I wanted to be able to generate wallpapers using your code, so I adapted it to generate rectangular images by splitting the "size" parameter to "size_x" and "size_y".
Also, do you think that by controlling the random bias and weights initialization with perlin noise the network could create a smooth animation ? I'd be very interested in that, but this is my fisrt experience with pytorch, so I'll wrangle with it later.
Hi, Really neat work, thank you !
I wanted to be able to generate wallpapers using your code, so I adapted it to generate rectangular images by splitting the "size" parameter to "size_x" and "size_y".
Also, do you think that by controlling the random bias and weights initialization with perlin noise the network could create a smooth animation ? I'd be very interested in that, but this is my fisrt experience with pytorch, so I'll wrangle with it later.
Alexis