Question on how template will be made available to modellers

[orichters]

For REMIND-MAgPIE to prepare a submission, we need to read the template into R. We currently use this function that can read xlsx files but also yaml files such as:

- Final Energy:
    unit: EJ/yr

But: It does not yet understand things such as:

- Emissions|{Tier-1 Species}:
-     unit: "{Tier-1 Species}"

How will the template be made available to modellers, such that we can prepare? Thanks!

[danielhuppmann]

Thank you for raising this issue. My preferred approach would be to use the R package reticulate to read and process the yaml files, and then convert to a suitable R object. Does that sound acceptable?

[orichters]

Yes, as long as we somehow get an R object containing the full variable names out of it, we should be able to convert it into something our scripts understands.

[FlorianLeblancDr]

Hi Oliver, Daniel, Do you know any equivalent of reticulate for python?

I am not sure the question was fully answered: wether each teams should use an appropriate script to convert the .yaml file to a full variable list, Or if it is here on github that the community (we) should provide automatic conversion of the .yaml full (i-e, expanding the Tag on variables).

[danielhuppmann]

Hi Florian, thank you for the reminder.

I wanted to write a tutorial for it with reticulate, just never got to it.

And @phackstock and I discussed usability for non-programmers: we are planning to create a full xlsx spreadsheet of all variables and regions including attributes, etc. for each Scenario Explorer instance and also for common-definitions. Just need a bit more time to implement it...

[danielhuppmann]

We now implemented a feature where the lists of variables and regions will be made available as an xlsx spreadsheet at https://files.ece.iiasa.ac.at/common-definitions/project-template.xlsx (and will be updated automatically to be in sync with the repository). The link to the file is already added to the Readme of this repository.

@orichters, is that sufficient for you?

[orichters]

Yes, that is perfect! loadIIASATemplate happily read the file and we can figure out how we automate the download.

[orichters]

Will the min, max specification, summation rules etc. also be exported?

[danielhuppmann]

Yes, all attributes in the yaml are included as columns. We still need to come up with conventions on attribute-names for min/max-attributes, but that’s a different discussion…

[orichters]

Thanks, excellent!

[orichters]

There is another small issue: The expansion of the tags does not work yet in the components field:

[danielhuppmann]

Yes, it's on the to-do list: https://github.com/IAMconsortium/nomenclature/issues/342