BDecoder

class torcheeg.models.BDecoder(in_channels: int = 64, out_channels: int = 4, grid_size: Tuple[int, int] = (9, 9))

The variational autoencoder consists of two parts, an encoder, and a decoder. The encoder compresses the input into the latent space. The decoder receives as input the information sampled from the latent space and produces it as similar as possible to ground truth. The latent vector should approach the gaussian distribution supervised by KL divergence based on the variation trick. This class implement the decoder part.

encoder = BEncoder(in_channels=4)
decoder = BDecoder(in_channels=64, out_channels=4)
mock_eeg = torch.randn(1, 4, 9, 9)
mu, logvar = encoder(mock_eeg)
std = torch.exp(0.5 * logvar)
eps = torch.randn_like(std)
z = eps * std + mu
fake_X = decoder(z)

Parameters

• in_channels (int) – The input feature dimension (of noise vectors). (defualt: 64)

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

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

forward(x: Tensor)

Parameters

x (torch.Tensor) – Given the mean and standard deviation vectors, the feature vector z obtained using the reparameterization technique. The shapes of the feature vector should be [n, 64]. Here, n corresponds to the batch size, and 64 corresponds to in_channels.

Returns

the decoded results, which should have the same shape as the input noise, i.e., [n, 4, 9, 9]. Here, n corresponds to the batch size, 4 corresponds to in_channels, and (9, 9) corresponds to grid_size.

Return type

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