Show HN: GAN Playground – Play with Generative Adversarial Neural Nets In-Browser

Re:Show HN: GAN Playground – Play with Generative Adversaria... by sp332 on Sunday November 05, 2017 @04:55PM (#15631886)

There's just no point getting fancy for MNIST. Using nothing but fully-connected layers and rmsprop, I got good results in less than one minute. https://imgur.com/a/7xzmQ

Hidden units for discriminator layers - 100, 40, 2.

Generator layers - 40, 100, 768, then "reshape" into 28x28x1.

[ Parent | on HN ]

Re:Show HN: GAN Playground – Play with Generative Adversaria... by reiinakano on Sunday November 05, 2017 @09:48PM (#15633178)

Unlike regular classification, it's hard to quantify the performance of a GAN by just looking at the loss curves. You really have to make a subjective evaluation by looking at the generated images.

By playing around with it, the results I got from fully connected were nowhere near as good as the results I got from convolutional.

[ Parent | on HN ]

Re:Show HN: GAN Playground – Play with Generative Adversaria... by EvgeniyZh on Sunday November 05, 2017 @05:43PM (#15632181)

what you're showing is basically perfect discriminator with awful generator, isn't it?

[ Parent | on HN ]

Re:Show HN: GAN Playground – Play with Generative Adversaria... by sp332 on Sunday November 05, 2017 @05:46PM (#15632196)

Yeah ok it's pretty misleading. And if you let it run longer the generator starts getting a little better. It took me a long time to figure out the UI.

[ Parent | on HN ]

Re:Show HN: GAN Playground – Play with Generative Adversaria... by EvgeniyZh on Sunday November 05, 2017 @06:00PM (#15632284)

well you have the example on the left of graphs.

[ Parent | on HN ]