-
I am working on integrating this repository with ComfyUI, so this project can get support for ControlNet, LoRA, etc. all for free from ComfyUI's infrastructure.
The main issue is, I need to underst…
-
Residual noise from DDIM and issues with GIF export are affecting the quality of AnimateDiff results. Update docs based on info from this issue:
https://github.com/huggingface/diffusers/issues/7548
-
I tried to use your notebook example for 1.4 but ran into all kinds of issues. Can you please provide an example for SDXL?
Many thanks.
-
### Your question
I know that the original two samplers are different, and the results are different in the other (a1111), etc.
However, comfyui outputs the same results from the same seed.
I teste…
ytoaa updated
2 weeks ago
-
Thank you for your excellent work!
May I ask if it's possible to release the trained models? (Class-conditional DDPM & DDIM)
-
I saw the respace.py file creates a DDIM skip time step, but when I set the , the code in the test.py file does not work because there is no definition of the method .
Is the DDIM code unfinished?…
-
Hi,
Thank you for sharing the code for your great work. I am curious how you implemented the untargeted attack you mentioned in your paper. There's no explanation in the paper either.
-
Hi. First of all, thank you for this wonderful repository. I try to run a training and have the following problem:
I downloaded a small part of the imagenet dataset (2Gb) and unzipped it. There wer…
-
Excellent work!
I have a question to ask, is this DDIM inversion necessary in the process of diffusion of video data? I tried to use DPM inversion and got seriously wrong results. I tried to use DDI…
-
Currently running the code, and after 1000 steps it starts the 'sampling loop time step' but says it will do this 3125 times and is estimated to take about a week. Is this a part of training, and why …