This is a wrapper for the get_recommendations() function, which provides a workaround for the limit of 5 seed tracks per recommendation call. The function splits a supplied vector of track IDs into subsets of length 5, then applies a get_recommendations() call, 5 tracks at a time. This should generate a data frame of recommended tracks, with length equal to the supplied vector of track ids.