zjukg / Ruleformer

[Paper][COLING2022] Ruleformer: Context-aware Rule Mining over Knowledge Graph
18 stars 2 forks source link

Data preprocessing #1

Closed MeycL closed 1 year ago

MeycL commented 1 year ago

Thank you very much for your work open source. I would like to ask about the data preprocessing there, for FB15K-237 dataset jump=1, for wn18rr dataset jump=2, for umls dataset jump=3, is the fact as I understand it? I am looking forward to your answer.

oier991215 commented 1 year ago

Hello, for the FB15K-237 dataset, what is the jump setting to reproduce the experimental results?

yep96 commented 1 year ago

Thanks for your interest in our work! The hyperparameter jump corresponds to the rule length T in the paper. For specific data, please refer to the T= in parentheses after the model name in the experimental results table.