Replies: 9 comments 10 replies
-
this code is 100% not mine and goes waaay back. but this code sounded fishy: if valid_step == math.floor(valid_step):
return int(valid_step) + 1 i've changed it to: if valid_step == math.floor(valid_step):
return min(int(valid_step) + 1, num_steps) |
Beta Was this translation helpful? Give feedback.
-
And to be honest, we can thank the Tiled VAE dev, as these were more or less obscured issues (this and the empty tensor PR you merged from me) - they didn't happen when not upscaling, and I like to use a high sample rate because it usually results in crazy good images. |
Beta Was this translation helpful? Give feedback.
-
@vladmandic now I get this...fun stuff!
|
Beta Was this translation helpful? Give feedback.
-
Would something like this fix it?
|
Beta Was this translation helpful? Give feedback.
-
digging a bit deeper in this code - it all comes down from fact that and higher number of steps, bigger the likelihood of timesteps becoming silly values - for example, with steps above 100 there is a non-trivial chance that some timestep will be interpolated as zero, so it can lead to division by zero (and that's just an example). so what can we do?
i'm leaning towards option (3). thoughts/opinions? |
Beta Was this translation helpful? Give feedback.
-
steps = gr.Slider(minimum=1, maximum=99, step=1, elem_id=f"{tabname}_steps", label="Sampling steps", value=20) done. although its applied only on fresh configs, so if you already have |
Beta Was this translation helpful? Give feedback.
-
is this the origin of the issue? huggingface/diffusers#2585 |
Beta Was this translation helpful? Give feedback.
-
Thanks, the funny thing is that I had already read the abstract of samplers being flawed when the paper was released and was aware of that "medium brightness" issue, I just had lost track on that topic until I saw their paper again. Have you seen their comparative report before/after their proposed fixes? |
Beta Was this translation helpful? Give feedback.
-
@TheOnlyHolyMoly do you mind creating a feature request for this so it doesn't fall off the radar? |
Beta Was this translation helpful? Give feedback.
-
When upscaling, including using tiled vae, I'm hitting an issue here in ldm/modules/diffusionmodules/util.py - I think that for whatever reason, the ddim_timesteps is one over the max number of indicies in alphacums:
IndexError: index 1000 is out of bounds for dimension 0 with size 1000
I made the following adjustment and it tries to proceed:
This is when sampling 150 (max) steps...DDIM says it's sampling 167 in the console, though...
Either way, doing fewer than 150 steps, or with the above code adjustment, the below error results so it may be a bug in this repo after all?
Beta Was this translation helpful? Give feedback.
All reactions