microsoft / microsoft/Graphormer
[Feature request] Supports Sparse Graph Representation
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 2.5k
- Forks
- 374
- PR merge metrics
- No merged PRs in 30d
Description
Summary
In torch_geometric, the adjacent matrix is represented as sparse edge index (graph connectivity in COO format with shape). There are two potential advantages of sparse edge index: 1. less memory utilization; 2. graphs in one minibatch are represented by one edge_index tensor. Considering to be compatible.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No file, test, or entry point is named. Start by locating how Graphormer accepts graph connectivity and how batched edge_index tensors are represented, then compare that path with torch_geometric's sparse COO convention. Done means sparse graph inputs are accepted with reduced memory use while preserving minibatch compatibility, with tests covering the supported representation.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100