Bugfix.

PiperOrigin-RevId: 486344068

Author: tensorflower-gardener

tensorflow_privacy/privacy/privacy_tests/membership_inference_attack/dataset_slicing.py

@@ -42,16 +42,19 @@ def _slice_data_by_indices(data: AttackInputData, idx_train,
   result.labels_train = _slice_if_not_none(data.labels_train, idx_train)
   result.loss_train = _slice_if_not_none(data.loss_train, idx_train)
   result.entropy_train = _slice_if_not_none(data.entropy_train, idx_train)
-  # Copy over sample weights if provided.
-  result.sample_weight_train = data.sample_weight_train
-  result.sample_weight_test = data.sample_weight_test
+  # Slice sample weights if provided.
+  result.sample_weight_train = _slice_if_not_none(data.sample_weight_train,
+                                                  idx_train)
 
   # Slice test data.
   result.logits_test = _slice_if_not_none(data.logits_test, idx_test)
   result.probs_test = _slice_if_not_none(data.probs_test, idx_test)
   result.labels_test = _slice_if_not_none(data.labels_test, idx_test)
   result.loss_test = _slice_if_not_none(data.loss_test, idx_test)
   result.entropy_test = _slice_if_not_none(data.entropy_test, idx_test)
+  # Slice sample weights if provided.
+  result.sample_weight_test = _slice_if_not_none(data.sample_weight_test,
+                                                 idx_test)
 
   # A slice has the same multilabel status as the original data. This is because
   # of the way multilabel status is computed. A dataset is multilabel if at

tensorflow_privacy/privacy/privacy_tests/membership_inference_attack/dataset_slicing_test.py

@@ -115,8 +115,8 @@ def __init__(self, methodname):
     loss_test = np.array([0.5, 3.5, 7, 4.5])
     entropy_train = np.array([0.4, 8, 0.6, 10])
     entropy_test = np.array([15, 10.5, 4.5, 0.3])
-    sample_weight_train = np.array([1.0, 0.5])
-    sample_weight_test = np.array([0.5, 1.0])
+    sample_weight_train = np.array([1.0, 0.2, 0.5, 0.8])
+    sample_weight_test = np.array([0.5, 1.0, 0.1, 0.8])
 
     self.input_data = AttackInputData(
         logits_train=logits_train,
@@ -175,8 +175,8 @@ def test_slice_by_class(self):
     # Check sample weights
     self.assertLen(output.sample_weight_train, 2)
     np.testing.assert_array_equal(output.sample_weight_train, [1.0, 0.5])
-    self.assertLen(output.sample_weight_test, 2)
-    np.testing.assert_array_equal(output.sample_weight_test, [0.5, 1.0])
+    self.assertLen(output.sample_weight_test, 1)
+    np.testing.assert_array_equal(output.sample_weight_test, [0.5])
 
   def test_slice_by_percentile(self):
     percentile_slice = SingleSliceSpec(SlicingFeature.PERCENTILE, (0, 50))