mujpy

A Python \(\mu\)SR data analysis based on classes, with a graphical interface designed for jupyter, released under the GPL-3 licence.

It aims at the power of musrfit with the user-friendly appearance of mulab.

Version 3.1. Main technical features:

• a model built on two-letter bricks: mg, for Gaussian-damped cosine, ml for Lorentzian-damped cosine etc.

• sequential fits by the same model, driven by a run list and a list of grouping dictionaries, for asymmetry definition

• global fits by user-defined parameters assigned to model parameters in a json file

• a mulab-like gui interface in jupyterlab that allows fit model and parameter editing, hopefully with a gentler learning curve than musrfit

• direct standalone gui web interface by `voila`, included

Try mujpy!

on linux, or on Windows by lightweight WSL2 (~10GB of ubuntu-in-Windows). Follow simple installation instructions. Both os come with python, you only need to: create a venv python -m venv ~/.mujpy-env, activate it source ~/.mujpy-env/bin/activate, and invoke pip install mujpy. Now try your freshly installed mujpy from command line, to demonstrate its capabilities. Just type python -m mujpy.tests.tests and 16 fully automatic cases will popup, in order of growing complexity (find a description in the Tutorial).

There is also a GUI and its demos, see basic instructions on ReaTheDocs. Choice of jupyter notebook %matplotlib qt for popup plots, or %matplotlib widget for Log-tab side-by-side log+plot.

Please email bugs to the author.