ignored.

 Consistency check on (some of) the parameters of two histograms (for operations).

 Set the axis # of bins and bin limits on dimension idim

 Set content of bin with coordinates "coord" to "v"

 Set error of bin with coordinates "coord" to "e", enable errors if needed

 Enable calculation of errors

 Combine the content of "group" neighboring bins into
 a new bin and return the resulting THnSparse.
 For group=2 and a 3 dimensional histogram, all "blocks"
 of 2*2*2 bins will be put into a bin.

 Change (i.e. set) the title.

 If title is in the form "stringt;string0;string1;string2 ..."
 the histogram title is set to stringt, the title of axis0 to string0,
 of axis1 to string1, of axis2 to string2, etc, just like it is done
 for TH1/TH2/TH3.
 To insert the character ";" in one of the titles, one should use "#;"
 or "#semicolon".

 Combine the content of "group" neighboring bins for each dimension
 into a new bin and return the resulting THnSparse.
 For group={2,1,1} and a 3 dimensional histogram, pairs of x-bins
 will be grouped.

 Clear the histogram

 Calculate the integral of the histogram

 Increment the bin content of "bin" by "w",
 return the bin index.

 Calculate errors (or not if "calc" == kFALSE)