AEBiGRUNetwork

class AEBiGRUNetwork(latent_space_dim=128, n_layers=None, n_units=None, activation='relu', temporal_latent_space=False)

A class to implement an Auto-Encoder based on Bidirectional GRUs.

Parameters:

latent_space_dimint, default=128

Dimension of the latent space.

n_layersint, default=None

Number of BiGRU layers. If None, defaults will be used.

n_unitslist

Number of units in each BiGRU layer. If None, defaults will be used.

activationUnion[list, str]

Activation function(s) to use in each layer. Can be a single string or a list.

temporal_latent_spacebool, default = False

Flag to choose whether the latent space is an MTS or Euclidean space.

Methods

build_network(input_shape, **kwargs)

Construct a network and return its input and output layers.

build_network(input_shape, **kwargs)

Construct a network and return its input and output layers.

Parameters:

input_shapetuple of shape = (n_timepoints (m), n_channels (d))

The shape of the data fed into the input layer.

Returns:

encodera keras Model.

decodera keras Model.