HaloFlow uses a machine learning approach to infer Mh and stellar mass, M∗, using grizy band magnitudes, morphological properties quantifying characteristic size, concentration, and asymmetry, total measured satellite luminosity, and number of satellites.