dataeval.utils.models.VAEDecoder¶

class dataeval.utils.models.VAEDecoder(input_shape, latent_dim, encoding_shape)¶

Variational Autoencoder Decoder.

Parameters:¶

input_shape : tuple[int, ...]¶

Shape of the original input data in CHW format.

latent_dim : int¶

The size of the latent space (encoding dimension).

encoding_shape : tuple[int, ...]¶

Shape of the encoded input data.

forward(z)¶

Perform a forward pass through the decoder.

Parameters:¶

z : torch.Tensor¶

The latent representation.

Returns:¶

The reconstructed output tensor.

Return type:¶

torch.Tensor