version=model_version)
response = grpc_stub.ModelReady(request)
print("model {}".format(response))
# Metadata
request = service_pb2.ServerMetadataRequest()
response = grpc_stub.ServerMetadata(request)
print("server metadata:\n{}".format(response))
request = service_pb2.ModelMetadataRequest(name=model_name,
version=model_version)
response = grpc_stub.ModelMetadata(request)
print("model metadata:\n{}".format(response))
# Configuration
request = service_pb2.ModelConfigRequest(name=model_name,
version=model_version)
response = grpc_stub.ModelConfig(request)
print("model config:\n{}".format(response))
# Infer
request = service_pb2.ModelInferRequest()
request.model_name = model_name
request.model_version = model_version
request.id = "my request id"
input = service_pb2.ModelInferRequest().InferInputTensor()
input.name = "input"
input.datatype = "FP32"
input.shape.extend([1, 299, 299, 3])
request.inputs.extend([input])
nargs='?',
default=None,
help='Input image / Input folder.')
FLAGS = parser.parse_args()
# Create gRPC stub for communicating with the server
channel = grpc.insecure_channel(FLAGS.url)
grpc_stub = service_pb2_grpc.GRPCInferenceServiceStub(channel)
# Make sure the model matches our requirements, and get some
# properties of the model that we need for preprocessing
metadata_request = service_pb2.ModelMetadataRequest(
name=FLAGS.model_name, version=FLAGS.model_version)
metadata_response = grpc_stub.ModelMetadata(metadata_request)
config_request = service_pb2.ModelConfigRequest(name=FLAGS.model_name,
version=FLAGS.model_version)
config_response = grpc_stub.ModelConfig(config_request)
max_batch_size, input_name, output_name, c, h, w, format, dtype = parse_model(
metadata_response, config_response.config)
supports_batching = max_batch_size > 0
if not supports_batching and FLAGS.batch_size != 1:
raise Exception("This model doesn't support batching.")
# Send requests of FLAGS.batch_size images. If the number of
# images isn't an exact multiple of FLAGS.batch_size then just
# start over with the first images until the batch is filled.
requests = []
responses = []
result_filenames = []
version=model_version)
response = grpc_stub.ModelReady(request)
print("model {}".format(response))
# Metadata
request = service_pb2.ServerMetadataRequest()
response = grpc_stub.ServerMetadata(request)
print("server metadata:\n{}".format(response))
request = service_pb2.ModelMetadataRequest(name="resnet_v1_50_graphdef",
version=model_version)
response = grpc_stub.ModelMetadata(request)
print("model metadata:\n{}".format(response))
# Configuration
request = service_pb2.ModelConfigRequest(name="resnet_v1_50_graphdef",
version=model_version)
response = grpc_stub.ModelConfig(request)
print("model config:\n{}".format(response))
# Infer
request = service_pb2.ModelInferRequest()
request.model_name = "resnet_v1_50_graphdef"
request.model_version = model_version
request.id = "my request id"
input = service_pb2.ModelInferRequest().InferInputTensor()
input.name = "input"
input.datatype = "FP32"
input.shape.extend([1, 224, 224, 3])
request.inputs.extend([input])