Can classifier learner app, take input for each samples as a matrix?

Hi there

I know How to use MATLAB classifer learner app. Now I have modified my data in sucha way that, each sample is transformed as 10XF where F is the number of features of predictors as said in documentation. As svm or any classifier in this app take input as a matrix or table, where each data samples has single row in matrix, But my data is modified as above can I use classifier learner (SVM) app here?

Regards

Answer 1

[Omega]

Hi Krishna,

The MATLAB Classifier Learner app is designed to take input data where each sample is represented as a single row in a matrix or table. It does not support samples where each observation is itself a matrix (like 10xF per sample). Each sample must be a 1xN vector of features.

To use the Classifier Learner app, you would need to reshape or preprocess your data so that each sample is represented as a single row vector. For example, you could flatten each 10xF matrix into a 1x(10*F) vector. This way, each sample will have a single row in the input matrix, which the app can handle. Alternatively, you could also extract summary statistics (mean, standard deviation, etc.) from the 10xF matrix to get a fixed-length feature vector.

I hope it helps!