R/value-functions.R:

CRAN
RGAN: Generative Adversarial Nets (GAN) in R

($D(G(z))$)
#'
#' @return The function returns a named list with the entries d_loss and g_loss

R/value-functions.R:

GITHUB
mneunhoe/RGAN: Generative Adversarial Nets (GAN) in R

($D(G(z))$)
#'
#' @return The function returns a named list with the entries d_loss and g_loss

GAN_value_fct: GAN Value Function

CRAN
RGAN: Generative Adversarial Nets (GAN) in R

examples ($D(x)$)
fake_scores
The discriminator scores on fake examples ($D(G(z))$)

KLWGAN_value_fct: KLWGAN Value Function

CRAN
RGAN: Generative Adversarial Nets (GAN) in R

fake_scores
The discriminator scores on fake examples ($D(G(z))$)
Value

WGAN_value_fct: WGAN Value Function

CRAN
RGAN: Generative Adversarial Nets (GAN) in R

The discriminator scores on fake examples ($D(G(z))$)
Value
The function returns a named list with the entries d_loss and g_loss

KLWGAN_value_fct: KLWGAN Value Function

GITHUB
mneunhoe/RGAN: Generative Adversarial Nets (GAN) in R

fake_scores
The discriminator scores on fake examples ($D(G(z))$)
Value

GAN_value_fct: GAN Value Function

GITHUB
mneunhoe/RGAN: Generative Adversarial Nets (GAN) in R

on real examples ($D(x)$)
fake_scores
The discriminator scores on fake examples ($D(G(z))$)

R/gan-trainer.R:

GITHUB
mneunhoe/RGAN: Generative Adversarial Nets (GAN) in R

if(track_loss & length(losses) > 0){
losses$g_loss <- c(losses$g_loss, step_loss$g_loss)
losses$d_loss <- c

WGAN_value_fct: WGAN Value Function

GITHUB
mneunhoe/RGAN: Generative Adversarial Nets (GAN) in R

The discriminator scores on fake examples ($D(G(z))$)
Value
The function returns a named list with the entries d_loss and g_loss