Xin Lv

I have the same issue. Do you find any solutions?

I have found the problems. These codes use a small trick [here](https://github.com/salesforce/MultiHopKG/blob/23747699aefe3a9f835f9875ce446a18de96dcb1/src/rl/graph_search/pn.py#L325). For a triple (h, r, t) in the dataset, this trick will mask some action with entity e_1,...

Yeah, `e_t` is not masked, but what if the action space do not contain `e_t`? Are all actions masked? I have print some vectors and find that only after function...

I have printed the tensor `sample_action_dist ` in [this code](https://github.com/salesforce/MultiHopKG/blob/23747699aefe3a9f835f9875ce446a18de96dcb1/src/rl/graph_search/pg.py#L191). I found that some rows in this tensor are all zeros, which can lead to `invalid multinomial distribution (sum of...

Yeah, I think it is OK :). The above trick only fails with a very small probability. And in most cases, it works fine.

Thank you for your interest in this work, I have helped others to run this code in the recent past and got good results. In order to get the correct...

在format_batch函数中，只有满足type(batch_e2[0]) is list或者type(batch_e1[0]) is list时才会调用，这种情况只出现在训练表示学习的模型阶段，不会涉及到pg.py代码。可以检查一下是不是错误的更改了config文件，建议参考README的教程运行代码。

我们这个项目一直有人维护更新，但是一个人的精力可能没办法覆盖到所有的ea相关的论文，欢迎大家来提issue。

To better identify the problem, could you tell me what dataset you are running on?

Could you give your PyTorch version? I redownload and run the code without encountering any errors. Using FB15K-237-20% as an example, make sure you run the following commands in order:...