bandwidth implementation

torch_sparse/__init__.py

@@ -47,6 +47,7 @@ from .matmul import matmul  # noqa
 from .cat import cat, cat_diag  # noqa
 from .rw import random_walk  # noqa
 from .metis import partition  # noqa
+from .bandwidth import reverse_cuthill_mckee  # noqa
 from .saint import saint_subgraph  # noqa
 from .padding import padded_index, padded_index_select  # noqa
 
@@ -90,6 +91,7 @@ __all__ = [
     'cat_diag',
     'random_walk',
     'partition',
+    'reverse_cuthill_mckee',
     'saint_subgraph',
     'padded_index',
     'padded_index_select',

torch_sparse/bandwidth.py

+import scipy.sparse as sp
+from typing import Tuple, Optional
+
+import torch
+from torch_sparse.tensor import SparseTensor
+from torch_sparse.permute import permute
+
+
+def reverse_cuthill_mckee(src: SparseTensor,
+                          is_symmetric: Optional[bool] = None
+                          ) -> Tuple[SparseTensor, torch.Tensor]:
+
+    if is_symmetric is None:
+        is_symmetric = src.is_symmetric()
+
+    if not is_symmetric:
+        src = src.to_symmetric()
+
+    sp_src = src.to_scipy(layout='csr')
+    perm = sp.csgraph.reverse_cuthill_mckee(sp_src, symmetric_mode=True).copy()
+    perm = torch.from_numpy(perm).to(torch.long).to(src.device())
+
+    out = permute(src, perm)
+
+    return out, perm
+
+
+SparseTensor.reverse_cuthill_mckee = reverse_cuthill_mckee