boost_rwnn {RWNN} | R Documentation |
Boosting random weight neural networks
Description
Use gradient boosting to create ensemble random weight neural network models.
Usage
boost_rwnn(
formula,
data = NULL,
n_hidden = c(),
lambda = NULL,
B = 100,
epsilon = 0.1,
method = NULL,
type = NULL,
control = list()
)
## S3 method for class 'formula'
boost_rwnn(
formula,
data = NULL,
n_hidden = c(),
lambda = NULL,
B = 100,
epsilon = 0.1,
method = NULL,
type = NULL,
control = list()
)
Arguments
formula |
A formula specifying features and targets used to estimate the parameters of the output layer. |
data |
A data-set (either a data.frame or a tibble) used to estimate the parameters of the output layer. |
A vector of integers designating the number of neurons in each of the hidden layers (the length of the list is taken as the number of hidden layers). | |
lambda |
The penalisation constant(s) passed to either rwnn or ae_rwnn (see |
B |
The number of levels used in the boosting tree. |
epsilon |
The learning rate. |
method |
The penalisation type passed to ae_rwnn. Set to |
type |
A string indicating whether this is a regression or classification problem. |
control |
A list of additional arguments passed to the control_rwnn function. |
Value
An ERWNN-object.
References
Friedman J.H. (2001) "Greedy function approximation: A gradrient boosting machine." The Annals of Statistics, 29, 1189-1232.
Examples
n_hidden <- 10
B <- 100
epsilon <- 0.1
lambda <- 0.01
m <- boost_rwnn(y ~ ., data = example_data, n_hidden = n_hidden,
lambda = lambda, B = B, epsilon = epsilon)