Greetings,
I've seen some posts about this question, but I couldn't conclude.
During selection, I'd like to select the k best individuals and then perform tournament selection "len(pop) - k" times, so the size of the intermediate population remains the same.
I tried using "selBest" and "selTournament" but couldn't set the parameters right. I got the error "multiple values for argument". I also tried a solution suggested here about registering a separate function that returns "selBest" + "selTournament", but got the same errors, so I must be missing something.
here's the relevant code:
I know that elitism increases the selection pressure, but that is intended. I usually use my own GA implementations and have successfully applied elitism. This time I'm doing GP and decided to use DEAP to help my case. Great tool, by the way.
Many thanks,