Make validation that labels are integers.

PiperOrigin-RevId: 326216555

tensorflow_privacy/privacy/membership_inference_attack/data_structures.py

@@ -101,6 +101,10 @@ class AttackType(enum.Enum):
     return '%s' % self.name
 
 
+def _is_integer_type_array(a):
+  return np.issubdtype(a.dtype, np.integer)
+
+
 @dataclass
 class AttackInputData:
   """Input data for running an attack.
@@ -175,6 +179,14 @@ class AttackInputData:
         self.logits_train is None):
       raise ValueError('At least one of labels, logits or losses should be set')
 
+    if self.labels_train is not None and not _is_integer_type_array(
+        self.labels_train):
+      raise ValueError('labels_train elements should have integer type')
+
+    if self.labels_test is not None and not _is_integer_type_array(
+        self.labels_test):
+      raise ValueError('labels_test elements should have integer type')
+
     # TODO(b/161366709): Add checks for equal sizes
 
 

tensorflow_privacy/privacy/membership_inference_attack/example.py

@@ -66,7 +66,7 @@ def generate_features_and_labels(samples_per_cluster=250, scale=0.1):
       np.ones(samples_per_cluster),
       np.ones(samples_per_cluster) * 2,
       np.ones(samples_per_cluster) * 3,
-  ))
+  )).astype("uint8")
 
   return (features, labels)
 
@@ -115,6 +115,8 @@ attack_results = mia.run_attacks(
     AttackInputData(
         labels_train=training_labels,
         labels_test=test_labels,
+        logits_train=training_pred,
+        logits_test=test_pred,
         loss_train=crossentropy(training_labels, training_pred),
         loss_test=crossentropy(test_labels, test_pred)),
     SlicingSpec(entire_dataset=True, by_class=True),