how to assign the output of trained Pattern Recognition Neural Network to a class for each instant

1 view (last 30 days)
muhammad usama
muhammad usama on 19 Nov 2021
Commented: muhammad usama on 26 Nov 2021
i have a trained pattern recognition neural network with 8 outputs ranges under [0 1] (e.g for instant [0,0.2,0.25,0.35,0.02,0.08,0.1,0] with the sum equal to 1 at the each (1e-6)). Now i would like to assign the max value of the array to a specified class and rest value of the array to be zero for each instant. what is the best posible way to do it. Your help will be appreciated
Best regards

Answers (1)

yanqi liu
yanqi liu on 26 Nov 2021
sir,may be find the max index to get the result,such as
% assign the max value of the array to a specified class and rest value of the array to be zero for each insta
p = net(data);
% such as
% p = [0,0.2,0.25,0.35,0.02,0.08,0.1,0];
[~, index] = max(p);
classes = {'A', 'B', 'C','D','E','F','G', 'H'};
result = classes{index}
  1 Comment
muhammad usama
muhammad usama on 26 Nov 2021
Thanks for the response. It’s much appreciated. I solve the problem by just using the [m,ind]=max(data from NN) And then using the IF functions to assign the class.

Sign in to comment.

Community Treasure Hunt

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

Start Hunting!