autogluon / autogluon-rag

Retrieval-Augmented Generation in 3 Lines of Code!
Apache License 2.0
23 stars 4 forks source link

Vector DB Module #10

Closed shreyash2106 closed 3 months ago

cheungdaven commented 3 months ago

It is important to make documents traversable when we chunk, embed, and build the vector database, to ensure users can easily trace the source of the information. This traceability enhances transparency and trust, as users can see where the generated content originated from.

We may need create some metadata such as "doc_id" and "chunk_id" for this purpose.

shreyash2106 commented 3 months ago

It is important to make documents traversable when we chunk, embed, and build the vector database, to ensure users can easily trace the source of the information. This traceability enhances transparency and trust, as users can see where the generated content originated from.

We may need create some metadata such as "doc_id" and "chunk_id" for this purpose.

Okay sounds good, I will add this into the module

cheungdaven commented 3 months ago

It is important to make documents traversable when we chunk, embed, and build the vector database, to ensure users can easily trace the source of the information. This traceability enhances transparency and trust, as users can see where the generated content originated from. We may need create some metadata such as "doc_id" and "chunk_id" for this purpose.

Okay sounds good, I will add this into the module

You can do this in a separate PR.