deepof.models.get_recurrent_decoder

deepof.models.get_recurrent_decoder(input_shape: tuple, latent_dim: int, gru_unroll: bool = False, bidirectional_merge: str = 'concat')

Return a recurrent neural decoder.

Builds a deep neural network capable of decoding the structured latent space generated by one of the compatible classes into a sequence of motion tracking instances, either reconstructing the original input, or generating new data from given clusters.

Parameters:

• input_shape (tuple) – shape of the input data

• latent_dim (int) – dimensionality of the latent space

• gru_unroll (bool) – whether to unroll the GRU layers. Defaults to False.

• bidirectional_merge (str) – how to merge the forward and backward GRU layers. Defaults to “concat”.

Returns:

a keras model that can be trained to decode the latent space into a series of motion tracking sequences.

Return type:

keras.Model