FIX: vaidation code

train.py

@@ -89,7 +89,7 @@ def build_test_loader(data_keys, args):
     test_loader = DataLoader(
         dataset=test_dataset,
         sampler=test_dist_sampler,
-        batch_size=4
+        batch_size=1
     )
     return test_loader
 
@@ -299,29 +299,25 @@ def valid_one_epoch(test_loader, model, device, args):
 
         with torch.no_grad():
             out, gt_count = model(img, kpoint)
-            if args.debug:
+            pred_count = torch.squeeze(out, 1)
-                print("out: {} | gt_count: {}".format(
+            gt_count   = torch.squeeze(gt_count, 1)
-                    out.shape, gt_count.shape
-                ))
-            count = torch.sum(out).item()
-            gt_count = torch.sum(gt_count).item()
-
-        mae += abs(gt_count - count)
-        mse += abs(gt_count - count) ** 2
-
-        # if i % 15 == 0:
-        print("[valid_one_epoch] {} Gt {:.2f} Pred {}".format(
-            fname[0], gt_count, count
-        ))
 
+        diff = torch.sum(torch.abs(gt_count - pred_count)).item()
+        mae += diff
+        mse += diff ** 2
         mae = mae * 1.0 / (len(test_loader) * batch_size)
         mse = np.sqrt(mse / (len(test_loader)) * batch_size)
 
-        nni.report_intermediate_result(mae)
+        if i % 5 == 0:
+            print("[valid_one_epoch] {}\t| Gt {:.2f} Pred {:.4f}\t| mae {:.4f} mse {:.4f} |".format(
+                fname[0], torch.sum(gt_count).item(), torch.sum(pred_count).item(),
+                mae, mse
+            ))
+
+    nni.report_intermediate_result()
     print("* MAE {mae:.3f} | MSE {mse:.3f} *".format(
         mae=mae, mse=mse
     ))
-
     return mae