huggingface / huggingface/transformers
DeepMind Retro
- Dominant language
- Python
- Stars
- 166k
- Forks
- 34.6k
- Avg merge
- 3d 9h
- Merged PRs (30d)
- 281
Description
# 🌟 New model addition
Basically a retrieval augmented model like RAG, but without expensive retriever end2end training
## Open source status
* [ ] the model implementation is available: Not available
* [ ] the model weights are available: Not available
* [ ] who are the authors:(https://arxiv.org/pdf/2112.04426.pdf)
Contributor guide
Research direction
Start by reading the linked DeepMind Retro paper to understand the proposed model and its requirements. The issue provides no implementation files or tests; done would require determining the model scope and making its implementation and weights available, as indicated by the unchecked status items.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100