Train multiple loras sequentially?

[PaperOrb]

I'm wondering if we can create a few kohya settings jsons and then have kohya load them sequentially so you can have it train multiple loras one after the other overnight?

edit: nm, found out that lora easy training scripts has this functionality.

[Kadah]

I'd like to see this functionality here. More so if it could handle a pool of more than one GPU.

A workaround I think would work is use the Print training command function and build a basic script to execute them outside of the gui.