Discriminator {RGAN}

Discriminator

Description

Provides a torch::nn_module with a simple fully connected neural net, for use as the default architecture for tabular data in RGAN.

Usage

Discriminator(
  data_dim,
  hidden_units = list(128, 128),
  dropout_rate = 0.5,
  sigmoid = FALSE
)

Arguments

data_dim

The number of columns in the data set

hidden_units

A list of the number of neurons per layer, the length of the list determines the number of hidden layers

dropout_rate

The dropout rate for each hidden layer

sigmoid

Switch between a sigmoid and linear output layer (the sigmoid is needed for the original GAN value function)

Value

A torch::nn_module for the Discriminator