Prediction of target data new input value that is not present in input data set, after training a model in nftool

After training a model in Neural Net Fitting tool in Matlab, how do I predict the target value for the new input data value that is not in the input data set that we trained in nftool. I am sharing the trained model and new input sample data in .mat format.

Answer 1

[Rahul]

I have no information whether you are performing classification or regression. It is also ambiguous that what type of training data you have and how many classes you have and what is the size of the feature set. However, please find the below lines of codes to help with your query.

load('s.mat') % load the sample data to test.
load('r1.mat') % load the trained model and training results.
net = r1.Network; % assigning a variable to the trained model.
test_out = net(sample) % test the result using trained model 'net' for the unknown test data 'sample'

You can also refer below examples given in the documentation page.

1. Fit Data with a Shallow Neural Network - MATLAB & Simulink - MathWorks India

2. Body Fat Estimation - MATLAB & Simulink - MathWorks India

[Mandru Vamsi Venkata Krishna]

Thank you

I got the answer and I got to know where I have made the mistake