This value-added catalog contains the results from applying the astroNN deep-learning code to APOGEE DR17 spectra to determine stellar parameters, individual stellar abundances (Leung & Bovy 2019a) retrained with ASPCAP DR17, distances (Leung & Bovy 2019b) retained with Gaia eDR3, and ages (Mackereth et al. 2019) retained with APOKASC-2 and low-metallicity asteroseismic ages from Montalbán et al. (2021). In addition, properties of the orbits in the Milky Way such as eccentricities, peri/apocenter radii, maximal disk height zmax, orbital actions, frequencies and angles (and their uncertainties) for all stars are computed using the fast method of Mackereth & Bovy (2018) assuming the MWPotential2014 gravitational potential from Bovy (2015). The scripts and retrained neural network models on APOGEE DR17, Gaia eDR3 used to make this catalog are available on GitHub here.
We recommend that you always use the latest updated catalog (DR17), but the earlier DR16 version of the APOGEE astroNN VAC is still available on the SAS here which uses models trained on APOGEE DR14 and Gaia DR2. The data file is row-matched to the allStar for easy access to the regular catalog data for each object. A previous version of the DR16 VAC (v0) contained a small number of special targets that were assigned incorrect IDs (see this page). This was corrected in v1.
