Final best solution is not the best found

[camillamelo]

In the end of each iteration and sometimes even in the end of the irace, the configuration that appears on the best-so-far ignores new configurations when they didn't run for all instances, even if it has budget to run missing instances.
I ran again experiment 3 as requested on the mail group, but it wasn't exactly the same because Iwasn't using the seed inside the python script running on the target-runner. But I think it had the same problems, it still had 24 experiments on budget but it stopped without running some of the elite configurations for all instances.
output3b.zip