ProdLDA

Module Contents

ProdLDA

Autoencoding Variational Inference For Topic Models. ICLR 2017

class ProdLDA(vocab_size, num_topics=50, en_units=200, dropout=0.4)

Bases: torch.nn.Module

Autoencoding Variational Inference For Topic Models. ICLR 2017

Akash Srivastava, Charles Sutton.

num_topics = 50
a
mu2
var2
fc11
fc12
fc21
fc22
mean_bn
logvar_bn
decoder_bn
fc1_drop
theta_drop
fcd1
get_beta()
get_theta(x)
reparameterize(mu, logvar)
encode(x)
decode(theta)
forward(x)
loss_function(x, recon_x, mu, logvar)