Thank you for sharing the code. In your paper, Rel and Rel+ are reported as metrics. I read an issue about relation F1. I think that for a standard reporting considering the overall performance(i.e., the performance of both NER and RE), e2e_f1 should be reported instead of f1. Is that correct?
Thank you for sharing the code. In your paper, Rel and Rel+ are reported as metrics. I read an issue about relation F1. I think that for a standard reporting considering the overall performance(i.e., the performance of both NER and RE), e2e_f1 should be reported instead of f1. Is that correct?