lala Posted July 7, 2015 Share Posted July 7, 2015 anyone get it to work yet? im trying ryan kennedys and its throwing an error.. win machine. others have got the same error and fixed it and i cant. damnn Link to comment Share on other sites More sharing options...
kaini Posted July 7, 2015 Share Posted July 7, 2015 try this guide: https://www.reddit.com/r/deepdream/comments/3c2s0v/newbie_guide_for_windows/ ryan's guide is for mac. Link to comment Share on other sites More sharing options...
lala Posted July 7, 2015 Share Posted July 7, 2015 yeah giving it a go now, cheers. there are people that overcame the problem but yeah, nice one. i always fail at docking Link to comment Share on other sites More sharing options...
lala Posted July 7, 2015 Share Posted July 7, 2015 try this guide: https://www.reddit.com/r/deepdream/comments/3c2s0v/newbie_guide_for_windows/ ryan's guide is for mac. stuck at step 13. "$env:Path += ";C:\Program Files (x86)\OpenSSH\bin" Doesnt even do anything when i type it in.. Link to comment Share on other sites More sharing options...
lala Posted July 7, 2015 Share Posted July 7, 2015 nada.. and ive done without the inverted comma at the beginning, thats a copy/paste error. beginning to get tiring now. ive been at it all day lol Link to comment Share on other sites More sharing options...
lala Posted July 7, 2015 Share Posted July 7, 2015 getting somewhere, you need to close powershell then start it again and it worked keep your fingers crossed for me guys i dont know wtf im doing Link to comment Share on other sites More sharing options...
lala Posted July 7, 2015 Share Posted July 7, 2015 stuck here. last step, im inside the virtual machine and trying to navigate to the right folder and i cant find it. so close! Link to comment Share on other sites More sharing options...
lala Posted July 7, 2015 Share Posted July 7, 2015 new gallery website thing, with noone in queue. get yer watmm on http://deepdream.akkez.ru/ Link to comment Share on other sites More sharing options...
Guest bitroast Posted July 7, 2015 Share Posted July 7, 2015 new gallery website thing, with noone in queue. get yer watmm on http://deepdream.akkez.ru/ lol Link to comment Share on other sites More sharing options...
Hugh Mughnus Posted July 7, 2015 Share Posted July 7, 2015 meow? mow Link to comment Share on other sites More sharing options...
Verdant Hickies Posted July 7, 2015 Share Posted July 7, 2015 Okay, next thing. /thread Link to comment Share on other sites More sharing options...
lala Posted July 7, 2015 Share Posted July 7, 2015 i just got the damn thing working on my pc. :angryface: Link to comment Share on other sites More sharing options...
Hugh Mughnus Posted July 7, 2015 Share Posted July 7, 2015 Okay, next thing. /thread yeah no hawt Link to comment Share on other sites More sharing options...
caze Posted July 7, 2015 Share Posted July 7, 2015 that's probably less disturbing than your previous avatar. Link to comment Share on other sites More sharing options...
Silent Member Posted July 7, 2015 Share Posted July 7, 2015 My iddems are still in queue. So slow. Sean pls Meh Link to comment Share on other sites More sharing options...
lala Posted July 7, 2015 Share Posted July 7, 2015 ive upped the RAM allocation and shiz to 15gigs and im trying to go up to a 3840x2560 image. see what happens Link to comment Share on other sites More sharing options...
Hugh Mughnus Posted July 7, 2015 Share Posted July 7, 2015 that's probably less disturbing than your previous avatar. lol it kind of is, you're right! Link to comment Share on other sites More sharing options...
YangYing Posted July 7, 2015 Share Posted July 7, 2015 man it so cool to see this idea developing, loving this stuff Link to comment Share on other sites More sharing options...
lala Posted July 7, 2015 Share Posted July 7, 2015 im going to stick all my artwork in a net and make an ai learn it if i can, then do some weird art with it. fingers crossed. ive got this bit nailed, had to re-do it 5 times so i know what im doing now :) the reddit thread works my computer is well fast now doing a 640x480 image. woop Link to comment Share on other sites More sharing options...
triachus Posted July 7, 2015 Share Posted July 7, 2015 did two too, using that russian site Link to comment Share on other sites More sharing options...
lala Posted July 7, 2015 Share Posted July 7, 2015 im trying to get the images so they dont look like jpeg quality=3 Link to comment Share on other sites More sharing options...
coax Posted July 7, 2015 Share Posted July 7, 2015 The highest layers of the DNN will ultimately be the training set images (approx), so you'll end up with a lot of dogs and buildings and eyes. The broader the training set is the more it can recognize. To get really trippy images you either need an abstract training set or use lower layers to detect edges and lines etc rather than complete images. Would be cool to mix in images of different layers in DNN of focus. So like one part uses layer 2 amplification, and another part uses layer 4 or whatever, in this way you get different levels of abstraction in the same image. Creativity in humans uses all layers at the same time, but ALSO takes into account what we know about the real world like gravity, objects function etc, then we could have real generation on par with humans. Right now it's just a passive pattern recognizer, not an active pattern recognizer with precision output (like our hands for drawing etc) with knowledge like humans have. Link to comment Share on other sites More sharing options...
lala Posted July 7, 2015 Share Posted July 7, 2015 yeah im speaking with someone whos trying to install the better model set, with less dogs. im going for quality. currently on inception 2 part 6, 22gigs full and no crash yet, 3840x2560. fingers crossed Link to comment Share on other sites More sharing options...
coax Posted July 7, 2015 Share Posted July 7, 2015 The problem is you are stuck within the training set precision or the abstract nonsense. IT will either detect objects we already know, or it'll just mangle abstractness. The really difficult part is creating either 1) meaningful objects of abstract lines or 2) creating meaningful compositions of trainingset data like dogs and buildings. You have a high chance of creating random combinations rather than precise meaningful ones. The problem is to create knowledge of the world etc Link to comment Share on other sites More sharing options...
Adieu Posted July 7, 2015 Share Posted July 7, 2015 Love the geriatric orgy in the gallery here http://deepdream.akkez.ru/ The watmmers are taking over Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.