Deprecated test output? arkane/statmech.py

[mefuller]

Describe the bug On running make test post install by cloning repository and using conda to create environment, output instructs user to paste results into arkane/statmech.py which does not exist

How to reproduce make test

Expected behavior The output, if critical should be automatically written to the file. However, there is no arkane/statmech.py currently (no arkane/ directory), nor is statmech.py to be found under arc/ or arc/statmech/, so it is not at all clear where this output should be copied (if at all)

Additional context Relevant output:

Level of theory: level1
Scale Factor for Zero-Point Energies     = 0.950
Scale Factor for Harmonic Frequencies    = 0.963
Scale Factor for Fundamental Frequencies = 0.925
(execution time: 3)



Level of theory: level2
The following species from the standard set did not converge at this level:
 ['CO2']
Scale Factor for Zero-Point Energies     = 0.980
Scale Factor for Harmonic Frequencies    = 0.994
Scale Factor for Fundamental Frequencies = 0.955
(execution time: 5)




You may copy-paste the following harmonic frequencies scaling factor/s to Arkane
(paste in the `freq_dict` under assign_frequency_scale_factor() in arkane/statmech.py):

                 'level1': 0.963,  # [4]

                 'level2': 0.994,  # [4]



Scaling factors calculation for 2 levels of theory completed (elapsed time: 8.5).

Miniconda 3 on Fedora Rawhide

[xiaoruiDong]

@mefuller Thank you for reporting that. As you described, the test.py is truly misleading and outdated.

1. /arkane mentioned in the cited blocks are actually pointing to the arkane directory under RMG-Py repository (we use RMG-Py and its database as backend).
2. According to the latest version, if you need any frequency factors that are not included in the RMG, these values should be added to /RMG-database/input/quantum_correlations/data.py's freq_dict variable.

We will update the test.py soon.