The code for our ICML 2022 paper Cycle Representation Learning for Inductive Relation Prediction

train.py: train the model and evaluate using the AUC-PR score

test_ranking.py: evaluate the model with Hits@10 score.

change the settings in the main function of train.py and test_ranking.py to adjust new datasets.

The data are available in https://github.com/kkteru/grail