two questions about code

[qpisuo]

Dear Author,

First of all, I would like to commend you on your excellent work. I have a few questions while using your code:

1. I noticed that in the command line arguments, the parameter alpha controls the weight of the skip-gram loss. If it is set to 0, does it mean that the skip-gram pretraining is not used? If so, does the model still obtain initial embeddings in this case?

2. Could you please clarify the purpose of add_padding_idx? Is it designed for non-uniform hypergraphs? For example, my hypergraph contains both 2-edge and 3-edge relations.

Looking forward to your reply. Thank you!

[zhangwen123123]

你好，你复现出来这两个值有这么小吗auc: 0.499, aupr: 0.166 还是我哪个环节出错了