Lightning-AI / Lightning-AI/pytorch-lightning

Help PyG to support convertig homogenous LightningModule GNNs into heterogenous ones

Open
#21,629 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

feature
Dominant language
Python
Stars
31.4k
Forks
3.8k
Avg merge
6d 7h
Merged PRs (30d)
6

Description

Description & Motivation

Currentlly it's not possible to use the internal to_hetero() conversion from PyG to turn a homogenous GNN into a heterogenous one, if the GNN is based on LightningModule .

If the PyG team decides it's worth supporting Lightning in that case, probably your help would be appreciated in that matter.

Pitch

No response

Alternatives

Provide an own wrapper utility to turn the GraphModule returned by PyG's to_hetero() back into a LightningModule.

Additional context

PyG issue: https://github.com/pyg-team/pytorch_geometric/issues/10653

cc @lantiga

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reading the linked PyG issue 10653 and the interaction between LightningModule-based GNNs and PyG's internal to_hetero() conversion. A complete contribution would need an agreed approach and documented support or wrapper behavior, with validation that homogeneous LightningModule GNNs can be converted to heterogeneous ones.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.