5 ms·
what's the state of GNN support elsewhere? does everyone else also roll their own, or are folks using Pytorch or something else?
by claytonjy 5y ago
what's the state of GNN support elsewhere? does everyone else also roll their own, or are folks using Pytorch or something else?
- rytill 5y agoThis is a big one: https://github.com/pyg-team/pytorch_geometric https://github.com/pyg-team/pytorch_geometric
- patagurbon 5y agoDGL is the other big one, it supports several frameworks (at least PyTorch and MXNet).
- UncleOxidant 5y agoGeometricFlux for GNNs in Julia: https://github.com/FluxML/GeometricFlux.jl https://github.com/FluxML/GeometricFlux.jl
- agentofoblivion 5y agoDeep Graph Library (DGL) is the big one, which can use either PyTorch, MXNet or Tensorflow as the backend and is developed by AWS. You also have PyTorch Geometric and Jraph, which is built on top of JAX and used mostly by researchers at DeepMind as far as I can tell.
- thecleaner 5y agoThe bottleneck in GNN computations is that the aggregation ops cant be expressed as matrix operations and require writing custom kernels. This problem was solved in PyTorch with torch-scatter. The other bottleneck is subsampling (e.g k-hop) which also dont benefit from GPU support. Other than that the embedding aspects can just be written as nn ops.