Caffe's second "top" of `"Accuracy"` layer

Question

Looking at the code of "Accuracy" layer, I see there is an option for a second output/"top" for this layer.
What does this second output produce?

Answer 1

Looking at accuracy_layer.hpp, where the number of outputs for the layer are defined, there's this comment:

// If there are two top blobs, then the second blob will contain
// accuracies per class.

So, the second "top" of the "Accuracy" layer simply reports per-class accuracies.

Just as a side note for layer Accuracy, the reported Accuracy is normalized by the number of "valid" predictions (same as using normalization: VALID for loss layers).