Navigation

Operators and Keywords

Function List:

C++ API

nnet

A feed forward multi-layer neural network.

Select category:

Creation

newff
`newff' create a feed-forward backpropagation network

Data preprocessing & postprocessing

mapstd
Map values to mean 0 and standard derivation to 1.
prestd
`prestd' preprocesses the data so that the mean is 0 and the standard deviation is 1.
poststd
`poststd' postprocesses the data which has been preprocessed by `prestd'.
trastd
`trastd' preprocess additional data for neural network simulation.

Simulation

sim
`sim' is usuable to simulate a before defined neural network.

Training

train
A neural feed-forward network will be trained with `train'

Transfer functions

logsig
`logsig' is a non-linear transfer function used to train neural networks.
purelin
`purelin' is a linear transfer function used by neural networks
radbas
Radial basis transfer function.
tansig
`tansig' is a non-linear transfer function used to train neural networks.

Utility

dividerand
Divide the vectors in training, validation and test group according to the informed ratios
ind2vec
`vec2ind' convert indices to vector
isposint
`isposint' returns true for positive integer values.
min_max
`min_max' returns variable Pr with range of matrix rows
saveMLPStruct
`saveStruct' saves a neural network structure to *.
subset
`subset' splits the main data matrix which contains inputs and targets into 2 or 3 subsets depending on the parameters.
vec2ind
`vec2ind' convert vectors to indices