Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Conversation

@kemingy
Copy link
Member

@kemingy kemingy commented Aug 1, 2025

No description provided.

@kemingy kemingy requested a review from Copilot August 1, 2025 07:38
Copy link
Contributor

Copilot AI left a comment

Choose a reason for hiding this comment

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

Pull Request Overview

This PR fixes graph search functionality for image/PDF files by adding a fake chunk reference and improving the search configuration. The changes address issues where graph entities and relations need chunk references that don't naturally exist for non-text input types.

  • Creates a fake chunk for image/PDF processing to provide necessary references for graph entities and relations
  • Adds a new similar_k parameter to control the number of similar entities/relationships retrieved during graph search
  • Makes topk parameters optional with default values and separates entity retrieval count from final result count
Comments suppressed due to low confidence (1)

vechord/pipeline.py:280

  • The parameter name 'Keyword' should be lowercase 'keyword' to follow Python naming conventions for parameter names.
        fake_chunk = Chunk(doc_id=doc.uid, text="", Keyword=None, vec=None)

@kemingy kemingy merged commit c1f3d95 into tensorchord:main Aug 1, 2025
7 checks passed
@kemingy kemingy deleted the graph_chunk branch August 1, 2025 07:39
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.

1 participant