Merge pull request #1586 from lorenzbaraldi/eval_loss

Put validation loss under amp_autocast

train.py

@@ -970,16 +970,16 @@ def validate(
 
             with amp_autocast():
                 output = model(input)
-            if isinstance(output, (tuple, list)):
-                output = output[0]
+                if isinstance(output, (tuple, list)):
+                    output = output[0]
 
-            # augmentation reduction
-            reduce_factor = args.tta
-            if reduce_factor > 1:
-                output = output.unfold(0, reduce_factor, reduce_factor).mean(dim=2)
-                target = target[0:target.size(0):reduce_factor]
+                # augmentation reduction
+                reduce_factor = args.tta
+                if reduce_factor > 1:
+                    output = output.unfold(0, reduce_factor, reduce_factor).mean(dim=2)
+                    target = target[0:target.size(0):reduce_factor]
 
-            loss = loss_fn(output, target)
+                loss = loss_fn(output, target)
             acc1, acc5 = utils.accuracy(output, target, topk=(1, 5))
 
             if args.distributed:

validate.py

@@ -294,9 +294,9 @@ def validate(args):
             with amp_autocast():
                 output = model(input)
 
-            if valid_labels is not None:
-                output = output[:, valid_labels]
-            loss = criterion(output, target)
+                if valid_labels is not None:
+                    output = output[:, valid_labels]
+                loss = criterion(output, target)
 
             if real_labels is not None:
                 real_labels.add_result(output)