scglue.models.scclue.SCCLUEModel

class scglue.models.scclue.SCCLUEModel(adatas, latent_dim=50, x2u_h_depth=2, x2u_h_dim=256, u2x_h_depth=2, u2x_h_dim=256, du_h_depth=2, du_h_dim=256, dropout=0.2, shared_batches=False, random_seed=0)[source]

Bases: Model

Methods

`adopt_pretrained_model`	rtype: `None`
`compile`	Prepare model for training
`cross_predict`	rtype: `AnnData`
`encode_data`	Compute data sample (cell) embedding
`fit`	Alias of `.trainer.fit`.

Attributes

`ALIGN_BURNIN_PRG`
`MAX_EPOCH_PRG`
`PATIENCE_PRG`
`REDUCE_LR_PATIENCE_PRG`
`logger`