split data in train and test sets in a very specific way

Hey guys!

I am trying to design a neural network (via fitcnet) to classify subjects into three classes. There is a txt file that consists of 1092 rows and 14 collums (1092 x 14). The 14th collumn has the class of the sets either A, B or C and I specificly want the first 450 x 14 of them to be the training set and the rest of them to be the test set. Is there any way to do that?

Thank you in advance!

Answer 1

[Cris LaPierre]

Follow this example: https://www.mathworks.com/help/stats/fitcnet.html#mw_b102d938-6d2a-4c06-a485-f743363fb79a

The change to make, then, is that you manually create your partition instead of using cvpartition.

dataTrain = mydata(1:450,:);
dataTest = mydata(451:end,:);

[AIKATERINI NINOU]

Omg thank you, it was easier than I thought !

What if i also want to manually creat a vallidation data set. For example the first 450 rows as training set, the next 50 rows as vallidation data set and the remainings as a testset. Is it based on the same reasoning?

[Cris LaPierre]

Follow the same process. If you need help with indexing an array, see Ch 5 of MATLAB Onramp.

[AIKATERINI NINOU]

thank you Cris!