(#3) Defense: Add processing logic for exformer and exmotd

src/defense/process.py

@@ -1,5 +1,5 @@
 from utils.logging import concat_batch_images
-from defense.models import Reformer, Denoiser, Motd
+from defense.models import Reformer, Denoiser, Motd, ExMotd
 
 import tensorflow as tf
 
@@ -30,7 +30,7 @@
         type=str,
         help="Defense method",
         required=True,
-        choices=["reformer", "denoiser", "motd"],
+        choices=["reformer", "exformer", "denoiser", "motd", "exmotd"],
     )
 
     parser.add_argument(
@@ -59,19 +59,32 @@
             input_shape=input_shape,
             intensity=args.intensity[0],
         )
+    elif args.defense == "exformer":
+        defense_model = Reformer(
+            "defense_exformer_cifar10",
+            input_shape=input_shape,
+            intensity=args.intensity[0],
+        )
     elif args.defense == "denoiser":
         defense_model = Denoiser(
             f"defense_denoiser_{args.dataset}",
             input_shape=input_shape,
             intensity=args.intensity[0],
         )
-    else:
+    elif args.defense == "motd":
         defense_model = Motd(
             f"defense_motd_{args.dataset}",
             input_shape=input_shape,
             dataset=args.dataset,
             intensities=args.intensity,
         )
+    else:
+        defense_model = ExMotd(
+            f"defense_exmotd_{args.dataset}",
+            input_shape=input_shape,
+            dataset=args.dataset,
+            intensities=args.intensity,
+        )
 
     defense_model.compile()
     defense_model.load()