def Predict(self, request, context):
self.logger.info(
"Received inference request: {}".format(request)[:256])
numpy_outputs = {}
try:
numpy_request_inputs: Dict[str] = {
k: tensor_proto_to_np(t)
for k, t in request.inputs.items()
}
numpy_outputs: Dict[str] = self.executable(**numpy_request_inputs)
# If TensorProto is returned, than pass it. If Numpy is returned, cast it to TensorProto
tensor_proto_outputs: Dict[str, Tensor] = {
k: (v if isinstance(v, Tensor) else np_to_tensor_proto(v))
for k, v in numpy_outputs.items()
}
result = PredictResponse(outputs=tensor_proto_outputs)
self.logger.info("Answer: {}".format(result)[:256])
return result
except ValueError as e:
self.logger.exception(
"Could not convert numpy output ({}) to tensor proto. {}".
format(numpy_outputs, e))
context.abort(grpc.StatusCode.OUT_OF_RANGE, repr(e))
except Exception as ex:
self.logger.exception(
"Function {} failed to handle request".format(
self.signature.signature_name))
context.abort(grpc.StatusCode.INTERNAL, repr(ex))
def test_int_scalar_tensor_to_np_scalar_back_to_tensor(self, shape):
tp_kwargs = {"int64_val": [1], "dtype": DT_INT64, "tensor_shape": shape}
original_tensor_proto = Tensor(**tp_kwargs)
np_representation = tensor_proto_to_np(original_tensor_proto)
restored_tensor_proto = np_to_tensor_proto(np_representation)
assert restored_tensor_proto == original_tensor_proto
def test_half_dtype_conversion(self, dtype):
tensor_shape = TensorShape(dims=[3,1])
tp_kwargs = {DTYPE_TO_FIELDNAME[dtype]: np.array([1.10, 2.20, 3.30], dtype=np.float16).view(np.uint16),
"dtype": dtype,
"tensor_shape": tensor_shape}
original_tensor_proto = Tensor(**tp_kwargs)
np_representation = tensor_proto_to_np(original_tensor_proto)
restored_tensor_proto = np_to_tensor_proto(np_representation)
assert restored_tensor_proto == original_tensor_proto
def test_tensor_to_np_array_to_tensor_float(self, dtype):
tensor_shape = TensorShape(dims=[3, 1])
tp_kwargs = {DTYPE_TO_FIELDNAME[dtype]: [1.10, 2.20, 3.30],
"dtype": dtype,
"tensor_shape": tensor_shape}
original_tensor_proto = Tensor(**tp_kwargs)
np_representation = tensor_proto_to_np(original_tensor_proto)
restored_tensor_proto = np_to_tensor_proto(np_representation)
assert restored_tensor_proto == original_tensor_proto
def test_np_scalar_to_tensor_to_np(self, np_dtype):
x = np.array([1.0], dtype=np_dtype)[0]
tensor_proto = np_to_tensor_proto(x)
x_restored = tensor_proto_to_np(tensor_proto)
assert x == x_restored
def test_complex_scalar_to_tensor_to_np(self, dt):
x = np.array([-1 - 1j], dtype=dt)[0]
tensor_proto = np_to_tensor_proto(x)
x_restored = tensor_proto_to_np(tensor_proto)
assert x == x_restored
def test_complex_np_to_tensor_to_np(self, dt):
x = np.array([-1 - 1j, -1 + 1j, +1 - 1j, +1 + 1j], dtype=dt)
tensor_proto = np_to_tensor_proto(x)
x_restored = tensor_proto_to_np(tensor_proto)
assert np.all(x == x_restored)
def test_str_scalar_to_tensor_and_back(self):
x = np.str("a")
tensor_proto = np_to_tensor_proto(x)
x_restored = tensor_proto_to_np(tensor_proto)
assert x == x_restored
def test_str_np_to_tensor_to_np(self):
x = np.array(["a", "b", "c"], dtype=np.str)
tensor_proto = np_to_tensor_proto(x)
x_restored = tensor_proto_to_np(tensor_proto)
assert np.all(x == x_restored)
def test_bool_scalar_to_tensor_and_back(self):
x = np.bool()
tensor_proto = np_to_tensor_proto(x)
x_restored = tensor_proto_to_np(tensor_proto)
assert x == x_restored
def test_bool_np_to_tensor_to_np(self):
x = np.array([True, False, True], dtype=np.bool)
tensor_proto = np_to_tensor_proto(x)
x_restored = tensor_proto_to_np(tensor_proto)
assert np.all(x == x_restored)
def test_unsupported_np_to_tensor_to_np(self, np_dtype):
x = np.array([1.0, 2.0, 3.0], dtype=np_dtype)
tensor_proto = np_to_tensor_proto(x)
x_restored = tensor_proto_to_np(tensor_proto)
assert np.all(x == x_restored)