operators.pdf.table.TableFeaturizer

class operators.pdf.table.TableFeaturizer(field='rich_doc_pdf_url', model='microsoft/table-transformer-structure-recognition', pages_field=None)

A featurizer that detects tables in PDF documents.

Parameters

Parameters

Name	Type	Default	Info
field	str	'rich_doc_pdf_url'	The name of the column containing the PDF URL paths.
model	str	'microsoft/table-transformer-structure-recognition'	The pretrained Table Transformer model to use for table detection.
pages_field	Optional[str]	None	The name of the column containing the page numbers on which to run the operator on. If None, the operator will run on all pages. Defaults to None.

Returns

Returns

A Table object containing the table metadata information.

Return type

Return type

{RichDocCols.TABLES}