Bayesian photometric redshifts of blended sources

blendz is a Python module for estimating photometric redshifts of (possibly) blended sources with an arbitrary number of intrinsic components. Using nested sampling, blendz gives you samples from the joint posterior distribution of redshift and magnitude of each component, plus the relative model probability to identify whether a source is blended.

blendz is easy to install using pip

pip install blendz

and can be run using either simple configuration files

pz = blendz.Photoz(config_path='path/to/config.txt')

or keyword arguments

pz = blendz.Photoz(data_path='path/to/data.txt',
                   mag_cols = [1, 2, 3, 4, 5],
                   sigma_cols = [6, 7, 8, 9, 10],
                   ref_band = 2,
                   filters=['sdss/u', 'sdss/g',
                            'sdss/r', 'sdss/i', 'sdss/z'])


to set the configuration.


If you use this code in your research, please attribute this paper:

         author = {{Jones}, D.~M. and {Heavens}, A.~F.},
         title = "{Bayesian photometric redshifts of blended sources}",
         journal = {ArXiv e-prints},
         archivePrefix = "arXiv",
         eprint = {1808.02846},
         keywords = {Astrophysics - Cosmology and Nongalactic Astrophysics, Astrophysics - Astrophysics of Galaxies},
         year = 2018,
         month = aug,
         adsurl = {},
         adsnote = {Provided by the SAO/NASA Astrophysics Data System}