I wrote this for my PhD. I haven’t found the actual experiment files, but the method can be recreated from the formulas if there’s any interest in that. I can probably refine it a little bit if I implement it again.