626_privacy/tensorflow_privacy
2
0
Fork 1
Code Issues Pull requests 1 Projects Releases Packages Wiki Activity Actions

Fix num_records in quantile_adaptive_clip_sum_query_test.

Browse source 
PiperOrigin-RevId: 292995170
This commit is contained in:
Galen Andrew 2020-02-03 13:46:05 -08:00 committed by Steve Chien
parent 9bb3c1e6d8
commit 945075a136
1 changed files with 35 additions and 34 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

69
tensorflow_privacy/privacy/dp_query/quantile_adaptive_clip_sum_query_test.py
View file

@ -273,10 +273,41 @@ class QuantileAdaptiveClipSumQueryTest(
('start_high_geometric', False, True))
def test_adaptation_linspace(self, start_low, geometric):
# 100 records equally spaced from 0 to 10 in 0.1 increments.
# Test that with a decaying learning rate we converge to the correct
# median value and bounce around it.
# Test that we converge to the correct median value and bounce around it.
num_records = 21
records = [tf.constant(x) for x in np.linspace(
0.0, 10.0, num=21, dtype=np.float32)]
0.0, 10.0, num=num_records, dtype=np.float32)]
query = quantile_adaptive_clip_sum_query.QuantileAdaptiveClipSumQuery(
initial_l2_norm_clip=(1.0 if start_low else 10.0),
noise_multiplier=0.0,
target_unclipped_quantile=0.5,
learning_rate=1.0,
clipped_count_stddev=0.0,
expected_num_records=num_records,
geometric_update=geometric)
global_state = query.initial_global_state()
for t in range(50):
_, global_state = test_utils.run_query(query, records, global_state)
actual_clip = global_state.sum_state.l2_norm_clip
if t > 40:
self.assertNear(actual_clip, 5.0, 0.25)
@parameterized.named_parameters(
('start_low_arithmetic', True, False),
('start_low_geometric', True, True),
('start_high_arithmetic', False, False),
('start_high_geometric', False, True))
def test_adaptation_all_equal(self, start_low, geometric):
# 20 equal records. Test that we converge to that record and bounce around
# it. Unlike the linspace test, the quantile-matching objective is very
# sharp at the optimum so a decaying learning rate is necessary.
num_records = 20
records = [tf.constant(5.0)] * num_records
learning_rate = tf.Variable(1.0)
@ -286,37 +317,7 @@ class QuantileAdaptiveClipSumQueryTest(
target_unclipped_quantile=0.5,
learning_rate=learning_rate,
clipped_count_stddev=0.0,
expected_num_records=2.0,
geometric_update=geometric)
global_state = query.initial_global_state()
for t in range(50):
tf.compat.v1.assign(learning_rate, 1.0 / np.sqrt(t + 1))
_, global_state = test_utils.run_query(query, records, global_state)
actual_clip = global_state.sum_state.l2_norm_clip
if t > 40:
self.assertNear(actual_clip, 5.0, 0.25)
@parameterized.named_parameters(
('arithmetic', False),
('geometric', True))
def test_adaptation_all_equal(self, geometric):
# 20 equal records. Test that with a decaying learning rate we converge to
# that record and bounce around it.
records = [tf.constant(5.0)] * 20
learning_rate = tf.Variable(1.0)
query = quantile_adaptive_clip_sum_query.QuantileAdaptiveClipSumQuery(
initial_l2_norm_clip=1.0,
noise_multiplier=0.0,
target_unclipped_quantile=0.5,
learning_rate=learning_rate,
clipped_count_stddev=0.0,
expected_num_records=2.0,
expected_num_records=num_records,
geometric_update=geometric)
global_state = query.initial_global_state()