Continue an execution from where the latter has ended

[Hyzbade]

When an evolutionary algorithm has ended with its computation, we should be able to launch a new execution from where the algorithm has stopped.

[samthiriot]

Obviously usefull; yet it has to be done in some clean way.

Each algo would require to be able to reinitialize its internal state for being able to restart from it.
So only the output files would not be enough, we also do need the internal state.

To be discussed ?

[Hyzbade]

Ok, serialization maybe?