We are excited to announce the release of PyG 2.7 🎉🎉🎉

PyG 2.7 is the culmination of work from 53 contributors who have worked on features and bug-fixes for a total of over 282 commits since torch-geometric==2.6.0.

Highlights

PyTorch 2.8 Support

PyG 2.7 is fully compatible with PyTorch 2.8 and supports the following combinations:

| PyTorch 2.8 | cpu | cu126 | cu128 | cu129 | |--------------|-------|---------|---------|---------| | Linux | ✅ | ✅ | ✅ | ✅ | | Windows | ✅ | ✅ | ✅ | ✅ | | macOS | ✅ | | | |

In addition, PyG 2.7 supports two previous PyTorch minor releases, PyTorch 2.7 and 2.6:

| PyTorch 2.7 | cpu | cu118 | cu126 | cu128 | |--------------|-------|---------|---------|---------| | Linux | ✅ | ✅ | ✅ | ✅ | | Windows | ✅ | ✅ | ✅ | ✅ | | macOS | ✅ | | | |

| PyTorch 2.6 | cpu | cu118 | cu124 | cu126 | |--------------|-------|---------|---------|---------| | Linux | ✅ | ✅ | ✅ | ✅ | | Windows | ✅ | ✅ | ✅ | ✅ | | macOS | ✅ | | | |

Breaking Changes

Dropped support for Python 3.9 (#10461)
Dropped support for PyTorch 1.11 - 2.5 (#10500, #10248, #10247)

Deprecation

Deprecated torch_geometric.distributed (#10411)

Bugfixes

Fixed ogbn_train_cugraph example for distributed cuGraph (#10439)
Added safe_onnx_export function with workarounds for onnx_ir.serde.SerdeError issues in ONNX export (#10422)
Fixed importing PyTorch Lightning in torch_geometric.graphgym and torch_geometric.data.lightning when using lightning instead of pytorch-lightning (#10404, #10417))
Fixed detach() warnings in example scripts involving tensor conversions (#10357)
Fixed non-tuple indexing to resolve PyTorch deprecation warning (#10389)
Fixed conversion to/from cuGraph graph objects by ensuring cudf column names are correctly specified (#10343)

Features

Added llm generated explanations to TAGDataset (#9918)
Added torch_geometric.llm and its examples (#10436)
Added support for negative weights in sparse_cross_entropy (#10432)
Added connected_components() method to Data and HeterData (#10388)
Added LPFormer Graph Transformer for Link Prediction (#9956)
Added BidirectionalSampler, which samples both forwards and backwards on graph edges (#10126)
Enable Sampling both forwards and reverse edges on NeighborSampler (#10126)
Added ability to merge together SamplerOutput objects ()

Changes

Adapt dgcnn_classification example to work with ModelNet and MedShapeNet Datasets (#9823)
Chained exceptions explicitly instead of implicitly (#10242)
Updated cuGraph examples to use buffered sampling which keeps data in memory and is significantly faster than the deprecated buffered sampling (#10079)
Updated Dockerfile to use latest from NVIDIA (#9794)
Dropped Python 3.8 support (#9696)
Added a check that confirms that custom edge types of NumNeighbors actually exist in the graph (#9807)
Automatic num_params in LLM + update GRetriever default llm (#9938)
Updated calls to NumPy's deprecated np.in1d to ()

New Contributors

@JacekDuszenko made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9676
@brs96 made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9686
@eurunuela made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9735
@nhadler made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9752
@wzm2256 made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9743
@darabos made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9773
@Aiik made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9776
@ybubnov made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9748
@ryoji-kubo made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9756
@valentingoelz made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9790
@abertics made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9813
@sbhavani made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9794
@kolmiw made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9905
@lukedoubleu made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9860
@erytheis made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9927
@RorryB made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9877
@caic99 made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9949
@fedelopez77 made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/10124
@tommyly201 made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/10105
@semihcanturk made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9018
@JeS24 made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/10208
@LeonResearch made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/10115
@ShadowDragon5 made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/10282
@jeffguy made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/10257
@jdhenaos made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/9823
@co63oc made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/10274
@michaelfortunato made their first contribution in https://github.com/pyg-team/pytorch_geometric/pull/10222

Full Changelog: https://github.com/pyg-team/pytorch_geometric/compare/2.6.0...2.7.0

pytorch_geometric

More Python Projects

AutoGPT

yt-dlp

markitdown

HelloGitHub