Skip to content

Releases: rusty1s/pytorch_sparse

0.6.18

06 Oct 08:51
7d22892
Compare
Choose a tag to compare

What's Changed

  • Support sparse element-wise multiplication in SparseTensor via sparse_mat1 * sparse_mat2 (#323)
  • Fix gradient computation in to_symmetric (#327)
  • Fix empty edge indices handling (#332)
  • Add mps Apple silicon GPU Acceleration support (#335)
  • Fix shape of node_weight in metis computation (#342)
  • Add PyTorch 2.1.0 support (#344)

New Contributors

Full Changelog: 0.6.17...0.6.18

0.6.17

22 Mar 13:48
2a73371
Compare
Choose a tag to compare
  • PyTorch 2.0 support (#317)
  • Integrated faster index_sort in case pyg-lib is installed as well (#306)
  • Added a balance_edge option to the METIS graph partitioning algorithm (#309)
  • Added SparseTensor.to_torch_sparse_csc_tensor functionality (#319)

0.6.16

22 Dec 14:51
Compare
Choose a tag to compare
  • Fix spspmm on newer CUDA versions/GPUs
  • PyTorch 1.13 support
  • torch.bfloat16 support in spmm
  • Use faster hash-map routine in CPP code paths

0.6.15

17 Aug 10:46
6143af2
Compare
Choose a tag to compare
  • Temporal sampling is now correctly performed in disjoint mode (#267)
  • Replace std::unordered_map with phmap::flat_hash_map for faster sampling (#266)
  • Neighborhood sampling on heterogeneous graphs is now fully-deterministic (#265)

0.6.14

30 Jun 11:25
587d1d0
Compare
Choose a tag to compare
  • Internal C++ method for sampling neighbors based on temporal constraints (#202, #225, #226)
  • Sampling operators now respect torch.manual_seed (#217)

0.6.13

11 Mar 15:19
Compare
Choose a tag to compare
  • SparseTensor: __eq__ functionality
  • SparseTensor: add functionality of two sparse matrices (#177)
  • SparseTensor: to_torch_csr_tensor and from_torch_csr_tensor functionality
  • SparseTensor: Allow indexing via np.array (#194)
  • SparseTensor: Skip unnecessary assertions and enable non-blocking data transfers (#195)
  • Allow loading of CPU wheels in a PyTorch CUDA installation

PyTorch 1.10 is now required.

0.6.12

08 Sep 09:32
Compare
Choose a tag to compare
  • (Internal) heterogeneous neighbor sampling support via torch.ops.torch_sparse.hetero_neighbor_sample
  • (Internal) heterogeneous graph transformer sampling support via torch.ops.torch_sparse.hgt_sample (thanks to @chantat)
  • Fixed a bug in set_diag in case SparseTensor does not hold any non-zero elements

0.6.11

29 Jul 17:31
Compare
Choose a tag to compare
  • Half-precision support (torch.half) for all operators in torch-sparse

0.6.10

17 Jun 07:00
Compare
Choose a tag to compare

This release brings PyTorch 1.9.0 and Python 3.9 support to torch-sparse.

Additional functionality

  • Added a check for row.max() < sparse_sizes[0] and col.max() < sparse_sizes[1] when creating a SparseTensor in order to avoid unexpected behavior (thanks to @Adam1679)
  • partition now supports the additional optional argument node_weight (thanks to @Spazierganger)

0.6.9

03 Mar 15:26
Compare
Choose a tag to compare
  • Reduced the size of shared library files
  • CUDA wheels can now also operate on CPU-only devices
  • spmm now supports torch.half
  • Added parallelization strategies for CPU functionalities
  • Fixed a bug in which sample_adj did not return a sparse matrix with sorted indices
  • Fixed a bug in spmm in case num_edges < num_nodes