How to use the onnxruntime.get_device function in onnxruntime

def supports_device(cls, device):
        """
        Check whether the backend is compiled with particular device support.
        In particular it's used in the testing suite.
        """
        return device in get_device()

name = datasets.get_example("logreg_iris.onnx")
model = load(name)

rep = backend.prepare(model, 'CPU')
x = np.array([[-1.0, -2.0, 5.0, 6.0],
              [-1.0, -2.0, -3.0, -4.0],
              [-1.0, -2.0, 7.0, 8.0]],
             dtype=np.float32)
label, proba = rep.run(x)
print("label={}".format(label))
print("probabilities={}".format(proba))

########################################
# The device depends on how the package was compiled,
# GPU or CPU.
print(get_device())

########################################
# The backend can also directly load the model
# without using *onnx*.

rep = backend.prepare(name, 'CPU')
x = np.array([[-1.0, -2.0, -3.0, -4.0],
              [-1.0, -2.0, -3.0, -4.0],
              [-1.0, -2.0, -3.0, -4.0]],
             dtype=np.float32)
label, proba = rep.run(x)
print("label={}".format(label))
print("probabilities={}".format(proba))

#######################################
# The backend API is implemented by other frameworks

model = load(name)

rep = backend.prepare(model, 'CPU')
x = np.array([[-1.0, -2.0]], dtype=np.float32)
try:
    label, proba = rep.run(x)
    print("label={}".format(label))
    print("probabilities={}".format(proba))
except (RuntimeError, InvalidArgument) as e:
    print(e)

########################################
# The device depends on how the package was compiled,
# GPU or CPU.
from onnxruntime import get_device
print(get_device())

########################################
# The backend can also directly load the model
# without using *onnx*.

rep = backend.prepare(name, 'CPU')
x = np.array([[-1.0, -2.0]], dtype=np.float32)
try:
    label, proba = rep.run(x)
    print("label={}".format(label))
    print("probabilities={}".format(proba))
except (RuntimeError, InvalidArgument) as e:
    print(e)

model = load(name)

rep = backend.prepare(model, 'CPU')
x = np.array([[-1.0, -2.0]], dtype=np.float32)
try:
    label, proba = rep.run(x)
    print("label={}".format(label))
    print("probabilities={}".format(proba))
except (RuntimeError, InvalidArgument) as e:
    print(e)

########################################
# The device depends on how the package was compiled,
# GPU or CPU.
from onnxruntime import get_device
print(get_device())

########################################
# The backend can also directly load the model
# without using *onnx*.

rep = backend.prepare(name, 'CPU')
x = np.array([[-1.0, -2.0]], dtype=np.float32)
try:
    label, proba = rep.run(x)
    print("label={}".format(label))
    print("probabilities={}".format(proba))
except (RuntimeError, InvalidArgument) as e:
    print(e)

def is_compatible(cls, model, device=None, **kwargs):
        """
        Return whether the model is compatible with the backend.

        :param model: unused
        :param device: None to use the default device or a string (ex: `'CPU'`)
        :return: boolean
        """
        if device is None:
            device = get_device()
        return cls.supports_device(device)

"""
        if isinstance(model, OnnxRuntimeBackendRep):
            return model
        elif isinstance(model, InferenceSession):
            return OnnxRuntimeBackendRep(model)
        elif isinstance(model, (str, bytes)):
            options = SessionOptions()
            for k, v in kwargs.items():
                if hasattr(options, k):
                    setattr(options, k, v)
            inf = InferenceSession(model, options)
            # backend API is primarily used for ONNX test/validation. As such, we should disable session.run() fallback
            # which may hide test failures.
            inf.disable_fallback()
            if device is not None and not cls.supports_device(device):
                raise RuntimeError("Incompatible device expected '{0}', got '{1}'".format(device, get_device()))
            return cls.prepare(inf, device, **kwargs)
        else:
            # type: ModelProto
            check_model(model)
            bin = model.SerializeToString()
            return cls.prepare(bin, device, **kwargs)