smoother.models.reduction._spanvi

Attributes

logger

Classes

`SPANVAE`	Add spatial loss to the latent representation in the SCANVAE model.
`SpatialANVI`	Spatially-aware ANnotation Variational Inference model.

Module Contents

smoother.models.reduction._spanvi.logger

class smoother.models.reduction._spanvi.SPANVAE(n_input: int, spatial_loss: smoother.SpatialLoss | None = None, lambda_spatial_loss=0.1, sp_loss_as_kl: bool = False, n_batch: int = 0, n_labels: int = 0, n_hidden: int = 128, n_latent: int = 10, n_layers: int = 1, n_continuous_cov: int = 0, n_cats_per_cov: Iterable[int] | None = None, dropout_rate: float = 0.1, dispersion: Literal['gene', 'gene-batch', 'gene-label', 'gene-cell'] = 'gene', log_variational: bool = True, gene_likelihood: Literal['zinb', 'nb'] = 'zinb', use_observed_lib_size: bool = True, y_prior: torch.Tensor | None = None, labels_groups: Sequence[int] = None, use_labels_groups: bool = False, linear_classifier: bool = False, classifier_parameters: dict | None = None, use_batch_norm: Literal['encoder', 'decoder', 'none', 'both'] = 'both', use_layer_norm: Literal['encoder', 'decoder', 'none', 'both'] = 'none', **vae_kwargs)

Bases: scvi.module.SCANVAE

Add spatial loss to the latent representation in the SCANVAE model.

Parameters:

spatial_loss – Spatial loss to apply on the latent representation. If None, no spatial loss.
lambda_spatial_loss – Weight of the spatial loss.
sp_loss_as_kl – Whether to treat the spatial loss as a KL divergence term. If True, the spatial loss will replace the original KL divergence term in the loss function. If False, the spatial loss will be added to the original loss function.

spatial_loss = None

l_sp_loss = 0.1

sp_loss_as_kl = False

diag_sp_inv_cov = None

loss(tensors, inference_outputs, generative_ouputs, kl_weight=1, labelled_tensors=None, classification_ratio=None): Compute the loss.

classmethod from_vae(vae_module: scvi.module.SCANVAE, spatial_loss: smoother.SpatialLoss | None = None, lambda_spatial_loss=0.1, sp_loss_as_kl: bool = False)

class smoother.models.reduction._spanvi.SpatialANVI(st_adata: anndata.AnnData, spatial_loss: smoother.SpatialLoss | None = None, lambda_spatial_loss=0.1, sp_loss_as_kl: bool = False, n_hidden: int = 128, n_latent: int = 10, n_layers: int = 1, dropout_rate: float = 0.1, dispersion: Literal['gene', 'gene-batch', 'gene-label', 'gene-cell'] = 'gene', gene_likelihood: Literal['zinb', 'nb', 'poisson'] = 'nb', latent_distribution: Literal['normal', 'ln'] = 'normal', use_observed_lib_size: bool = True, linear_classifier: bool = False, datamodule: lightning.LightningDataModule | None = None, **model_kwargs)

Bases: scvi.model.SCANVI

Spatially-aware ANnotation Variational Inference model.

_data_splitter_cls

_module_cls

_training_plan_cls

_train_runner_cls

module

_model_summary_string

init_params_

dr_logs

train(max_epochs: int | None = None, accelerator: str = 'auto', devices: int | List[int] | str = 'auto', plan_kwargs: dict | None = None, **trainer_kwargs): Unsupervised training of the spatial model without minibatches.

classmethod from_rna_model(st_adata: anndata.AnnData, sc_model: scvi.model.SCANVI, spatial_loss: smoother.SpatialLoss | None = None, lambda_spatial_loss=0.1, sp_loss_as_kl: bool = False, unfrozen: bool = False, freeze_dropout: bool = False, freeze_expression: bool = True, freeze_decoder_first_layer: bool = True, freeze_batchnorm_encoder: bool = True, freeze_batchnorm_decoder: bool = False, freeze_classifier: bool = True, **spanvae_kwargs)

Alternate constructor for exploiting a pre-trained model on RNA-seq data.

Note that because of the dropout layer, even though the new instance is initialized with the same parameters as the pre-trained model, new_instance.get_latent_representation() may not return the same latent representation as the pre-trained model.

Parameters:

st_adata – registed anndata object
sc_model – pretrained SCANVI model