Fixes

jingdu7560 · Aug 25, 2019 · 21efc67 · 21efc67
1 parent 2ccb69d
commit 21efc67
Show file tree

Hide file tree

Showing 2 changed files with 9 additions and 10 deletions.
diff --git a/keras/engine/training.py b/keras/engine/training.py
@@ -680,8 +680,10 @@ def _prepare_total_loss(self, masks=None):
                         y_true, y_pred, sample_weight=sample_weight)
 
                 if len(self.outputs) > 1:
-                    update_ops = self._output_loss_metrics[i].update_state(output_loss)
-                    self._metric_updates += update_ops
+                    # TODO
+                    # update_ops = self._output_loss_metrics[i].update_state(output_loss)
+                    # self._metric_updates += update_ops
+                    self._output_loss_metrics[i](output_loss)
 
                 if total_loss is None:
                     total_loss = loss_weight * output_loss

diff --git a/tests/keras/metrics_test.py b/tests/keras/metrics_test.py
@@ -30,8 +30,8 @@ def test_sum(self):
         assert K.eval(m.total) == 100
 
         # check update_state() and result() + state accumulation + tensor input
-        K.eval(m.update_state([1, 5]))
-        assert np.isclose(K.eval(m.result()), 106)
+        result = m([1, 5])
+        assert np.isclose(K.eval(result), 106)
         assert K.eval(m.total) == 106  # 100 + 1 + 5
 
         # check reset_states()
@@ -97,9 +97,8 @@ def test_mean(self):
         assert K.eval(m.count) == 1
 
         # check update_state() and result()
-        update_op = m.update_state([1, 5])
-        K.eval(update_op)
-        assert np.isclose(K.eval(m.result()), 106 / 3)
+        result = m([1, 5])
+        assert np.isclose(K.eval(result), 106 / 3)
         assert K.eval(m.total) == 106  # 100 + 1 + 5
         assert K.eval(m.count) == 3
 
@@ -195,9 +194,7 @@ def test_unweighted(self):
         y_true = ((0, 1, 0, 1, 0), (0, 0, 1, 1, 1), (1, 1, 1, 1, 0), (0, 0, 0, 0, 1))
         y_pred = ((0, 0, 1, 1, 0), (1, 1, 1, 1, 1), (0, 1, 0, 1, 0), (1, 1, 1, 1, 1))
 
-        update_op = mse_obj.update_state(y_true, y_pred)
-        K.eval(update_op)
-        result = mse_obj.result()
+        result = mse_obj(y_true, y_pred)
         np.isclose(0.5, K.eval(result), atol=1e-5)
 
     def test_weighted(self):