Erwan Pannier

I did some calculations to get insights on how to improve database loading & RAM use : see [radis-benchmark/manual_benchmarks/test_database_loading.ipynb](https://github.com/radis/radis-benchmark/blob/master/manual_benchmarks/test_database_loading.ipynb)

Featured in GSOC 2021 project [Reduce Memory Usage 🚀 ](https://openastronomy.org/gsoc/gsoc2021/#/projects?project=memory_usage)

Hello @gagan-aryan. Many questions here: 1. I don't think of anything else right now. List of parameters can be found in the [Hitran parser](https://github.com/radis/radis/blob/develop/radis/io/hitran.py#L64), they are all independants (except from...

New benchmark test on extra columns to remove + use of vaex (see previous benchmark for CO : https://github.com/radis/radis/pull/306#issuecomment-896426701 ) Test code: ```python import radis from radis import SpectrumFactory import...

@dcmvdbekerom let's check this sometime !

Agreed, and already implemented. ``` Spectrum.update("radiance_noslit") ``` It can do so from abscoeff, and will require emisscoeff or an equivalent spectral array under nnoequilibrium. And actually, starting from a recent...

@dcmvdbekerom is this fixed by #514 ?

Good remark ! I don't know how to settle on one or the other. Asking the community : Vote for : 1. `s.get()` should always return 'nm' or 'cm-1' sorted...

@dcmvdbekerom it does in both cases, but with either X either Y in reverse order if you changed 'nm' to 'cm-1'

4:0, seems everyone agree ! Let's make it. Issue open if someone wants to take it.