[Misc] Fix validation dataset's usage in RGCN link example (#7308)

examples/pytorch/rgcn/README.md

@@ -58,4 +58,4 @@ Summary
 ### Link Prediction
 | Dataset       | Best MRR
 | ------------- | -------
-| FB15k-237     | ~0.2439
+| FB15k-237     | ~0.2397

examples/pytorch/rgcn/link.py

@@ -223,11 +223,9 @@ def perturb_and_get_filtered_rank(
     return torch.LongTensor(ranks)
 
 
-def calc_mrr(
-    emb, w, test_mask, triplets_to_filter, batch_size=100, filter=True
-):
+def calc_mrr(emb, w, mask, triplets_to_filter, batch_size=100, filter=True):
     with torch.no_grad():
-        test_triplets = triplets_to_filter[test_mask]
+        test_triplets = triplets_to_filter[mask]
         s, r, o = test_triplets[:, 0], test_triplets[:, 1], test_triplets[:, 2]
         test_size = len(s)
         triplets_to_filter = {
@@ -249,7 +247,7 @@ def train(
     dataloader,
     test_g,
     test_nids,
-    test_mask,
+    val_mask,
     triplets,
     device,
     model_state_file,
@@ -284,7 +282,7 @@ def train(
             model.eval()
             embed = model(test_g, test_nids)
             mrr = calc_mrr(
-                embed, model.w_relation, test_mask, triplets, batch_size=500
+                embed, model.w_relation, val_mask, triplets, batch_size=500
             )
             # save best model
             if best_mrr < mrr:
@@ -309,6 +307,7 @@ if __name__ == "__main__":
     test_g = get_subset_g(g, g.edata["train_mask"], num_rels, bidirected=True)
     test_g.edata["norm"] = dgl.norm_by_dst(test_g).unsqueeze(-1)
     test_nids = torch.arange(0, num_nodes)
+    val_mask = g.edata["val_mask"]
     test_mask = g.edata["test_mask"]
     subg_iter = SubgraphIterator(train_g, num_rels)  # uniform edge sampling
     dataloader = GraphDataLoader(
@@ -328,7 +327,7 @@ if __name__ == "__main__":
         dataloader,
         test_g,
         test_nids,
-        test_mask,
+        val_mask,
         triplets,
         device,
         model_state_file,