def test_count_returns_sensible_result(self):
with TestPipeline() as pipeline:
# Arrange
col = [(u, "pk1") for u in range(30)]
pcol = pipeline | 'Create produce' >> beam.Create(col)
# Use very high epsilon and delta to minimize noise and test
# flakiness.
budget_accountant = budget_accounting.NaiveBudgetAccountant(
total_epsilon=800, total_delta=0.999)
private_collection = (
pcol | 'Create private collection' >> private_beam.MakePrivate(
budget_accountant=budget_accountant,
privacy_id_extractor=lambda x: x[0]))
count_params = aggregate_params.CountParams(
noise_kind=pipeline_dp.NoiseKind.GAUSSIAN,
max_partitions_contributed=2,
max_contributions_per_partition=3,
budget_weight=1,
partition_extractor=lambda x: x[1])
# Act
result = private_collection | private_beam.Count(
count_params=count_params)
budget_accountant.compute_budgets()
# Assert
# This is a health check to validate that the result is sensible.
# Hence, we use a very large tolerance to reduce test flakiness.
beam_util.assert_that(
result,
beam_util.equal_to([("pk1", 30.0)],
equals_fn=lambda e, a: PrivateBeamTest.
value_per_key_within_tolerance(e, a, 5.0)))
def test_count_calls_with_public_partitions_returns_sensible_result(self):
# Arrange
col = [(u, "pubK1") for u in range(30)]
col += [(u, "privK1") for u in range(30)]
dist_data = PrivateRDDTest.sc.parallelize(col)
# Use very high epsilon and delta to minimize noise and test
# flakiness.
budget_accountant = budget_accounting.NaiveBudgetAccountant(
total_epsilon=800, total_delta=0.999)
def privacy_id_extractor(x):
return x[0]
prdd = private_spark.make_private(dist_data, budget_accountant,
privacy_id_extractor)
count_params = agg.CountParams(
noise_kind=pipeline_dp.NoiseKind.GAUSSIAN,
max_partitions_contributed=2,
max_contributions_per_partition=3,
budget_weight=1,
public_partitions=["pubK1", "pubK2"],
partition_extractor=lambda x: x[1])
# Act
actual_result = prdd.count(count_params)
budget_accountant.compute_budgets()
# Assert
# This is a health check to validate that the result is sensible.
# Hence, we use a very large tolerance to reduce test flakiness.
expected_result_dict = {"pubK2": 0.0, "pubK1": 30.0}
actual_result_dict = self.to_dict(actual_result.collect())
for pk, count in actual_result_dict.items():
self.assertTrue(
self.value_per_key_within_tolerance(count,
expected_result_dict[pk],
5.0))
def test_count_calls_aggregate_with_correct_params(self, mock_aggregate):
# Arrange
dist_data = PrivateRDDTest.sc.parallelize([(1, "pk1"), (2, "pk1")])
mock_aggregate.return_value = PrivateRDDTest.sc.parallelize([(2,
["pk1"])])
budget_accountant = budget_accounting.NaiveBudgetAccountant(1, 1e-10)
def privacy_id_extractor(x):
return x[0]
prdd = private_spark.make_private(dist_data, budget_accountant,
privacy_id_extractor)
count_params = agg.CountParams(
noise_kind=pipeline_dp.NoiseKind.GAUSSIAN,
max_partitions_contributed=2,
max_contributions_per_partition=3,
budget_weight=1,
public_partitions=None,
partition_extractor=lambda x: x[1])
# Act
actual_result = prdd.count(count_params)
# Assert
mock_aggregate.assert_called_once()
args = mock_aggregate.call_args[0]
rdd = dist_data.map(lambda x: (privacy_id_extractor(x), x))
self.assertListEqual(args[0].collect(), rdd.collect())
params = pipeline_dp.AggregateParams(
noise_kind=pipeline_dp.NoiseKind.GAUSSIAN,
metrics=[pipeline_dp.Metrics.COUNT],
max_partitions_contributed=count_params.max_partitions_contributed,
max_contributions_per_partition=count_params.
max_contributions_per_partition,
public_partitions=count_params.public_partitions)
self.assertEqual(args[1], params)
self.assertEqual(actual_result.collect(), [(2, "pk1")])
def test_count_calls_aggregate_with_params(self, mock_aggregate):
runner = fn_api_runner.FnApiRunner()
with beam.Pipeline(runner=runner) as pipeline:
# Arrange
pcol = pipeline | 'Create produce' >> beam.Create(
[1, 2, 3, 4, 5, 6])
budget_accountant = budget_accounting.NaiveBudgetAccountant(
total_epsilon=1, total_delta=0.01)
private_collection = (
pcol | 'Create private collection' >> private_beam.MakePrivate(
budget_accountant=budget_accountant,
privacy_id_extractor=PrivateBeamTest.privacy_id_extractor))
count_params = aggregate_params.CountParams(
noise_kind=pipeline_dp.NoiseKind.GAUSSIAN,
max_partitions_contributed=2,
max_contributions_per_partition=3,
budget_weight=1,
partition_extractor=lambda x: f"pk:{x // 10}")
# Act
transformer = private_beam.Count(count_params=count_params)
private_collection | transformer
# Assert
self.assertEqual(transformer._budget_accountant, budget_accountant)
mock_aggregate.assert_called_once()
args = mock_aggregate.call_args[0]
params = pipeline_dp.AggregateParams(
noise_kind=pipeline_dp.NoiseKind.GAUSSIAN,
metrics=[pipeline_dp.Metrics.COUNT],
max_partitions_contributed=count_params.
max_partitions_contributed,
max_contributions_per_partition=count_params.
max_contributions_per_partition,
public_partitions=count_params.public_partitions)
self.assertEqual(args[1], params)