Why is the 10% of neural/sim output totally crazy?
Info
This question is closed. Reopen it to edit or answer.
Show older comments
I train my NN using the default train and newff functions with a 834X5 input matrix and 85 hidden neurons.
When I try use sim so as to forward test my data,around 10% is far beyond the targetted range.
Why is this? Should I ignore these outputs as outliers?
thanks
Answers (0)
This question is closed.
Closed:
on 20 Aug 2021
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
