It is currently hard to cache results with pickled objects. For example, if you want to save the fitness value for subsequent usage of the same PyGMO problem class (this class pickles functions), you have to rerun the fitness function. This behaviour should be documented (and if possible resolved).
It is currently hard to cache results with pickled objects. For example, if you want to save the fitness value for subsequent usage of the same PyGMO problem class (this class pickles functions), you have to rerun the fitness function. This behaviour should be documented (and if possible resolved).