has anyone tried Mathematica's Neural network tool? Do you rate if for fast throughput for training? I can find this study so far: https://grey.colorado.edu/emergent/index.php/Comparison_of_Neural_Network_Simulators
Perhaps a pool of researchers in this area can pitch in together and can keep a 'living' catalogue... I am happy to put to do my bit..
Open Source alternatives can be found in the R system for statistical computation, e.g. packages "nnet" and "RSNNS". Package nnet is implemented in C and RSNNS in C++, both with an R interface. More, see:
Package nnet:
Venables, W. N. & Ripley, B. D. (2002) Modern Applied Statistics with S. Fourth Edition. Springer, New York. ISBN 0-387-95457-0
http://cran.r-project.org/web/packages/nnet/
Christoph Bergmeir, Jose M. Benitez (2012). Neural Networks in R Using the Stuttgart Neural Network Simulator: RSNNS. Journal of Statistical Software, 46(7), 1-26. http://www.jstatsoft.org/v46/i07/
http://cran.r-project.org/web/packages/RSNNS/
... but there are even more packages, see the CRAN task view about machine learning:
Thankyou. there seem to be too many options. I have chosen to work with R as I have used it before and it also features classical regression to compare against. Here is a worked example for those interested: http://gekkoquant.com/2012/05/26/neural-networks-with-r-simple-example/
https://mandymejia.wordpress.com/2014/08/18/three-ways-to-use-matlab-from-r/ because total migration is a headache and looks like nowadays everyone needs to have a few bridges in between somewhat suitable software environments...
I would recommend R, using the RStudio environment. There are lots of packages and good online support/forums. Try RSNNS for various machine learning tools.
Or, try the Anaconda Python distribution with includes SciKit-Learn. If you can get Anaconda Accelerate you can use parallel processing / GPU for faster processing.