Add `encoding="latin1"` to mnelab's read function?

💬 Support & Discussions
1

I have some eeg files from users in South America that use extended characters. When I try to open them with mnelab (0.8.5), I receiver an error message suggesting I should try encoding="latin1" . That is what I do when using e.g. mne.io.read_raw.

What would be the best way to handle this for mnelab?

Thanks.

Full error message:

Traceback (most recent call last):
  File "/home/john/anaconda3/envs/neurogram/lib/python3.10/site-packages/mne/io/edf/edf.py", line 1623, in _read_annotations_edf
    triggers = re.findall(pat, tals.decode(encoding))
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xe7 in position 287: invalid continuation byte

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/home/john/anaconda3/envs/neurogram/lib/python3.10/site-packages/mnelab/mainwindow.py", line 588, in open_data
    self.model.load(fname)
  File "/home/john/anaconda3/envs/neurogram/lib/python3.10/site-packages/mnelab/model.py", line 36, in wrapper
    f(*args, **kwargs)
  File "/home/john/anaconda3/envs/neurogram/lib/python3.10/site-packages/mnelab/model.py", line 110, in load
    data = read_raw(fname, *args, **kwargs, preload=True)
  File "/home/john/anaconda3/envs/neurogram/lib/python3.10/site-packages/mnelab/io/readers.py", line 82, in read_raw
    return readers[ext](fname, *args, **kwargs)
  File "/home/john/anaconda3/envs/neurogram/lib/python3.10/site-packages/mne/io/edf/edf.py", line 1412, in read_raw_edf
    return RawEDF(input_fname=input_fname, eog=eog, misc=misc,
  File "<decorator-gen-244>", line 12, in __init__
  File "/home/john/anaconda3/envs/neurogram/lib/python3.10/site-packages/mne/io/edf/edf.py", line 181, in __init__
    onset, duration, desc = _read_annotations_edf(
  File "/home/john/anaconda3/envs/neurogram/lib/python3.10/site-packages/mne/io/edf/edf.py", line 1625, in _read_annotations_edf
    raise Exception(
Exception: Encountered invalid byte in at least one annotations channel. You might want to try setting "encoding='latin1'".
2

You are right, I think there should be an option to set this. I have created a feature request here: Add encoding option for reader · Issue #371 · cbrnr/mnelab · GitHub

1 Like