Junsong Chen
Junsong Chen
> thanks! also a note, can we change the type for scheduler to `KarrasDiffusionSchedulers`? this here indicates that sigma pipeline only works with DPM https://github.com/lawrence-cj/diffusers/blob/d5f4c217b5f6ac4fd23b4ee627611e0fbe0707a7/src/diffusers/pipelines/pixart_alpha/pipeline_pixart_sigma.py#L191C20-L191C47 Cool. Changed.
> for the failing test, can you add the new doc page here? > > https://github.com/lawrence-cj/diffusers/blob/d5f4c217b5f6ac4fd23b4ee627611e0fbe0707a7/docs/source/en/_toctree.yml#L308 Fixed all the request here. Cc: @sayakpaul @yiyixuxu
Gentle ping here. @yiyixuxu
Sry to disturb you, but I found I can't push a commit to this branch, any suggestions? @sayakpaul
Agree. I can help adding the make style and test part. Is it possible to add me as a collaborator to your forked diffusers repo? @eightmusic
ok！thank you so muchSent from my iPhoneOn Mar 16, 2023, at 21:44, Yi Wei ***@***.***> wrote: It will take about 2.5 days, which is similar to BEVFormer's training time. —Reply...
Could you try this more specific version of toolkit? https://github.com/NVlabs/Sana/blob/e08d42879faa1117a3588a5e41e9c2041d43175d/environment_setup.sh#L12
What's your flow matching model's schedule? Could you provide more info about it?
Got it. Actually, I haven't tested about order=3 cases. We only use order=2 in our experiments. I'll look into it later, also if you have any insight on it, let...
Sry, but we don't meet this problem. Maybe the `huggingface_hub=0.26.1` version is not the same with ours.