scglue.models.sc.DataEncoder.forward

DataEncoder.forward(x, xrep, lazy_normalizer=True)

Encode data to sample latent distribution

Parameters:

• x (torch.Tensor) – Input data

• xrep (torch.Tensor) – Alternative input data

• lazy_normalizer (bool) – Whether to skip computing x normalizer (just return None) if xrep is non-empty

Return type:

typing.Tuple[torch.distributions.normal.Normal, typing.Optional[torch.Tensor]]

Returns:

• u – Sample latent distribution

• normalizer – Data normalizer

Note

Normalization is always computed on x. If xrep is empty, the normalized x will be used as input to the encoder neural network, otherwise xrep is used instead.