lambda functions assigned to variables are replaced by acutal functio…

…n definitions in test validate metrics. Resolved E731 linting error

tests/test_tools.py

@@ -528,31 +528,40 @@ def test_validate_ctf_pass(self):
             model, image_shape=(1, 1, *model.image_shape), device=DEVICE
         )
 
+    # Metric validation tests
     def test_validate_metric_inputs(self):
-        metric = lambda x: x
+        def identity_metric(x):
+            return x
+
         with pytest.raises(TypeError, match="metric should be callable and accept two"):
-            po.tools.validate.validate_metric(metric, device=DEVICE)
+            po.tools.validate.validate_metric(identity_metric, device=DEVICE)
 
     def test_validate_metric_output_shape(self):
-        metric = lambda x, y: x - y
+        def difference_metric(x, y):
+            return x - y
+
         with pytest.raises(
             ValueError, match="metric should return a scalar value but output"
         ):
-            po.tools.validate.validate_metric(metric, device=DEVICE)
+            po.tools.validate.validate_metric(difference_metric, device=DEVICE)
 
     def test_validate_metric_identical(self):
-        metric = lambda x, y: (x + y).mean()
+        def mean_metric(x, y):
+            return (x + y).mean()
+
         with pytest.raises(
             ValueError, match="metric should return <= 5e-7 on two identical"
         ):
-            po.tools.validate.validate_metric(metric, device=DEVICE)
+            po.tools.validate.validate_metric(mean_metric, device=DEVICE)
 
     def test_validate_metric_nonnegative(self):
-        metric = lambda x, y: (x - y).sum()
+        def sum_metric(x, y):
+            return (x - y).sum()
+
         with pytest.raises(
             ValueError, match="metric should always return non-negative"
         ):
-            po.tools.validate.validate_metric(metric, device=DEVICE)
+            po.tools.validate.validate_metric(sum_metric, device=DEVICE)
 
     @pytest.mark.parametrize("model", ["frontend.OnOff.nograd"], indirect=True)
     def test_remove_grad(self, model):