xwhan / DeepPath

code and docs for my EMNLP paper "DeepPath: A Reinforcement Learning Method for Knowledge Graph Reasoning"
533 stars 135 forks source link

Fix bug on fact_prediction_eval.py (line 315) #11

Closed jimmywangheng closed 5 years ago

jimmywangheng commented 5 years ago

Line 315, evaluating on transD, should be rel_vec_D, not rel_vec