I had a meeting with Mike today to discuss the splitting of the model. He is worried about the compute costs of rebuilding the model every time. I showed him the demos that I had worked on and we realised that just the predict part using an existing model was using just as much compute time as the build-predict model and therefore didn’t need splitting.
Mike has now asked me to improve that dashboard so that he can show potential funders. I’m not sure the data is robust enough for this yet, but he feels that it is, so that will be my work over the next few days.