Skip to content

fraserw/trippy

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

TRIPPy: Python based Trailed Source Photometry

(published in the Astrophysical Journal)

TRIPPy is a python package aimed to perform all the steps required to measure accurate photometry of both trailed and non-trailed (stationary) astronomical sources. This includes the ability to generate stellar and trailed point source functions, and to use circular and pill shaped apertures to measure photometry and estimate appropriate aperture corrections.

Citation: If you use TRIPPy in your science works, please cite Fraser, W. et al., 2016, To appear in AJ. DOI at Zenodo: http://dx.doi.org/10.5281/zenodo.48694

Update to version 1.1

TRIPPy has been updated to version 1.1, which includes many changes around accuracy and performance. Performance increases are all over, and you'll see as you use it.

A bug was found in the github verion prior to v1.0 which lightly affected the stellar source aperture correction estimates. This did not affect the first released version, and has been fixed for v1.0 and later.

A few notable feature improvements are the ability to pass more than one radius as an array format when calling phot, a new space-saving PSF format, and significant improvements in the interaction with the star selector panel.

Installation

TRIPPy is compatible with python 2.7 and 3.5.

Dependencies

TRIPPy depends on a few separate, equally useful python packages. These packages are:

  • numpy
  • scipy
  • matplotlib
  • astropy
  • emcee (optional)
  • numba (optional, very very useful for big speed improvements. Highly recommended!!)

An additional optional piece of software is the SExtractor package. This can often be installed with through standard linux repository systems (yum, apt-get, etc.) or can be simply downloaded and compiled. [Source is available here].(http://www.astromatic.net/software/sextractor)

If you are using anaconda python, basically all of the above packages can be installed with conda install commands. Otherwise use pip.

Technically pip should automatically install all of the above requirements. In practice however, pip falls over quiet hilariously. Rather, the most reliable thing to do is install these requirements first.

All packages can be install using pip. eg. pip install numpy . Alternatively, the first four requirements can be installed (and maintained!) with an anaconda python installation. Instructions to install anaconda are available here.

Finally, optional packages, numba, and emcee (MC Hammer) can be installed. This will provide robust source fitting utilities in the MCMCfit module, clustering utilities, and MASSIVE speed improvements that are unavailable if numda isn't installed.

Test that these modules are available by attempting to import each one:

python -c "import numpy; import scipy; import matplotlib; import astropy"

If the optional emcee, or sklearn packages were installed

python -c "import emcee"
python -c "import numba"

TRIPPy installation

Once dependencies are installed, trippy can be installed with a simple pip invocation

pip install trippy

Accesibility can be checked with a few import commands

python -c "import trippy; from trippy import bgFinder; from trippy import psfStarChooser; from trippy import scamp"

and if the optional emcee package was installed

python -c "from trippy import MCMCfit"

Internal (unit) tests

Once you have installed trippy, please run

python test.py

in the tests directory. This will run 23 separate tests to confirm that your install is functional. If something fails, contact Wes!

Usage

I have provided an ipython notebook which shows most of the functionality of the trippy package including PSF/TSF generation, photometry, and source fitting and removal. Please checkout the notebook.