SVM Model not working properly on test data

33 views (last 30 days)
Adam Hasse
Adam Hasse on 8 Jan 2018
Commented: G.N. Cobbold on 14 Dec 2020
Hello all, I'm using fitcsvm to classify two sets of data, Class 1 and Class 2. When I train the model on a dataset and then check how it did using predict, it seems to be working perfectly. However, when I input a new dataset, it only ever classifies it in Class 2, even though I use the cost modifier to state that classifying Class 1 in Class 2 is three times worse than misclassifying Class 2 as a Class 1.
SVM Model:
SVMModel = fitcsvm(data, classLabel, 'KernelFunction', 'gaussian',...
'Standardize',true, 'ClassNames', {'Class 1','Class 2'},...
'Cost',[0 3; 1 0]);
Verification of Training set:
[svmLabel, score] = predict(SVMModel, data);
Test Set:
[newsvmLabel, score] = predict(SVMModel, testData);
I think I'm using it correctly, so any advice on what's going on would be much appreciated.
  1 Comment
the cyclist
the cyclist on 8 Jan 2018
Can you upload the data, or a small sample that exhibits the problem? That way we can run your code and see for ourselves.

Sign in to comment.

Answers (3)

Bernhard Suhm
Bernhard Suhm on 23 Feb 2018
You are applying the predict function correctly, so there must be something wrong with your testData, probably the type in one column doesn't match the training data. Without providing an example - as "the cyclist" request - impossible to help you further.

Roushanak Haji Hassani
Roushanak Haji Hassani on 1 Oct 2018
did you solve the issue? I have a similar problem.
  1 Comment
Bernhard Suhm
Bernhard Suhm on 1 Oct 2018
Without the data we are just guessing at what the problem may be. Actually you may need to look at your training data as well, maybe that's highly imbalanced and the cost function didn't work.

Sign in to comment.

Roushanak Haji Hassani
Roushanak Haji Hassani on 1 Oct 2018
Edited: Roushanak Haji Hassani on 1 Oct 2018
I have attached the train data set and test here. The first column in the train dataset is the labels and the others are extracted features. I have applied Linear SVM classifier and I got 92.5 % accuracy on train data. But when I am applying on the test data the result is not good at all. I would appreciate if you let me know what is wrong with data.
G.N. Cobbold
G.N. Cobbold on 14 Dec 2020
Hi Salwa
Did you solve the issue? I have a similar problem

Sign in to comment.

Community Treasure Hunt

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

Start Hunting!

Translated by