sfaira.models.embedding.ModelKerasAe¶

class sfaira.models.embedding.ModelKerasAe(in_dim, latent_dim: Tuple = (64, 32, 64), l2_coef: float = 0.0, l1_coef: float = 0.0, dropout_rate: float = 0.0, input_dropout: float = 0.0, batchnorm: bool = True, activation='relu', init='glorot_uniform', output_layer='nb')¶

Combines the encoder and decoder into an end-to-end model for training.

Attributes

version

Methods

`predict_embedding`(x, **kwargs)
`predict_reconstructed`(x, **kwargs)