pyg-team/pytorch_geometric

HANConv link prediction example

Open

#4,221 opened on Mar 8, 2022

View on GitHub
聽(4 comments)聽(0 reactions)聽(0 assignees)Python聽(19,985 stars)聽(3,514 forks)batch import
1 - Priority P1documentationfeaturehelp wanted

Description

馃摎 Describe the documentation issue

There is an example for how to use HANConv to do node classification: https://github.com/pyg-team/pytorch_geometric/blob/master/examples/hetero/han_imdb.py, and we also have an example for how to do link prediction with heterogeneous graph: https://github.com/pyg-team/pytorch_geometric/blob/master/examples/hetero/hetero_link_pred.py.

However, we have to do different transformations in the two examples. It is tricky to figure out how to do link prediction with HANConv.

Suggest a potential alternative/fix

I would suggest to have an example showing how to do link prediction with HANConv, especially how to do T.RandomLinkSplit and T.AddMetaPaths together.

Contributor guide

Tech stack
python
Domain
documentationmachine learning
Issue type
documentation
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
3-5 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
fresh
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
clear
Prerequisites
Understand HANConvRandomLinkSplitAddMetaPathsPyTorch Geometric basics
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
75
Research direction
Study the existing examples for HANConv node classification and heterogeneous link prediction. Understand how to apply T.RandomLinkSplit and T.AddMetaPaths together in a single workflow. Create a new example file in the examples/hetero directory that demonstrates link prediction using HANConv, ensuring the transformations are correctly sequenced. Refer to issue comments for any additional guidance.
HANConv link prediction example 路 pyg-team/pytorch_geometric#4221 | Good First Issue