Propagate ref particle in PG

[FredericWantiez]

I think we need to propagate the reference particle in PG but sweep! is missing a ref parameter

[coveralls]

Pull Request Test Coverage Report for Build 933715358

Warning: This coverage report may be inaccurate.

This pull request's base commit is no longer the HEAD commit of its target branch. This means it includes changes from outside the original pull request, including, potentially, unrelated coverage changes.

• For more information on this, see Tracking coverage changes with pull request builds.
• To avoid this issue with future PRs, see these Recommended CI Configurations.
• For a quick fix, rebase this PR at GitHub. Your next report should be accurate.

Details

• 4 of 4 (100.0%) changed or added relevant lines in 2 files are covered.
• 1 unchanged line in 1 file lost coverage.
• Overall coverage decreased (-0.3%) to 55.163%

---

Files with Coverage Reduction	New Missed Lines	%
src/container.jl	1	90.98%

Totals
Change from base Build 895050782:	-0.3%
Covered Lines:	203
Relevant Lines:	368

---

💛 - Coveralls

[codecov]

Codecov Report

Merging #29 (5c5723e) into master (176640e) will increase coverage by 0.27%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master      #29      +/-   ##
==========================================
+ Coverage   55.43%   55.70%   +0.27%     
==========================================
  Files           5        5              
  Lines         368      368              
==========================================
+ Hits          204      205       +1     
+ Misses        164      163       -1     

Impacted Files	Coverage Δ
src/container.jl	92.62% <100.00%> (+0.81%)	⬆️
src/smc.jl	97.22% <100.00%> (ø)

---

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 176640e...5c5723e. Read the comment docs.

[devmotion]

Isn't this handled in

AdvancedPS.jl/src/smc.jl

Lines 104 to 113 in 176640e

	# Create a new set of particles.
	nparticles = sampler.nparticles
	x = map(1:nparticles) do i
	if i == nparticles
	# Create reference trajectory.
	forkr(state.trajectory)
	else
	Trace(model)
	end
	end

?

[FredericWantiez]

Was reading through resample_propagate, looks like we need to set ref to replace the selected particle in the last spot:
https://github.com/TuringLang/AdvancedPS.jl/blob/master/src/container.jl#L225-L229

[devmotion]

Should also fix https://github.com/TuringLang/Turing.jl/blob/57e0512e3b63e4622c755dc6d933b51eb76d4483/src/inference/AdvancedSMC.jl#L304 when a new release of AdvancedPS is available.

[yebai]

Many thanks @FredericWantiez - looks correct to me!