Skip to content

Conversation

@ljeagle
Copy link

@ljeagle ljeagle commented Jul 7, 2023

Upgrade the AwaDB from 0.3.5 to 0.3.6

  1. Support get, delete and update batch documents
  2. Add the implementation of the max_marginal_relevance_search
  3. Change the default vectors' distance computation type from L2 to InnerProduct
    Please review @rlancemartin, @eyurtsev @hwchase17. Thanks!

awa-ai and others added 4 commits July 8, 2023 00:53
1. support get, delete and update batch documents
2. add the implementation of max_marginal_relevance_search
3. change the vector's default distance from L2 to InnerProduct
Explicitly list requires_reference in function (langchain-ai#7357)
delete unnecessary annotation
@vercel
Copy link

vercel bot commented Jul 7, 2023

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name Status Preview Comments Updated (UTC)
langchain ✅ Ready (Inspect) Visit Preview 💬 Add feedback Jul 8, 2023 3:39am

@vercel vercel bot temporarily deployed to Preview July 7, 2023 17:22 Inactive
adjust the code format of awadb.py for the awadb vectorstore
@vercel vercel bot temporarily deployed to Preview July 8, 2023 00:32 Inactive
@vercel vercel bot temporarily deployed to Preview July 8, 2023 00:40 Inactive
Copy link
Contributor

@efriis efriis left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Great changes! Main requested change is to align docstring on similarity_search_with_relevance_scores with the referenced types.

Looks like you removed the reference to relevance scores even though they're still present.

**kwargs: Any,
) -> List[Document]:
"""Return docs most similar to query."""
"""The most k similar documents of the specified text query.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
"""The most k similar documents of the specified text query.
"""Returns the k most similar documents to the specified text query.

nit

Args:
query: Text query
k: The most k similar documents of the text query
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
k: The most k similar documents of the text query
k: The maximum number of documents to return

kwargs: Any possible extend parameters in the future.
Returns:
The most k similar documents of the specified text query
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
The most k similar documents of the specified text query
The k most similar documents to the specified text query

0 is dissimilar, 1 is most similar.
Returns:
List of Documents the most similar to the text query.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
List of Documents the most similar to the text query.
List of (Document, relevance_score) tuples the most similar to the text query.

@ljeagle
Copy link
Author

ljeagle commented Jul 8, 2023

@efriis I have added and changed the notes about relevance scores. please check, thanks!

@vercel vercel bot temporarily deployed to Preview July 8, 2023 03:26 Inactive
@efriis
Copy link
Contributor

efriis commented Jul 8, 2023

@ljeagle looks like the formatter/lint check failed - could you update that and I can merge it in?

@ljeagle
Copy link
Author

ljeagle commented Jul 8, 2023

@ljeagle looks like the formatter/lint check failed - could you update that and I can merge it in?

ok, thanks

@vercel vercel bot temporarily deployed to Preview July 8, 2023 03:39 Inactive
@efriis efriis merged commit fb6e63d into langchain-ai:master Jul 8, 2023
@efriis
Copy link
Contributor

efriis commented Jul 8, 2023

Thanks @ljeagle !

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants