def _function_calls(self, query, include_implicit, include_metatypes):
exec = ex.TraversalExecutor(self._tx)
schema_concept_types = exec.get_schema_concept_types(
query,
include_implicit=include_implicit,
include_metatypes=include_metatypes)
labels = trv.labels_from_types(schema_concept_types)
return labels
def test_integration(self):
client = grakn.client.GraknClient(uri=TEST_URI)
session = client.session(keyspace=TEST_KEYSPACE)
tx = session.transaction().write()
print("================= THINGS ======================")
te = ex.TraversalExecutor(tx)
schema_concept_types = te.get_schema_concept_types(
encode.GET_THING_TYPES_QUERY,
include_implicit=True,
include_metatypes=False)
labels = trv.labels_from_types(schema_concept_types)
print(list(labels))
schema_concept_types = te.get_schema_concept_types(
encode.GET_THING_TYPES_QUERY,
include_implicit=True,
include_metatypes=False)
super_types = trv.get_sups_labels_per_type(schema_concept_types,
include_self=True,
include_metatypes=False)
print("==== super types ====")
[print(type, super_types) for type, super_types in super_types.items()]
print("================= ROLES ======================")
schema_concept_types = te.get_schema_concept_types(
encode.GET_ROLE_TYPES_QUERY,
include_implicit=True,
include_metatypes=False)
labels = trv.labels_from_types(schema_concept_types)
print(list(labels))
schema_concept_types = te.get_schema_concept_types(
encode.GET_ROLE_TYPES_QUERY,
include_implicit=True,
include_metatypes=False)
super_types = trv.get_sups_labels_per_type(schema_concept_types,
include_self=True,
include_metatypes=False)
print("==== super types ====")
[print(type, super_types) for type, super_types in super_types.items()]
def test_integration(self):
client = grakn.Grakn(uri="localhost:48555")
session = client.session(keyspace="test_schema")
tx = session.transaction(grakn.TxType.WRITE)
print("================= THINGS ======================")
te = ex.TraversalExecutor(tx)
schema_concept_types = te.get_schema_concept_types(
encode.GET_THING_TYPES_QUERY,
include_implicit=True,
include_metatypes=False)
labels = trv.labels_from_types(schema_concept_types)
print(list(labels))
schema_concept_types = te.get_schema_concept_types(
encode.GET_THING_TYPES_QUERY,
include_implicit=True,
include_metatypes=False)
super_types = trv.get_sups_labels_per_type(schema_concept_types,
include_self=True,
include_metatypes=False)
print("==== super types ====")
[print(type, super_types) for type, super_types in super_types.items()]
print("================= ROLES ======================")
schema_concept_types = te.get_schema_concept_types(
encode.GET_ROLE_TYPES_QUERY,
include_implicit=True,
include_metatypes=False)
labels = trv.labels_from_types(schema_concept_types)
print(list(labels))
schema_concept_types = te.get_schema_concept_types(
encode.GET_ROLE_TYPES_QUERY,
include_implicit=True,
include_metatypes=False)
super_types = trv.get_sups_labels_per_type(schema_concept_types,
include_self=True,
include_metatypes=False)
print("==== super types ====")
[print(type, super_types) for type, super_types in super_types.items()]
def __init__(self, schema_tx):
################################################################################################################
# Schema Traversals
################################################################################################################
# This depends upon the schema being the same for the keyspace used in training vs eval and predict
schema_traversal_executor = schema_ex.TraversalExecutor(schema_tx)
# THINGS
thing_schema_traversal = trav.traverse_schema(
schema_traversal_executor, GET_THING_TYPES_QUERY)
# ROLES
role_schema_traversal = trav.traverse_schema(schema_traversal_executor,
GET_ROLE_TYPES_QUERY)
role_schema_traversal['has'] = ['has']
############################################################################################################
# Encoders Initialisation
############################################################################################################
with tf.name_scope('encoding_init') as scope:
thing_encoder = schema.MultiHotSchemaTypeEncoder(
thing_schema_traversal, name='thing_encoder')
role_encoder = schema.MultiHotSchemaTypeEncoder(
role_schema_traversal, name='role_encoder')
# In case of issues https://github.com/tensorflow/hub/issues/61
string_encoder = tf_hub.TensorFlowHubEncoder(
"https://tfhub.dev/google/nnlm-en-dim128-with-normalization/1",
128)
data_types = list(neighbour.DATA_TYPE_NAMES)
data_types.insert(
0, NO_DATA_TYPE
) # For the case where an entity or relation is encountered
data_types_traversal = {
data_type: data_types
for data_type in data_types
}
# Later a hierarchy could be added to data_type meaning. e.g. long and double are both numeric
data_type_encoder = schema.MultiHotSchemaTypeEncoder(
data_types_traversal, name='data_type_encoder')
self._encoders = {
'role_type':
role_encoder,
'role_direction':
lambda x: tf.to_float(x, 'role_dir_to_float'),
'neighbour_type':
thing_encoder,
'neighbour_data_type':
lambda x: data_type_encoder(tf.convert_to_tensor(x)),
'neighbour_value_long':
lambda x: tf.to_float(x, name='long_to_float'),
'neighbour_value_double':
lambda x: x,
'neighbour_value_boolean':
lambda x: tf.to_float(boolean.one_hot_boolean_encode(
x, 'boolean_1_hot'),
name='boolean_to_float'),
'neighbour_value_date':
lambda x: tf.nn.l2_normalize(tf.to_float(x, name='date_to_float')),
'neighbour_value_string':
string_encoder
}