generative-ai icon indicating copy to clipboard operation
generative-ai copied to clipboard

Fix display error in refine method and enhance similarity search index with document metadata for document qa with langchain

Open ravenouse opened this issue 2 years ago • 4 comments

This PR made two changes to the language/examples/document-qa/question_answering_large_documents_langchain.ipynb notebook:

  1. It fixed the display error of the refine method outputs. The original code displayed outputs from the map reduce method instead of the refine method.

  2. It changed the similarity search index part in the Q&A with similarity search section. This modification allows the map_reduce_embeddings_outputs to include source information such as page numbers, similar to the previous map_reduce_outputs and refine_outputs.

Following the Contributing Guide, I have submitted the CLA and ran the Code Quality Checks.

I kindly request you to review the changes at your convenience. Please let me know if there is anything else I can do to assist with the review process or provide further clarification.

Thank you so much.

ravenouse avatar Jun 12 '23 00:06 ravenouse

Check out this pull request on  ReviewNB

See visual diffs & provide feedback on Jupyter Notebooks.


Powered by ReviewNB

@inardini can you check this PR?

lavinigam-gcp avatar Jun 14 '23 03:06 lavinigam-gcp

Hi @ravenouse, thanks for the PR. It's quite difficult to review this as I see the entire file has changed. Is there any chance you could re-submit the files with just the changes you made to make this easier?

polong-lin avatar Jun 23 '23 11:06 polong-lin

Hi @polong-lin, thank you so much for the feedback! I have re-submitted the changes. Please let me know what you think about it!

ravenouse avatar Jun 26 '23 14:06 ravenouse

@ravenouse Sorry for the delay in review/updates. There was a repository restructuring that made your PR difficult to merge, but it should be fine now.

holtskinner avatar Jan 05 '24 20:01 holtskinner

Hi @holtskinner, thank you so much for your time on this! Happy New Year!

ravenouse avatar Jan 05 '24 20:01 ravenouse