Hacker News new | ask | show | jobs
by muzakthings 3383 days ago
You realize that softmax scores aren't probabilities, right?

It's just a relative measure of confidence, scaled such that they all sum to 1.0.

2 comments

That's kinda true, but (regularization aside) for standard loss functions it's minimized at the point it's well calibrated, right? Given the scores in the image (97% animal, 90% tiger, etc) they seem to be binary classifiers e.g. "is this a tiger?" So of all scores in the neighborhood of 90%, 90% should be "yes it is," making it a measure of confidence compatible with probability.

Please someone correct me if I'm wrong, but I'm pretty sure that's how it works, just like how logistic regression gives you a probability.

You can't add any of the numbers in the picture to equal 1.0 (or 100)
What you need to do is to take the top prediction and see how accurate it is compared to a test set. The scores on the picture represent confidence not accuracy.