Tianyu Zhao

> 提交文件是预测结果文件的压缩包，提交格式可以参考：https://www.biendata.xyz/competition/hgb-1/evaluation/ 每次提交四个数据集，一共二十个文件的话，这二十个文件的score都会有吗，还是一个数据集一个score，或者这四个数据集只有一个score

It is a bug, which is fixed in https://github.com/BUPT-GAMMA/OpenHGNN/commit/f8067afc92443a994f12926a70b9c354171a1cfc. However, it is still hard to use to plot ranking and distribution. We will give some details demos in the future.

In a way, it's a bug. `self.hidden_dim = self.h_dim * self.num_heads`. Maybe we should remove one of the parameters to fix the bug. Or we move the code L232 after...

Hello, First, the model `DeepWalk` is a homogeneous GNN algorithm, which will not be included in OpenHGNN. The reason why DeepWalk appears in `trainer` is that `metapath2vec` use sthe trainer...

Good question! 1. It seems that the experiments of `mp2vec` only use one meta-path. 2. In section 4.2.2 `Setting the Fusion Function` of `herec`, it offers three functions used in...

For now, we just offer [a demo to support custom dataset](https://openhgnn.readthedocs.io/en/latest/install/usage.html#evaluate-a-new-dataset). Besides, we are developing an easier API to support custom dataset, like hongyi said.