Compute performance measures from neural network

5 views (last 30 days)
How to compute
Coefficient of Determination - R^2
Root Mean Squared Error - RMSE
Mean Absolute Percentage Error - MAPE
from
net = train(net,X,T);
Y = net(X);

Accepted Answer

KSSV
KSSV on 8 Jun 2022
Edited: KSSV on 8 Jun 2022
net = train(net,X,T);
Yp = net(X); % predicted from Net and Let Yt be your true value
R = regression(Yp,Yt) ; % Regression
RMSE = sqrt(mean((Yt - Yp).^2)); % RMSE
% MAPE
pre_MAPE = abs((yp-Yt)./Yt);
MAPE = mean(pre_MAPE);
  4 Comments
KSSV
KSSV on 9 Jun 2022
Off course, yes. Why you got such a doubt?
Yesterday what I gave is a regression (R). For R square: refer: https://in.mathworks.com/help/matlab/data_analysis/linear-regression.html
Elysi Cochin
Elysi Cochin on 9 Jun 2022
I saw coefficient of correlation and coefficient of determination, so got a doubt, which the regression function computes. Thank you Sir for clearing

Sign in to comment.

More Answers (0)

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!