MATLAB Answers

Message: "You cannot fit posterior probabilities when all scores are NaNs

1 view (last 30 days)
Uerm
Uerm on 13 Feb 2020
Hi,
I have attached a code where I use several classifiers with several different parameters. When I try to run the code with a feature matrix as input, I get the following error:
Error using fitSVMPosterior (line 155)
You cannot fit posterior probabilities when all scores are NaNs.
Error in ClassificationSVM/fitPosterior (line 602)
[obj,trans] = fitSVMPosterior(obj,varargin{:});
Error in trainClassifier128ADASYN (line 154)
[~,scoresLinSVM128ADASYN] = resubPredict(fitPosterior(classificationLinearSVM128ADASYN));
I have never had issues when I ran the code previously and I have tried it a lot of times. I have only had this problem now.
Can anyone guess what the problem is?
Thanks!

Answers (0)

Community Treasure Hunt

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

Start Hunting!