If I have two train cycles, but only want to train a certain parameter based on the previous cycle optif, how could that be done, since in train optif is always reset to 1.

In other words, I only want to train exit parameters (SL) in case the previous cycle without SL did return optif > 1.


What is the goal, prevent overfitting by training and don't make unprofitable traincycles / parametervalues profitable by the exit only.

Maybe in better words, traincycle 1 without stoploss finds from 20 parameters 15 profitable, traincycle 2 should be possible to train the SL only for the previously found profitable ones (and omit the the 5 unprofitbale ones, to prevent making them profitable by SL only).