diff --git a/stargan/solver.py b/stargan/solver.py
index 2863d1e..4c1c54e 100644
--- a/stargan/solver.py
+++ b/stargan/solver.py
@@ -571,25 +571,26 @@ class Solver(object):
 
     def test_attack(self):
         """Translate images using StarGAN trained on a single dataset."""
-        # Load the trained generator.
-        self.restore_model(self.test_iters)
         
-        # Set data loader.
-        if self.dataset == 'CelebA':
-            data_loader = self.celeba_loader
-        elif self.dataset == 'RaFD':
-            data_loader = self.rafd_loader
-
-        # Initialize Metrics
-        l1_error = 0.0
-        l2_error = 0.0
-        perceptual_error = 0.0
-        n_samples = 0
-
-        # 11 layers
-        # layer_num_orig = 1
-
         for layer_num_orig in range(11):
+            # Load the trained generator.
+            self.restore_model(self.test_iters)
+            
+            # Set data loader.
+            if self.dataset == 'CelebA':
+                data_loader = self.celeba_loader
+            elif self.dataset == 'RaFD':
+                data_loader = self.rafd_loader
+
+            # Initialize Metrics
+            l1_error = 0.0
+            l2_error = 0.0
+            perceptual_error = 0.0
+            n_samples = 0
+
+            # 11 layers
+            # layer_num_orig = 1
+
             print('Layer ', layer_num_orig)
             for i, (x_real, c_org) in enumerate(data_loader):
                 # Black image
@@ -641,9 +642,9 @@ class Solver(object):
                 if i == 199:
                     break
         
-        # Print metrics
-        print('{} images. L1 error: {}. L2 error: {}. Perceptual error: {}.'.format(n_samples, l1_error / n_samples, l2_error / n_samples, 
-        perceptual_error / n_samples))
+            # Print metrics
+            print('{} images. L1 error: {}. L2 error: {}. Perceptual error: {}.'.format(n_samples, l1_error / n_samples, l2_error / n_samples, 
+            perceptual_error / n_samples))
 
     def test_multi(self):
         """Translate images using StarGAN trained on multiple datasets."""