Full-epochs decoding
Based on N=1 subjects. Each dot represents the mean cross-validation score for a single subject. The dashed line is expected chance performance.
Time generalization: AdvanceTempo vs. DelayTempo
Time generalization (generalization across time, GAT): each classifier is trained on each time point, and tested on all other time points. The results were averaged across N=1 subjects.
CSP decodingAdvanceTempo vs. DelayTempo
Mean decoding scores. Error bars represent bootstrapped 95% confidence intervals.
CSP TF decodingAdvanceTempo vs. DelayTempo
Found 0 clusters with p < 0.05 (clustering bins with absolute t-values > nan).
  """Mobile brain body imaging (MoBI) gait adaptation experiment.

See ds001971 on OpenNeuro: https://github.com/OpenNeuroDatasets/ds001971
"""

from mne_bids_pipeline.typing import ArbitraryContrast

bids_root = "~/mne_data/ds001971"
deriv_root = "~/mne_data/derivatives/mne-bids-pipeline/ds001971"

task = "AudioCueWalkingStudy"
interactive = False
ch_types = ["eeg"]
reject = {"eeg": 150e-6}
conditions = ["AdvanceTempo", "DelayTempo"]
contrasts = [
    ArbitraryContrast(
        name="AdvanceMinusDelay",
        conditions=["AdvanceTempo", "DelayTempo"],
        weights=[1.0, -1.0],
    ),
]

subjects = ["001"]
runs = ["01"]
epochs_decim = 5  # to 100 Hz

# This is mostly for testing purposes!
decode = True
decoding_time_generalization = True
decoding_time_generalization_decim = 2
decoding_csp = True
decoding_csp_freqs = {
    "beta": [13, 20, 30],
}
decoding_csp_times = [-0.2, 0.0, 0.2, 0.4]

# Just to test that MD5 works
memory_file_method = "hash"


  Platform             Linux-6.8.0-1040-aws-x86_64-with-glibc2.35
Python               3.13.11 (main, Dec  6 2025, 01:10:48) [GCC 11.4.0]
Executable           /home/circleci/.pyenv/versions/3.13.11/bin/python3.13
CPU                  Intel(R) Xeon(R) Platinum 8124M CPU @ 3.00GHz (36 cores)
Memory               4.0 GiB

Core
├☑ mne               1.12.0.dev117+g39ee092e1 (unable to check for latest version on GitHub, unknown error: HTTP Error 403: rate limit exceeded)
├☑ numpy             2.4.3 (OpenBLAS 0.3.31.dev with 2 threads)
├☑ scipy             1.17.1
└☑ matplotlib        3.10.8 (backend=agg)

Numerical (optional)
├☑ sklearn           1.8.0
├☑ numba             0.64.0
├☑ nibabel           5.4.2
├☑ pandas            3.0.1
├☑ h5io              0.2.5
├☑ h5py              3.16.0
└☐ unavailable       nilearn, dipy, openmeeg, cupy

Visualization (optional)
├☑ pyvista           0.47.1 (OpenGL 4.5 (Core Profile) Mesa 23.2.1-1ubuntu3.1~22.04.3 via llvmpipe (LLVM 15.0.7, 256 bits))
├☑ pyvistaqt         0.11.3
├☑ vtk               9.6.0
├☑ qtpy              2.4.3 (PySide6=6.10.2)
└☐ unavailable       ipympl, pyqtgraph, mne-qt-browser, ipywidgets, trame_client, trame_server, trame_vtk, trame_vuetify

Ecosystem (optional)
├☑ mne-bids          0.19.0.dev28+g6dcc45d4c
├☑ mne-icalabel      0.8.1
├☑ mne-bids-pipeline 1.10.0.dev171+g19a09516e
├☑ eeglabio          0.1.3
├☑ edfio             0.4.13
├☑ curryreader       0.1.2
├☑ pybv              0.7.6
├☑ defusedxml        0.7.1
└☐ unavailable       mne-nirs, mne-features, mne-connectivity, neo, mffpy, pymef, antio