Build a supervised VAE regression model (VAEReg) — vae_build • soilVAE

Build a supervised VAE regression model (VAEReg)

Usage

vae_build(
  input_dim,
  hidden_enc = c(512L, 256L),
  hidden_dec = c(256L, 512L),
  latent_dim = 32L,
  dropout = 0.1,
  lr = 0.001,
  beta_kl = 1,
  alpha_y = 1
)

Arguments

input_dim: integer
hidden_enc: integer vector
hidden_dec: integer vector
latent_dim: integer
dropout: numeric
lr: numeric learning rate
beta_kl: numeric
alpha_y: numeric

Value

A fitted 'Python' 'Keras' model object (VAEReg).

Examples

if (FALSE) { # \dontrun{
  vae_configure()
  model <- vae_build(input_dim = 100L)
} # }