Range for neural network random weights
Topics
surrogate_models
Specification
Alias: None
Arguments: REAL
Description
Controls the range of the input layer random weights in the neural network model. The default range is 2.0, resulting in weights in (-1, 1). These weights are applied after the training inputs have been scaled into [-0.8, 0.8].
© 2022 National Technology and Engineering Solutions of Sandia, LLC. | Questions & Comments | Privacy & Security
Sandia National Laboratories is a multimission laboratory managed and operated by National Technology and Engineering Solutions of Sandia, LLC., a wholly owned subsidiary of Honeywell International, Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA-0003525.