The trigger words are "pillory", "stocks" and "bondage" I suggest using them at 1.2 to 1.3
Description
FAQ
Comments (10)
It would be cool if there was a view from behind and side :)
i might do a model for that but I can't add it into the same model
@hunter98 I second side view. That would be great.
Half the time it doesn't work correctly.
not sure how to fix that do you have any ideas?
@hunter98 more data?
@fansay i used almost 200 pics
@hunter98 @hunter98 did you tag each picture properly? if so then I suggest maybe (if you haven't already) to retrain with a few more epochs say that this lora is 8 epochs retrain to 12-15 epochs and use "save lora every 1 epoch" then simply XYZ the same prompt with the Loras 9-15 and see which one is best.
this is technically an easy concept, with 200 pictures you should have no problems having it to work properly. make sure you did not have an oversight when entering the various settings. and it should work.
@fansay I think I see where your gens went wrong you are setting the lora to stuff like 1.2 to 1.5 you need to keep the lora at 1 and set the keywords "pillory, stocks, bondage" to 1.2 to 1.3 that should work better
It's there a way to makes these like handcuffs