def create_tf_dataset_for_client(self, client_id):
tf_dataset = self._create_tf_dataset_fn(client_id)
tensor_utils.check_nested_equal(tf_dataset.output_types,
self._output_types)
tensor_utils.check_nested_equal(tf_dataset.output_shapes,
self._output_shapes)
return tf_dataset
def create_tf_dataset_for_client(self, client_id):
tf_dataset = self._create_dataset(client_id)
tensor_utils.check_nested_equal(
tf.compat.v1.data.get_output_types(tf_dataset), self._output_types)
tensor_utils.check_nested_equal(
tf.compat.v1.data.get_output_shapes(tf_dataset), self._output_shapes)
return tf_dataset
def create_tf_dataset_for_client(self, client_id):
if client_id not in self.client_ids:
raise ValueError(
"ID [{i}] is not a client in this ClientData. See "
"property `client_ids` for the list of valid ids.".format(
i=client_id))
tf_dataset = self._create_dataset(client_id)
tensor_utils.check_nested_equal(tf_dataset.element_spec,
self._element_type_structure)
return tf_dataset
def create_tf_dataset_for_client(self, client_id: str) -> tf.data.Dataset:
"""Creates a new `tf.data.Dataset` containing the client training examples.
This function will create a dataset for a given client if `client_id` is
contained in the `client_ids` property of the `FilePerUserClientData`.
Unlike `self.serializable_dataset_fn`, this method is not serializable.
Args:
client_id: The string identifier for the desired client.
Returns:
A `tf.data.Dataset` object.
"""
if client_id not in self.client_ids:
raise ValueError(
'ID [{i}] is not a client in this ClientData. See '
'property `client_ids` for the list of valid ids.'.format(
i=client_id))
client_dataset = self.serializable_dataset_fn(tf.constant(client_id))
tensor_utils.check_nested_equal(client_dataset.element_spec,
self._element_type_structure)
return client_dataset
def test_check_nested_equal(self):
nested_dict = {
'KEY1': {
'NESTED_KEY': 0
},
'KEY2': 1,
}
nested_list = [('KEY1', ('NESTED_KEY', 0)), ('KEY2', 1)]
flat_dict = {
'KEY1': 0,
'KEY2': 1,
}
nested_dtypes = {
'x': [tf.int32, tf.float32],
'y': tf.float32,
}
nested_shapes = {
# N.B. tf.TensorShape([None]) == tf.TensorShape([None])
# returns False, so we can't use a None shape here.
'x': [[1], [3, 5]],
'y': [1],
}
# Should not raise an exception.
tensor_utils.check_nested_equal(nested_dict, nested_dict)
tensor_utils.check_nested_equal(nested_list, nested_list)
tensor_utils.check_nested_equal(flat_dict, flat_dict)
tensor_utils.check_nested_equal(nested_dtypes, nested_dtypes)
tensor_utils.check_nested_equal(nested_shapes, nested_shapes)
with self.assertRaises(TypeError):
tensor_utils.check_nested_equal(nested_dict, nested_list)
with self.assertRaises(ValueError):
# Different nested structures.
tensor_utils.check_nested_equal(nested_dict, flat_dict)
# Same as nested_dict, but using float values. Equality still holds for
# 0 == 0.0 despite different types.
nested_dict_different_types = {
'KEY1': {
'NESTED_KEY': 0.0
},
'KEY2': 1.0,
}
nest.assert_same_structure(nested_dict, nested_dict_different_types)
# Same as nested_dict but with one different value
nested_dict_different_value = {
'KEY1': {
'NESTED_KEY': 0.5
},
'KEY2': 1.0,
}
with self.assertRaises(ValueError):
tensor_utils.check_nested_equal(nested_dict,
nested_dict_different_value)
tensor_utils.check_nested_equal([None], [None])
def always_neq(x, y):
del x, y
return False
with self.assertRaises(ValueError):
tensor_utils.check_nested_equal([1], [1], always_neq)
def create_tf_dataset_for_client(self, client_id):
tf_dataset = self._create_dataset(client_id)
tensor_utils.check_nested_equal(
tf.data.experimental.get_structure(tf_dataset),
self._element_type_structure)
return tf_dataset
def create_tf_dataset_for_client(self, client_id):
tf_dataset = self._create_tf_dataset_fn(client_id)
tensor_utils.check_nested_equal(tf_dataset.element_spec,
self._element_type_structure)
return tf_dataset
