NOTE - this package is still under development. API may change!
See the documentation at http://turbustat.readthedocs.org/
This package is aimed at facilitating comparisons between spectral line data cubes. Included in this package are several techniques described in the literature which aim to describe some property of a data cube (or its moments/integrated intensity). We extend these techniques to be used as comparisons.
Ideally, we require a distance metric to satisfy several properties. A full description is shown in Yeremi et al. (2014). The key properties are:
- cubes with similar physics should have a small distance
- unaffected by coordinate shifts
- sensitive to differences in physical scale
- independent of noise levels in the data
Not all of the metrics satisfy the idealized properties. A full description of all statistics in this package will be shown in Koch et al. (in prep).
Currently, the only way install TurbuStat is to clone the repository and run
python setup.py install
Due to conflicts with the dependencies, TurbuStat will NOT automatically install all dependencies (only numpy and astropy). To check if your version of python has all the dependencies installed, run:
python setup.py check_deps
Requires:
- matplotlib-1.2
- scipy-0.12
- sklearn-0.13.0
- pandas-0.13
- statsmodels-0.4.0
- astrodendro-dev
- signal-id
- spectral-cube
This package was developed by: