search

NicMcPhee / XO-bias-study

Results and write-up of our genetic programming crossover bias study
MIT License
0 stars 0 forks source link

Look at impact of XO bias on tree size #19

Open NicMcPhee opened 9 years ago

NicMcPhee commented 9 years ago

It would be interesting to see what impact (if any) XO bias has on average tree size over time (an idea from Lee Spector). I don't think ECJ spits out sizes by default, but it must be capable of doing so given how much the GMU people have looked at bloat, so we'd have to figure out how to get that data.

NicMcPhee commented 9 years ago

I think I figured out how to get ECJ to give us tree size data, but when I run things on the cluster I seem to lose/not get those files. It would be nice to look at this for at least one or two problems, but I don't know that we want to go back and re-run all the previous problems to get this data for everything.

NicMcPhee commented 9 years ago

I take that back – I do seem to have things working so that we get size data. The Pagie-1 runs have it, and I suspect that any future runs will have it as well. I'm not sure we want to go back and re-do past runs to get that data, but we should certainly have a look at the data we do have just in case there's something interesting going on.