Voplica Git

Does anyone know how to convert `ultralytics` (YOLO) masks to the `input_mask` for SAM? I tried both above approaches (reusing [_compute_logits_from_mask](https://github.com/facebookresearch/segment-anything/issues/169#issuecomment-1779583246) and reusing [reference_to_sam_mask](https://github.com/facebookresearch/segment-anything/issues/169#issuecomment-1551120325)) but seems those are for different...

DoRa training is now supported in `kohya-ss`: https://github.com/kohya-ss/sd-scripts/issues/1127 As the discussion suggests, DoRa inference is also supported in WebUI

Sorry. Wrong repository. Opened it in the GUI issues: https://github.com/bmaltais/kohya_ss/issues/2647

Actually I realized that the issue comes from `sdxl_train.py` which is `sd-scripts`. So I believe the issue was opened correctly. Thus, reopening it.

> conditioning_data_dir cannot be used with is_reg. (Due to architectural issues Thank you for the information. However, if I remove `conditioning_data_dir` from the second subset (DreamBooth) then it fails with...

> Thank you for reporting the issue. I think you are using masked loss with the dataset with `conditioning_data_dir`. Unfortunately, `conditioning_data_dir` is not supported with `is_reg` option. I will update...

> I didn't reproduce it just now with my attempt at forcing the model I saw on a traceback. Thanks a lot for taking time to verify this issue! Would...