General Toolkit for Modeling Radial Velocities.
Written by BJ Fulton and Erik Petigura, Fulton and Petigura (in prep.).
- Documentation is available on ReadTheDocs.org: http://radvel.readthedocs.io
- Object-oriented (i.e. models, likelihoods, priors, and posteriors are defined as objects)
- Extensible (i.e. naturally define new likelihoods, priors, parameterizatoins)
- Convenient API to fix/float parameters
- Easily plugs in to the the suite of
scipy.optimize
routines for max-likelihood fitting - Works with
emcee
MCMC - parameters are represented as dicts not arrays
- Can handle data from multiple telescopes
- Easily convert between different parameterizations
- Computation of Kepler's equation (numerically intensive) written in C
- Bundle up examples in convenient test cases
- PERF: Optimizations for low eccentricity orbits
- Streamline API
Follow examples in
radvel/tests/SyntheticData.ipynb
radvel/tests/EPIC-2037_Fitting+MCMC.ipynb
radvel/tests/164922_Fitting+MCMC.ipynb
You'll need the following dependencies
- emcee
- corner
- pandas (to read in hdf5)
- matplotlib-1.5.0
- cython (tested with 0.22)
- pdflatex installed and in your system's path