Ruining Li
Ruining Li
> @RuiningLi I don't understand how this is related to #6395 ? > > > Fixes #6395 Sorry, I meant #6359 .
> thanks! I would like to understand the use case a little bit more Why do you need to use `EulerDiscreteScheduler` for training? Stable Video Diffusion by default uses `EulerDiscreteScheduler`,...
@yiyixuxu Not yet. Also, SVD itself uses `EulerDiscreteScheduler` for fine-tuning, and it also uses a continuous sigma sampler rather than a discrete one.
> EulerDiscreteScheduler is used by **internal** training done by HuggingFace, for example, the SDXL ControlNet models were trained with `--use_euler`, but this means you can't train v-prediction models with that.....
> that's not the point i was making, they use it internally on _epsilon_ prediction models. it's just that training on Euler isn't unordinary and it's even done inside HF...
I encountered the same issue in a customized environment. Any progress on this?
Thank you for your prompt reply Zhengyi! Do you also plan to release the script for generating orthographic CCMs as used to construct the training data? That would be super...
Hi, do you plan to release the training code of the reconstruction model as well? (Since now only the Multi-view diffusion's training codes are released :)