pyg-nightly 2.7.0.dev20250304__py3-none-any.whl → 2.7.0.dev20250306__py3-none-any.whl

{pyg_nightly-2.7.0.dev20250304.dist-info → pyg_nightly-2.7.0.dev20250306.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pyg-nightly
-Version: 2.7.0.dev20250304
+Version: 2.7.0.dev20250306
 Summary: Graph Neural Network Library for PyTorch
 Keywords: deep-learning,pytorch,geometric-deep-learning,graph-neural-networks,graph-convolutional-networks
 Author-email: Matthias Fey <matthias@pyg.org>

{pyg_nightly-2.7.0.dev20250304.dist-info → pyg_nightly-2.7.0.dev20250306.dist-info}/RECORD

@@ -1,4 +1,4 @@
-torch_geometric/__init__.py,sha256=xYe0gimmmtBMAoZna0JCHDvfFNOZ3HxQgae8H78U-Cg,1978
+torch_geometric/__init__.py,sha256=kP_Bjpvq2DPS6AF3gNyHlfeD4qWNVIFQrsHUpMuLDr8,1978
 torch_geometric/_compile.py,sha256=f-WQeH4VLi5Hn9lrgztFUCSrN_FImjhQa6BxFzcYC38,1338
 torch_geometric/_onnx.py,sha256=V9ffrIKSqhDw6xUZ12lkuSfNs48cQp2EeJ6Z19GfnVw,349
 torch_geometric/backend.py,sha256=lVaf7aLoVaB3M-UcByUJ1G4T4FOK6LXAg0CF4W3E8jo,1575
@@ -290,7 +290,7 @@ torch_geometric/loader/temporal_dataloader.py,sha256=AQ2QFeiXKbPp6I8sUeE8H7br-1_
 torch_geometric/loader/utils.py,sha256=f27mczQ7fEP2HpTsJGJxKS0slPu0j8zTba3jP8ViNck,14901
 torch_geometric/loader/zip_loader.py,sha256=3lt10fD15Rxm1WhWzypswGzCEwUz4h8OLCD1nE15yNg,3843
 torch_geometric/metrics/__init__.py,sha256=3krvDobW6vV5yHTjq2S2pmOXxNfysNG26muq7z48e94,699
-torch_geometric/metrics/link_pred.py,sha256=wGQG-Fl6BQYJMLZe_L_iIl4ixj6TWgLkkuHyMMraWBA,30480
+torch_geometric/metrics/link_pred.py,sha256=t2YHbEYc8Jbj_4Sb-Wdk5T5uzsSErpjBpUiSqOSf-NM,30729
 torch_geometric/nn/__init__.py,sha256=kQHHHUxFDht2ztD-XFQuv98TvC8MdodaFsIjAvltJBw,874
 torch_geometric/nn/data_parallel.py,sha256=lDAxRi83UNuzAQSj3eu9K2sQheOIU6wqR5elS6oDs90,4764
 torch_geometric/nn/encoding.py,sha256=QNjwWczYExZ1wRGBmpuqYbn6tB7NC4BU-DEgzjhcZqw,3115
@@ -458,7 +458,7 @@ torch_geometric/nn/models/re_net.py,sha256=pz66q5b5BoGDNVQvpEGS2RGoeKvpjkYAv9r3W
 torch_geometric/nn/models/rect.py,sha256=2F3XyyvHTAEuqfJpiNB5M8pSGy738LhPiom5I-SDWqM,2808
 torch_geometric/nn/models/rev_gnn.py,sha256=1b6wU-6YTuLsWn5p8c5LXQm2KugEAVcEYJKZbWTDvgQ,11796
 torch_geometric/nn/models/schnet.py,sha256=0aaHrVtxApdvn3RHCGLQJW1MbIb--CSYUrx9O3hDOZM,16656
-torch_geometric/nn/models/sgformer.py,sha256=V-F3J-yEdjuhkSsCkQePG_ByNHAu0BcKpknVxjCI3KY,6761
+torch_geometric/nn/models/sgformer.py,sha256=3NDzkEVRtM1QmeJsXSq7FBhGGchyUvyX1SDPKYg9F70,6875
 torch_geometric/nn/models/signed_gcn.py,sha256=J40CnedFIqtKI1LhW1ITSEFRbA_XiJZL6lASrKwUEAI,9841
 torch_geometric/nn/models/tgn.py,sha256=kEGdfLJybkbMT4UMoAh2nCzfX3_nDjfm1cicuPHEwAM,11878
 torch_geometric/nn/models/visnet.py,sha256=97OFMCsPDEI5BCSi7RhoRcU2CNRp7zck2tEzrltFZj4,43192
@@ -633,7 +633,7 @@ torch_geometric/utils/undirected.py,sha256=H_nfpI0_WluOG6VfjPyldvcjL4w5USAKWu2x5
 torch_geometric/visualization/__init__.py,sha256=PyR_4K5SafsJrBr6qWrkjKr6GBL1b7FtZybyXCDEVwY,154
 torch_geometric/visualization/graph.py,sha256=ZuLPL92yGRi7lxlqsUPwL_EVVXF7P2kMcveTtW79vpA,4784
 torch_geometric/visualization/influence.py,sha256=CWMvuNA_Nf1sfbJmQgn58yS4OFpeKXeZPe7kEuvkUBw,477
-pyg_nightly-2.7.0.dev20250304.dist-info/licenses/LICENSE,sha256=ic-27cMJc1kWoMEYncz3Ya3Ur2Bi3bNLWib2DT763-o,1067
-pyg_nightly-2.7.0.dev20250304.dist-info/WHEEL,sha256=_2ozNFCLWc93bK4WKHCO-eDUENDlo-dgc9cU3qokYO4,82
-pyg_nightly-2.7.0.dev20250304.dist-info/METADATA,sha256=u-LjJTVVl8r1lbo_4EADaStEERbBn8JGD2vJBzZO734,63021
-pyg_nightly-2.7.0.dev20250304.dist-info/RECORD,,
+pyg_nightly-2.7.0.dev20250306.dist-info/licenses/LICENSE,sha256=ic-27cMJc1kWoMEYncz3Ya3Ur2Bi3bNLWib2DT763-o,1067
+pyg_nightly-2.7.0.dev20250306.dist-info/WHEEL,sha256=_2ozNFCLWc93bK4WKHCO-eDUENDlo-dgc9cU3qokYO4,82
+pyg_nightly-2.7.0.dev20250306.dist-info/METADATA,sha256=dgxfCImotLlU9VF4RUJG86cqJndisPrw49Qzj5dpTsE,63021
+pyg_nightly-2.7.0.dev20250306.dist-info/RECORD,,

torch_geometric/__init__.py

@@ -31,7 +31,7 @@ from .lazy_loader import LazyLoader
 contrib = LazyLoader('contrib', globals(), 'torch_geometric.contrib')
 graphgym = LazyLoader('graphgym', globals(), 'torch_geometric.graphgym')
 
-__version__ = '2.7.0.dev20250304'
+__version__ = '2.7.0.dev20250306'
 
 __all__ = [
     'Index',

torch_geometric/metrics/link_pred.py

@@ -221,8 +221,8 @@ class LinkPredMetric(_LinkPredMetric):
             self.add_state('accum', torch.tensor(0.), dist_reduce_fx='sum')
             self.add_state('total', torch.tensor(0), dist_reduce_fx='sum')
         else:
-            self.register_buffer('accum', torch.tensor(0.))
-            self.register_buffer('total', torch.tensor(0))
+            self.register_buffer('accum', torch.tensor(0.), persistent=False)
+            self.register_buffer('total', torch.tensor(0), persistent=False)
 
     def update(
         self,
@@ -523,10 +523,11 @@ class LinkPredNDCG(LinkPredMetric):
         discount = torch.arange(2, k + 2, dtype=dtype).log2()
 
         self.discount: Tensor
-        self.register_buffer('discount', discount)
+        self.register_buffer('discount', discount, persistent=False)
 
         if not weighted:
-            self.register_buffer('idcg', cumsum(1.0 / discount))
+            self.register_buffer('idcg', cumsum(1.0 / discount),
+                                 persistent=False)
         else:
             self.idcg = None
 
@@ -617,7 +618,7 @@ class LinkPredCoverage(_LinkPredMetric):
         if WITH_TORCHMETRICS:
             self.add_state('mask', mask, dist_reduce_fx='max')
         else:
-            self.register_buffer('mask', mask)
+            self.register_buffer('mask', mask, persistent=False)
 
     def update(
         self,
@@ -673,11 +674,11 @@ class LinkPredDiversity(_LinkPredMetric):
             self.add_state('accum', torch.tensor(0.), dist_reduce_fx='sum')
             self.add_state('total', torch.tensor(0), dist_reduce_fx='sum')
         else:
-            self.register_buffer('accum', torch.tensor(0.))
-            self.register_buffer('total', torch.tensor(0))
+            self.register_buffer('accum', torch.tensor(0.), persistent=False)
+            self.register_buffer('total', torch.tensor(0), persistent=False)
 
         self.category: Tensor
-        self.register_buffer('category', category)
+        self.register_buffer('category', category, persistent=False)
 
     def update(
         self,
@@ -740,7 +741,7 @@ class LinkPredPersonalization(_LinkPredMetric):
             self.add_state('total', torch.tensor(0), dist_reduce_fx='sum')
         else:
             self.preds: List[Tensor] = []
-            self.register_buffer('total', torch.tensor(0))
+            self.register_buffer('total', torch.tensor(0), persistent=False)
 
     def update(
         self,
@@ -829,11 +830,11 @@ class LinkPredAveragePopularity(_LinkPredMetric):
             self.add_state('accum', torch.tensor(0.), dist_reduce_fx='sum')
             self.add_state('total', torch.tensor(0), dist_reduce_fx='sum')
         else:
-            self.register_buffer('accum', torch.tensor(0.))
-            self.register_buffer('total', torch.tensor(0))
+            self.register_buffer('accum', torch.tensor(0.), persistent=False)
+            self.register_buffer('total', torch.tensor(0), persistent=False)
 
         self.popularity: Tensor
-        self.register_buffer('popularity', popularity)
+        self.register_buffer('popularity', popularity, persistent=False)
 
     def update(
         self,

torch_geometric/nn/models/sgformer.py

@@ -91,6 +91,8 @@ class SGModule(torch.nn.Module):
     def forward(self, x: Tensor, batch: Tensor):
         # to dense batch expects sorted batch
         batch, indices = batch.sort(stable=True)
+        rev_perm = torch.empty_like(indices)
+        rev_perm[indices] = torch.arange(len(indices), device=indices.device)
         x = x[indices]
         x, mask = to_dense_batch(x, batch)
         layer_ = []
@@ -114,7 +116,7 @@ class SGModule(torch.nn.Module):
 
         x_mask = x[mask]
         # reverse the sorting
-        unsorted_x_mask = x_mask[indices.argsort()]
+        unsorted_x_mask = x_mask[rev_perm]
         return unsorted_x_mask