def test_initializer(self):
X = np.array([[1, 2], [3, 4]]).astype(np.float32)
Y = np.array([[1, 2], [3, 4]]).astype(np.float32)
weight = np.array([[1, 0], [0, 1]])
graph_def = make_graph(
[
make_node("Add", ["X", "Y"], ["Z0"]),
make_node("Cast", ["Z0"], ["Z"], to="float"),
make_node("Mul", ["Z", "weight"], ["W0"]),
make_node("Tanh", ["W0"], ["W1"]),
make_node("Sigmoid", ["W1"], ["W2"]),
make_node("Scale", ["W2"], ["W3"], scale=-1.0)
],
name="test_initializer",
inputs=[
make_tensor_value_info("X", onnx.TensorProto.FLOAT, (2, 2)),
make_tensor_value_info("Y", onnx.TensorProto.FLOAT, (2, 2)),
make_tensor_value_info("weight", onnx.TensorProto.FLOAT,
(2, 2)),
],
outputs=[
make_tensor_value_info("W3", onnx.TensorProto.FLOAT, (2, 2))
],
initializer=[
make_tensor("weight", onnx.TensorProto.FLOAT, [2, 2],
weight.flatten().astype(float))
])
def sigmoid(x):
return 1 / (1 + np.exp(-x))
W_ref = -sigmoid(np.tanh((X + Y) * weight))
c2_rep = c2.prepare(make_model(graph_def))
output = c2_rep.run({"X": X, "Y": Y})
np.testing.assert_almost_equal(output["W3"], W_ref)
def test_onnx_to_caffe2(self):
onnx_model = tempfile.NamedTemporaryFile()
output = tempfile.NamedTemporaryFile()
init_net_output = tempfile.NamedTemporaryFile()
node_def = helper.make_node(
"Relu", ["X"], ["Y"])
graph_def = helper.make_graph(
[node_def],
"test",
[helper.make_tensor_value_info("X", onnx_pb2.TensorProto.FLOAT, (2, 2))],
[helper.make_tensor_value_info("Y", onnx_pb2.TensorProto.FLOAT, (2, 2))],
initializer=[helper.make_tensor("X",
onnx_pb2.TensorProto.FLOAT,
[2, 2],
np.zeros((2, 2)).flatten().astype(float))])
model_def = make_model(graph_def, producer_name='onnx-to-caffe2-test')
onnx_model.write(model_def.SerializeToString())
onnx_model.flush()
result = self._run_command(
onnx_to_caffe2, [
onnx_model.name,
'--output', output.name,
'--init-net-output', init_net_output.name,
])
caffe2_net = caffe2_pb2.NetDef()
caffe2_net.ParseFromString(output.read())
self.assertEqual(len(caffe2_net.op), 1)
self.assertEqual(caffe2_net.op[0].type, 'Relu')
caffe2_init_net = caffe2_pb2.NetDef()
caffe2_init_net.ParseFromString(init_net_output.read())
self.assertEqual(len(caffe2_init_net.op), 1)
def caffe2_net_to_onnx_model(cls, *args, **kwargs):
model = make_model(cls.caffe2_net_to_onnx_graph(*args, **kwargs))
opset_id = OperatorSetIdProto()
opset_id.domain = '' # ONNX
opset_id.version = cls._target_opset_version
model.opset_import.extend([opset_id])
checker.check_model(model)
return model
def test_relu_graph(self):
X = np.random.randn(3, 2).astype(np.float32)
Y_ref = np.clip(X, 0, np.inf)
node_def = make_node(
"Relu", ["X"], ["Y"])
output = c2.run_node(
node_def, {"X": X})
np.testing.assert_almost_equal(output.Y, Y_ref)
graph_def = make_graph(
[node_def],
name="test",
inputs=[make_tensor_value_info("X", onnx.TensorProto.FLOAT, [3, 2])],
outputs=[make_tensor_value_info("Y", onnx.TensorProto.FLOAT, [3, 2])])
c2_rep = c2.prepare(make_model(graph_def))
output = c2_rep.run(X)
np.testing.assert_almost_equal(output.Y, Y_ref)
def caffe2_net_to_onnx_model(*args, **kwargs):
model_def = make_model(caffe2_net_to_onnx_graph(*args, **kwargs))
checker.check_model(model_def)
return model_def
def caffe2_net_to_onnx_model(cls, *args, **kwargs):
model = make_model(cls.caffe2_net_to_onnx_graph(*args, **kwargs))
checker.check_model(model)
return model
