examples\controlnet add only_mid_control=true and sd_locked=false

[geroldmeisinger]

Is your feature request related to a problem? Please describe. The original controlnet training tutorial mentions the options

• only_mid_control=True: This can be helpful when your computation power is limited and want to speed up the training, or when you want to facilitate the "global" context learning.
• sd_locked=False: This will unlock some layers in SD and you will train them as a whole.

Describe the solution you'd like Please make the option available as well.

Describe alternatives you've considered Going back to the original training script without any optimizations implemented.

[patrickvonplaten]

cc @williamberman @patil-suraj

[williamberman]

Yeah these are neat ideas! I think these would be good options for community PRs, happy to open up to the community :)

[kelisiya]

Have any updates about this project ?

[jiangyuhangcn]

How to Implement SD_ Locked=False in Diffusers？

[ValMystletainn]

I would like to try this issue. Seemingly, It can be done by set different gradient recording switch and keep it compatible with accelerate. @williamberman @patrickvonplaten @patil-suraj

[jon-chuang]

Seems pretty straightforward, just add an option to freeze/unfreeze certain weights.

[UmerHA]

Hi @ValMystletainn, are you still on it? :) Otherwise I'd tackle it.

[UmerHA]

Okay, I'd tackle this :) Eta 1 week.

[Night1099]

This is not an option still in current controlnet train script