Currently the default behavior is to treat the nucleus as spin 3/2 by importing f32 from nqr_parser6. A good fix would be to use the nuclear_data files to determine the appropriate function to compute nqr frequency from Cq and eta.