feat(rag): implement recall-fusion-rerank pipeline- Introduce LLM-bas… #247
Conversation
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 18d33b177f
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
spoon_ai/rag/config.py
Outdated
| ) rag_dir=rag_dir, | ||
| ) |
There was a problem hiding this comment.
The return statement in get_default_config has an extra rag_dir=rag_dir, and closing ) after the constructor call, which makes spoon_ai/rag/config.py invalid Python. This will raise a SyntaxError on import, so any code path that imports the RAG config (e.g., retriever initialization) will fail immediately at startup.
Useful? React with 👍 / 👎.
…ed Reranker (zero-dependency via Requests) to rescore candidates.- Update RagRetriever to use RRF (Reciprocal Rank Fusion) for merging Vector and BM25 results.- Enhance BM25 tokenization with regex for better code retrieval.- Add RAG_RERANK_PROVIDER configuration support (OpenAI/OpenRouter).
yueliao11
force-pushed
the
inntegration/rag
branch
from
18d33b1 to
2d2c4e0
Compare
1 participant
…ed Reranker (zero-dependency via Requests) to rescore candidates.- Update RagRetriever to use RRF (Reciprocal Rank Fusion) for merging Vector and BM25 results.- Enhance BM25 tokenization with regex for better code retrieval.- Add RAG_RERANK_PROVIDER configuration support (OpenAI/OpenRouter).