Adds functionaility for logl and prior_transform args and kwargs

kbarbary / nestle

Pure Python, MIT-licensed implementation of nested sampling algorithms for evaluating Bayesian evidence.

http://kbarbary.github.io/nestle

Other

75 stars 27 forks source link

Adds functionaility for logl and prior_transform args and kwargs #24

Closed GregoryAshton closed 6 years ago

GregoryAshton commented 6 years ago

The motivation for this, in particular, is one usually wants to pass some data into the loglikelihood, rather than (as in the examples) allowing it to feed in from the outer scope (not sure about the each terminology there).

coveralls commented 6 years ago

Coverage decreased (-0.9%) to 80.769% when pulling 44314e5363eb6726da7959a6ab973aa6ac8c51b2 on GregoryAshton:add-args-and-kwargs into bbe78126d4bc15210ed0b3c16ff13b8e3c3d6e38 on kbarbary:master.

kbarbary commented 6 years ago

Thanks! Can you make the following changes:

Rename _function_wrapper to _FunctionWrapper (to make clear that it's a class)
have the defaults for args and kwargs be None, and use the self.args = [] if args is None else args pattern (as in emcee)? It doesn't matter here but it's best practice not to use mutable objects as defaults.

With those changes, this looks good.

GregoryAshton commented 6 years ago

I've pushed a squashed commit with those changes :)

bek0s commented 5 years ago

Thanks a lot for implementing this. Since it is an API change, it would be great if someone could push the current GitHub version to PyPI. Thanks!