def setUp(self):
state = np.random.RandomState(0)
X = generate_dummy_dataset_alltypes(state, N_ROWS, DIM_REG, DIM_BOOL,
DIM_ORD, DIM_CAT).X
temps = state.random((N_ROWS, 1))
self.data = X
self.network = VaeNet(config, dtype=tf.dtypes.float64)
def __init__(self, config: VAE.Config, **kwargs):
tf.keras.Model.__init__(self, **kwargs)
self.config = config
self.network = VaeNet(config, **kwargs)
self.lossnet = VaeLossNet(latent_eps=1e-6, prefix="loss", **kwargs)
self.weight_getter = Vae.CoolingRegime(config, dtype=self.dtype)
AutoencoderModelBaseMixin.__init__(
self,
self.weight_getter,
self.network,
self.config.get_latent_parser_type(),
self.config.get_fake_output_getter(),
)
DIM_X = DIM_REG + DIM_BOOL + sum(DIM_ORD) + sum(DIM_CAT)
EMB_DIM = 10
LAT_DIM = 5
NCATS = 5
desciminatorConfig = DescriminatorNet.Config(embedding_dimensions=[EMB_DIM])
vaeConfig = VaeNet.Config(
input_dimensions=MultipleObjectiveDimensions(
regression=DIM_REG,
boolean=DIM_BOOL,
ordinal=DIM_ORD,
categorical=DIM_CAT,
),
output_dimensions=MultipleObjectiveDimensions(
regression=DIM_REG,
boolean=DIM_BOOL,
ordinal=DIM_ORD,
categorical=DIM_CAT,
),
encoder_embedding_dimensions=[EMB_DIM],
decoder_embedding_dimensions=[EMB_DIM],
latent_dim=LAT_DIM,
)
config0 = VaeNet.Config(
input_dimensions=MultipleObjectiveDimensions(
regression=DIM_REG,
boolean=DIM_BOOL,
ordinal=DIM_ORD,
categorical=DIM_CAT,