PyRSF is a rate-and-state friction (RSF) modelling package written in Python (2.7 and 3+ compatible). It features:
- Forward RSF modelling
- Inversion of the RSF parameters to data
- Adaptive or dense time-stepping procedures for forward modelling and inversion/inference
- Bayesian inference / Markov Chain Monte Carlo (MCMC)
- A modified RSF formulation with cut-off velocity
- User-defined state evolution functions
- Stick-slip simulations with stable limit cycles for K < Kc, facilitated by radiation damping
There are various other RSF modelling tools out there, such as Chris Marone's xlook and John Leeman's rsfmodel toolkit. PyRSF is complementary to these.
PyRSF depends on various numerical and scientific libraries, including SciPy and emcee, as well as the plotting libraries matplotlib and seaborn.
- Make a to-do list
- Bayesian inference and uncertainty estimation
- Adaptive time-stepping
- iPython notebooks
- Inversion of the RSF framework to stick-slip data
- Re-introduction of multiple state parameters