FBCNet

class dpeeg.models.FBCNet.FBCNet(nCh: int, nTime: int, nCls: int, bands: int = 9, m: int = 32, stride: int = 4)

FBCNet: A Multi-view Convolutional Neural Network for Brain-Computer Interface (FBCNet).

FBCNet [1] employs a multi-view data representation followed by spatial filtering to extract spectro-spatially discriminative features. In FBCNet, a novel Variance layer is proposed to effectively aggregate the EEG time- domain information.

Parameters:

• nCh (int) – Number of electrode channels.

• nTime (int) – Number of data sampling points. For example, a 4-second data input with a sampling rate of 250 Hz is 1000.

• nCls (int) – Number of classification categories.

• bands (int) – The filter dimension of the input multi-view data.

• m (int) – Number of sptatial filters.

• stride (int) – Number of time windows. Must evenly divide nTime.

References

[1]

R. Mane et al., “FBCNet: A Multi-view Convolutional Neural Network for Brain-Computer Interface,” arXiv.org. Accessed: Mar. 09, 2023. [Online]. vailable: https://arxiv.org/abs/2104.01233v1

forward(x)

Forward pass function that processes the input EEG data and produces the decoded results.

Parameters:

x (Tensor) – Input EEG data, shape (batch_size, bands, nCh, nTime).

Returns:

cls_prob – Predicted class probability, shape (batch_size, nCls).

Return type:

Tensor