ann_d_log_activ — derivative of logistic activation function
This function is the derivative of default neuronal activation function.
Any other, user defined, function should have the same input and output format for variables.
Particularly note that the derivative is expressed in terms of activation, not total neuronal input.
z Matrix containing the derivative of activation, one pattern per column.
For each pattern (column): z(i) = y(i) * [1 - y(i)] y Matrix containing the current neuronal activation, one pattern per column.
For each pattern: each y(i) for the corresponding i-th neuron (on the current layer).