About the value of spread of GRNN
4 views (last 30 days)
Show older comments
Hi, everyone! I encountered some problem when using GRNN in neural network tool box. Spread is the only variable need to be determined in GRNN. I have tried different values for spread, the curve fitting capability for training data is good, but the interpolation for regions where training data is unavailiable is very poor. How can I improve the interpolation capability by using GRNN? Thanks!
2 Comments
Greg Heath
on 10 Aug 2013
Not enough information:
size(input)?
size(target)?
How are input and target scaled?
What range of spread did you use?
Accepted Answer
Greg Heath
on 10 Aug 2013
See my previous answer.
Thank you for formally accepting my answer
0 Comments
More Answers (0)
See Also
Categories
Find more on Deep Learning Toolbox 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!