class scvi.external.cellassign.CellAssignModule(n_genes, rho, basis_means, b_g_0=None, random_b_g_0=True, n_batch=0, n_cats_per_cov=None, n_continuous_cov=0)[source]#

Bases: BaseModuleClass

Model for CellAssign.

  • n_genes (int) – Number of input genes

  • n_labels – Number of input cell types

  • rho (Tensor) – Binary matrix of cell type markers

  • basis_means (Tensor) – Basis means numpy array

  • b_g_0 (Optional[Tensor] (default: None)) – Base gene expression tensor. If None, use randomly initialized b_g_0.

  • random_b_g_0 (bool (default: True)) – Override to enforce randomly initialized b_g_0. If True, use random default, if False defaults to b_g_0.

  • n_batch (int (default: 0)) – Number of batches, if 0, no batch correction is performed.

  • n_cats_per_cov (Optional[Iterable[int]] (default: None)) – Number of categories for each extra categorical covariate

  • n_continuous_cov (int (default: 0)) – Number of continuous covariates

Attributes table#


Methods table#

generative(x, size_factor[, design_matrix])

Run the generative model.


Run the recognition model.

loss(tensors, inference_outputs, ...[, n_obs])

Compute the loss.

sample(tensors[, n_samples, library_size])

Sample from the posterior distribution.

Attributes# bool#


CellAssignModule.generative(x, size_factor, design_matrix=None)[source]#

Run the generative model.


Run the recognition model.

In the case of variational inference, this function will perform steps related to computing variational distribution parameters. In a VAE, this will involve running data through encoder networks.

This function should return a dictionary with str keys and Tensor values.

CellAssignModule.loss(tensors, inference_outputs, generative_outputs, n_obs=1.0)[source]#

Compute the loss.

CellAssignModule.sample(tensors, n_samples=1, library_size=1)[source]#

Sample from the posterior distribution.