If it is searchable pdf file I would suggest using Camelot. If it is not searchable or a scan, I would use Tesseract OCR to extract text from the table. Let me know if it helps or if you need more info regarding either Tesseract or Camelot.