ParentChildChunkerConfig#
- class scikitplot.corpus.ParentChildChunkerConfig(parent_size=2000, child_size=400, child_overlap=0, separators=('\n\n', '\n', '. ', ' ', ''), emit_parents=True)[source]#
Configuration for
ParentChildChunker.- Parameters:
- parent_sizeint, default=2000
Largest parent, in characters.
- child_sizeint, default=400
Largest child, in characters; smaller than
parent_size.- child_overlapint, default=0
Context characters of a child, as
chunk_overlapofRecursiveChunkerConfig, within its parent.- separatorstuple of str, default=(”nn”, “n”, “. “, “ “, “”)
Separators for both levels, coarsest first, ending with
"".- emit_parentsbool, default=True
Emit each parent before its children.
Falseemits children only (each still records its parent’s span inparent_span).
- Parameters: