a set of C++ library classes
for neural networks development



Dependence among outputs and output errors

Classes for evaluating the dependence of output and output errors of multi input multi ouptut (MIMO) learning machines
o MutualInfo2
Class for evaluating mutual information between two random variables
Random variables are discretized dividing their range values in bins.
o MIError2
Class for evaluating the correlation between two random variables representing errors
Random variables are discretized dividing their range values in bins.
o MutualMatrix
Class for building mutual information among all the couples of a set of random variables
This class builds matrices of mutual information, joint entropy, and mutual information error (if the random variables represent errors).
o OUTERR_EXT
Suffix of the output file of errors
o Outerr
Class for generating error files of the outputs of the ecoc nets
It reads files out of the outputs of an ECOC-MLP or file pout od the PND and outputs the error file computing the absolute error between the outputs of the net and the correct output.
Classes for evaluating the dependence of output and output errors of multi input multi ouptut (MIMO) learning machines

Alphabetic index Hierarchy of classes


Last Updated February 2001
For comments and suggestions mail to Giorgio Valentini