scvi.model.CondSCVI.get_vamp_prior

CondSCVI.get_vamp_prior(adata=None, p=50)[source]

Return an empirical prior over the cell-type specific latent space (vamp prior) that may be used for deconvolution.

Parameters
adata : AnnData | NoneOptional[AnnData] (default: None)

AnnData object with equivalent structure to initial AnnData. If None, defaults to the AnnData object used to initialize the model.

p : intint (default: 50)

number of components in the mixture model underlying the empirical prior

Return type

ndarrayndarray

Returns

mean_vprior: np.ndarray

(n_labels, p, D) array

var_vprior

(n_labels, p, 3) array