Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Necessity of torchvision dependency #1644

Closed
niazarak opened this issue Jul 20, 2022 · 5 comments
Closed

Necessity of torchvision dependency #1644

niazarak opened this issue Jul 20, 2022 · 5 comments

Comments

@niazarak
Copy link

Hi, thanks for your work, we've been using sbert for quite some time.

We noticed that sentence-transformers depend on torchvision, and while it was necessary for CLIP implementation in previous versions, now sentence-transformers use CLIP from HF (commit).

Is there any reason to keep this dependency now?

@niazarak niazarak changed the title Necessity of torchvision Necessity of torchvision dependency Jul 20, 2022
@nreimers
Copy link
Member

I think the HF clip model also requires torchvision

@iamhatesz
Copy link

How about moving torchvision to extras dependencies? I have a completely non-CV setup and this only complicates things for me. See e.g.: pytorch/vision#5393

@ecly
Copy link

ecly commented Oct 27, 2022

Agreed with @iamhatesz that having it as optional would be ideal, as torchvision sometimes can be a headache.

@glemaitre
Copy link

@tomaarsen I think this issue can be closed after the merge of #1881

@tomaarsen
Copy link
Collaborator

I agree, thanks for pointing this out!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

6 participants