How to converting cell to matrix for image based classification?

4 views (last 30 days)
To train my data using SVM classifier, I am using 'fitcecoc' function, the input argument it takes must be matrix or table.
I am trying basic classification using only images not doing any feature extractiion. The feature vector that I created consist of nested cell which I am not able to convert into matrix. Please help
  2 Comments
KALYAN ACHARJYA
KALYAN ACHARJYA on 31 Aug 2021
Edited: KALYAN ACHARJYA on 31 Aug 2021
Have you tried with cell2mat of train vector, then apply the same?
Nikhat Ali
Nikhat Ali on 31 Aug 2021
Yes I tried using that, it converts the cell to matrix. But still having error as training label and train vector not have same size.

Sign in to comment.

Accepted Answer

Robert U
Robert U on 1 Sep 2021
Hi Nikhat Ali,
Please have a look into documentation fitcecoc(). The way you call the function is not working since data format and data dimensions is not correct.
"trainvector" must be data type "double" or "single". Plus, the matrix is supposed to be sparse or full, and the length of "trainingLabels" must correspond to the number of observations.
I suppose, neither data type nor number of observations fit.
Kind regards,
Robert

More Answers (0)

Products


Release

R2021a

Community Treasure Hunt

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

Start Hunting!