Denormalization output neural network
24 views (last 30 days)
Show older comments
I used the equation of the neural network ("y = b2 + LW * tansig (b1 + IW * x)") to an Excel, so you can calculate the output values of two input values and one target of a network of two neurons and one layer, values generated by the equation "y = b2 + LW * tansig (IW + b1 * x)" (all done in excel) are normalize, give me exactly equal to the coefficient of determination delivered by Matlab but it gives me different values between output Excel and output Matlab because one is normalize (excel).
Someone can write the equation for me to denormalize my output values, I have sought "mapminmax" command but I can not understand how denormalize (the variables of the equation). Is there a website where explain the command mapminmax , but can not understand how denormalize (to do it in excel).
I know the equation of normalize:
y = (ymax-ymin)*(x-xmin)/(xmax-xmin) + ymin; where :
ymax= max. value of specified range in this case = 1 ( which is default for mapminmax command)
ymin= min. value of specified range in this case=-1 ( also default value)
x= value to be scaled.
xmax= max. value of numbers to be scaled
xmin= min. value of numbers to be scaled
but i need to know the equation of denormalization to use in Excel. How dat use for max and min value in denormalization.
please help me
1 Comment
Greg Heath
on 10 Oct 2015
Edited: Greg Heath
on 10 Oct 2015
If you have questions on any command, use the commands help, doc and type. For example
help mapminmax
doc mapminmax
type mapminmax
Hope this helps.
Greg
Accepted Answer
Greg Heath
on 10 Oct 2015
SEARCH THE NEWSGROUP
GREG ANALYTIC TUTORIAL
Hope this helps.
Thank you for formally accepting my answer
Greg
2 Comments
More Answers (0)
See Also
Categories
Find more on Image Data Workflows 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!