search

OxfordIonTrapGroup / oitg

Python package of helper routines (result loading, fitting, etc) for the Oxford Ion-Trap Group (OITG).
https://oxfordiontrapgroup.github.io/oitg/
13 stars 8 forks source link

Global Fit Optimisation Routine #24

Open pathfinder49 opened 4 years ago

pathfinder49 commented 4 years ago

We may want a routine for global fit optimisation of binomial (or other) distributed data.

This should give the global optimum parameter estimates and their covariances. Ideally, it should be capable of producing a plot of P(parameters|data).

Broken out from this issue

pathfinder49 commented 4 years ago

Markov Chain Monte Carlo might me the appropreate method for this.

There are library implementations in PyMC and PyStan. A simple introduction is here A discussion can be found here

hartytp commented 4 years ago

FWIW I'm not convinced this would be a useful addition to oitg. The aim should be to cover 95% of cases well using robust simple code, and leave the tricky cases to users to fit offline. Writing a toolkit that can robustly do global optimization without significant amounts of per-case user input is a very hard problem.