BGenerator

class torcheeg.models.BGenerator(in_channels: int = 128, out_channels: int = 4, grid_size: Tuple[int, int] = (9, 9))

TorchEEG provides an EEG feature generator based on CNN architecture and GAN for generating EEG grid representations of different frequency bands based on a given class label.

import torch

from torcheeg.models.gan.bgan import BGenerator

g_model = BGenerator(in_channels=128)
z = torch.normal(mean=0, std=1, size=(1, 128))
fake_X = g_model(z)

Parameters:

• in_channels (int) – The input feature dimension (of noise vectors). (default: 128)

• out_channels (int) – The generated feature dimension of each electrode. (default: 4)

• grid_size (tuple) – Spatial dimensions of grid-like EEG representation. (default: (9, 9))

forward(x: Tensor)

Parameters:

x (torch.Tensor) – a random vector, the ideal input shape is [n, 128]. Here, n corresponds to the batch size, and 128 corresponds to in_channels.

Returns:

the generated fake EEG signals. Here, 4 corresponds to the out_channels, and (9, 9) corresponds to the grid_size.

Return type:

torch.Tensor[n, 4, 9, 9]