SVAE

This module is the implementation of the Hyperspherical VAE proposed in (https://arxiv.org/abs/1804.00891). This models uses a Hyperspherical latent space.

Available samplers

HypersphereUniformSampler

Sampling from uniform distribution on hypersphere.
class pythae.models.SVAEConfig(input_dim=None, latent_dim=10, uses_default_encoder=True, uses_default_decoder=True, reconstruction_loss='mse')[source]

\(\mathcal{S}\)-VAE model config config class

Parameters

  • input_dim (tuple) – The input_data dimension.

  • latent_dim (int) – The latent space dimension in which lives the hypersphere. Default: None.

  • reconstruction_loss (str) – The reconstruction loss to use [‘bce’, ‘mse’]. Default: ‘mse’

class pythae.models.SVAE(model_config, encoder=None, decoder=None)[source]

\(\mathcal{S}\)-VAE model.

Parameters

  • model_config (SVAEConfig) – The Variational Autoencoder configuration setting the main

  • model. (parameters of the) –

  • encoder (BaseEncoder) – An instance of BaseEncoder (inheriting from torch.nn.Module which plays the role of encoder. This argument allows you to use your own neural networks architectures if desired. If None is provided, a simple Multi Layer Preception (https://en.wikipedia.org/wiki/Multilayer_perceptron) is used. Default: None.

  • decoder (BaseDecoder) – An instance of BaseDecoder (inheriting from torch.nn.Module which plays the role of decoder. This argument allows you to use your own neural networks architectures if desired. If None is provided, a simple Multi Layer Preception (https://en.wikipedia.org/wiki/Multilayer_perceptron) is used. Default: None.

Note

For high dimensional data we advice you to provide you own network architectures. With the provided MLP you may end up with a MemoryError.

forward(inputs, **kwargs)[source]

The VAE model

Parameters

inputs (BaseDataset) – The training dataset with labels

Returns

An instance of ModelOutput containing all the relevant parameters

Return type

ModelOutput

get_nll(data, n_samples=1, batch_size=100)[source]

Function computed the estimate negative log-likelihood of the model. It uses importance sampling method with the approximate posterior distribution. This may take a while.

Parameters

  • data (torch.Tensor) – The input data from which the log-likelihood should be estimated. Data must be of shape [Batch x n_channels x …]

  • n_samples (int) – The number of importance samples to use for estimation

  • batch_size (int) – The batchsize to use to avoid memory issues