optimal sampling - Githubissues

keflavich / imf

Simple tools to work with the Initial Mass Function

MIT License

44 stars 17 forks source link

optimal sampling #36

Closed richardson-t closed 1 year ago

richardson-t commented 1 year ago

refactor optimal sampling + integrate into make_cluster + edit documentation

keflavich commented 1 year ago

If you're ready for final review on this, let me know, and show some of your examples & test cases. It's especially helpful if you can put examples into a notebook and add that, or into the docs.

richardson-t commented 1 year ago

ready for review

now incorporates new KK09 + root finding handling

keflavich commented 1 year ago

The code looks good, as far as I can see. Could you please include an example? I'm pretty sure you made at least one example plot during the debugging process.

richardson-t commented 1 year ago

this PR contains all the new code that I've introduced, which covers:

implementation of optimal sampling -new functions in imf.py -refactoring of make_cluster to include optimal sampling -documentation -explanatory notebook: examples/opt_sampling.ipynb
revamped KoenConvolvedPowerLaw -new implementation based on precomputed lookup tables -new distribution in distributions.py -refactoring of previous KoenConvolvedPowerLaw into SpotKoenConvolvedPowerLaw -documentation -explanatory notebook: examples/koen.ipynb

I think this PR is ready for final review of the new content; figuring out how to integrate it into the base repository can come once the code is OK

keflavich commented 1 year ago

@richardson-t I left some significant wording changes (made everything terser, but also formatted some things) in https://github.com/keflavich/imf/pull/36/commits/5db4432bbca724047c2a174b950f39cdd074f86a

Could you add the missing equation? And check over the text? I am pretty sure everything I thinned out was not necessary, but best to be sure