Thanks :)

I actually have had inactivity detection implemented at some point, but I've resigned of this in favor for user-defined stop condition. But you are right, I will bring that back.

It's actually expected that in some cases it might take many generations to learn, as the population is rather small. In other attempts they can do this in like 5 generations, so please, tray to restart the whole process.

Keeping one track untill at least one car is able to finish it is not as good idea as it seems due to ovefritting. The will simply memorize all the moves depending on some insigificant factors and they won't be albe to finish any other track. I have tested that :D

However, maybe you will consider the request described here:

https://news.ycombinator.com/item?id=12397375

For research purposes :)

Yeah, sure, why not :) I will have some spare time at the weekend, so then I will try to implement evenything that has been proposed in the feedback that i've got.

Changing mutation rate to 0.1 and crossover rate to 0.3 seems to improve the learnig..

0.5 mutation rate is way too much..

Yeah, that might be true, and that's what this editor is for :D I didn't want to give users the best solution at the beggining, so they might search for it themselves. It's actally only the third combination of parameters that I've tested.

On the other hand 0.5 is not so large, when you look at the mutation procedure. If a speciman is going to be mutated, only 20% of it's weights are changed and only within a range [-0.1,...,0.1]. But fill free to experiment :)

Took me to gen 11 and I only lost 1 car after that they all completed.