Am using Model Maker and am throwing this error:
/usr/local/lib/python3.7/dist-packages/tensorflow/python/util/compat.py in as_text(bytes_or_text, encoding)
103 return bytes_or_text
104 elif isinstance(bytes_or_text, bytes):
--> 105 return bytes_or_text.decode(encoding)
106 else:
107 raise TypeError('Expected binary or unicode string, got %r' % bytes_or_text)
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xb0 in position 37: invalid start byte
My guessing is the xml annotation files may not be correctly coded in UTF-8, but they do look pretty ok to me.
Has anyone else had this issue? Any solutions to this error? Using approximately 16,000 images across 16 classes.
Was using exact methodology code plus also annotation of images with approx 1,600 images and everything ran smoothly.
Any help is greatly appreciated.
Cheers,
Will