pyg-team/pytorch_geometric

Weighted aggregations `torch_geometric.nn.aggr.Aggregation`

Open

#6,867 opened on Mar 7, 2023

View on GitHub
 (10 comments) (0 reactions) (0 assignees)Python (3,514 forks)batch import
1 - Priority P1featurehelp wantednn

Repository metrics

Stars
 (19,985 stars)
PR merge metrics
 (Avg merge 35d 1h) (14 merged PRs in 30d)

Description

🚀 The feature, motivation and pitch

The current API of torch_geometric.nn.aggr.Aggregation seems to be unnecessarily limited with respect to edge weights. At least some architectures introduce edge weights, with GCN being the most prominent one. For mean aggregation, it is simple to work around this with a custom message passing / convolution that multiplies features with edge weights. However, for most other weighted aggregations (e.g. median, quantiles, standard deviation, ...) it is not.

Alternatives

Add edge weights to aggregations API

Additional context

No response

Contributor guide

Research direction
Investigate the existing Aggregation API and propose a design for weighted aggregations by adding an edge weight parameter or a separate weighted aggregation class. Look at how GCN handles edge weights and extend to other aggregations like median, quantiles, etc.
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.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
PythonPyTorchGNN basics
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
65