open-clip-torch 2.26.1

Open reproduction of consastive language-image pretraining (CLIP) and related.

Stars: 9985, Watchers: 9985, Forks: 962, Open Issues: 116

The mlfoundations/open_clip repo was created 3 years ago and the last code push was 3 days ago.
The project is extremely popular with a mindblowing 9985 github stars!

How to Install open_clip_torch

You can install open_clip_torch using pip

pip install open_clip_torch

or add it to a project with poetry

poetry add open_clip_torch

Package Details

Author

Gabriel Ilharco, Mitchell Wortsman, Romain Beaumont

License

MIT

Homepage

https://github.com/mlfoundations/open_clip

PyPi:

https://pypi.org/project/open-clip-torch/

GitHub Repo:

https://github.com/mlfoundations/open_clip

Classifiers

• Scientific/Engineering
• Scientific/Engineering/Artificial Intelligence
• Software Development
• Software Development/Libraries
• Software Development/Libraries/Python Modules

Related Packages

Errors

A list of common open_clip_torch errors.

Code Examples

Here are some open_clip_torch code examples and snippets.

GitHub Issues

The open_clip_torch package has 116 open issues on GitHub

• Support for training on spot instances
• Pretrained weights not found
• Convnext not found
• Intermediate Checkpoints
• This machine is not connected to the Internet, how to adapt the code to prevent the pre-model from being downloaded online.
• Save intermediate checkpoints when sampling without replacement (take 2)
• Table in README
• Implement Locking of Text Tower for CLIP Models
• Entry Not Found Error for JSON file when using OpenCLIP ViT-H/14
• How to use the open_clip model I have trained on my own dataset?
• Add text-text (audio) CLIP
• [WIP] Testing the lion optimizer
• Add video support
• [WIP] Support FSDP
• Add TextTextCLIP

See more issues on GitHub