Jih/model selection

[jonathanishhorowicz]

• Implemented simulation-based model seleciton according to the "ABC SMC algorithm for model selection" at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2658655/ (top of page 190).

• Some arguments added to ABC functions so that they could be used for this model selection algorithm. They have default values that maintain the original behavior. Main ones are:

-- added suppress_maxiter_warnings to ABC functions since each is called with max_iter=1. -- added normalise_weights to iterateABCSMC! so that the weights can be normalised at the end of a model selection population rather than after a single ABC-SMC run with max_iter=1.

[codecov-io]

Codecov Report

Merging #25 into master will decrease coverage by 8.62%. The diff coverage is 14.92%.

@@            Coverage Diff             @@
##           master      #25      +/-   ##
==========================================
- Coverage   71.55%   62.93%   -8.63%     
==========================================
  Files          17       18       +1     
  Lines         777      901     +124     
==========================================
+ Hits          556      567      +11     
- Misses        221      334     +113

Impacted Files	Coverage Δ
src/abc/simulation.jl	100% <ø> (ø)	:arrow_up:
src/abc/io.jl	16.47% <ø> (ø)	:arrow_up:
src/util/emulation_helpers.jl	46.42% <0%> (-7.74%)	:arrow_down:
src/abc/model_selection.jl	0% <0%> (ø)
src/abc/smc.jl	86.09% <70%> (-1.5%)	:arrow_down:
src/abc/rejection.jl	90.62% <76.47%> (-2.32%)	:arrow_down:

---

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 9ee9f14...e6e9135. Read the comment docs.