Class LayoutBasedChunkingConfig (0.11.11)

LayoutBasedChunkingConfig(mapping=None, *, ignore_unknown_fields=False, **kwargs)

Configuration for the layout based chunking.

Attributes

NameDescription
chunk_size int
The token size limit for each chunk. Supported values: 100-500 (inclusive). Default value: 500.
include_ancestor_headings bool
Whether to include appending different levels of headings to chunks from the middle of the document to prevent context loss. Default value: False.