Gen_NN2 and Gen_NN3 are used to generate training sets from maps. Gen_NN2 uses two input maps and Gen_NN3 three. SAMT selects randomly from the inputs and the highlighted output map a training set containing about 2000 data sets. SAMT-NN can open this sets and build a model using neural network technology.

The export to R function is an easy interface to the statistic package R (http://www.r-project.org). It works like Gen_NNx excluding it uses a hdf filesystem to generate the data. That allows the usage of more than three inputs.