lucidrains/x-clip
A PyTorch implementation of CLIP, a multi-modal model that learns to associate images with text using contrastive learning.
Not currently ranked — collecting fresh signals.
star history
The repository provides a complete implementation of CLIP from OpenAI with additional experimental improvements from recent research papers. It includes support for fine-grained contrastive learning (FILIP), decoupled contrastive learning (DCL), extra latent projections (CLOOB), visual self-supervised learning, and masked language modeling (MLM) on text. The implementation allows configuring text and image encoders with customizable depth, heads, and patch sizes.
Frequently asked
- What is lucidrains/x-clip?
- A PyTorch implementation of CLIP, a multi-modal model that learns to associate images with text using contrastive learning.
- Is x-clip open source?
- Yes — lucidrains/x-clip is open source, released under the MIT license.
- What language is x-clip written in?
- lucidrains/x-clip is primarily written in Python.
- How popular is x-clip?
- lucidrains/x-clip has 724 stars on GitHub.
- Where can I find x-clip?
- lucidrains/x-clip is on GitHub at https://github.com/lucidrains/x-clip.