onnx.backend

Backend

class onnx.backend.base.Backend

classmethod run_node(node: NodeProto, inputs: Any, device: str = 'CPU', outputs_info: Optional[Sequence[Tuple[dtype, Tuple[int, ...]]]] = None, **kwargs: Dict[str, Any]) → Optional[Tuple[Any, ...]]

Simple run one operator and return the results. :param outputs_info: a list of tuples, which contains the element type and :param shape of each output. First element of the tuple is the dtype: :param and: :param the second element is the shape. More use case can be found in: :param https: //github.com/onnx/onnx/blob/main/onnx/backend/test/runner/__init__.py

classmethod supports_device(device: str) → bool

Checks whether the backend is compiled with particular device support. In particular it’s used in the testing suite.

BackendRep

class onnx.backend.base.BackendRep

collect_snippets

Device

class onnx.backend.base.Device(device: str)

Describes device type and device id syntax: device_type:device_id(optional) example: ‘CPU’, ‘CUDA’, ‘CUDA:1’

DeviceType

class onnx.backend.base.DeviceType

load_model_tests

Runner

class onnx.backend.test.runner.Runner(backend: Type[Backend], parent_module: Optional[str] = None)

property test_cases: Dict[str, Type[TestCase]]

List of test cases to be applied on the parent scope Example usage:

globals().update(BackendTest(backend).test_cases)

property test_suite: TestSuite

TestSuite that can be run by TestRunner Example usage:

unittest.TextTestRunner().run(BackendTest(backend).test_suite)

property tests: Type[TestCase]

One single unittest.TestCase that hosts all the test functions Example usage:

onnx_backend_tests = BackendTest(backend).tests

TestCase

class onnx.backend.test.case.test_case.TestCase(name: str, model_name: str, url: Optional[str], model_dir: Optional[str], model: Optional[onnx.onnx_ml_pb2.ModelProto], data_sets: Optional[Sequence[Tuple[Sequence[numpy.ndarray], Sequence[numpy.ndarray]]]], kind: str, rtol: float, atol: float, __test__: bool = False)