controlNet training for dstilled models

Is it possible to train controlNet for distilled model (Block-removed Knowledge-distilled Stable Diffusion, https://github.com/Nota-NetsPresso/BK-SDM)?

When I used scripts in example folder for distilled model, it returns error, similar to "unknown mid_block_type ...." , probably because of unet structure changed. Are there any ways to train controlNet for distilled model?

Start by reproducing the error with the training scripts in the example folder and the BK-SDM distilled model. Read the reported “unknown mid_block_type” failure alongside the changed UNet structure, then determine whether ControlNet training can support this model; done means a documented compatibility result or a working training path.

Vom Indexierungsmodell aus dem Issue-Text verfasst.