about the cache_directories.zip and the weighting_factor_DF_full_map_39kft.pkl

[mygeotail]

I'm interested in the dose model and want to run on my laptop. But I can not download the cache_directories.zip and the weighting_factor_DF_full_map_39kft.pkl in the dir of notebooks_and_data_and_figures_for_paper. I try the git lfs clone, it's something wrong.

And the magnetocomics is another problem. I've the old execution version of magnetocomics .It works, but the igrf tale is old. Could you share with the new executable version magnetocosmics?

[ChrisSWDavis]

Hi @mygeotail, thanks for finding and flagging the issues, I've tested the file downloading issue on my local machine, and it seems like the problem is fairly straightforward - the ssc-maire Git Large File Storage is out of space, so it won't let anyone download files from the ssc-maire Git LFS. Unfortunately, I don't have an obvious fix myself as I'm not working at Surrey Space Center at the moment, but @drflei and @FraserBaird may be able to find a proper solution to the issue. I did see a possible temporary workaround online that involves forking the repository and doing some sort of stuff related to archiving the repository? In any case if you do happen to find any solutions then it would be great if you could post them here.

My understanding (if I remember correctly) is that you SHOULD still be able to run the various notebook scripts yourself (although if I'm wrong you may need to re-write the notebooks a little bit) without necessarily needing weighting_factor_DF_full_map_39kft.pkl and cache_directories.zip, but its just that the first set of simulations you do will take a lot longer, as the cache files will need to be re-generated. If the files can't be automatically be regenerated though, and if they're indeed important to your work, its possible that we may be able to send you the files directly, so let us know if you need them.

With regards to the old magnetocosmics version this is a problem we've frequently run into as well, that its currently hard to find accessable versions of more modern magnetocosmics online - @drflei may be able to help you with this and give you with more info, as he knows more about this side of things than I do.

Kind regards,
Dr Chris Davis

[mygeotail]

Thanks you for providing the requested information. I've try the forking the repository,but it does not work.

You're right that most of python scripts could run without the weighting_factor_DF_full_map_39kft.pkl. I'm just want to replicate the original result.
Anyway, Thanks a lot for you answer.

Best regards,
Jianguang Guo

[ChrisSWDavis]

To update on at least the Magnetocosmics side of this issue, you may be happy to know @mygeotail I've now added an option to now run AniMAIRE without needing to use Magnetocosmics! Now you can run OTSO to pre-generate asymptotic directions, and then provide the asymptotic direction files to OTSO, bypassing the need to have Magnetocosmics installed. I would be really interested in any thoughts you or others might have about the new system, but I hope this will now allow you to run AniMAIRE as you need to.