Reproduce Performance Metrics

[berrahmadi]

Is a script to compute the performancee metrics reported in the paper (i.e.MRR, Hit@1, and Hit@3) is missing? Can you please clarify, thank you!

[berrahmadi]

I found some related code in model.py, method evaluate_generalization.

This method is called in the main methods of the respective models. I have tried to run the main method in one of the models' scripts, i.e. q2p.py, but these embeddings files _eventuality_embeddings.pkl and _relation_embeddings.pkl are missing (and they are not generated by any code in this repo). Do you still have them? Otherwise, how did you generate them?