Feature Request Automatic Embedding Download from Hugging Face #9998
FlowDownTheRiver
started this conversation in
Show and tell
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
Description : This is a script that is reading embeddings list from hugging face and downloads the embeddings,updates if necessary.
The code is originally written here: https://gist.github.com/kermit4977/6087a23f81d3184c99e952f878d2802a Credits to kermit4977
and thanks to NMKD,he edited the code for a categorized download as objects,styles.
I have made a bat file to install bs4,download the embeddings and clean the traces of bs4. bs4 aka beautiful soup.
get_embeddings.zip
So with the -embeddings-dir argument I tried to pass multiple folders, "embeddings/object" and "embeddings/style" I failed.
I tried to edit shared.py like this : parser.add_argument("--embeddings-dir", type=str, nargs='+', default=os.path.join(script_path, 'embeddings'), help="embeddings directory for textual inversion (default: embeddings)") and I have failed.
Btw the reason embbedings are categorized so we know what is an object what is a style,but how can we load them all in the launch time from multi directories?
Anyway I leave it up to you to add this feature or not. Files are in the zip to be tested.
....and if possible can we have a drop down menu of embeddings, like saving style and adding style.Then something like which can read the embeddings from the folder and records them into a csv or something and load them on launch,when we press the embeddings dropdown we can see all available styles and add them to our prompt in a very convenient way. This can be very nice.
Beta Was this translation helpful? Give feedback.
All reactions