Interpretable Evaluation for (Almost) All NLP Tasks - View it on GitHub