validate_inputs: Give error messages for invalid inputs in exported functions.

In ML2Pvae: Variational Autoencoder Models for IRT Parameter Estimation

Give error messages for invalid inputs in exported functions.

Description

Give error messages for invalid inputs in exported functions.

Usage

validate_inputs(
  num_items,
  num_skills,
  Q_matrix,
  model_type = 2,
  mean_vector = rep(0, num_skills),
  covariance_matrix = diag(num_skills),
  enc_hid_arch = c(ceiling((num_items + num_skills)/2)),
  hid_enc_activations = rep("sigmoid", length(enc_hid_arch)),
  output_activation = "sigmoid",
  kl_weight = 1,
  learning_rate = 0.001
)

Arguments

num_items	the number of items on the assessment; also the number of nodes in the input/output layers of the VAE
num_skills	the number of skills being evaluated; also the size of the distribution learned by the VAE
Q_matrix	a binary, num_skills by num_items matrix relating the assessment items with skills
model_type	either 1 or 2, specifying a 1 parameter (1PL) or 2 parameter (2PL) model
mean_vector	a vector of length num_skills specifying the mean of each latent trait
covariance_matrix	a symmetric, positive definite, num_skills by num_skills, matrix giving the covariance of the latent traits
enc_hid_arch	a vector detailing the number an size of hidden layers in the encoder
hid_enc_activations	a vector specifying the activation function in each hidden layer in the encoder; must be the same length as enc_hid_arch
output_activation	a string specifying the activation function in the output of the decoder; the ML2P model alsways used 'sigmoid'
kl_weight	an optional weight for the KL divergence term in the loss function
learning_rate	an optional parameter for the adam optimizer

ML2Pvae documentation built on May 23, 2022, 9:05 a.m.