This data set includes photometric features as measured by the Sloan Digital Sky Survey (SDSS; sdss.org) that can be used as a training set in a machine learning model to separate galaxies and stars. It is used as part of a problem developed for the LSSTC Data Science Fellowship Program. The query of the SDSS data base to produce this data set was: SELECT TOP 20000 p.psfMag_r, p.fiberMag_r, p.fiber2Mag_r, p.petroMag_r, p.deVMag_r, p.expMag_r, p.modelMag_r, p.cModelMag_r, s.class FROM PhotoObjAll AS p JOIN specObjAll s ON s.bestobjid = p.objid WHERE p.mode = 1 AND s.sciencePrimary = 1 AND p.clean = 1 AND s.class != 'QSO' ORDER BY p.objid ASC The training set does not includes QSOs. The "blind" test set used at the end of the problems includes QSO data.

Last modified

• 03/07/2022

Creator

• Miller, Adam

DOI

• https://doi.org/10.21985/n2-84vp-2x17

Publisher

• Northwestern University

Date created

• 2018-02-21

Resource type

• Dataset

License

• Open Data Commons Attribution License (ODC-By)

Total Size

• 0 Bytes