MATLAB: When to use mapstd, mapminmax

Deep Learning Toolboxmapminmaxmapstdnormalizationstandardization

Mapstd is used for standardization, mapminmax is for normalization. But When to use mapstd and mapminmax? I want to do a regression(function fitting) with neural network. While I am creating the nn with nftool, it normalize the input data and target(mapminmax) automatically. So I was wondered that.

Best Answer

I prefer standardization for 2 reasons
 1. Easier to detect outliers for removal or modification
 2. Typically, 0<= MSE <=1 so I use the MSE training goals 0.01 
(regression/pattern-recognition) and 0.005 (time-series).
HOWEVER,
 MATLAB uses MAPMINMAX as a default. Since it is too much a pain 
in the buttocks to replace it with MAPSTD, I just accept it but 
use ZSCORE standardization before training to check for outliers.
I have posted zillions of examples in both NEWSGROUP and ANSWERS.
Hope this helps.
Thank you for formally accepting my answer
Greg
Related Question