The my_fitter class is an example of how to use the gml_generic_fitter class
![]() | BestLogLikelihood A variable to hold an important quantity |
![]() | Data A pointer which holds an array of my_data structures |
![]() | bestLL (void) Calculate the absolute minimum best fit |
![]() | data (register IML_ULINT I) An interface function that must be defined |
![]() | evaluate ( register gml_double_vec &PVector) The only interface function that must absolutely be defined |
![]() | model_error (register IML_ULINT I) An interface function that must be defined |
![]() | my_fitter (IML_USINT DCnt) The only constructor |
![]() | weight (register IML_ULINT I) An interface function that must be defined |
![]() | ~my_fitter () The only destructor |
A: What to do in your derived (child) class.
B: Fitting, convergence and parameter configuration.
C: Calculating standard fit statistics.
D: Unweighted fit statistics.
E: Weighted fit statistics.
F: Calculating standard error bounds on parameters.
G: Calculating robust error bounds via error-surface search.
H: Informational variables.
The my_fitter class is an example of how to use the gml_generic_fitter class. We will us it to fit a Weibull function to psychometric data. The class is derived from the gml_generic_fitter class, as described in the documentation. All I need to do is tell gml_generic_fitter that I need two parameters, make sure to set the DataCount variable that I inherit from gml_generic_fitter, then initialize my own stuff.
~my_fitter()
virtual IML_DOUBLE model_error(register IML_ULINT I)
virtual IML_DOUBLE data(register IML_ULINT I)
virtual IML_DOUBLE weight(register IML_ULINT I)
virtual IML_DOUBLE evaluate( register gml_double_vec &PVector)
generated by doc++