Fail to convert new version(opset=12) of onnx model to Relay

Questions
#1

For onnx’s new version of model, for example Alexnet-12, its opset is 12.
when compiling new onnx model, an error appears in relay.frontend.from_onnx,
TVMError: Check failed: type_code_ == kDLFloat (8 vs. 2) : expected float but get Object.
demo code:

MODEL_PATH = 'model/AlexNet/bvlcalexnet-12.onnx'
input_name = "data_0"
batch, channel, height, weight = 1, 3, 224, 224
I_np = np.random.uniform(size=(batch, channel, height, weight)).astype(dtype)
onnx_model = onnx.load(MODEL_PATH)
shape_dict = {input_name: I_np.shape}
mod, params = relay.frontend.from_onnx(onnx_model, shape_dict)

TVM discussion has an old issue report about it.
Any advance?