I have noticed that the HP Neural Node provides property "Target Standardization" (page 6.10 of course text): in what instances should a target variable be standardised?
I cannot generalize this. Always dependent on your data and the NN model structure and options. Because there is also None available as an option try with None as the baseline. Then evaluate other methods and pick the one suitable for your data.
Of course; would a standadisation of an interval target help improve prediction estimates, for instance, if the target had a highly skewed distribution?
I cannot generalize this. Always dependent on your data and the NN model structure and options. Because there is also None available as an option try with None as the baseline. Then evaluate other methods and pick the one suitable for your data.
This is a knowledge-sharing community for learners in the Academy. Find answers to your questions or post here for a reply. To ensure your success, use these getting-started resources: