Skip to content

Simple interface to MLP models via keras3

Source: R/keras3.R
keras3_mlp.Rd

Instead of building a keras3 model sequentially, keras3_mlp can be used to create a feedforward network with a single hidden layer. Regularization is via either weight decay or dropout.

Usage

keras3_mlp(
  x,
  y,
  hidden_units = 5,
  penalty = 0,
  dropout = 0,
  epochs = 20,
  activation = "softmax",
  seed = sample.int(10^5, size = 1),
  ...
)

Arguments

x

A data frame or matrix of predictors.

y

A vector (factor or numeric) or matrix (numeric) of outcome data.

hidden_units

An integer for the number of hidden units.

penalty

A non-negative real number for the amount of weight decay. Either this parameter or dropout can be specified.

dropout

The proportion of parameters to set to zero. Either this parameter or penalty can be specified.

epochs

An integer for the number of passes through the data.

activation

A character string for the type of activation function between layers.

seed

A single positive integer to control randomness.

...

Additional named arguments to pass to keras3::compile() or keras3::fit().

Value

A keras3 model object.