ValueError: zero-size array to reduction operation maximum which has no identity

Hello! During the recognition process the following error is thrown and the recognition effectively stops proceeding, while the Status still reads “Status: ERROR: The process is still running” (OCR4all ver 0.3.0, LAREX ver 0.3.1):

Process ForkProcess-3:
Traceback (most recent call last):
  File "/usr/lib/python3.6/multiprocessing/process.py", line 258, in _bootstrap
    self.run()
  File "/usr/lib/python3.6/multiprocessing/process.py", line 93, in run
    self._target(*self._args, **self._kwargs)
  File "/usr/local/lib/python3.6/dist-packages/calamari_ocr-1.0.5-py3.6.egg/calamari_ocr/ocr/datasets/input_dataset.py", line 99, in run
    out = self.apply_single(*data)
  File "/usr/local/lib/python3.6/dist-packages/calamari_ocr-1.0.5-py3.6.egg/calamari_ocr/ocr/datasets/input_dataset.py", line 119, in apply_single
    line, params = self.params.data_processor.apply([line], 1, False)[0]
  File "/usr/local/lib/python3.6/dist-packages/calamari_ocr-1.0.5-py3.6.egg/calamari_ocr/ocr/data_processing/data_preprocessor.py", line 19, in apply
    processes=processes, progress_bar=progress_bar, max_tasks_per_child=max_tasks_per_child)
  File "/usr/local/lib/python3.6/dist-packages/calamari_ocr-1.0.5-py3.6.egg/calamari_ocr/utils/multiprocessing.py", line 32, in parallel_map
    out = list(map(f, d))
  File "/usr/local/lib/python3.6/dist-packages/calamari_ocr-1.0.5-py3.6.egg/calamari_ocr/ocr/data_processing/data_preprocessor.py", line 50, in _apply_single
    data, params = proc._apply_single(data)
  File "/usr/local/lib/python3.6/dist-packages/calamari_ocr-1.0.5-py3.6.egg/calamari_ocr/ocr/data_processing/data_preprocessor.py", line 50, in _apply_single
    data, params = proc._apply_single(data)
  File "/usr/local/lib/python3.6/dist-packages/calamari_ocr-1.0.5-py3.6.egg/calamari_ocr/ocr/data_processing/center_normalizer.py", line 15, in _apply_single
    out, params = self.normalize(data, cval=np.amax(data))
  File "<__array_function__ internals>", line 6, in amax
  File "/usr/local/lib/python3.6/dist-packages/numpy/core/fromnumeric.py", line 2668, in amax
    keepdims=keepdims, initial=initial, where=where)
  File "/usr/local/lib/python3.6/dist-packages/numpy/core/fromnumeric.py", line 90, in _wrapreduction
    return ufunc.reduce(obj, axis, dtype, out, **passkwargs)
ValueError: zero-size array to reduction operation maximum which has no identity

I have attached the PAGEXML and image files where the error occurs. I couldn’t find anything suspicious here.

0125.txt

[alexander-winkler]

I hope this helps …

0125.zip

[maxnth]

Thank you for the files. I sadly couldn’t reproduce the broken coordinates when converting to latest with the supplied legacy files but we’ll keep looking into this issue.