Kind Sirs:
Some platforms create an Optimal Signal which shows the best buy and sell points by looking back at the historical data within the training set.
For training purposes, this Optimal Signal is given to the NN as the Desired Output, while of course, you give the NN any number of inputs. Sometimes, you can add some small percentage of noise to the Optimal Signal during training for robustness.
I am scouring the Manual but I can't seem to find how Zorro handles this matter of Optimal Signals (target output) for Neural Nets..
