Why need model updating (training) in test function

xwhan / DeepPath

code and docs for my EMNLP paper "DeepPath: A Reinforcement Learning Method for Knowledge Graph Reasoning"

533 stars 135 forks source link

Open zhaohan-xi opened 2 years ago

zhaohan-xi commented 2 years ago

Hi Dear Author, can I ask why there is a model parameter updating during testing (here)