Founding Engineer
Creator of PyG (PyTorch Geometric) - Founding Engineer @ kumo.ai - PhD from TU Dortmund University - Interested in Graph Representation Learning
- Dortmund, Germany
- https://rusty1s.github.io
- @rusty1s
Highlights
- Pro
Block or Report
Block or report rusty1s
Report abuse
Contact GitHub support about this user’s behavior. Learn more about reporting abuse.
Report abusePinned
-
-
snap-stanford/ogb Public
Benchmark datasets, data loaders, and evaluators for graph machine learning
-
-
pytorch_sparse Public
PyTorch Extension Library of Optimized Autograd Sparse Matrix Operations
-
pyg_autoscale Public
Implementation of "GNNAutoScale: Scalable and Expressive Graph Neural Networks via Historical Embeddings" in PyTorch
-
Implementation of "Deep Graph Matching Consensus" in PyTorch
4,981 contributions in the last year
Less
More
Activity overview
Contributed to
pyg-team/pytorch_geometric,
pyg-team/pyg-lib,
rusty1s/pytorch_spline_conv
and 15 other
repositories
Contribution activity
June 2023
Created 23 commits in 2 repositories
Created a pull request in pyg-team/pytorch_geometric that received 1 comment
Opened 9 other pull requests in 1 repository
pyg-team/pytorch_geometric
9
merged
- Disable graph break test
-
Add number of sampled nodes/edges information to
LinkLoader -
Added the
HMpersonalized fashion recommendation dataset -
Fail gracefully in case
GNNExplainerwill not compute gradients for node features or edges -
Fix
GCN2example fortorch.sparse_csrlayout - Remove some warnings
-
Fix
disable_dynamic_shapesdecorator in combination with TorchScript - Fix full tests on master
-
cudfsupport inmap_index
Reviewed 22 pull requests in 4 repositories
pyg-team/pytorch_geometric
18 pull requests
-
Enable
trim_to_layerwith hetero CSR flow -
Add
NodeEncoderfrom the GraphMixer paper - Fix typo in documentation
-
Introduce
disable_dynamic_shapesexperimental flag; adding its use intoto_dense_batch function -
Add option to override the
segmmheuristic decision inHeteroLinear -
Use
GraphSAGEmodel in Reddit example -
Use
NeighborLoaderinstead ofNeighborSamplerin Cluster-GCN example - Add conda build number 1 for republishing with Python 3.11
- Publish Pyg 2.3.1 conda packages
-
Fix
torch_geometric.utils.scatter_argmax -
Making
RevGNNexample work w/otorch-sparse -
Make
TGNusetorchnative whentorch-scatternot available -
Add
MovieLens-1Mheterogeneous dataset - Enable Python 3.11 conda packages
- Amazon book dataset
- [Benchmark] Add compile option in benchmark kernel, citation and points
-
Add
LocalDatasetfor distributed training -
Move the
scalertensor inGeneralConvto the correct device
rusty1s/pytorch_scatter
2 pull requests
rusty1s/pytorch_sparse
1 pull request
pytorch/pytorch
1 pull request
Answered 1 discussion in 1 repository
pyg-team/pytorch_geometric
pyg-team/pytorch_geometric
-
DataLoader shuffle affects loss and accuracy on evaluation
This contribution was made on Jun 1
20
contributions
in private repositories
Jun 1 – Jun 5






