Electron migration in molecules is the progenitor of chemical reactions and biological functions after light-matter interaction. Following this ultrafast dynamics, however, has been an enduring endeavor. Recently, it has been suggested that high-harmonic spectroscopy (HHS) is able to probe dynamics with attosecond temporal and sub-˚angstr¨om spatial resolution. Still, real-time visualization of single-molecule dynamics continues to be a great challenge because experimental harmonic spectra are due to the coherent averages of light emission from individual molecules of different alignments. Here we demonstrate that the uniting of machine learning (ML) algorithm and HHS in two-color laser pulses enables us to retrieve the complex amplitude and phase of harmonics from single fixed-in-space molecule. From the complex single-molecule dipoles for different harmonics, we construct a movie of electron migration after tunnel ionization of the molecules at time steps of 50 attoseconds. Moreover, the angular dependence of molecular charge migration is fully resolved. By examining the movie, we observe that electron holes do not just “migrate” along the laser direction, but may “swirl” around the atom centers. Our ML-based HHS establishes a general reconstruction scheme for studying ultrafast charge migration in molecules, paving a way for further advance in tracing and controlling photochemical reactions by femtosecond lasers.