| Neural Network Toolbox | Search Help Desk |
| dmae | Examples See Also |
Mean absolute error performance derivative function
dPerf_dE = dmae('e',E,X,perf,PP)
dPerf_dX = dmae('x',E,X,perf,PP)
dmae is the derivative function for mae.
dmae('d',E,X,PERF,PP) takes these arguments,
E - Matrix or cell array of error vector(s).
X - Vector of all weight and bias values.
perf - Network performance (ignored).
PP - Performance parameters (ignored).
dPerf/dE.
dmae('x',E,X,PERF,PP) returns the derivative dPerf/dX.
Here we define E and X for a network with one 3-element output and six weight and bias values.
E = {[1; -2; 0.5]};
X = [0; 0.2; -2.2; 4.1; 0.1; -0.2];
Here we calculate the network's mean absolute error performance, and derivatives of performance.
perf = mae(E)
dPerf_dE = dmae('e',E,X)
dPerf_dX = dmae('x',E,X)
Note that mae can be called with only one argument and dmae with only three arguments because the other arguments are ignored. The other arguments exist so that mae and dmae conform to standard performance function argument lists.
mae