[Bugfix] Improve CompGCN (#3663)

Co-authored-by: Mufei Li <mufeili1996@gmail.com>

examples/pytorch/compGCN/README.md

@@ -9,9 +9,9 @@ This example was implemented by [zhjwy9343](https://github.com/zhjwy9343) and [K
 
 Dependencies
 ----------------------
-- pytorch 1.7.1
-- dgl 0.6.0
-- numpy 1.19.4
+- pytorch 1.9.0
+- dgl 0.7.1
+- numpy 1.20.3
 - ordered_set 4.0.2
 
 Dataset
@@ -67,11 +67,11 @@ Performance
 | Dataset |        FB15k-237         |          WN18RR          |
 |---------| ------------------------ | ------------------------ |
 |  Metric |    Paper   /  ours (dgl) |    Paper   /  ours (dgl) |
-|   MRR   |    0.355   /    0.349    |    0.479   /    0.471    |
-|   MR    |     197    /     208     |    3533    /     3550    |
-| Hit@10  |    0.535   /   0.526     |    0.546   /    0.532    |
-|  Hit@3  |    0.390   /    0.381    |    0.494   /    0.480    |
-|  Hit@1  |    0.264   /    0.260    |    0.443   /    0.438    |
+|   MRR   |    0.355   /    0.348    |    0.479   /    0.466    |
+|   MR    |     197    /     208     |    3533    /     3542    |
+| Hit@10  |    0.535   /   0.527     |    0.546   /    0.525    |
+|  Hit@3  |    0.390   /    0.380    |    0.494   /    0.476    |
+|  Hit@1  |    0.264   /    0.259    |    0.443   /    0.435    |
 
 
 

examples/pytorch/compGCN/main.py

@@ -120,7 +120,7 @@ def main(args):
 		
             # compute loss
             tr_loss = loss_fn(logits, label)
-            train_loss.append(tr_loss)
+            train_loss.append(tr_loss.item())
 
             # backward
             optimizer.zero_grad()
@@ -142,7 +142,7 @@ def main(args):
             print("saving model...")
         else:
             kill_cnt += 1
-            if kill_cnt > 25:
+            if kill_cnt > 100:
                 print('early stop.')
                 break
         print("In epoch {}, Train Loss: {:.4f}, Valid MRR: {:.5}\n, Train time: {}, Valid time: {}"\
@@ -164,7 +164,7 @@ if __name__ == '__main__':
     parser.add_argument('--score_func', dest='score_func', default='conve', help='Score Function for Link prediction')
     parser.add_argument('--opn', dest='opn', default='ccorr', help='Composition Operation to be used in CompGCN')
 
-    parser.add_argument('--batch', dest='batch_size', default=128, type=int, help='Batch size')
+    parser.add_argument('--batch', dest='batch_size', default=1024, type=int, help='Batch size')
     parser.add_argument('--gpu', type=int, default='0', help='Set GPU Ids : Eg: For CPU = -1, For Single GPU = 0')
     parser.add_argument('--epoch', dest='max_epochs', type=int, default=500, help='Number of epochs')
     parser.add_argument('--l2', type=float, default=0.0, help='L2 Regularization for Optimizer')

examples/pytorch/compGCN/utils.py

@@ -32,7 +32,7 @@ def ccorr(a, b):
 	-------
 	Tensor, having the same dimension as the input a.
 	"""
-	return th.irfft(com_mult(conj(th.rfft(a, 1)), th.rfft(b, 1)), 1, signal_sizes=(a.shape[-1],))
+	return th.fft.irfftn(th.conj(th.fft.rfftn(a, (-1))) * th.fft.rfftn(b, (-1)), (-1))
 
 #identify in/out edges, compute edge norm for each and store in edata
 def in_out_norm(graph):