mne.io.Raw.filter Error with NIRS data

[Unknolwn]

Description of the problem

Hello, I’m working with NIRS data have trouble updating my code to MNE latest version because of the following MNE filtering function stops working when I do. mne.io.Raw — MNE 1.2.2 documentation

MNE version: 1.0.3
MNE-BIDS version : 0.10
MNE-NIRS version : 0.5.0
operating system: Windows 10

Steps to reproduce

Here is an example of a working code with MNE 1.0.3 but with MNE 1.1 or 1.3 I have a warning with same function : 
import mne
import mne_bids
import mne_nirs
#steps to import data
BidsPath="C:/datafolder"
dataset = mne_bids.BIDSPath(root=BidsPath, task=mne_bids.get_entity_vals(BidsPath, 'task')[0], datatype="nirs", suffix="nirs", extension=".snirf")
subjects = mne_bids.get_entity_vals(BidsPath, 'subject')
sessions = mne_bids.get_entity_vals(BidsPath,'session')
bids_path=dataset.update(subject=subjects[0])
bids_path=dataset.update(session=sessions[0])
raw_data=mne_bids.read_raw_bids(bids_path=bids_path, extra_params=None, verbose=False)
#convert data
OD=mne.preprocessing.nirs.optical_density(raw_data)
Conc=mne.preprocessing.nirs.beer_lambert_law(OD,ppf=6)
#filter data
Filtered=Conc.copy().filter(0.05,0.1, l_trans_bandwidth=0.01, h_trans_bandwidth=0.01, verbose=False)

Link to data

You can find NIRS dataset here: https://osf.io/b4wck/

Expected results

Filtered data with only this warning: mne\filter.py:312: DeprecationWarning: Keyword argument 'nyq' is deprecated in favour of 'fs' and will be removed in SciPy 1.12.0. this_h = firwin(this_N, (prev_freq + this_freq) / 2.,

Actual results

Result with MNE 1.3:

File "", line 1, in File "", line 10, in read_raw_bids File "C:\environment\lib\site-packages\mne_bids\read.py", line 713, in read_raw_bids raw = _handle_channels_reading(channels_fname, raw) File "C:\environment\lib\site-packages\mne_bids\read.py", line 552, in _handle_channels_reading raw.rename_channels({raw_ch_name: bids_ch_name}) File "", line 12, in rename_channels File "C:\environment\lib\site-packages\mne\channels\channels.py", line 414, in rename_channels rename_channels(self.info, mapping, allow_duplicates) File "", line 12, in rename_channels File "C:\environement\lib\site-packages\mne\channels\channels.py", line 1112, in rename_channels raise ValueError('New channel names are not unique, renaming failed') ValueError: New channel names are not unique, renaming failed

Additional information

Platform: Windows-10-10.0.19044-SP0 mne: 1.0.3 numpy: 1.22.3 {} scipy: 1.10.1 matplotlib: 3.7.1 {backend=TkAgg}

sklearn: 1.2.2 numba: Not found nibabel: 5.1.0 nilearn: 0.10.0 dipy: Not found cupy: Not found pandas: 2.0.0 pyvista: Not found pyvistaqt: Not found ipyvtklink: Not found vtk: Not found PyQt5: Not found ipympl: Not found pooch: v1.7.0

mne_bids: 0.10 mne_nirs: 0.5.0

[welcome[bot]]

Hello! 👋 Thanks for opening your first issue here! ❤️ We will try to get back to you soon. 🚴

[larsoner]

the following MNE filtering function stops working when I do

To me it looks like it's the reading function that fails much earlier than the filtering function. Can you confirm that the failure is actually up at the read_raw_bids step?

[hoechenberger]

To me it looks like it's the reading function that fails

Yes, according to the traceback, this issue has nothing to do with filtering. The issue should be transferred over to mne-tools/mne-bids

[Unknolwn]

Oh, okay, I'm going to do that

[Unknolwn]

I found this issue (https://github.com/mne-tools/mne-bids/issues/1117) on MNE_BIDS that looks like mine but the conclusion seems to lead back to MNE-python.

[larsoner]

In https://github.com/mne-tools/mne-bids/issues/1117 I think the conclusion was that MNE-BIDS needs to take care of ordering differences between the data and channels.tsv. Not sure if it's the same problem, will have to test...