class moabb.datasets.MunichMI[source][source]

Munich Motor Imagery dataset.

Motor imagery dataset from Grosse-Wentrup et al. 2009 [1].

A trial started with the central display of a white fixation cross. After 3 s, a white arrow was superimposed on the fixation cross, either pointing to the left or the right. Subjects were instructed to perform haptic motor imagery of the left or the right hand during display of the arrow, as indicated by the direction of the arrow. After another 7 s, the arrow was removed, indicating the end of the trial and start of the next trial. While subjects were explicitly instructed to perform haptic motor imagery with the specified hand, i.e., to imagine feeling instead of visualizing how their hands moved, the exact choice of which type of imaginary movement, i.e., moving the fingers up and down, gripping an object, etc., was left unspecified. A total of 150 trials per condition were carried out by each subject, with trials presented in pseudorandomized order.

Ten healthy subjects (S1–S10) participated in the experimental evaluation. Of these, two were females, eight were right handed, and their average age was 25.6 years with a standard deviation of 2.5 years. Subject S3 had already participated twice in a BCI experiment, while all other subjects were naive to BCIs. EEG was recorded at M=128 electrodes placed according to the extended 10–20 system. Data were recorded at 500 Hz with electrode Cz as reference. Four BrainAmp amplifiers were used for this purpose, using a temporal analog high-pass filter with a time constant of 10 s. The data were re-referenced to common average reference offline. Electrode impedances were below 10 kΩ for all electrodes and subjects at the beginning of each recording session. No trials were rejected and no artifact correction was performed. For each subject, the locations of the 128 electrodes were measured in three dimensions using a Zebris ultrasound tracking system and stored for further offline analysis.



Grosse-Wentrup, Moritz, et al. “Beamforming in noninvasive brain–computer interfaces.” IEEE Transactions on Biomedical Engineering 56.4 (2009): 1209-1219.


data_path(subject[, path, force_update, …])

Get path to local copy of a subject data.

download([subject_list, path, force_update, …])

Download all data from the dataset.


Return the data correspoonding to a list of subjects.

data_path(subject, path=None, force_update=False, update_path=None, verbose=None)[source][source]

Get path to local copy of a subject data.

  • subject (int) – Number of subject to use

  • path (None | str) – Location of where to look for the data storing location. If None, the environment variable or config parameter MNE_DATASETS_(dataset)_PATH is used. If it doesn’t exist, the “~/mne_data” directory is used. If the dataset is not found under the given path, the data will be automatically downloaded to the specified folder.

  • force_update (bool) – Force update of the dataset even if a local copy exists.

  • update_path (bool | None **Deprecated**) – If True, set the MNE_DATASETS_(dataset)_PATH in mne-python config to the given path. If None, the user is prompted.

  • verbose (bool, str, int, or None) – If not None, override default verbose level (see mne.verbose()).


path – Local path to the given data file. This path is contained inside a list of length one, for compatibility.

Return type

list of str