Poisson likelihood

[frazane]

Adds the Poisson likelihood, along with an example notebook demonstrating a GP regression with count data.

Pull request type

Please check the type of change your PR introduces:

• Bugfix
• Feature
• Code style update (formatting, renaming)
• Refactoring (no functional changes, no api changes)
• Build related changes
• Documentation content changes
• Other (please describe):

[thomaspinder]

Nice work! Two things that would be nice:

1. A test to check that a non-conjugate GP is returned from taking the product of the prior and Poisson likelihood.
2. A notebook demonstrating this. Simulated data would be fine and it could heavily draw on the MCMC routine used in the classification notebook.

[codecov-commenter]

Codecov Report

Merging #231 (b9acd0d) into v0.6 (4853ebf) will increase coverage by 0.07%.
The diff coverage is 100.00%.

📣 This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more

@@            Coverage Diff             @@
##             v0.6     #231      +/-   ##
==========================================
+ Coverage   87.04%   87.11%   +0.07%     
==========================================
  Files          51       51              
  Lines        1961     1964       +3     
  Branches      183      184       +1     
==========================================
+ Hits         1707     1711       +4     
+ Misses        187      186       -1     
  Partials       67       67              

Flag	Coverage Δ
unittests	87.11% <100.00%> (+0.07%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
gpjax/__init__.py	100.00% <100.00%> (ø)
gpjax/likelihoods.py	96.15% <100.00%> (+0.50%)	⬆️

... and 3 files with indirect coverage changes