Etie trained at 1024x1024, I think the quality increase is pretty nice, but doesn't always play well with "pose, etc" LoRAs that require certain dimensions. I'm using the Tiled Diffusion/VAE thing to gen at 768x1024 then upscale with hires 1.5x. Took the 16th epoch.
Testing out training at this along with 512x512. Will still release, 512 as usual, but might do this as well given I got time (took an hour to train instead of ~10 min). Also trying out new training method with cosine+restarts, instead of constant.
Etie trained at 1024x1024, I think the quality increase is pretty nice, but doesn't always play well with "pose, etc" LoRAs that require certain dimensions. I'm using the Tiled Diffusion/VAE thing to gen at 768x1024 then upscale with hires 1.5x. Took the 16th epoch.
Testing out training at this along with 512x512. Will still release, 512 as usual, but might do this as well given I got time (took an hour to train instead of ~10 min). Also trying out new training method with cosine+restarts, instead of constant.