Training of Neural Networks

Training of neural networks using backpropagation, resilient backpropagation with (Riedmiller, 1994) or without weight backtracking (Riedmiller and Braun, 1993) or the modified globally convergent version by Anastasiadis et al. (2005). The package allows flexible settings through custom-choice of error and activation function. Furthermore, the calculation of generalized weights (Intrator O & Intrator N, 1993) is implemented.


Version 1.44.2
  • Make compute() return the same as in previous versions
  • Soft-deprecate compute() for now
Version 1.44.1
  • Add support for dots in formula
  • Allow direct multiclass outcomes with factors
Version 1.44.0
  • New maintainer: Marvin N.Wright
  • Add a predict() function to replace compute()
  • Use Deriv package for symbolic derivation
  • Many internal changes to improve code quality
  • Bug fixes

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


1.44.2 by Marvin N. Wright, 3 years ago

Report a bug at

Browse source code at

Authors: Stefan Fritsch [aut] , Frauke Guenther [aut] , Marvin N. Wright [aut, cre] , Marc Suling [ctb] , Sebastian M. Mueller [ctb]

Documentation:   PDF Manual  

GPL (>= 2) license

Imports grid, MASS, grDevices, stats, utils, Deriv

Suggests testthat

Imported by EventDetectR, FSinR, FWRGB, LilRhino, Modeler, RSDA, SignacX, nnfor, regressoR, trackdem, traineR.

Depended on by MARSANNhybrid, quarrint.

Suggested by NeuralNetTools, NeuralSens, fscaret, innsight, mcboost, mlr, nnetpredint, plotmo, vip.

See at CRAN