Time for a good 'ol fashion grump-off.
Wait, you're calling bumbeh grumpy? When we have real, authentic grumpsters (stop autocorrecting me stupid iPad!) like CML and Lucre on here?
Bumbeh's cooperative cube-building thread has brought a lot of new blood as well, all posting relevant and irrelevant content. He clearly didn't think things through when he decided to host that thing over here
Oh dear god, they found us Quanta. Well at least I know who has been posting the irrelevant content o_OI think two or three people stuck around
Anyone going to GP Lille?
Holy crap, this is so awesome!
http://googleresearch.blogspot.co.uk/2015/06/inceptionism-going-deeper-into-neural.html
I have no clue, but probably a pretty substantial amount? There's an awful lot of different animals in that cloud picture!So cool! How much data do you think these neural networks need to build up to start recognising the things they were trained on?
How do they structure this data? (probably a huge question - tell me to google it if you like)
I'll start making more creative output once I settle into my new apartment. I can't promise it's going to be for RipLab btw. Oh, I'm going to be on a TV show starting September. Did I ever mention that here?
Cool! Comedy Gig?
You can use that network that generates Magic cards. You can just feed it The Sun headlines instead of Magic cards to train it. The default training set is Shakespeare textsAnybody know a good place to start with neural networks? I want to feed one with headlines from The Sun and see what it generates. So is there some software I can run, or something to read that will help me program one?