Robert Gieseke
Robert Gieseke
If the speed-up is needed the caching could probably also be part of a local pipeline.
Yeah, sorry for being unclear, i meant that if 0.4s were needed (due being re-loaded in a loop or something), the caching could be done in some way as part...
I used the approach of creating Python modules from YAML or CSV in a few projects, i think it's the best if there is some automatic checking that it's not...
Works for me :-) See #32
Some options: - JOSS - JORS - GMD
https://github.com/BorealisAI/lite_tracer might be worth checking. > LiteTracer acts as a drop-in replacement for argparse, and it can generate unique identifiers for experiments in addition to what argparse already does. Along...
@danielhuppmann @znicholls @lewisjared @swillner Maybe that has been discussed already and i missed it, but couldn't we call these columns `scm` and `iam`? If it's confusing us few having `model`...
> -1, because we have a potential user base of energy (or sector-specific) modellers, who wouldn’t consider themselves as IAMs. Good point, i wasn't aware of that!
Recent description of the SCM in the SI https://iopscience.iop.org/1748-9326/12/2/024022/media/erlaa5b0a_suppdata.pdf of this paper https://iopscience.iop.org/article/10.1088/1748-9326/aa5b0a
Hi @maritsandstad - that's great to hear! @swillner and @znicholls are still figuring out the first model integrations for e.g. Dice so the API definition is still under development, Cicero-SCM...