Feature/marginalise

[goldingn]

An interface for marginalisation of random variables

Initially just scalar discrete variables, and multivariate normal variables (via Laplace approximation).

[lintr-bot]

R/dag_class.R:6:1: style: functions should have cyclomatic complexity of less than 50, this has 51.

​dag_class <- R6Class(
^

R/marginalise.R:172:1: style: Lines should not be more than 80 characters.

​        # x will be the target; a tf constructor function for the distribution being
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[codecov]

Codecov Report

Merging #344 into master will increase coverage by 0.23%.
The diff coverage is 97.36%.

@@            Coverage Diff             @@
##           master     #344      +/-   ##
==========================================
+ Coverage   94.94%   95.18%   +0.23%     
==========================================
  Files          30       32       +2     
  Lines        4728     5023     +295     
==========================================
+ Hits         4489     4781     +292     
- Misses        239      242       +3

Impacted Files	Coverage Δ
R/dag_class.R	97.72% <100%> (+0.03%)	⬆️
R/utils.R	85.97% <95.12%> (+0.83%)	⬆️
R/marginalise.R	95.18% <95.18%> (ø)
R/marginalisers.R	98.79% <98.79%> (ø)
R/node_types.R	97.48% <0%> (+1.67%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 18c581c...fe40342. Read the comment docs.