Helper Module for Deep Learning.

Variational Deep Embedding (VaDE).

class pynet.models.vae.vade.VaDENet(n_classes, input_dim, latent_dim, hidden_dims=[500, 500, 2000], binary=True)[source]¶

Variational Deep Embedding (VaDE) Network.

Variational Deep Embedding: An Unsupervised and Generative Approach to Clustering, Zhuxi Jiang, arXiv 2017.

__init__(n_classes, input_dim, latent_dim, hidden_dims=[500, 500, 2000], binary=True)[source]¶

Init classs.

Parameters

n_classes: int

the number of clusters.

input_dim: int

the dimension of observed data.

latent_dim: int

the dimension of latent space.

hidden_dims: list of int, default [500, 500, 2000]

the network hidden dimensions.

binary: bool, default False

apply sigmoid to get binary output.

data: Tensor (N, M), default None

data that can be used during display callbacks.

labels: list of int (N, ), default None

labels that can be used during display callbacks.

decode(z)[source]¶

encode(x)[source]¶

forward(x)[source]¶

get_gamma(z, z_mu, z_logvar)[source]¶

init_gmm(dataloader, device)[source]¶

init_layers(layer_dims, activation='relu', dropout=0.5)[source]¶

load(path)[source]¶

predict(x)[source]¶

reparameterize(mu, logvar)[source]¶

save(path)[source]¶

property weights¶

class pynet.models.vae.vade.VaDEPreTrainNet(model)[source]¶

Auto-Encoder for pretraining VaDE.

Variational Deep Embedding: An Unsupervised and Generative Approach to Clustering, Zhuxi Jiang, arXiv 2017.

__init__(model)[source]¶

Init class.

Parameters

model: VaDE

the model.

decode(z)[source]¶

encode(x)[source]¶

forward(x)[source]¶