r-validann

Methods and tools for analysing and validating the outputs and modelled functions of artificial neural networks (ANNs) in terms of predictive, replicative and structural validity. Also provides a method for fitting feed-forward ANNs with a single hidden layer.