Skip to content

K2 Systematics Correction

Python package for K2 systematics correction using Gaussian processes.

Contributors

Hannu Parviainen, Suzanne Aigrain, Benjamin Pope

Installation

git clone https://github.com/OxES/k2sc.git
cd k2sc
python setup.py install --user

Basic usage - lightkurve interface

The easiest way to use k2sc is together with the standard Kepler/K2/TESS data analysis package lightkurve. See tutorial to the left!

Basic usage - Command Line

A MAST K2 light curve can be detrended by calling

k2sc <filename>

where <filename> is either a MAST light curve filename, list of files, or a directory.

Useful flags

  • --flux-type can be either pdc or sap
  • --de-max-time <ss> maximum time (in seconds) to run global GP hyperparameter optimization (differential evolution) before switching to local optimization.
  • --de-npop <nn> size of the de population, can be set to 50 to speed up the optimization.
  • --save-dir <path> defines where to save the detrended files
  • --logfile <filename>

MPI

K2SC supports MPI automatically (requires MPI4Py.) Call k2sc as

mpirun -n N k2sc <files>

where <files> is a list of files or a directory to be detrended (for example, path/to/ktwo*.fits).

Requires

  • NumPy, SciPy, astropy, George, MPI4Py

Citing

If you use K2SC in your research, please cite

Aigrain, S., Parviainen, H. & Pope, B. (2016, accepted to MNRAS), arXiv:1603.09167

or use this ready-made BibTeX entry

@article{Aigrain2016,
    arxivId = {1603.09167},
    author = {Aigrain, Suzanne and Parviainen, Hannu and Pope, Benjamin},
    keywords = {data analysis,methods,photometry,planetary systems,techniques},
    title = {{K2SC: Flexible systematics correction and detrending of K2 light curves using Gaussian Process regression}},
    url = {http://arxiv.org/abs/1603.09167},
    year = {2016}
}