-
Notifications
You must be signed in to change notification settings - Fork 7
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add support for QBOi variables #179
base: master
Are you sure you want to change the base?
Conversation
@jinboxie here is a development branch for supporting QBOi variables. And i started a template for one variable. I will check if you have permission for this repo and push codes. |
Codecov ReportBase: 28.98% // Head: 28.93% // Decreases project coverage by
Additional details and impacted files@@ Coverage Diff @@
## master #179 +/- ##
==========================================
- Coverage 28.98% 28.93% -0.06%
==========================================
Files 93 93
Lines 3981 3988 +7
==========================================
Hits 1154 1154
- Misses 2827 2834 +7
Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here. ☔ View full report at Codecov. |
@jinboxie I updated the utendogw handler. But ran into the issue in cmor: I ran the code on Cori, with
|
Hi Jill. Yes, the data wanted zonal mean instead of 3D. I reprocessed the data now, please try again. |
I made some updates, now it complains about missing lat_bnds. I think somehow latitude bounds are missing from zonal averaging in your processing. |
@chengzhuzhang Hi Jill. I created a new file during this postprocess zonal mean calculation. It is now added the new bnds and every variables from the previous files. It should be good now. |
Hi @jinboxie it looks like the longitude related data are still present in the data files. I think we picked a hard variable to start, e3sm_to_cmip currently doesn't support zonal mean variables, which is not standard CMIP requests. And implementing it elegantly will take sometime. Another difficulty I have is to actually understand how cmor interact with those QBOi tables, it seems to me cmor is always try to pick up CMIP6 tables, I haven't figured out how to force cmor to pick up other tables. I have to table this after AMS. At the mean time, I think you can generate a mapping of the conversion, In any case that would be useful either we use e3sm_to_cmip or directly use cmor. |
@chengzhuzhang ok! |
Hi @jinboxie you could look here https://github.com/E3SM-Project/e3sm_to_cmip/pull/179/files for code change I made to temporarily make You could follow below commands to install
get this and following is my command to run on the utendogw variable. e3sm_to_cmip -i /global/cscratch1/sd/xie7/forJill/data -o ~/tests -v qboi_utendogw -t /global/homes/c/chengzhu/qboi-cmor-tables/Tables/ -u /global/homes/c/chengzhu/e3sm_to_cmip/e3sm_to_cmip/resources/qboi_template.json You could try patch the lib.py script and fit the qboi variables.. |
Thank you Jill. |
@jinboxie no problem. I think the lib.py script is not easy to read and perhaps needs some refactoring. I'm also tagging @tomvothecoder here and we can probably think about a re-write in the long run. |
Hi Jill @chengzhuzhang . I'm having a problem with the command. this was even the same with ua for some reason 0%| | 0/1 [00:00<?, ?it/s] |
@jinboxie , based on our discussion. It is best to start from standard e3sm_to_cmip use case, so please download the CMIP tables from here: https://github.com/PCMDI/cmip6-cmor-tables and point to downloaded tables by using |
Hi Jill @chengzhuzhang , please refer to https://github.com/E3SM-Project/e3sm_to_cmip/tree/jinboxie_qboi |
Thank you, @jinboxie ! |
This PR is to support variables requested by QBOi project, using the customized qboi cmor tables. https://gitlab.com/JamesAnstey/qboi-cmor-tables