In the training process, loss decreases ，but recall remains unchanged - Githubissues

xiangwang1223 / neural_graph_collaborative_filtering

Neural Graph Collaborative Filtering, SIGIR2019

MIT License

781 stars 261 forks source link

In the training process, loss decreases ，but recall remains unchanged #48

Open songiu opened 3 years ago