Spaces:

sd-concepts-library

/

stable-diffusion-conceptualizer

like 574

Running on A10G

I never could get this to work with Automatic1111. Fact is none of the public ones work only ones you make.

have you read this? it gives some clear directions
https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Features#textual-inversion

edit:
inversion is also failing for me with webui and the sd-concepts repo.
I think you need to create inversions from this repo, to be able to use with webui. but i could be wrong.
https://github.com/rinongal/textual_inversion

have you read this? it gives some clear directions
https://github.com/AUTOMATIC1111/stable-diffusion-webui/wiki/Features#textual-inversion

Sure did, and I watched two tutorials, but neither creator used the community ones only ones they made. I can't make them as I tried the colab and just as I knew would happen did with 15 mins to go out of four hours it turned my gpu off and I lost all the work.

it is a known issue, that (this repos) textual inversion does not work (at the moment) with webui.
it is worked on the diffusers library, and the developers will be making some modifications or conversion tool for webui inversions.
https://github.com/huggingface/diffusers/issues/468
"
patrickvonplaten commented 12 days ago
Interesting!

cc @apolinario @patil-suraj

Which webui tool are you using @boyjunqiang ? Maybe we could see if we could integrate diffusers into the codebase or alternatively we could add some conversion scripts here.
"

you could point the developers of automatic111 webui to that specific error that you are facing.

I tried the webui with ---embeddings-dir to the special folder from my inversion here, where the .bin file lies. i think the https://github.com/rinongal/textual_inversion and this repo here, are using somewhat different models or versions of pytorch, or tensorflow, and therefore they dont work together.

edit:
i opened a issue on the webui repo
https://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/1023

Interesting. Some (very few actually) can I get to work after days of trying various stuff. What I found I had to do was start the prompt with (((blah-blah-emb style))) then the prompt for it to work.

@valhalla

@GeneralAwareness
it resolved here: https://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/1023
resolution: install git-lfs and clone the embeddings with that. if you only clone with git, only a lfs link is cloned and not the actual file.

!apt install git-lfs -y
!git-lfs clone https://huggingface.co/spaces/sd-concepts-library

log:
txt2img:
Loaded a total of 1 text inversion embeddings.

@GeneralAwareness
it resolved here: https://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/1023
resolution: install git-lfs and clone the embeddings with that. if you only clone with git, only a lfs link is cloned and not the actual file.

!apt install git-lfs -y
!git-lfs clone https://huggingface.co/spaces/sd-concepts-library

log:
txt2img:
Loaded a total of 1 text inversion embeddings.

I am on Windows so do not have apt.

Use
git lfs clone https://etc

WARNING: git lfs clone is deprecated and will not be updated
with new flags from git clone

Asking me for hugging face credentials so I put them in.

git clone has been updated in upstream Git to have comparable
speeds to git lfs clone.
Cloning into 'sd-concepts-library'...
remote: Repository not found
fatal: repository 'https://huggingface.co/spaces/sd-concepts-library/' not found
Error(s) during clone:
git clone failed: exit status 128

try this:
git lfs clone https://hf_username: hf_password@huggingface.co/sd-concepts-library/
or that:
git lfs clone https://hf_username:hf_token@huggingface.co/sd-concepts-library/
or this:
git lfs clone https://huggingface.co/sd-concepts-library/
replace hf_username, hf_password and hf_token with your huggingface credentials