[docs] Adapters - #14842
[docs] Adapters#14842stevhliu wants to merge 5 commits into
Conversation
github-actions
Bot
added
size/L
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
sayakpaul
left a comment
sayakpaul
left a comment
There was a problem hiding this comment.
Thanks. I left some comments.
| [DreamBooth](https://huggingface.co/papers/2208.12242) is a training technique that updates the entire diffusion model by training on just a few images of a subject or style. It works by associating a special word in the prompt with the example images. | ||
|
|
||
| To load a trained checkpoint for inference, see [Load a DreamBooth adapter for inference](../using-diffusers/dreambooth). | ||
| To load a trained checkpoint for inference, see [Load a DreamBooth adapter for inference](../using-diffusers/legacy_adapters#dreambooth). |
There was a problem hiding this comment.
DreamBooth, by no means, is a legacy technique. I would keep it out of the legacy section.
| pipeline("A bowl of ramen shaped like a cute kawaii bear, by Feng Zikai", cross_attention_kwargs={"scale": 1.0}).images[0] | ||
| pipeline("A bowl of ramen shaped like a cute kawaii bear, by Feng Zikai").images[0] |
There was a problem hiding this comment.
Because the adapter_weights was set above?
| ```text | ||
| control image --> ControlNet --> residuals | ||
| | | ||
| text --> encoder --------------+ | | ||
| v v | ||
| denoiser blocks | ||
| (frozen base) | ||
| ``` |
There was a problem hiding this comment.
Same. I don't think it's adding much. For example, what is "v"?
There was a problem hiding this comment.
v is supposed to be an arrow showing it inserted into the denoiser block, but we can remove :)
|
thanks @stevhliu I agree with @sayakpaul dreambooth its still very popular and used, Textual Inversion, and T2I-Adapter on the other hand are rarely used anymore so I think it is fine to archive them. |
This branch has not been deployed
4 participants
Updates the Adapters and personalization docs so they read as a more coherent current guide: