model_data2

           Use separate 2-dimensional data in distinct group as fit model.

           CAFE

           fitmodel

           x         - Input x value array. Should be double precision.
           parameter - Parameter to compute model. Must contain 1
                       value.

           data:grp - group using to get data. !! THIS
                      PARAMETER MUST BE FIXED AND INTEGER !!!
                      (Otherwise the fit algorithm tries to
                      vary the group - would be quite interesting...)
                      It is prudent not to define a model for
                      the group which contains these data.

           y = data[x].
           To allow arbitrary x-values 2-dimensional interpolation will be
           used. There are implemented different forms of
           interpolation (see SETUP).

           The way of interpolation can be defined via setup
           parameter (use set,key=val). The command prefix is "data2d".
           intlin : (default) The interpolation is done by first
                    creating Delaunay triangles for the model data
                    points. For each input x-value (2-tuples) the
                    enclosing triangle is searched and a linear
                    interpolation between the data points
                    performed. For x-values outside the  hull of the
                    data set z-values of closest data point are used.
                    Don't expect high accuracy (not recomended for
                    fitting).
              tps : Use thin plate spline interpolation (function
                    min_cur_surf(../tsp)).
           curmin : Use minimal curvature interpolation (function
                    min_cur_surf).

           None

           $Id: cafe_model_data2.pro,v 1.6 2004/06/08 17:09:37 goehler Exp $

[Home Page] [Software, Documentation] [IDL Documentation] [Quick Reference] [Feedback]