i want to extract tables from a pdf using img2table[paddle] but i have a weird issue

This is my code:

import os
from img2table.ocr import PaddleOCR
paddle_ocr = PaddleOCR(lang="fr", kw={"use_dilation": True})
from img2table.document import PDF
pdf = PDF(src=pdf_path)
print(pdf)

pdf.to_xlsx(output_path,
                    ocr=paddle_ocr,
                    implicit_rows=True,
                    borderless_tables=True,
                    min_confidence=50)

but i get this error:

 RuntimeWarning: Mean of empty slice.
  return _methods._mean(a, axis=axis, dtype=dtype,
Erreur inattendue: division by zero