Shannon

Ahh, thank you for the extra information; that helps immensely, in addition to the intuition for possibly less stable training processes given the larger images. So @bartolsthoorn for the images...

@bartolsthoorn I ran dcgan with the following arguments: `python main.py --cuda --dataset folder --dataroot /images --outf /output` I tried changing the `nc = 3` value to `nc = 1` since...

Managed to override the default image loader in torchvision so it properly pulls the images in with grayscale format and changed the `nc = 1`--seems to be running nicely now...

No improvement, though I guess it's a little easier to see that it's not *pure* noise in the fake images. Still looks like static, though. 

Ha! Isn't all of practical ML just tricks :) Haven't had a chance yet--behind on my teaching. Hoping to get back to this within the next week though, and will...

Thank you @davidplowman! I'll implement this fix and report back. As an aside, I'm curious about that `stride` parameter. I'm coming from a machine learning / AI background so I...

This does, thank you. My concept of stride is at a much higher level--image convolutions, which happen well after image acquisition--and the fact that this particular stride implementation is happening...

(best guess on why the process is being killed: out of memory)

@d-weber Same and same. Would love to bypass PyQt altogether.

This makes absolutely no sense to me considering this line seems to evaluate the `NOGUI` variable correctly https://github.com/raspberrypi/picamera2/blob/main/setup.py#L12 ``` > export NOGUI="1" > python -c "import os; print(os.getenv('NOGUI', '0') ==...