[GraphBolt] Change default value from `coo` to `csc` in `FusedCSCSamplingGraph`. (#6833)

Co-authored-by: Ubuntu <ubuntu@ip-172-31-0-133.us-west-2.compute.internal>

python/dgl/graphbolt/impl/fused_csc_sampling_graph.py

@@ -325,7 +325,7 @@ class FusedCSCSamplingGraph(SamplingGraph):
         self,
         nodes: Union[torch.Tensor, Dict[str, torch.Tensor]],
         # TODO: clean up once the migration is done.
-        output_cscformat=False,
+        output_cscformat=True,
     ) -> Union[FusedSampledSubgraphImpl, SampledSubgraphImpl]:
         """Return the subgraph induced on the inbound edges of the given nodes.
 
@@ -555,7 +555,7 @@ class FusedCSCSamplingGraph(SamplingGraph):
         replace: bool = False,
         probs_name: Optional[str] = None,
         # TODO: clean up once the migration is done.
-        output_cscformat=False,
+        output_cscformat=True,
     ) -> Union[FusedSampledSubgraphImpl, SampledSubgraphImpl]:
         """Sample neighboring edges of the given nodes and return the induced
         subgraph.
@@ -745,7 +745,7 @@ class FusedCSCSamplingGraph(SamplingGraph):
         replace: bool = False,
         probs_name: Optional[str] = None,
         # TODO: clean up once the migration is done.
-        output_cscformat=False,
+        output_cscformat=True,
     ) -> Union[FusedSampledSubgraphImpl, SampledSubgraphImpl]:
         """Sample neighboring edges of the given nodes and return the induced
         subgraph via layer-neighbor sampling from the NeurIPS 2023 paper

tests/python/pytorch/graphbolt/impl/test_fused_csc_sampling_graph.py

@@ -706,7 +706,7 @@ def test_in_subgraph_node_pairs_homogeneous():
 
     # Extract in subgraph.
     nodes = torch.LongTensor([4, 1, 3])
-    in_subgraph = graph.in_subgraph(nodes)
+    in_subgraph = graph.in_subgraph(nodes, output_cscformat=False)
 
     # Verify in subgraph.
     assert torch.equal(
@@ -778,7 +778,7 @@ def test_in_subgraph_node_pairs_heterogeneous():
         "N0": torch.LongTensor([1]),
         "N1": torch.LongTensor([2, 1]),
     }
-    in_subgraph = graph.in_subgraph(nodes)
+    in_subgraph = graph.in_subgraph(nodes, output_cscformat=False)
 
     # Verify in subgraph.
     assert torch.equal(
@@ -991,7 +991,7 @@ def test_sample_neighbors_homo(labor, indptr_dtype, indices_dtype):
 
     # 2. Sample with nodes in matched dtype with graph's indices.
     nodes = torch.tensor([1, 3, 4], dtype=indices_dtype)
-    subgraph = sampler(nodes, fanouts)
+    subgraph = sampler(nodes, fanouts, output_cscformat=False)
 
     # Verify in subgraph.
     sampled_num = subgraph.node_pairs[0].size(0)
@@ -1064,14 +1064,14 @@ def test_sample_neighbors_hetero(labor, indptr_dtype, indices_dtype):
             "Data type of nodes must be consistent with indices.dtype"
         ),
     ):
-        _ = sampler(nodes, fanouts)
+        _ = sampler(nodes, fanouts, output_cscformat=False)
 
     # 2. Sample with nodes in matched dtype with graph's indices.
     nodes = {
         "n1": torch.tensor([0], dtype=indices_dtype),
         "n2": torch.tensor([0], dtype=indices_dtype),
     }
-    subgraph = sampler(nodes, fanouts)
+    subgraph = sampler(nodes, fanouts, output_cscformat=False)
 
     # Verify in subgraph.
     expected_node_pairs = {
@@ -1111,11 +1111,11 @@ def test_sample_neighbors_hetero(labor, indptr_dtype, indices_dtype):
             "Data type of nodes must be consistent with indices.dtype"
         ),
     ):
-        _ = sampler(nodes, fanouts)
+        _ = sampler(nodes, fanouts, output_cscformat=False)
 
     # 2. Sample with nodes in matched dtype with graph's indices.
     nodes = {"n1": torch.tensor([0], dtype=indices_dtype)}
-    subgraph = sampler(nodes, fanouts)
+    subgraph = sampler(nodes, fanouts, output_cscformat=False)
 
     # Verify in subgraph.
     expected_node_pairs = {
@@ -1194,7 +1194,7 @@ def test_sample_neighbors_fanouts(
     nodes = {"n1": torch.LongTensor([0]), "n2": torch.LongTensor([0])}
     fanouts = torch.LongTensor(fanouts)
     sampler = graph.sample_layer_neighbors if labor else graph.sample_neighbors
-    subgraph = sampler(nodes, fanouts)
+    subgraph = sampler(nodes, fanouts, output_cscformat=False)
 
     # Verify in subgraph.
     assert (
@@ -1250,7 +1250,7 @@ def test_sample_neighbors_replace(
 
     nodes = {"n1": torch.LongTensor([0]), "n2": torch.LongTensor([0])}
     subgraph = graph.sample_neighbors(
-        nodes, torch.LongTensor([4]), replace=replace
+        nodes, torch.LongTensor([4]), replace=replace, output_cscformat=False
     )
 
     # Verify in subgraph.
@@ -1288,7 +1288,9 @@ def test_sample_neighbors_return_eids_homo(labor):
 
     # Generate subgraph via sample neighbors.
     nodes = torch.LongTensor([1, 3, 4])
-    subgraph = graph.sample_neighbors(nodes, fanouts=torch.LongTensor([-1]))
+    subgraph = graph.sample_neighbors(
+        nodes, fanouts=torch.LongTensor([-1]), output_cscformat=False
+    )
 
     # Verify in subgraph.
     expected_reverse_edge_ids = edge_attributes[gb.ORIGINAL_EDGE_ID][
@@ -1344,7 +1346,7 @@ def test_sample_neighbors_return_eids_hetero(labor):
     nodes = {"n1": torch.LongTensor([0]), "n2": torch.LongTensor([0])}
     fanouts = torch.tensor([-1, -1])
     sampler = graph.sample_layer_neighbors if labor else graph.sample_neighbors
-    subgraph = sampler(nodes, fanouts)
+    subgraph = sampler(nodes, fanouts, output_cscformat=False)
 
     # Verify in subgraph.
     expected_reverse_edge_ids = {
@@ -1402,6 +1404,7 @@ def test_sample_neighbors_probs(replace, labor, probs_name):
         fanouts=torch.tensor([2]),
         replace=replace,
         probs_name=probs_name,
+        output_cscformat=False,
     )
 
     # Verify in subgraph.
@@ -1448,6 +1451,7 @@ def test_sample_neighbors_zero_probs(replace, labor, probs_or_mask):
         fanouts=torch.tensor([5]),
         replace=replace,
         probs_name="probs_or_mask",
+        output_cscformat=False,
     )
 
     # Verify in subgraph.
@@ -1959,6 +1963,7 @@ def test_sample_neighbors_homo_pick_number(fanouts, replace, labor, probs_name):
         fanouts=torch.LongTensor(fanouts),
         replace=replace,
         probs_name=probs_name if probs_name != "none" else None,
+        output_cscformat=False,
     )
     sampled_num = subgraph.node_pairs[0].size(0)
 
@@ -2050,6 +2055,7 @@ def test_sample_neighbors_hetero_pick_number(
         fanouts=torch.LongTensor(fanouts),
         replace=replace,
         probs_name=probs_name if probs_name != "none" else None,
+        output_cscformat=False,
     )
 
     if probs_name == "none":

tests/python/pytorch/graphbolt/impl/test_ondisk_dataset.py

@@ -1143,6 +1143,7 @@ def test_OnDiskDataset_preprocess_homogeneous():
         subgraph = fused_csc_sampling_graph.sample_neighbors(
             torch.arange(num_samples),
             torch.tensor([fanout]),
+            output_cscformat=False,
         )
         assert len(subgraph.node_pairs[0]) <= num_samples