ncem.models.ModelCVAEncem

class ncem.models.ModelCVAEncem(input_shapes, latent_dim: int = 10, intermediate_dim_enc: int = 128, intermediate_dim_dec: int = 128, depth_enc: int = 1, depth_dec: int = 1, dropout_rate: float = 0.1, l2_coef: float = 0.0, l1_coef: float = 0.0, cond_type: str = 'gcn', cond_depth: int = 1, cond_dim: int = 8, cond_dropout_rate: float = 0.1, cond_activation: Union[str, keras.engine.base_layer.Layer] = 'relu', cond_l2_reg: float = 0.0, cond_use_bias: bool = True, use_domain: bool = False, scale_node_size: bool = False, use_type_cond: bool = True, use_batch_norm: bool = False, transform_input: bool = False, output_layer: str = 'gaussian', **kwargs)

Model class for NCEM conditional variational autoencoder with graph layer IND (MAX) or GCN.

Methods