Submit demcmc to be an affiliated package

[dstansby]

Package Details

• Package name: demcmc
• Maintainer(s): @dstansby
• Repository: https://github.com/dstansby/demcmc
• Documentation link: https://demcmc.readthedocs.io
• PyPI link (if applicable): https://pypi.org/project/demcmc/
• Continuous integration link(s): https://github.com/dstansby/demcmc/actions
• Link to Code of Conduct or page describing adopted Code of Conduct: https://github.com/dstansby/demcmc/blob/7b88b4f5fb169b22922dcaf7e6aa4aada1e2219b/CODE_OF_CONDUCT.md

Description of Package

A package for estimating differential emission measures (DEMs) from observations of the Sun using Monte Carlo methods.

demcmc provides an interface for collecting information on emission lines together in one place (a list of EmissionLine objects), and running a DEM inversion on them (using predict_dem_emcee()). Each EmissionLine object contains the intensity of an observed line, and the contribution function of that line. The inversion is carried out using Markov chain Monte carlo (MCMC) methods, with the emcee package used to run the sampling.

Package Review

Editor Submission Checklist

• Is the submission appropriate (compatible with the SunPy CoC)?
• Does the project have an appropriate license?
• Is the project in an online version control system?
• Does the project provide a Python interface?
• Is the project on PyPI?
• Is the project useful to the solar physics community?
• Version that was reviewed (tag or commit hash)

Instructions to Reviewer

Please copy the following and select the ranking for each criteria, the full review criteria can be found here:

* Functionality           : General Package / Specialized Package / Not Relevant
* Integration             : Well integrated / Partially Integrated / Minimal Integration
* Documentation           : Extensive / Some / Little
* Testing                 : Excellent / Good / Needs Work
* Duplication             : None / Some / Major
* Community               : Excellent / Good / Needs Work
* Development Status      : Stable / Subject to Change / Low Activity / Needs Work

[wtbarnes]

@Cadair are you up for reviewing this?

[Cadair]

Yep, I can review this. I have no idea how or why I missed the notifications for this.

[ayshih]

Does the first m in demcmc stand for "measure" or "Markov"? =) I keep wanting to inject an additional m into the name of this package.

[Cadair]

Review

Version reviewed: 1.1.0

Scores

• Functionality:
• Integration: ¹
• Documentation:
• Testing:
• Duplication:
• Community: ²
• Development Status: ³

Comments

Overall this looks like a great package, which definitely has utility to those working with DEMs. Some points / justifications for the ratings above:

1. I have put full integration, but seemingly there is none at all with core or any other affiliated package? Is there scope for this in the future?
2. I only put good here because I felt there wasn't enough evidence to justify an excellent, but as the package is new this is understandable.
3. I have put low activity here because it seems to have been very quiet for the last few months. Do you feel this is fair, given the current level of developer effort? (i.e. do you think telling people this is a low activity package will reflect reality in the next year?).

Outcome

Accepted.

[dstansby]

Thanks for taking the time to review!

Comments

Overall this looks like a great package, which definitely has utility to those working with DEMs. Some points / justifications for the ratings above:

1. I have put full integration, but seemingly there is none at all with core or any other affiliated package? Is there scope for this in the future?

There is scope in the future, specifically to

• take GenericMaps as input and automatically parallelise the calculation over pixels.
• integrate closely with an emission measure calculation software (e.g. fiasco) if/when there is one of those that gains affiliated status

Unfortunately I didn't have time/funding to look at these yet.

2. I only put good here because I felt there wasn't enough evidence to justify an excellent, but as the package is new this is understandable.

👍 this is fair, especially since it was a time limited project and I don't have further time/funding to build a community.

3. I have put low activity here because it seems to have been very quiet for the last few months. Do you feel this is fair, given the current level of developer effort? (i.e. do you think telling people this is a low activity package will reflect reality in the next year?).

👍 also very fair, without anyone else proactively contributing I won't be working on this more in the near future.

[Cadair]

You want to open the PR to add this to the list or shall I?

[dstansby]

I'm having a sunpy day, so I can do it later 👍