Consider a scenario where you are provided with KnownLabel Matrix and PredictedLabel matrix. I would like to measure the goodness of the PredictedLabel matrix against the KnownLabel Matrix. But the challenge here is that KnownLabel Matrix have few rows only one 1 and other few rows have many 1's (those instance are multi labeled). An example of KnownLabel Matrix is given below. A =[1 0 0 0 0 1 0 0
![What are the measure for accuracy of multilabel data?](https://cdn-ak-scissors.b.st-hatena.com/image/square/efe2a4b85715116c366abc9aba8c10b9f28dffb9/height=288;version=1;width=512/https%3A%2F%2Fcdn.sstatic.net%2FSites%2Fstats%2FImg%2Fapple-touch-icon%402.png%3Fv%3D344f57aa10cc)