def test_deserialize_federated_value_with_incompatible_member_types_raises(
self):
x = 10
x_type = computation_types.to_type(tf.int32)
int_member_proto, _ = value_serialization.serialize_value(x, x_type)
y = 10.
y_type = computation_types.to_type(tf.float32)
float_member_proto, _ = value_serialization.serialize_value(y, y_type)
fully_specified_type_at_clients = type_serialization.serialize_type(
computation_types.at_clients(tf.int32))
unspecified_member_federated_type = computation_pb2.FederatedType(
placement=fully_specified_type_at_clients.federated.placement,
all_equal=False)
federated_proto = serialization_bindings.Federated(
type=unspecified_member_federated_type,
value=[int_member_proto, float_member_proto])
federated_value_proto = serialization_bindings.Value(
federated=federated_proto)
self.assertIsInstance(int_member_proto, serialization_bindings.Value)
self.assertIsInstance(float_member_proto, serialization_bindings.Value)
self.assertIsInstance(federated_value_proto, serialization_bindings.Value)
with self.assertRaises(TypeError):
value_serialization.deserialize_value(federated_value_proto)
def test_create_selection(self):
executor = executor_bindings.create_reference_resolving_executor(
executor_bindings.create_tensorflow_executor())
expected_type_spec = TensorType(shape=[3], dtype=tf.int64)
value_pb, _ = value_serialization.serialize_value(
tf.constant([1, 2, 3]), expected_type_spec)
value = executor.create_value(value_pb)
self.assertEqual(value.ref, 0)
# 1. Create a struct from duplicated values.
struct_value = executor.create_struct([value.ref, value.ref])
self.assertEqual(struct_value.ref, 1)
materialized_value = executor.materialize(struct_value.ref)
deserialized_value, type_spec = value_serialization.deserialize_value(
materialized_value)
struct_type_spec = computation_types.to_type(
[expected_type_spec, expected_type_spec])
type_test_utils.assert_types_equivalent(type_spec, struct_type_spec)
deserialized_value = type_conversions.type_to_py_container(
deserialized_value, struct_type_spec)
self.assertAllClose([(1, 2, 3), (1, 2, 3)], deserialized_value)
# 2. Select the first value out of the struct.
new_value = executor.create_selection(struct_value.ref, 0)
materialized_value = executor.materialize(new_value.ref)
deserialized_value, type_spec = value_serialization.deserialize_value(
materialized_value)
type_test_utils.assert_types_equivalent(type_spec, expected_type_spec)
deserialized_value = type_conversions.type_to_py_container(
deserialized_value, struct_type_spec)
self.assertAllClose((1, 2, 3), deserialized_value)
def test_create_value(self):
executor = executor_bindings.create_reference_resolving_executor(
executor_bindings.create_tensorflow_executor())
# 1. Test a simple tensor.
expected_type_spec = TensorType(shape=[3], dtype=tf.int64)
value_pb, _ = value_serialization.serialize_value([1, 2, 3],
expected_type_spec)
value = executor.create_value(value_pb)
self.assertIsInstance(value, executor_bindings.OwnedValueId)
self.assertEqual(value.ref, 0)
self.assertEqual(str(value), '0')
self.assertEqual(repr(value), r'<OwnedValueId: 0>')
materialized_value = executor.materialize(value.ref)
deserialized_value, type_spec = value_serialization.deserialize_value(
materialized_value)
type_test_utils.assert_types_identical(type_spec, expected_type_spec)
self.assertAllEqual(deserialized_value, [1, 2, 3])
# 2. Test a struct of tensors, ensure that we get a different ID.
expected_type_spec = StructType([
('a', TensorType(shape=[3], dtype=tf.int64)),
('b', TensorType(shape=[], dtype=tf.float32))
])
value_pb, _ = value_serialization.serialize_value(
collections.OrderedDict(a=tf.constant([1, 2, 3]),
b=tf.constant(42.0)), expected_type_spec)
value = executor.create_value(value_pb)
self.assertIsInstance(value, executor_bindings.OwnedValueId)
# Assert the value ID was incremented.
self.assertEqual(value.ref, 1)
self.assertEqual(str(value), '1')
self.assertEqual(repr(value), r'<OwnedValueId: 1>')
materialized_value = executor.materialize(value.ref)
deserialized_value, type_spec = value_serialization.deserialize_value(
materialized_value)
# Note: here we've lost the names `a` and `b` in the output. The output
# is a more _strict_ type.
self.assertTrue(expected_type_spec.is_assignable_from(type_spec))
deserialized_value = type_conversions.type_to_py_container(
deserialized_value, expected_type_spec)
self.assertAllClose(deserialized_value,
collections.OrderedDict(a=(1, 2, 3), b=42.0))
# 3. Test creating a value from a computation.
@tensorflow_computation.tf_computation(tf.int32, tf.int32)
def foo(a, b):
return tf.add(a, b)
value_pb, _ = value_serialization.serialize_value(foo)
value = executor.create_value(value_pb)
self.assertIsInstance(value, executor_bindings.OwnedValueId)
# Assert the value ID was incremented again.
self.assertEqual(value.ref, 2)
self.assertEqual(str(value), '2')
self.assertEqual(repr(value), '<OwnedValueId: 2>')
def test_create_tuple_of_value_sequence(self):
datasets = (tf.data.Dataset.range(5), tf.data.Dataset.range(5))
executor = executor_bindings.create_tensorflow_executor()
struct_of_sequence_type = StructType([
(None, SequenceType(datasets[0].element_spec)),
(None, SequenceType(datasets[0].element_spec))
])
arg_value_pb, _ = value_serialization.serialize_value(
datasets, struct_of_sequence_type)
arg = executor.create_value(arg_value_pb)
@tensorflow_computation.tf_computation(struct_of_sequence_type)
def preprocess(datasets):
def double_value(x):
return 2 * x
@tf.function
def add_preprocessing(ds1, ds2):
return ds1.map(double_value), ds2.map(double_value)
return add_preprocessing(*datasets)
comp_pb = executor_pb2.Value(
computation=preprocess.get_proto(preprocess))
comp = executor.create_value(comp_pb)
result = executor.create_call(comp.ref, arg.ref)
output_pb = executor.materialize(result.ref)
result, result_type_spec = value_serialization.deserialize_value(
output_pb, type_hint=struct_of_sequence_type)
type_test_utils.assert_types_identical(result_type_spec,
struct_of_sequence_type)
def test_deserialize_federated_value_promotes_types(self):
x = [10]
smaller_type = computation_types.StructType([
(None, computation_types.to_type(tf.int32))
])
smaller_type_member_proto, _ = value_serialization.serialize_value(
x, smaller_type)
larger_type = computation_types.StructType([
('a', computation_types.to_type(tf.int32))
])
larger_type_member_proto, _ = value_serialization.serialize_value(
x, larger_type)
type_at_clients = type_serialization.serialize_type(
computation_types.at_clients(tf.int32))
unspecified_member_federated_type = computation_pb2.FederatedType(
placement=type_at_clients.federated.placement, all_equal=False)
federated_proto = executor_pb2.Value.Federated(
type=unspecified_member_federated_type,
value=[larger_type_member_proto, smaller_type_member_proto])
federated_value_proto = executor_pb2.Value(federated=federated_proto)
_, deserialized_type_spec = value_serialization.deserialize_value(
federated_value_proto)
type_test_utils.assert_types_identical(
deserialized_type_spec, computation_types.at_clients(larger_type))
def test_serialize_deserialize_variable_as_tensor_value(self): x = tf.Variable(10.0) type_spec = TensorType(tf.as_dtype(x.dtype), x.shape) value_proto, value_type = value_serialization.serialize_value(x, type_spec) type_test_utils.assert_types_identical(value_type, TensorType(tf.float32)) y, type_spec = value_serialization.deserialize_value(value_proto) type_test_utils.assert_types_identical(type_spec, TensorType(tf.float32)) self.assertAllEqual(x, y)
def test_serialize_deserialize_tensor_value_with_different_dtype(self):
x = tf.constant(10)
value_proto, value_type = value_serialization.serialize_value(
x, TensorType(tf.float32))
type_test_utils.assert_types_identical(value_type, TensorType(tf.float32))
y, type_spec = value_serialization.deserialize_value(value_proto)
type_test_utils.assert_types_identical(type_spec, TensorType(tf.float32))
self.assertEqual(y, 10.0)
def test_serialize_deserialize_nested_tuple_value_without_names(self): x = (10, 20) x_type = computation_types.to_type((tf.int32, tf.int32)) value_proto, value_type = value_serialization.serialize_value(x, x_type) type_test_utils.assert_types_identical(value_type, x_type) y, type_spec = value_serialization.deserialize_value(value_proto) type_test_utils.assert_types_equivalent(type_spec, x_type) self.assertEqual(y, structure.from_container((10, 20)))
def test_serialize_deserialize_value_proto_for_generic(self, np_generic):
type_spec = computation_types.TensorType(
dtype=np_generic.dtype, shape=np_generic.shape)
value_proto = value_serialization._value_proto_for_np_array(
np_generic, type_spec)
deserialized_array, deserialized_type = value_serialization.deserialize_value(
value_proto)
type_test_utils.assert_type_assignable_from(type_spec, deserialized_type)
self.assertAllEqual(np_generic, deserialized_array)
def get_value(self, value_id: str):
"""Retrieves a value using the `Compute` endpoint."""
response = self._stub.Compute(
executor_pb2.ComputeRequest(
executor=self._executor_pb,
value_ref=executor_pb2.ValueRef(id=value_id)))
py_typecheck.check_type(response, executor_pb2.ComputeResponse)
value, _ = value_serialization.deserialize_value(response.value)
return value
def test_serialize_deserialize_tensor_value_without_hint(
self, x, serialize_type_spec):
value_proto, value_type = value_serialization.serialize_value(
x, serialize_type_spec)
type_test_utils.assert_types_identical(value_type, serialize_type_spec)
y, type_spec = value_serialization.deserialize_value(value_proto)
type_test_utils.assert_types_identical(type_spec, serialize_type_spec)
self.assertEqual(y.dtype, serialize_type_spec.dtype.as_numpy_dtype)
self.assertAllEqual(x, y)
async def _compute(self, value_ref, type_spec):
self._check_has_executor_id()
py_typecheck.check_type(value_ref, executor_pb2.ValueRef)
request = executor_pb2.ComputeRequest(
executor=self._executor_id, value_ref=value_ref)
response = self._stub.compute(request)
py_typecheck.check_type(response, executor_pb2.ComputeResponse)
value, _ = value_serialization.deserialize_value(response.value, type_spec)
return value
def test_serialize_deserialize_federated_at_server(self):
x = 10
x_type = computation_types.at_server(tf.int32)
value_proto, value_type = value_serialization.serialize_value(x, x_type)
type_test_utils.assert_types_identical(
value_type, computation_types.at_server(tf.int32))
y, type_spec = value_serialization.deserialize_value(value_proto)
type_test_utils.assert_types_identical(type_spec, x_type)
self.assertEqual(y, 10)
def test_serialize_deserialize_value_proto_for_array_undefined_shape(self):
ndarray = np.array([10])
type_spec = computation_types.TensorType(dtype=ndarray.dtype, shape=[None])
value_proto = value_serialization._value_proto_for_np_array(
ndarray, type_spec)
deserialized_array, deserialized_type = value_serialization.deserialize_value(
value_proto)
type_test_utils.assert_type_assignable_from(type_spec, deserialized_type)
self.assertAllEqual(ndarray, deserialized_array)
def test_serialize_deserialize_tensor_value_with_nontrivial_shape(self):
x = tf.constant([10, 20, 30])
value_proto, value_type = value_serialization.serialize_value(
x, TensorType(tf.int32, [3]))
type_test_utils.assert_types_identical(value_type,
TensorType(tf.int32, [3]))
y, type_spec = value_serialization.deserialize_value(value_proto)
type_test_utils.assert_types_identical(type_spec, TensorType(tf.int32, [3]))
self.assertAllEqual(x, y)
def test_serialize_deserialize_tensor_value_unk_shape_without_hint(self):
x = np.asarray([1., 2.])
serialize_type_spec = TensorType(tf.float32, [None])
value_proto, value_type = value_serialization.serialize_value(
x, serialize_type_spec)
type_test_utils.assert_type_assignable_from(value_type, serialize_type_spec)
y, type_spec = value_serialization.deserialize_value(value_proto)
type_test_utils.assert_type_assignable_from(serialize_type_spec, type_spec)
self.assertEqual(y.dtype, serialize_type_spec.dtype.as_numpy_dtype)
self.assertAllEqual(x, y)
def test_serialize_deserialize_sequence_of_scalars(self, dataset_fn):
ds = tf.data.Dataset.range(5).map(lambda x: x * 2)
ds_repr = dataset_fn(ds)
value_proto, value_type = value_serialization.serialize_value(
ds_repr, computation_types.SequenceType(tf.int64))
type_test_utils.assert_types_identical(
value_type, computation_types.SequenceType(tf.int64))
y, type_spec = value_serialization.deserialize_value(value_proto)
type_test_utils.assert_types_identical(
type_spec, computation_types.SequenceType(tf.int64))
self.assertAllEqual(list(y), [x * 2 for x in range(5)])
def test_serialize_deserialize_tensor_value(self):
x = tf.constant(10.0)
type_spec = computation_types.TensorType(tf.as_dtype(x.dtype), x.shape)
value_proto, value_type = value_serialization.serialize_value(x, type_spec)
self.assertIsInstance(value_proto, serialization_bindings.Value)
self.assert_types_identical(value_type,
computation_types.TensorType(tf.float32))
y, type_spec = value_serialization.deserialize_value(value_proto)
self.assert_types_identical(type_spec,
computation_types.TensorType(tf.float32))
self.assertAllEqual(x, y)
def test_serialize_deserialize_sequence_of_scalars_graph_mode(self):
with tf.Graph().as_default():
ds = tf.data.Dataset.range(5).map(lambda x: x * 2)
value_proto, value_type = value_serialization.serialize_value(
ds, computation_types.SequenceType(tf.int64))
type_test_utils.assert_types_identical(
value_type, computation_types.SequenceType(tf.int64))
y, type_spec = value_serialization.deserialize_value(value_proto)
type_test_utils.assert_types_identical(
type_spec, computation_types.SequenceType(tf.int64))
self.assertAllEqual(list(y), [x * 2 for x in range(5)])
def test_serialize_deserialize_tensor_value_with_different_dtype(self):
x = tf.constant(10.0)
value_proto, value_type = value_serialization.serialize_value(
x, computation_types.TensorType(tf.int32))
self.assertIsInstance(value_proto, serialization_bindings.Value)
self.assert_types_identical(value_type,
computation_types.TensorType(tf.int32))
y, type_spec = value_serialization.deserialize_value(value_proto)
self.assert_types_identical(type_spec,
computation_types.TensorType(tf.int32))
self.assertEqual(y, 10)
def test_serialize_deserialize_tensor_value_with_nontrivial_shape(self):
x = tf.constant([10, 20, 30])
value_proto, value_type = value_serialization.serialize_value(
x, computation_types.TensorType(tf.int32, [3]))
self.assertIsInstance(value_proto, serialization_bindings.Value)
self.assert_types_identical(value_type,
computation_types.TensorType(tf.int32, [3]))
y, type_spec = value_serialization.deserialize_value(value_proto)
self.assert_types_identical(type_spec,
computation_types.TensorType(tf.int32, [3]))
self.assertAllEqual(x, y)
def test_serialize_deserialize_sequence_of_scalars(self):
ds = tf.data.Dataset.range(5).map(lambda x: x * 2)
value_proto, value_type = value_serialization.serialize_value(
ds, computation_types.SequenceType(tf.int64))
self.assertIsInstance(value_proto, serialization_bindings.Value)
self.assert_types_identical(value_type,
computation_types.SequenceType(tf.int64))
y, type_spec = value_serialization.deserialize_value(value_proto)
self.assert_types_identical(type_spec,
computation_types.SequenceType(tf.int64))
self.assertAllEqual(list(y), [x * 2 for x in range(5)])
def test_serialize_deserialize_federated_at_clients(self):
x = [10, 20]
x_type = computation_types.at_clients(tf.int32)
value_proto, value_type = value_serialization.serialize_value(x, x_type)
self.assertIsInstance(value_proto, serialization_bindings.Value)
self.assert_types_identical(value_type,
computation_types.at_clients(tf.int32))
y, type_spec = value_serialization.deserialize_value(value_proto)
self.assert_types_identical(type_spec,
computation_types.at_clients(tf.int32))
self.assertEqual(y, [10, 20])
def test_serialize_deserialize_sequence_of_ragged_tensors(self, dataset_fn):
self.skipTest('b/235492749')
ds = tf.data.Dataset.from_tensor_slices(tf.strings.split(['a b c', 'd e']))
ds_repr = dataset_fn(ds)
value_proto, value_type = value_serialization.serialize_value(
ds_repr, computation_types.SequenceType(element=ds.element_spec))
expected_type = computation_types.SequenceType(ds.element_spec)
type_test_utils.assert_types_identical(value_type, expected_type)
_, type_spec = value_serialization.deserialize_value(value_proto)
# Only checking for equivalence, we don't have the Python container
# after deserialization.
type_test_utils.assert_types_equivalent(type_spec, expected_type)
def test_serialize_deserialize_string_value(self, x):
tf_type = tf.as_dtype(x.dtype)
type_spec = TensorType(tf_type, x.shape)
value_proto, value_type = value_serialization.serialize_value(x, type_spec)
type_test_utils.assert_types_identical(value_type,
TensorType(tf_type, x.shape))
y, type_spec = value_serialization.deserialize_value(
value_proto, type_hint=type_spec)
type_test_utils.assert_types_identical(type_spec,
TensorType(tf_type, x.shape))
self.assertIsInstance(y, bytes)
self.assertAllEqual(x, y)
def test_call_no_arg(self):
executor = executor_bindings.create_tensorflow_executor()
@computations.tf_computation
def foo():
return tf.constant(123.0)
comp_pb = serialization_bindings.Value(computation=foo.get_proto(foo))
comp = executor.create_value(comp_pb)
result = executor.create_call(comp.ref, None)
result_value_pb = executor.materialize(result.ref)
result_tensor, _ = value_serialization.deserialize_value(result_value_pb)
self.assertEqual(result_tensor, 123.0)
def test_serialize_deserialize_tensor_value_without_hint_graph_mode(
self, x, serialize_type_spec):
# Test serializing non-eager tensors.
with tf.Graph().as_default():
tensor_x = tf.convert_to_tensor(x)
value_proto, value_type = value_serialization.serialize_value(
tensor_x, serialize_type_spec)
type_test_utils.assert_types_identical(value_type, serialize_type_spec)
y, deserialize_type_spec = value_serialization.deserialize_value(
value_proto)
type_test_utils.assert_types_identical(deserialize_type_spec,
serialize_type_spec)
self.assertEqual(y.dtype, serialize_type_spec.dtype.as_numpy_dtype)
self.assertAllEqual(x, y)
def test_serialize_deserialize_nested_tuple_value_with_names(self):
x = collections.OrderedDict(
a=10, b=[20, 30], c=collections.OrderedDict(d=40))
x_type = computation_types.to_type(
collections.OrderedDict(
a=tf.int32,
b=[tf.int32, tf.int32],
c=collections.OrderedDict(d=tf.int32)))
value_proto, value_type = value_serialization.serialize_value(x, x_type)
type_test_utils.assert_types_identical(value_type, x_type)
y, type_spec = value_serialization.deserialize_value(value_proto)
# Don't assert on the Python container since it is lost in serialization.
type_test_utils.assert_types_equivalent(type_spec, x_type)
self.assertEqual(y, structure.from_container(x, recursive=True))
def test_roundtrip_sequence_of_sparse_tensors(self):
self.skipTest('b/225927855')
tensors = (tf.constant([(1, 1)], dtype=tf.int64), (1.,),
tf.constant((3, 3), dtype=tf.int64))
ds = tf.data.Dataset.from_tensors(tensors)
ds = ds.map(lambda i, v, s: tf.SparseTensor(i, v, s)) # pylint: disable=unnecessary-lambda
serialized_value, _ = value_serialization.serialize_value(
ds, computation_types.SequenceType(ds.element_spec))
deserialized_ds, _ = value_serialization.deserialize_value(serialized_value)
self.assertEqual(ds.element_spec, deserialized_ds.element_spec)
self.assertAllEqual(
list(deserialized_ds), [tf.sparse.SparseTensor([(1, 1)], (1.), (3, 3))])
def test_serialize_deserialize_computation_value(self):
@tensorflow_computation.tf_computation
def comp():
return tf.constant(10)
value_proto, value_type = value_serialization.serialize_value(comp)
self.assertEqual(value_proto.WhichOneof('value'), 'computation')
type_test_utils.assert_types_identical(
value_type,
computation_types.FunctionType(parameter=None, result=tf.int32))
comp, type_spec = value_serialization.deserialize_value(value_proto)
# self.assertIsInstance(comp, computation_pb2.Computation)
type_test_utils.assert_types_identical(
type_spec,
computation_types.FunctionType(parameter=None, result=tf.int32))
