def test_lrn():
input_image_shape = (2, 3, 2, 1)
input_image = np.arange(int(
np.prod(input_image_shape))).reshape(input_image_shape).astype('f')
runtime = get_runtime()
model = ng.lrn(ng.constant(input_image),
alpha=1.0,
beta=2.0,
bias=1.0,
size=3)
computation = runtime.computation(model)
result = computation()
assert np.allclose(
result,
np.array([[[[0.0], [0.05325444]], [[0.03402646], [0.01869806]],
[[0.06805293], [0.03287071]]],
[[[0.00509002], [0.00356153]], [[0.00174719], [0.0012555]],
[[0.00322708], [0.00235574]]]],
dtype=np.float32))
# Test LRN default parameter values
model = ng.lrn(ng.constant(input_image))
computation = runtime.computation(model)
result = computation()
assert np.allclose(
result,
np.array([[[[0.0], [0.35355338]], [[0.8944272], [1.0606602]],
[[1.7888544], [1.767767]]],
[[[0.93704253], [0.97827977]], [[1.2493901], [1.2577883]],
[[1.5617375], [1.5372968]]]],
dtype=np.float32))
def LRN(onnx_node,
ng_inputs): # type: (NodeWrapper, List[NgraphNode]) -> NgraphNode
"""Carry out Local Region Normalization.
:param onnx_node: The ONNX node representation of LRN.
:param ng_inputs: The input data node.
:return: LRN output node.
"""
data = ng_inputs[0]
alpha = onnx_node.get_attribute_value('alpha', 1e-4)
beta = onnx_node.get_attribute_value('beta', 0.75)
bias = onnx_node.get_attribute_value('bias', 1.0)
size = onnx_node.get_attribute_value('size')
if size is None:
raise ValueError('LRN node (%s): required `size` attribute is missing',
onnx_node.name)
return ng.lrn(data, alpha, beta, bias, size)
