scvi.nn.DecoderSCVI#

class scvi.nn.DecoderSCVI(n_input, n_output, n_cat_list=None, n_layers=1, n_hidden=128, inject_covariates=True, use_batch_norm=False, use_layer_norm=False, scale_activation='softmax', **kwargs)[source]#

Bases: Module

Decodes data from a latent space of n_input dimensions into n_output dimensions.

Uses a fully-connected neural network of n_hidden layers.

Parameters:
n_input int

The dimensionality of the input (latent space)

n_output int

The dimensionality of the output (data space)

n_cat_list Iterable[int] (default: None)

A list containing the number of categories for each category of interest. Each category will be included using a one-hot encoding

n_layers int (default: 1)

The number of fully-connected hidden layers

n_hidden int (default: 128)

The number of nodes per hidden layer

dropout_rate

Dropout rate to apply to each of the hidden layers

inject_covariates bool (default: True)

Whether to inject covariates in each layer, or just the first (default).

use_batch_norm bool (default: False)

Whether to use batch norm in layers

use_layer_norm bool (default: False)

Whether to use layer norm in layers

scale_activation Literal['softmax', 'softplus'] (default: 'softmax')

Activation layer to use for px_scale_decoder

**kwargs

Keyword args for FCLayers.

Attributes table#

Methods table#

forward(dispersion, z, library, *cat_list)

The forward computation for a single sample.

Attributes#

DecoderSCVI.training: bool#

Methods#

DecoderSCVI.forward(dispersion, z, library, *cat_list)[source]#

The forward computation for a single sample.

  1. Decodes the data from the latent space using the decoder network

  2. Returns parameters for the ZINB distribution of expression

  3. If dispersion != 'gene-cell' then the value for that param will be None

Parameters:
dispersion str

One of the following

  • 'gene' - dispersion parameter of NB is constant per gene across cells

  • 'gene-batch' - dispersion can differ between different batches

  • 'gene-label' - dispersion can differ between different labels

  • 'gene-cell' - dispersion can differ for every gene in every cell

z Tensor

tensor with shape (n_input,)

library Tensor

library size

cat_list int

list of category membership(s) for this sample

Returns:

4-tuple of torch.Tensor parameters for the ZINB distribution of expression