Allowing resume_from to grab the previous state

[joelkandiah]

fixes #1033

[github-actions]

Benchmark Report for Commit e720bea

Computer Information

Julia Version 1.11.6
Commit 9615af0f269 (2025-07-09 12:58 UTC)
Build Info:
  Official https://julialang.org/ release
Platform Info:
  OS: Linux (x86_64-linux-gnu)
  CPU: 4 × AMD EPYC 7763 64-Core Processor
  WORD_SIZE: 64
  LLVM: libLLVM-16.0.6 (ORCJIT, znver3)
Threads: 1 default, 0 interactive, 1 GC (on 4 virtual cores)

Benchmark Results

|                 Model | Dimension |  AD Backend |      VarInfo Type | Linked | Eval Time / Ref Time | AD Time / Eval Time |
|-----------------------|-----------|-------------|-------------------|--------|----------------------|---------------------|
| Simple assume observe |         1 | forwarddiff |             typed |  false |                  9.2 |                 1.5 |
|           Smorgasbord |       201 | forwarddiff |             typed |  false |                666.4 |                46.0 |
|           Smorgasbord |       201 | forwarddiff | simple_namedtuple |   true |                435.3 |                56.2 |
|           Smorgasbord |       201 | forwarddiff |           untyped |   true |               1214.5 |                30.3 |
|           Smorgasbord |       201 | forwarddiff |       simple_dict |   true |               6863.0 |                27.0 |
|           Smorgasbord |       201 | reversediff |             typed |   true |               1497.4 |                29.0 |
|           Smorgasbord |       201 |    mooncake |             typed |   true |               1021.7 |                 4.7 |
|    Loop univariate 1k |      1000 |    mooncake |             typed |   true |               5988.3 |                 4.3 |
|       Multivariate 1k |      1000 |    mooncake |             typed |   true |               1023.8 |                 9.0 |
|   Loop univariate 10k |     10000 |    mooncake |             typed |   true |              67027.3 |                 4.0 |
|      Multivariate 10k |     10000 |    mooncake |             typed |   true |               9331.2 |                 9.6 |
|               Dynamic |        10 |    mooncake |             typed |   true |                135.6 |                12.7 |
|              Submodel |         1 |    mooncake |             typed |   true |                 13.1 |                 5.4 |
|                   LDA |        12 | reversediff |             typed |   true |               1203.3 |                 2.4 |

[codecov]

Codecov Report

❌ Patch coverage is 66.66667% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 82.25%. Comparing base (1d52c13) to head (e720bea).
⚠️ Report is 2 commits behind head on main.

Files with missing lines	Patch %	Lines
src/sampler.jl	66.66%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1034      +/-   ##
==========================================
- Coverage   82.26%   82.25%   -0.02%     
==========================================
  Files          38       38              
  Lines        3947     3950       +3     
==========================================
+ Hits         3247     3249       +2     
- Misses        700      701       +1     

☔ View full report in Codecov by Sentry.
📢 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.

[coveralls]

Pull Request Test Coverage Report for Build 17429847333

Details

• 2 of 3 (66.67%) changed or added relevant lines in 1 file are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage decreased (-0.01%) to 82.504%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
src/sampler.jl	2	3	66.67%

Totals
Change from base Build 17349688556:	-0.01%
Covered Lines:	3249
Relevant Lines:	3938

---

💛 - Coveralls

[yebai]

Quick thought: could this be unified with the initialisation strategy (see, eg, #985) that @penelopeysm is working on?

[penelopeysm]

Hmmmm it's related, although not quite the same, as this deals with initial_state whereas the InitContext stuff deals with initial_params. Perhaps there's a discussion to be had there about the interface although right now they're separate parts of the codebase.

[penelopeysm]

fixed by #1035 and other patches in MCMCChains + Turing