Every year, 250K+ peeps across the globe take a college-level E&M physics class and hate it. We’re going to make it more fun with less pondering. Well… that was the plan, but then there was a hurricane on the east coast and my cat got sick and this and that. So, I’ll be doing the physics later on my own.
No hurricanes next year, m'kay?
libraries: express, socket, processingjs
oops: this was for the part that just didn’t work google font library,MathJax
Voting is now closed.
(1)
therazorblade
charlesbeelerMaybe its just because I liked the shapes, but I like the concept of being able to show preferences based on time spent with an object. Lots could be done from this and obviously not “complete”, but a clever idea and fun to play with (I like seeing numbers change I guess) | |||
elizabethchaInteresting, but it took a long time to figure out what was going on. I didn’t immediately recognize the images as faces and then it seemed to take at least 10 tries to get visually different results. Would prefer more instant gratification. Maybe allowing users to choose multiple faces among the options and then generate the result? From a user experience, it would have been nice if the faces were larger and didn’t requiring hovering to feed them into the system. | |||
ryanspoonFun application of genetic algorithms. It takes some time to converge, though, and you need to “upvote” a lot of faces per iteration for it to find a pattern. Perhaps it should be a bit faster, and with a different interface? You should also keep in mind that humans can be intransitive (i.e. state that A > B and that B > C, but then state that C > A) One interesting thing would be a finer control over the algorithm, such as choosing the mutation, crossover and selection methods, changing the population and elite size (if there’s an elite), etc. | |||
That’s pretty cool. Gets boring fast though. I wish there was a way to make each iteration faster. I don’t want to keep my mouse over something for an arbitrarily long amount of time. | |||
christkvFun little project :) | |||
I liked the idea behind the create-a-face and genetic algo, but I didn’t feel I was seeing results fast enough. Maybe some sort of idea of how much time helps focus the generator? Also, maybe next time try clicking on the shape to focus the generator and just go click-by-click | |||
jasonh
I don’t get it. I understand that based on the patterns I like, you create a new set that derives from it, but beyond that, not sure what’s the point.