Required Files for Astro Amase

Required database and model files for running Astro AMASE (https://github.com/zfried/astro_amase), a Python package for automated molecular line identification in radio astronomical spectra. Included files: all_cdms_final_official.csv and all_jpl_final_official.csv (local copies of the CDMS and JPL molecular spectroscopy databases); cdms_pkl and jpl_pkl (pre-computed molsim Molecule objects for each catalog entry); transitions_database.pkl.gz (compressed database of molecular transitions used for candidate lookup); vlsr_database_logint.pkl.gz (compressed database used for automatic VLSR determination); model_weights_two_256.pth (trained weights for the SubgroupMLP fragment prediction model); and grammar.txt and grammar_smiles.txt (Group SELFIES grammar files defining the 326 molecular fragments used by the molecule prediction pipeline).