Sign up to take part
Registered users can ask their own questions, contribute to discussions, and be part of the Community!
Registered users can ask their own questions, contribute to discussions, and be part of the Community!
I am creating a RAG approach in DataIKU and I do not want to use Langchain. Is there any way I can run the similarity search on a knowledge bank without the use of Langchain as vectorstore or is it the only way possible?
Hi @ranjodh55 ,
The methods available via the LLM Mesh in DSS are detailed here;
https://developer.dataiku.com/latest/tutorials/machine-learning/genai/nlp/llm-mesh-rag/index.html#ru...
https://developer.dataiku.com/latest/concepts-and-examples/llm-mesh.html
Thanks
Hi @AlexT ,
Is there a way we can filter a knowledge bank based on metadata we choose? As an example ,if a knowledge bank contains data of 1000 documents and I want to retrieve on a specific document when an LLM is queried. Is it possible?
Thanks,
Thanks Alex