Normalised stats

[lukashergt]

This PR introduces a norm keyword for the NestedSamples.stats method, that allows passing a previously computed NestedSamples.stats instance as a normalisation reference, producing additional columns of differences [Delta_logZ, ...].

Typical cosmology use-case would be when you have multiple models (e.g. LCDM, wCDM, w0waCDM) and you want to have evidence and KL-divergence values relative to one particular model (e.g. LCDM). I was getting annoyed having to type this up every time myself...

Checklist:

• I have performed a self-review of my own code
• My code is PEP8 compliant (flake8 anesthetic tests)
• My code contains compliant docstrings (pydocstyle --convention=numpy anesthetic)
• New and existing unit tests pass locally with my changes (python -m pytest)
• I have added tests that prove my fix is effective or that my feature works
• I have appropriately incremented the semantic version number in both README.rst and anesthetic/_version.py

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 100.00%. Comparing base (8c972f5) to head (fa919c1).
Report is 1 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff            @@
##            master      #396   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           36        36           
  Lines         3060      3069    +9     
=========================================
+ Hits          3060      3069    +9     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[AdamOrmondroyd]

I like the idea, comments inline

[lukashergt]

Thanks @AdamOrmondroyd, all inline comments addressed.

[AdamOrmondroyd]

Why do you need a seed here?

[lukashergt]

I managed to get failing tests in what follows, because of changes to the seed from the tests I added above... Nothing dramatic, just bad chance in combination with the tests checking against 3*std or similar. I could have fiddled around with tolerances, but just trying a new seed there was enough, so I did not dig deeper...

[AdamOrmondroyd]

ok, I'll reapprove when you've done the version number ofc this is a minor bump

[AdamOrmondroyd]

lgtm