I have input and output data set for "ANFIS modeling in MATLAB", and I am getting some negative predicted values of output in testing. However, the predicted values of output in training are perfectly closer to the observed values. How to solve this issue. Thank you in advance.
Niaz Muhammad Shahani
I think there is a thread on this here
https://www.researchgate.net/post/Prevent_anfis_from_negative_prediction
Dear Niaz Muhammad Shahani,
First you have to check the consistency of the input and output data, then adapt the parameters relating to the ANFIS algorithm implemented in the "Neurofuzzy" toolbox. Also you can normalized your data, we can thus avoid negative values.
In addition you can use Microsfot Azur to normalize the data or use max-min normalization for data input, here is the link :
https://msdn.microsoft.com/fr-fr/library/azure/dn905838.aspx
Best regards
- Badges
- Science topic