sfaira.models.embedding.ModelKerasVae¶

class sfaira.models.embedding.ModelKerasVae(in_dim, latent_dim=(128, 64, 2, 64, 128), dropout_rate=0.1, l1_coef: float = 0.0, l2_coef: float = 0.0, batchnorm: bool = False, activation='tanh', init='glorot_uniform', output_layer='nb')¶

Attributes

version

Methods

`predict_embedding`(x[, variational])
`predict_reconstructed`(x)