I am working on a binary classification problem and the training results look somewhat
The blue curve is validation set while the red is the training set.
I wanted to ask few things:-
- Initially the training score is less than the validation score. My guess is that since the validation set has a size less than the training, it may be because of that. But I am not entirely sure about that.
- The training set does not reach 100% score. Is this an expected behaviour? I thought training set would reach 100% accuracy soon but as you can see, that is not happening.
- Is the graph showing that I should stop the training now? I have heard about the idea of early stopping but I want to know if the graph is showing that it has reached the point already.
PS: I was not able to post the score curve as I was allowed to post only one image. However, I will try to put that in the comment