Man pages for ML2Pvae
Variational Autoencoder Models for IRT Parameter Estimation

build_hidden_encoderBuild the encoder for a VAE
build_vae_correlatedBuild a VAE that fits to a normal, full covariance N(m,S)...
build_vae_independentBuild a VAE that fits to a standard N(0,I) latent...
correlation_matrixSimulated latent abilities correlation matrix
diff_trueSimulated difficulty parameters
disc_trueSimulated discrimination parameters
dot-onLoadDisplay a message upon loading package
get_ability_parameter_estimatesFeed forward response sets through the encoder, which outputs...
get_item_parameter_estimatesGet trainable variables from the decoder, which serve as item...
ML2PvaeML2Pvae: A package for creating a VAE whose decoder recovers...
q_1pl_constraintA custom kernel constraint function that forces nonzero...
q_constraintA custom kernel constraint function that restricts weights...
q_matrixSimulated Q-matrix
responsesResponse data
sampling_correlatedA reparameterization in order to sample from the learned...
sampling_independentA reparameterization in order to sample from the learned...
theta_trueSimulated ability parameters
train_modelTrains a VAE or autoencoder model. This acts as a wrapper for...
vae_loss_correlatedA custom loss function for a VAE learning a multivariate...
vae_loss_independentA custom loss function for a VAE learning a standard normal...
validate_inputsGive error messages for invalid inputs in exported functions.
ML2Pvae documentation built on May 23, 2022, 9:05 a.m.