camel icon indicating copy to clipboard operation
camel copied to clipboard

[Feature Request] Chunker module enabling custom chunking strategy

Open AveryYay opened this issue 9 months ago • 0 comments

Required prerequisites

  • [x] I have searched the Issue Tracker and Discussions that this hasn't already been reported. (+1 or comment there if it has.)
  • [ ] Consider asking first in a Discussion.

Motivation

Currently, the chunking mechanism in VectorRetriever is predefined and lacks flexibility for users who need different chunking strategies based on their specific use cases. Providing a Chunker module that allows users to define and implement their own chunking strategies will enhance usability and flexibility.

Solution

Introduce a Chunker class:

  • BaseChunker
  • CodeChunker
  • FixedLengthChunker
  • TitleBasedChunker etc.

Alternatives

No response

Additional context

No response

AveryYay avatar Mar 01 '25 01:03 AveryYay