Conversation
There was a problem hiding this comment.
This looks good for the modeling component of our phase 1b deliverables.
Below are some suggestions for the next iteration(s) (update: moved to ccmmf/organization#118):
-
README Clarifications:
- Describe all files and directories
- What goes in
tools/vs the higher level directory? - I found it helpful for scripts to be numbered as they were in 1a, why are they no longer numbered?
- What goes in
- Many of the 'tools/' are not mentioned in README (make_site_info, read_dwr_planting, slurm_array_submit, run_getERA5)
- Describe the overall workflow.
- Describe all files and directories
-
Scripts
- It would be helpful to identify functions that are intended to be added to PEcAn, and which are ccmmf or script specific.
- Note: The function
read_ncdiroverlaps with line 79 ff indownscale/02_extract_sipnet_output.R(see PR #2).
- Note: The function
- Are separate versions of
install_pecan.shandinstall_sipnet.shnecessary?
- It would be helpful to identify functions that are intended to be added to PEcAn, and which are ccmmf or script specific.
root dir: Does one distinct step of the workflow; probably called once per model run.
Wasn't obvious as I developed how many separate steps there would be, nor whether all of them would be mandatory to run each time (e.g. as currently written the inputs tarball contains some of the files created by
Easy: If it's here, I think it's CCMMF-specific. Otherwise I would have added to PEcAn to start with. If you see any candidates for relocation, flag them and we can consider.
Not essential, but copying them was an easy way to preserve the principle that one workflow directory contains everything you need to run the workflow. I think that's valuable enough to accept some duplication of simple files. |
Co-authored-by: Chris Black <chris@ckblack.org>
Scripts and documentation for the model execution portion of 1b. See README for details.