how to set goal in training parameter in nntool?

network type- feed forward back propagation training function- trainlm adapting learning function- learngdm

Answer 1

[Jayanti]

Hi Santosh,

From MATLAB R2022a "nntool" has been removed. Instead "nnstart" provides graphical interfaces that allow you to design and deploy fitting, pattern recognition, clustering, and time-series neural networks.

But as per MATLAB R2025a, I cannot find any direct way to set the goal through the GUI interface. But you can set the "goal" manually by generating the training code and modifying it:

Refer to the below steps for more details:

• In the app, click "Generate Code" in the export section of toolstrip.

• This opens a MATLAB script with all the network creation and training steps.
• In that script, add below code to set "goal" before training the network.

% Set your desired goal value
net.trainParam.goal = 0.01;  

You can also refer to the below link to check the version history of "nntool":

https://www.mathworks.com/help/deeplearning/ref/nntool.html#:~:text=networks%20and%20data.-,Version%20History,-Introduced%20before%20R2006a