Usage of cross entropy loss

Sorry, I meant multi-label classification.

Can you tell how can I define the accuracy function for above problem?
My label vector has ones at the classes which are there in the feature.