agav1 Module Reference

Module containing the implementation of the algorithm. More...

Functions/Subroutines
subroutine	Algorithm (xobs, yobs, ysigma, q, delta, npar, ndata)
	THE algorithm.

---

Detailed Description

Module containing the implementation of the algorithm.

It has to be called from a main program that serves as an interface if used with a experimental/observational data.

Todo:

Debug OpenMP implementation

---

Function/Subroutine Documentation

subroutine agav1::Algorithm	(	real,dimension(:),allocatable	xobs,
		real,dimension(:),allocatable	yobs,
		real,dimension(:),allocatable	ysigma,
		real,dimension(:),allocatable	q,
		real,dimension(:),allocatable	delta,
		integer	npar,
		integer	ndata
	)

THE algorithm.

This is the implementation of the Asexual Genetic Algorithm (AGA), it requires that the experimental/observational data have been previously loaded (this is done in the main program, and was left out of the main algorithm to improve flexibility), and if MPI is enabled it also needs to be prevously loaded. The algorithm loads the rest of the runtime parameters and return the fittest individual and an estimate of the errors.

Parameters:

[in]	xobs(ndata)	: Vector of data abscissae
[in]	yobs(ndata)	: Vector of data ordinates
[in]	ysigma(ndata)	: Vector of uncertainties in the ordinate
[out]	q(npar)	: Vector of parameters to be adjusted, at the end returns the fittest individul
[out]	delta(npar)	: Window (hyperbox) half-size, at the end the variable returns the uncertainty of q
[in]	ndata	: Number of data points

Definition at line 46 of file aga-v1.f90.

References func::deltabox(), mpi_module::err, func::gaussian(), input::InitCond(), mpi_module::master, func::med(), exfunc::meritfunc(), numgaus::ngaus, mpi_module::np, func::pickbest(), func::randdata(), mpi_module::rank, func::sigm(), and func::spawn().

Referenced by test().