How to see confusion matrix , training curve and ROC plot in newrb MATLAB?
3 views (last 30 days)
Show older comments
I am using classification using newrb function matlab (nntoolbox).When I train my network , it is showing training MSE vs no of neurons curve as output result. I want to see other results like confusion matrix , training curve and ROC plot in this function. I already got the result using patternnet but still can get all these results in newrb function. input dataset :70X186 target dataset : 2X186. If anybody has any idea , please let me know. Thanks a lot.
0 Comments
Accepted Answer
Greg Heath
on 7 Nov 2015
Edited: Greg Heath
on 7 Nov 2015
1. NEWRB does not have
a. Designer weight initialization
b. Datadivision or validation stop options.
c. Non-equal RADBAS spreads
Therefore, it would probably be more productive to just use PATTERNNET with RADBAS hidden nodes.
However, it might be possible to somehow use NEWRB RADBAS functions in PATTERNET (e.g., for weight initialization).
Hope this gives you food for thought.
Greg
0 Comments
More Answers (0)
See Also
Categories
Find more on Define Shallow Neural Network Architectures in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!