Ranter
Join devRant
Do all the things like
++ or -- rants, post your own rants, comment on others' rants and build your customized dev avatar
Sign Up
Pipeless API
From the creators of devRant, Pipeless lets you power real-time personalized recommendations and activity feeds using a simple API
Learn More
Comments
-
Cyborg8378y@Setherizor I've made a string finding one recently. That's what got me thinking. Music generation sounds cool but I dont know where to start.
I was thinking maybe I could apply these concepts on, say, a simulation where a creature learns to do a specific task through evolution. But I'm at a loss at finding an environment for the said creature and deciding on the task it should learn. -
Cyborg8378yMaking my own game would take a great deal of time. I was thinking of flappy bird myself and I've even found a nice pygame implementation of it too.
But the genome can't really be a direct input to the game can it? Since in FB pipes are randomly generated, so the agent will need to observe the surrounding too right? Maybe genome could be a matrix that dictates what should be done when the bird is (x, y) away from the pipes. But I'm afraid the state space could then become very large. Have you done something similar before? -
@jawadsh123 what do you mean with string finding? I want to start to genetic algorithm from somewhere too. Well I have actually made something using processing library of java. Made some rocket like objects learn to move towards a target. A new project wouldn't be bad.
-
lumy3558yWikipedia will explain it way better than me https://en.m.wikipedia.org/wiki/..., we used it as a school project, school ended, still trying to find time to work on it. Fun game
-
Cyborg8378y@Shubidubapp following Daniel Shiffman?
It was a simple algo that would start from a population of random string of alphabets, apply concepts of GA like reproduction, mutation and like, and try to arrive to the string you gave as a target.
It was a starter project.
Here's the repo: https://github.com/jawadsh123/... -
lumy3558yNop, best scores was 50% and i think we had a wrong way of counting score.(it work but wasn't the most representative for our algorithm) :/
-
Cyborg8378y@lumy would love to go through some examples of that. Is that project of yours open source?
-
lumy3558yIt is on my Github account (but I not proud of yet through, and I'm not really good into gen-algo yet), and you can find probably alot with name like that. (https://github.com/lumy/EternityII)
-
lumy3558y@jawadsh123 if you want help me (maybe take the lead?) I'm looking for it :p (it's pretty documented!)
-
Cyborg8378y@lumy Thanks for the share. I'm gonna go through the code tomorrow morn, but my god is the documentation beautiful. Why doesn't everyone document there code like this. ( Then again, neither do I 😅)
-
Cyborg8378yHey @lumy , I tried running your algo today. But it seems I require an Office 360 account to download those images mentioned in the repo. Do you guys have those images hosted somewhere else too? Like drive or something?
Any cool ideas for a side project in Genetic Algorithms?
undefined