Language is needed for a better semantic analysis of the flow of
paragraphs. If you want you
can force a specific language.
Return tables in the format that you need for you preprocessing pipelines.
If you want to maximize chunk length set this parameter to true.
Small
chunks will be merged
together.
If you want to include the title of the section in each chunk in
which
the
section will be
split, flag it.
If you want to include the heading of the table in each chunk in
which
the
table will be
split, flag it.
If you want the text of the images to be included in the chunks
set this parameter to true.
If you want to keep the footers' text in the chunks set this
parameter to true.
Remove headers by default, remove only the irrelevant ones, or
keep them all.