get_item_parameter_estimates: Get trainable variables from the decoder, which serve as item...

In ML2Pvae: Variational Autoencoder Models for IRT Parameter Estimation

Get trainable variables from the decoder, which serve as item parameter estimates.

Description

Get trainable variables from the decoder, which serve as item parameter estimates.

Usage

get_item_parameter_estimates(decoder, model_type = 2)

Arguments

decoder	a trained keras model; can either be the decoder or vae returned from build_vae_independent() or build_vae_correlated
model_type	either 1 or 2, specifying a 1 parameter (1PL) or 2 parameter (2PL) model; if 1PL, then only the difficulty parameter estimates (output layer bias) will be returned; if 2PL, then the discrimination parameter estimates (output layer weights) will also be returned

Value

a list which contains item parameter estimates; the length of this list is equal to model_type - the first entry in the list holds the difficulty parameter estimates, and the second entry (if 2PL) contains discrimination parameter estimates

Examples

Q <- matrix(c(1,0,1,1,0,1,1,0), nrow = 2, ncol = 4)
models <- build_vae_independent(4, 2, Q, model_type = 2)
decoder <- models[[2]]
item_parameter_estimates <- get_item_parameter_estimates(decoder, model_type = 2)
difficulty_est <- item_parameter_estimates[[1]]
discrimination_est <- item_parameter_estimates[[2]]

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