DCGAN_Generator: DCGAN Generator

In RGAN: Generative Adversarial Nets (GAN) in R

DCGAN Generator

Description

Provides a torch::nn_module with a simple deep convolutional neural net architecture, for use as the default architecture for image data in RGAN. Architecture inspired by: https://pytorch.org/tutorials/beginner/dcgan_faces_tutorial.html

Usage

DCGAN_Generator(
  noise_dim = 100,
  number_channels = 3,
  ngf = 64,
  dropout_rate = 0.5
)

Arguments

noise_dim	The length of the noise vector per example
number_channels	The number of channels in the image (RGB is 3 channels)
ngf	The number of feature maps in generator
dropout_rate	The dropout rate for each hidden layer

Value

A torch::nn_module for the DCGAN Generator

RGAN documentation built on March 30, 2022, 1:07 a.m.