Error in Matlab Genetic Algorithm optimization

1 visualización (últimos 30 días)
Iro
Iro el 3 de Abr. de 2014
Editada: Iro el 6 de Abr. de 2014
Hi,
I am using GA App in Matlab but by writing my own functions for Population creation, Crossover and Mutation. My population individuals are nxn matrices with 0 and 1 as elements. These matrices are evaluated by my Fitness function which returns a vector of scalar scores ( Vectorized = 'On' ).
In the GA App I choose the options ' Proportional ' for FitnessScaling and ' Roulette ' for Selection .
My problem is that after the scores calculation at the FitnessFcn the Optimization stops with the error " Conversion to double from cell is not possible " . And I do not understand where this error come from, either during Scaling or Selection . Is there anyway to identify the error source? I suspect that it is related to the fact that my population are matrices stored in cell arrays.
Does this mean that I also have to write my own Scaling and Selection functions? Or is there any other way to tackle it?
Thanks!

Respuesta aceptada

Sean de Wolski
Sean de Wolski el 3 de Abr. de 2014
dbstop if error
And look around
  3 comentarios
Sean de Wolski
Sean de Wolski el 3 de Abr. de 2014
Just run the above at the command line. It turns on the debugger to stop automatically whenever an error is thrown. You can then see everything as it is at the time of the error. The bad function line/variable will probably stand out at this point.
Iro
Iro el 6 de Abr. de 2014
Editada: Iro el 6 de Abr. de 2014
Thanks Sean!
I wonder if you also have a hint to this question:
Best!
Iro

Iniciar sesión para comentar.

Más respuestas (0)

Categorías

Más información sobre Genetic Algorithm en Help Center y File Exchange.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by