Hi, every one.
I’m reading the docs in PyTorch vision but what is the meaning of Top-1 error and Top-5 error present in this page.
You would increase the Top-1 error if the “top-1” prediction, i.e. the prediction of your model with the highest score, does not match the target.
On the other hand, you would increase the Top-5 error if none of the “top-5” predictions, i.e. the 5 predictions with the highest score, matches the target.