so just testing an AI learning algorithm... right now there is a glitch displaying the proper costume which results in a lil error... I think I fixed it though. basically l.kjhg,ftdsyexrd67tyvn\what it does is it rewards the system when it does a green costume (or whichever costume you want it to do) which makes it more likely to use that costume in the future. Please note that it can take a little while for it to figure out what color it should switch to, and even after that it will sometimes try to get rewarded for a different color! Its not perfect, but it works pretty well for a scratch game! https://scratch.mit.edu/studios/27986962/
