2-D Array Error in ICALabel

Support & Discussions
1

Hello,
I am using MNE 17.1 and mne-icalabel 0.6.0 in R (through reticulate). I am getting the following error:

RuntimeWarning: invalid value encountered in scalar multiply
g = np.square(d) * (np.log(d) - 1) # Green’s function.
Error in py_call_impl(callable, call_args$unnamed, call_args$named) :
ValueError: array must be at least 2-d

Here is my code.

ica ← mne$preprocessing$ICA(method=‘infomax’, fit_params=dict(extended=TRUE))
ica$fit(eeg)
clean_eeg ← mne_iclabel$label_components(eeg, ica, method=“iclabel”)

Any idea on what the problem might be?
I don’t think it has anything to do with me running MNE through R, because everything else seems to be working just fine.

2

Hello,

No idea what the problem could be. If you could provide a minimal working example, i.e. a python code snippet which can be run and reproduces the error, we could have a look.

Mathieu

1 Like
3

Here is a minimal working example:

import mne
import mne_icalabel
data_name = “GX_01_2019-09-24_15-45-53.set”
data = mne.io.read_raw_eeglab(data_name)
eeg_channels = data.ch_names[0:32]
data.pick_channels(eeg_channels)
data.crop(tmax=60)
ica = mne.preprocessing.ICA(
max_iter=“auto”,
method=“infomax”,
fit_params=dict(extended=True),
)
ica.fit(data)
clean_data = mne_icalabel.label_components(data,ica,method=‘iclabel’)

Here are the necessary files: Upload Files | Free File Upload and Transfer Up To 20 GB.
To get the .set file I just opened the .cnt file through EEGLAB and saved it to .set because MNE doesn’t work with .cnt from ANT Neuro.

I also tried a few other files I had (.edf and .bdf), They seem to work fine, so I am starting to think that either conversion from .cnt to .set created a problem or the original dataset (online dataset) has something wrong with it…

4

This file is failing during the computation of the topographic map. A topographic map requires a DigMontage to specify the channel location. Let’s have a look at the montage in this file:

data.get_montage().dig

Gives me:

[<DigPoint |        LPA : (-0.0, 0.0, 0.0) mm       : head frame>,
 <DigPoint |     Nasion : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |        RPA : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |     EEG #1 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |     EEG #2 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |     EEG #3 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |     EEG #4 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |     EEG #5 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |     EEG #6 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |     EEG #7 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |     EEG #8 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |     EEG #9 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |    EEG #10 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |    EEG #11 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |    EEG #12 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |    EEG #13 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |    EEG #14 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |    EEG #15 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |    EEG #16 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |    EEG #17 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |    EEG #18 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |    EEG #19 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |    EEG #20 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |    EEG #21 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |    EEG #22 : (0.0, 0.0, 0.0) mm        : head frame>,
...
 <DigPoint |    EEG #28 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |    EEG #29 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |    EEG #30 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |    EEG #31 : (0.0, 0.0, 0.0) mm        : head frame>,
 <DigPoint |    EEG #32 : (0.0, 0.0, 0.0) mm        : head frame>]

All sensors are positioned at the same point, the origin of the head coordinate frame. This montage is definitely not valid.

Mathieu

2 Likes
5

Hi ,

Thanks for the help!

data.set_montage(“easycap-M1”)

before running the code fixed it

6

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.