braindecode.preprocessing.RemoveDCOffset#
- class braindecode.preprocessing.RemoveDCOffset(*, force_dtype=None)[source]#
Remove the DC offset from the EEG data by subtracting the per-channel median.
This preprocessor mainly exists because some EEG data (depending on the electrical characteristics of the hardware) can have such a large DC offset that highpass filters do not necessarily fully remove it, unless some care is taken with filter settings (noted in EEGLAB documentation [Delorme2004]).
The operation performed is:
\[X'_{c,t} = X_{c,t} - \text{median}_t(X_{c,t})\]where \(c\) indexes the channel and \(t\) indexes time.
- Parameters:
force_dtype (
dtype|str|None) – Convert the EEG data to this dtype before and after processing. Defaults to None, which does not force a dtype conversion.
References
[Delorme2004]Delorme, A. and Makeig, S., 2004. EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. Journal of Neuroscience Methods, 134(1), pp.9-21.
Methods