Merge pull request #2274 from kieranfraser/dev_autoattack_metadata

Printable representation of Attack objects [JATIC-I2-IBM]

Author: beat-buesser

art/attacks/attack.py

@@ -186,6 +186,19 @@ def is_estimator_valid(estimator, estimator_requirements) -> bool:
                 return False
         return True
 
+    def __repr__(self):
+        """
+        Returns a string describing the attack class and attack_params
+        """
+        param_str = ""
+        for param in self.attack_params:
+            if hasattr(self, param):
+                param_str += f"{param}={getattr(self, param)}, "
+            elif hasattr(self, "_attack"):
+                if hasattr(self._attack, param):
+                    param_str += f"{param}={getattr(self._attack, param)}, "
+        return f"{type(self).__name__}({param_str})"
+
 
 class EvasionAttack(Attack):
     """

art/attacks/evasion/projected_gradient_descent/projected_gradient_descent_numpy.py

@@ -55,7 +55,7 @@ class ProjectedGradientDescentCommon(FastGradientMethod):
     | Paper link: https://arxiv.org/abs/1706.06083
     """
 
-    attack_params = FastGradientMethod.attack_params + ["max_iter", "random_eps", "verbose"]
+    attack_params = FastGradientMethod.attack_params + ["decay", "max_iter", "random_eps", "verbose"]
     _estimator_requirements = (BaseEstimator, LossGradientsMixin)
 
     def __init__(

tests/attacks/test_attack.py

@@ -0,0 +1,38 @@
+# MIT License
+#
+# Copyright (C) The Adversarial Robustness Toolbox (ART) Authors 2023
+#
+# Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated
+# documentation files (the "Software"), to deal in the Software without restriction, including without limitation the
+# rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit
+# persons to whom the Software is furnished to do so, subject to the following conditions:
+#
+# The above copyright notice and this permission notice shall be included in all copies or substantial portions of the
+# Software.
+#
+# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE
+# WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
+# TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+# SOFTWARE.
+import pytest
+
+
+@pytest.mark.skip_framework("tensorflow1", "tensorflow2v1", "keras", "non_dl_frameworks", "mxnet", "kerastf")
+def test_attack_repr(image_dl_estimator):
+
+    from art.attacks.evasion import ProjectedGradientDescentNumpy
+
+    classifier, _ = image_dl_estimator(from_logits=True)
+
+    attack = ProjectedGradientDescentNumpy(
+        estimator=classifier,
+        targeted=True,
+        decay=0.5,
+    )
+    print(repr(attack))
+    assert repr(attack) == (
+        "ProjectedGradientDescentNumpy(norm=inf, eps=0.3, eps_step=0.1, targeted=True, "
+        + "num_random_init=0, batch_size=32, minimal=False, summary_writer=None, decay=0.5, "
+        + "max_iter=100, random_eps=False, verbose=True, )"
+    )