VEncoder

class aixd.mlmodel.architecture.encoders.VEncoder(*args: Any, **kwargs: Any)[source]

Bases: Encoder

Encoder module of a conditional variational autoencoder with additional parameters for the latent distribution.

Parameters:
  • in_heads (Dict[str, nn.Module]) – Dictionary of input heads, where the keys are strings and the values are PyTorch modules.

  • out_heads (Dict[str, nn.Module]) – Dictionary of output heads, where the keys are strings and the values are PyTorch modules.

  • splits (Dict[str, Tuple[int, int]]) – Start and end indices for each feature in the input data vectors.

  • layer_widths (List[int]) – List of integer values representing the widths of the hidden layers in the encoder.

  • latent_dim (int) – Integer value representing the latent dimension of the model.

  • activation (str) – String representing the activation function to be used in the hidden layers of the encoder.

Methods

forward

Forward pass of the encoder.

Inherited Methods

summary

Prints a summary of a PyTorch model, including the number of parameters, the layers, their names, and the dimensionality.