search

biaslab / ForneyLab.jl

Julia package for automatically generating Bayesian inference algorithms through message passing on Forney-style factor graphs.
MIT License
149 stars 33 forks source link

Conjugate variational Inference #192

Closed ThijsvdLaar closed 2 years ago

ThijsvdLaar commented 2 years ago

This PR implements conjugate variational inference (#183) as a linear node approximation method.

codecov-commenter commented 2 years ago

Codecov Report

Merging #192 (c47eb0c) into master (30933f2) will increase coverage by 0.33%. The diff coverage is 80.33%.

Impacted file tree graph

@@            Coverage Diff             @@
##           master     #192      +/-   ##
==========================================
+ Coverage   86.72%   87.06%   +0.33%     
==========================================
  Files          95       97       +2     
  Lines        4461     4614     +153     
==========================================
+ Hits         3869     4017     +148     
- Misses        592      597       +5
Impacted Files Coverage Δ
src/ForneyLab.jl 100.00% <ø> (ø)
...c/engines/julia/update_rules/nonlinear_sampling.jl 83.45% <ø> (+20.41%) :arrow_up:
.../engines/julia/update_rules/nonlinear_conjugate.jl 69.56% <69.56%> (ø)
src/factor_nodes/categorical.jl 77.19% <100.00%> (ø)
src/factor_nodes/gaussian.jl 100.00% <100.00%> (ø)
src/factor_nodes/nonlinear.jl 97.87% <100.00%> (+0.19%) :arrow_up:
src/helpers.jl 64.60% <100.00%> (+2.33%) :arrow_up:
src/update_rules/nonlinear_conjugate.jl 100.00% <100.00%> (ø)
src/update_rules/nonlinear_sampling.jl 100.00% <100.00%> (ø)
src/factor_nodes/gamma.jl 87.50% <0.00%> (+2.50%) :arrow_up:
... and 1 more

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 30933f2...c47eb0c. Read the comment docs.