Dataset for "From Halos to Galaxies. VI. Improved Halo Mass Estimation for SDSS Groups and Measurement of the Halo Mass Function"

Zhao et al. (2025) developed a simulation-trained machine-learning (ML) framework to improve dark-matter halo mass estimates for SDSS galaxy groups, addressing systematic biases of abundance-matching halo masses (notably differences between groups with star-forming vs. passive centrals). They apply the calibrated ML model to the widely used Yang et al. SDSS group catalog and recommend using a model trained with Gaussian noise matched to observational measurement errors when predicting halo masses for SDSS groups.This Zenodo record provides a minimal crosswalk table (three columns) that lets you attach the Zhao+2025 noise-matched ML halo mass to galaxies identified in NYU-VAGC / Yang groups:NYU_VAGC_ID — NYU-VAGC object ID (row number in the NYU-VAGC files).Group_ID — Yang et al. SDSS group identifier (model-magnitude based group catalog); galaxies in the same group share the same Group_ID.ML_halo_mass_2 — ML halo mass estimate from Zhao et al. (2025), using the recommended noise-matched training strategy; units: $M_\odot/h$.