pyg-team/pytorch_geometric
View on GitHubCustomizing Aggregations within Message Passing
Open
#7901 opened on Aug 18, 2023
0 comments (0 comments)0 reactions (0 reactions)0 assignees (0 assignees)Python19,985 stars (19,985 stars)3,514 forks (3,514 forks)batch import
documentationhelp wantednn
Description
This issue does not include a description.
Contributor guide
- Tech stack
- pythonpytorch
- Domain
- machine learning
- Issue type
- feature
- 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.
- 1-3 hours
- Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
- stale
- ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
- mostly clear
- Prerequisites
- None
- Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
- 20
- Research direction
- The issue requests the ability to customize aggregation functions in message passing layers. Contributors should review the existing message passing base class (e.g., torch geometric.nn.conv.MessagePassing) to understand current aggregation options (add, mean, max, etc.). A possible implementation could involve allowing users to pass custom aggregation functions as arguments to layers, similar to how aggregation is currently handled via 'aggr' parameter. Maintainers likely expect a discussion on design before implementation.