scvi.module.AutoZIVAE

class scvi.module.AutoZIVAE(n_input, alpha_prior=0.5, beta_prior=0.5, minimal_dropout=0.01, zero_inflation='gene', **kwargs)

Bases: VAE

Implementation of the AutoZI model [Clivio et al., 2019].

Parameters:

• n_input (int) – Number of input genes

• alpha_prior (float | None (default: 0.5)) – Float denoting the alpha parameter of the prior Beta distribution of the zero-inflation Bernoulli parameter. Should be between 0 and 1, not included. When set to None, will be set to 1 - beta_prior if beta_prior is not None, otherwise the prior Beta distribution will be learned on an Empirical Bayes fashion.

• beta_prior (float | None (default: 0.5)) – Float denoting the beta parameter of the prior Beta distribution of the zero-inflation Bernoulli parameter. Should be between 0 and 1, not included. When set to None, will be set to 1 - alpha_prior if alpha_prior is not None, otherwise the prior Beta distribution will be learned on an Empirical Bayes fashion.

• minimal_dropout (float (default: 0.01)) – Float denoting the lower bound of the cell-gene ZI rate in the ZINB component. Must be non-negative. Can be set to 0 but not recommended as this may make the mixture problem ill-defined.

• zero_inflation (One of the following) –

  • 'gene' - zero-inflation Bernoulli parameter of AutoZI is constant per gene across cells

  • 'gene-batch' - zero-inflation Bernoulli parameter can differ between different batches

  • 'gene-label' - zero-inflation Bernoulli parameter can differ between different labels

  • 'gene-cell' - zero-inflation Bernoulli parameter can differ for every gene in every cell

See VAE docstring (scvi/models/vae.py) for more parameters. reconstruction_loss should not be specified.

Examples

>>> gene_dataset = CortexDataset()
>>> autozivae = AutoZIVAE(
        gene_dataset.nb_genes, alpha_prior=0.5, beta_prior=0.5, minimal_dropout=0.01
    )

Attributes table

training

Methods table

compute_global_kl_divergence()	Compute global KL divergence.
generative(z, library[, batch_index, y, ...])	Run the generative model.
get_alphas_betas([as_numpy])	Get the parameters of the Bernoulli beta prior and posterior distributions.
get_reconstruction_loss(x, px_rate, px_r, ...)	Compute the reconstruction loss.
loss(tensors, inference_outputs, ...[, ...])	Compute the loss.
rescale_dropout(px_dropout[, eps_log])	Rescale dropout rate.
reshape_bernoulli(bernoulli_params[, ...])	Reshape Bernoulli parameters to match the input tensor.
sample_bernoulli_params([batch_index, y, ...])	Sample Bernoulli parameters from the posterior distribution.
sample_from_beta_distribution(alpha, beta[, ...])	Sample from a beta distribution.

Attributes

AutoZIVAE.training: bool

Methods

AutoZIVAE.compute_global_kl_divergence()

Compute global KL divergence.

Return type:

Tensor

AutoZIVAE.generative(z, library, batch_index=None, y=None, size_factor=None, cont_covs=None, cat_covs=None, n_samples=1, eps_log=1e-08, transform_batch=None)

Run the generative model.

Return type:

dict[str, Tensor]

AutoZIVAE.get_alphas_betas(as_numpy=True)

Get the parameters of the Bernoulli beta prior and posterior distributions.

Return type:

dict[str, Tensor | ndarray]

AutoZIVAE.get_reconstruction_loss(x, px_rate, px_r, px_dropout, bernoulli_params, eps_log=1e-08, **kwargs)

Compute the reconstruction loss.

Return type:

Tensor

AutoZIVAE.loss(tensors, inference_outputs, generative_outputs, kl_weight=1.0, n_obs=1.0)

Compute the loss.

Return type:

tuple[Tensor, Tensor, Tensor]

AutoZIVAE.rescale_dropout(px_dropout, eps_log=1e-08)

Rescale dropout rate.

Return type:

Tensor

AutoZIVAE.reshape_bernoulli(bernoulli_params, batch_index=None, y=None)

Reshape Bernoulli parameters to match the input tensor.

Return type:

Tensor

AutoZIVAE.sample_bernoulli_params(batch_index=None, y=None, n_samples=1)

Sample Bernoulli parameters from the posterior distribution.

Return type:

Tensor

AutoZIVAE.sample_from_beta_distribution(alpha, beta, eps_gamma=1e-30, eps_sample=1e-07)

Sample from a beta distribution.

Return type:

Tensor