Problem running pdf converter

[sohanasarah]

I am trying to use the pdf to text converter. I installed Haystack 1.15.0 as suggested in the discussion: #4531. But now I'm getting the following error:

---

ValueError Traceback (most recent call last)
Cell In [32], line 4
1 from haystack.nodes import PDFToTextConverter
3 converter = PDFToTextConverter(remove_numeric_tables=True, valid_languages=["en"])
----> 4 doc_pdf = converter.convert(file_path="data/Introduction to mobile network engineering _ GSM, 3G-WCDMA, LTE.pdf", meta=None)[0]

File C:\Program Files\python\lib\site-packages\haystack\nodes\file_converter\pdf.py:171, in PDFToTextConverter.convert(self, file_path, meta, remove_numeric_tables, valid_languages, encoding, id_hash_keys, start_page, end_page, keep_physical_layout, sort_by_position, ocr, ocr_language, multiprocessing)
168 raise ValueError("The ocr parameter must be either 'auto' or 'full'.")
169 self._check_tessdata()
--> 171 pages = self._read_pdf(
172 file_path,
173 sort_by_position=sort_by_position,
174 start_page=start_page,
175 end_page=end_page,
176 ocr=ocr,
177 ocr_language=ocr_language,
178 multiprocessing=multiprocessing,
179 )
181 cleaned_pages = []
182 for page in pages:

File C:\Program Files\python\lib\site-packages\haystack\nodes\file_converter\pdf.py:299, in PDFToTextConverter._read_pdf(self, file_path, ocr_language, sort_by_position, start_page, end_page, ocr, multiprocessing)
296 parts = divide(cpu, page_list)
297 pages_mp = [(i, file_path, parts, sort_by_position, ocr, ocr_language) for i in range(cpu)]
--> 299 with ProcessPoolExecutor(max_workers=cpu) as pool:
300 results = pool.map(self._get_text_parallel, pages_mp)
301 for page in results:

File C:\Program Files\python\lib\concurrent\futures\process.py:609, in ProcessPoolExecutor.init(self, max_workers, mp_context, initializer, initargs)
606 raise ValueError("max_workers must be greater than 0")
607 elif (sys.platform == 'win32' and
608 max_workers > _MAX_WINDOWS_WORKERS):
--> 609 raise ValueError(
610 f"max_workers must be <= {_MAX_WINDOWS_WORKERS}")
612 self._max_workers = max_workers
614 if mp_context is None:

ValueError: max_workers must be <= 61

I need help fixing the issue. Thank you.

[sohanasarah]

I resolved the issue by setting multiprocessing to False.

[priyabratadasHAPPY]

Can you plz share colab or code with me? I stuck from 2weeks for this task -- At-- priyabrat2000das@gmail.com