fix: add type=table filter to prioritize table retrieval#828
Open
octo-patch wants to merge 1 commit intoCinnamon:mainfrom
Open
fix: add type=table filter to prioritize table retrieval#828octo-patch wants to merge 1 commit intoCinnamon:mainfrom
octo-patch wants to merge 1 commit intoCinnamon:mainfrom
Conversation
…mon#752) The extra document retrieval in DocumentRetrievalPipeline was fetching all documents on matching pages, not just table documents. Adding the type='table' metadata filter ensures only table documents are returned when the 'Prioritize Table' option is enabled.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Fixes #752
Problem
When the "Prioritize Table" option is enabled,
DocumentRetrievalPipelineretrieves extra documents from pages that contain relevant content. However, the metadata filter only matched byfile_nameandpage_label, which caused it to fetch all document chunks on those pages — including non-table documents like regular text paragraphs.As a result, the table documents that should be prioritized were not being reliably retrieved: the secondary retrieval was pulling in all document types, not just tables.
Solution
Added
{"type": {"$eq": "table"}}to the metadata filter in the extra-table retrieval query. This ensures only documents withtype="table"are fetched in the secondary retrieval pass, aligning with the intent of the "Prioritize Table" feature.Testing
doc.metadata.get("type", "") == "table"used inreasoning/react.py,reasoning/rewoo.py, andqa/format_context.py)