def TensorRepresentations(self) -> tensor_adapter.TensorRepresentations:
result = tensor_rep_util.GetTensorRepresentationsFromSchema(
self._schema)
if result is None:
result = (tensor_rep_util.InferTensorRepresentationsFromSchema(
self._schema))
return result
def testInferTensorRepresentationsFromSchema(
self,
ascii_proto,
expected,
generate_legacy_feature_spec=False,
schema_is_mixed=False):
if not _IS_LEGACY_SCHEMA and generate_legacy_feature_spec:
raise self.skipTest('This test exersizes legacy inference logic, but the '
'schema is not legacy schema.')
schema = text_format.Parse(ascii_proto, schema_pb2.Schema())
if _IS_LEGACY_SCHEMA:
schema.generate_legacy_feature_spec = generate_legacy_feature_spec
expected_protos = {
k: text_format.Parse(pbtxt, schema_pb2.TensorRepresentation())
for k, pbtxt in expected.items()
}
if not schema_is_mixed:
self.assertEqual(
expected_protos,
tensor_representation_util.InferTensorRepresentationsFromSchema(
schema))
self.assertEqual(
expected_protos,
tensor_representation_util.InferTensorRepresentationsFromMixedSchema(
schema))
def testInferTensorRepresentationsFromSchemaInvalidSchema(
self, ascii_proto, error_msg, generate_legacy_feature_spec=False):
if not _IS_LEGACY_SCHEMA and generate_legacy_feature_spec:
print('Skipping test case: ', self.id(), file=sys.stderr)
return
schema = text_format.Parse(ascii_proto, schema_pb2.Schema())
if _IS_LEGACY_SCHEMA:
schema.generate_legacy_feature_spec = generate_legacy_feature_spec
with self.assertRaisesRegex(ValueError, error_msg):
tensor_representation_util.InferTensorRepresentationsFromSchema(schema)
def testInferTensorRepresentationsFromSchemaInvalidSchema(
self, ascii_proto, error_msg, generate_legacy_feature_spec=False):
if not _IS_LEGACY_SCHEMA and generate_legacy_feature_spec:
raise self.skipTest('This test exersizes legacy inference logic, but the '
'schema is not legacy schema.')
schema = text_format.Parse(ascii_proto, schema_pb2.Schema())
if _IS_LEGACY_SCHEMA:
schema.generate_legacy_feature_spec = generate_legacy_feature_spec
with self.assertRaisesRegex(ValueError, error_msg):
tensor_representation_util.InferTensorRepresentationsFromSchema(schema)
with self.assertRaisesRegex(ValueError, error_msg):
tensor_representation_util.InferTensorRepresentationsFromMixedSchema(
schema)
def testInferTensorRepresentationsFromSchema(
self, ascii_proto, expected, generate_legacy_feature_spec=False):
# Skip a test if it's testing legacy logic but the schema is not the
# legacy schema.
if not _IS_LEGACY_SCHEMA and generate_legacy_feature_spec:
print('Skipping test case: ', self.id(), file=sys.stderr)
return
schema = text_format.Parse(ascii_proto, schema_pb2.Schema())
if _IS_LEGACY_SCHEMA:
schema.generate_legacy_feature_spec = generate_legacy_feature_spec
expected_protos = {
k: text_format.Parse(pbtxt, schema_pb2.TensorRepresentation())
for k, pbtxt in expected.items()
}
self.assertEqual(
expected_protos,
tensor_representation_util.InferTensorRepresentationsFromSchema(
schema))
