mne-bids-pipeline takes forever

💬 Support & Discussions
,
1
  • MNE version: mne-1.3.0
  • operating system: NAME=“Ubuntu” VERSION=“18.04.1 LTS (Bionic Beaver)”
mne_bids_pipeline --config config_tuab_eeg.py --n_jobs=40 --steps=preprocessing

Here is my config file.
And the output is:

[21:05:31] ╶╴👋 Welcome aboard the MNE BIDS Pipeline!
[21:05:31] ╶╴🧾 Using configuration: config_tuab_eeg.py
[21:05:31] ╶╴❌ Overriding config.n_jobs = 40
[21:05:31] ┌╴🚀 init/_01_init_derivatives_dir Now running  👇
[21:35:37] └╴🎉 init/_01_init_derivatives_dir Done running  👆 [30m 7s]
[21:35:37] ┌╴🚀 init/_02_find_empty_room Now running  👇
[21:35:37] │ ⏩ init/_02_find_empty_room Skipping, empty-room data only relevant for MEG …
[21:35:37] └╴🎉 init/_02_find_empty_room Done running  👆 [1s]
[21:35:37] ┌╴🚀 preprocessing/_01_data_quality Now running  👇

Hi, I’m running mne_bids_pipeline with the provided config and getting the above output. However, after 24 hours, there are no other updates! No warning! No errors! How can I find the problem?

2

Hi there! I have the identical issue - I made a stack overflow post with more details and code here.

I tried to recreate the matching_pennies experiment. While it worked for me in minutes on Windows, I did not get it to work on Linux. A more detailed description can be found in the stack-overflow post.

Does someone know what Mohammad and I are currently doing wrong? I highly appreciate any advice!

3

Since you’re only processing EEG data, the data quality step should actually only consist of opening a report and adding some info. Do you know if reading the input files manually actually works in MNE-BIDS?

Best wishes,
Richard