I would like to ask for help regarding my model.

I have a dataset of preprocessed images and I performed a binary classification with SVM on Python. I tuned the value of the c parameter from 0.001 to 1 with 10-fold CV and I obtained the following results.

enter image description here

My questions are:

  • Does this mean that the optimal value of C is 0.001?

  • The rest means that there is overfitting?

The first value is not 0.000 but 0.001

Thank you.


Essentially, you only want to look at your test accuracy. The training accuracy is only relevant in comparison with the test accuracy. In your situation, it seems that your test accuracy is roughly the same across the board beside a slight peak around 0.620. So I would pick this as the best value for C.

My answer is related to the one I gave on this post, here is a summary:

  • Overfitting is not when your train accuracy is really high (or even 100%). It is when your train accuracy increases and your test accuracy decreases is low.

  • it is not abnormal that your train accuracy is higher than your test accuracy. After all, your model has an advantage with the train set since it's been given the correct answer already.

  • $\begingroup$ Thank you very much for your help. $\endgroup$ May 29 '20 at 12:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.