Create custom mlflow PythonModel with `predict()` function that takes no arguments

Question

I'd like to create a custom mlflow PythonModel whose predict function does not take any arguments. It looks like mlflow.pyfunc.load_model (which I use to reover the custom moel) returns a PyFuncModel whose .predict method requires a data argument. Is there any clever way to get around that?

Side note: I know it's odd to have a model whose predict function doesn't require arguments, but I have my reasons!