[shravanngoswamii]

BUGS defines step(e) as 1 if e >= 0 (OpenBUGS manual, Appendix II), and so does JAGS.
It matters when the argument is a count. Eye Tracking's cl[j] <- step(sumSC[j] - 1) skipped every cluster with one member, so K came out low.

[github-actions]

JuliaBUGS.jl documentation for PR #572 is available at:
https://TuringLang.github.io/JuliaBUGS.jl/previews/PR572/

[github-actions]

Benchmark Results

Julia 1.13.4 on AMD EPYC 7763 64-Core Processor (Linux)

Ratio = JuliaBUGS/Stan (lower is better for JuliaBUGS)

# Params shows Stan / JuliaBUGS when parameter counts differ.

Log Density uses generated evaluation. Gradient columns measure log density + gradient.

McRvs = Mooncake reverse mode with generated evaluation; FwdDiff = ForwardDiff with graph evaluation.

Model	# Params	Log Density Ratio	McRvs Ratio	FwdDiff Ratio
rats	65	0.44x	1.96x	95.05x
pumps	12	1.93x	2.51x	20.52x
bones	13	0.46x	4.21x	0.22x
oxford	244	2.09x	5.58x	657.94x
epil	303	0.61x	2.10x	968.24x
lsat	1006	0.76x	2.84x	1966.13x
schools	133	0.26x	1.68x	191.71x
beetles	2	2.61x	5.07x	10.74x
air	5	2.05x	1.91x	14.27x

Note: Stan benchmarks use hand-optimized Stan models, not direct BUGS translations. Comparison is illustrative only.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 88.08%. Comparing base (5fd788c) to head (f71244e).

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #572      +/-   ##
==========================================
- Coverage   88.49%   88.08%   -0.42%     
==========================================
  Files          30       30              
  Lines        3851     3851              
==========================================
- Hits         3408     3392      -16     
- Misses        443      459      +16     

Flag	Coverage Δ
callbacks	28.50% <0.00%> (ø)
compilation_model	50.00% <100.00%> (ø)
compilation_source_gen	44.85% <100.00%> (ø)
distributed_sampling	38.55% <0.00%> (ø)
elementary	27.94% <100.00%> (+0.05%)	⬆️
frontend	47.02% <0.00%> (ø)
gibbs	39.63% <0.00%> (ø)
graphs	21.14% <0.00%> (ø)
inference_chains	37.07% <0.00%> (ø)
inference_flexichains	?
inference_hmc	?
inference_marginalization	31.10% <0.00%> (ø)
inference_mh	35.46% <0.00%> (ø)
inference_slice_sampling	26.27% <0.00%> (ø)
log_density	59.15% <0.00%> (ø)
model_operations	43.01% <0.00%> (ø)
parallel_sampling	?
params_with_stats	38.18% <0.00%> (ø)
serialization	34.28% <0.00%> (ø)

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

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.