Open nbardy opened 7 months ago
cc @patil-suraj wdty?
This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.
Please note that issues that do not follow the contributing guidelines are likely to be ignored.
Hey @nbardy ! This would be awesome to have, to start with we could add a new script under research_projects
.
This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.
Please note that issues that do not follow the contributing guidelines are likely to be ignored.
@nbardy still happy to accept a contribution of your script to research_projects
. If the community finds it to be more and more useful we can think of graduating it to be an officially supported script.
We will of course include your script in the documentation and communicate about it.
This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.
Please note that issues that do not follow the contributing guidelines are likely to be ignored.
gentle pin @nbardy let us know if you're still interested in adding this to research folder
Thanks for the ping. I have a version of this on a fork I can clean up for upstream
This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.
Please note that issues that do not follow the contributing guidelines are likely to be ignored.
We are still very interested in this @nbardy!
This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.
Please note that issues that do not follow the contributing guidelines are likely to be ignored.
None of the diffusers training scripts support multiple crops or aspect ratios in the data loader. This is mentioned in the tech report for SDXL as a part of their training process.
In my tests diffusion model struggle to do high resolution details. I did this for base SD 1.5 and was planning to port the code to the SDXL SCM scripts.
@sayakpaul I know you wanted to keep the base training scripts simple. Wanted to open a discussion here on whether to upstream. Is this something I should upstream? or better as something like a community script?
The SCM models are fantastic, but full short of the base SDXL model on high resolution details. Hoping this can fix some of that up.
References
SCM vs LDMs inference report ( Need to add multiple aspect ratios to inference report) https://wandb.ai/nbardy-facet/LCM-Inference-Test/reports/SDXL--Vmlldzo2MDAzNDU1
SCM LORA 4 steps
Base SDXL 30 steps
SDXL technical report aspect ratios.