dmlc / dmlc/dgl

How to consider edge features inside a heterogeneous graph? Or how to use EdgeGATConv() in HeteroGraphConv()?

Open
#7,872 2 comments 0 reactions 0 assignees View on GitHub
stale-issue
Dominant language
Python
Stars
14.3k
Forks
3.1k
PR merge metrics
No merged PRs in 30d

Description

## ❓ Questions and Help

Before proceeding, please note that we recommend
using our discussion forum (https://discuss.dgl.ai) for
general questions. As a result, this issue will
likely be CLOSED shortly.

I am now considering a heterogeneous graph containing 3 types of nodes, each connected to each other. I consider using the HeteroGraphConv() function to implement message passing, but find that the HeteroGraphConv() function is unable to input edge features.

I want to use EdgeGATConv as a sub-module of HeteroGraphConv(), but I don't understand the usage of mod_args and mod_kwargs, how can I properly use EdgeGATConv in HeteroGraphConv()?
Or, is there any other way to implement message passing in heterogeneous graphs that takes into account edge features?
Below is my current HeteroGraphConv() construction and I don't know how to construct the **forward** function.

Contributor guide

No contributing guide indexed for this repository

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.