caret (version 6.0-72)

avNNet: Neural Networks Using Model Averaging

Description

Aggregate several neural network models

Usage

avNNet(x, ...)
"avNNet"(formula, data, weights, ..., repeats = 5, bag = FALSE, allowParallel = TRUE, seeds = sample.int(1e+05, repeats), subset, na.action, contrasts = NULL)
"avNNet"(x, y, repeats = 5, bag = FALSE, allowParallel = TRUE, seeds = sample.int(1e+05, repeats), ...)
"print"(x, ...)
"predict"(object, newdata, type = c("raw", "class", "prob"), ...)

Arguments

x
matrix or data frame of x values for examples.
formula
A formula of the form class ~ x1 + x2 + ...
data
Data frame from which variables specified in formula are preferentially to be taken.
weights
(case) weights for each example -- if missing defaults to 1.
repeats
the number of neural networks with different random number seeds
bag
a logical for bagging for each repeat
allowParallel
if a parallel backend is loaded and available, should the function use it?
seeds
random number seeds that can be set prior to bagging (if done) and network creation. This helps maintain reproducibility when models are run in parallel.
subset
An index vector specifying the cases to be used in the training sample. (NOTE: If given, this argument must be named.)
na.action
A function to specify the action to be taken if NAs are found. The default action is for the procedure to fail. An alternative is na.omit, which leads to rejection of cases with missing values on any required variable. (NOTE: If given, this argument must be named.)
contrasts
a list of contrasts to be used for some or all of the factors appearing as variables in the model formula.
y
matrix or data frame of target values for examples.
object
an object of class avNNet as returned by avNNet.
newdata
matrix or data frame of test examples. A vector is considered to be a row vector comprising a single case.
type
Type of output, either: raw for the raw outputs, code for the predicted class or prob for the class probabilities.
...
arguments passed to nnet

Value

For avNNet, an object of "avNNet" or "avNNet.formula". Items of interest in #' the output are:

Details

Following Ripley (1996), the same neural network model is fit using different random number seeds. All the resulting models are used for prediction. For regression, the output from each network are averaged. For classification, the model scores are first averaged, then translated to predicted classes. Bagging can also be used to create the models.

If a parallel backend is registered, the foreach package is used to train the networks in parallel.

References

Ripley, B. D. (1996) Pattern Recognition and Neural Networks. Cambridge.

See Also

nnet, preProcess

Examples

Run this code
data(BloodBrain)
## Not run: 
# modelFit <- avNNet(bbbDescr, logBBB, size = 5, linout = TRUE, trace = FALSE)
# modelFit
# 
# predict(modelFit, bbbDescr)
# ## End(Not run)

Run the code above in your browser using DataCamp Workspace