inst/examples/aerwnn_example.R

In svilsen/RWNN: Random Weight Neural Networks

n_hidden <- c(20, 15, 10, 5)
lambda <- c(2, 0.01)

## Using L1-norm in the auto-encoder (sparse solution)
m <- ae_rwnn(y ~ ., data = example_data, n_hidden = n_hidden, lambda = lambda, method = "l1")

## Using L2-norm in the auto-encoder (dense solution)
m <- ae_rwnn(y ~ ., data = example_data, n_hidden = n_hidden, lambda = lambda, method = "l2")