Bugfix/3301 stan csv reader skip warmup

[mitzimorris]

Submission Checklist

• Run unit tests: ./runTests.py src/test/unit
• Run cpplint: make cpplint
• Declare copyright holder and open-source license: see below

Summary

Added logic to skip warmup draws, if any present, more checks on consistency between config and output.

Intended Effect

see #3301

How to Verify

Unit tests

Side Effects

N/A

Documentation

N/A

Copyright and Licensing

Please list the copyright holder for the work you are submitting (this will be you or your assignee, such as a university or company):
Columbia University

By submitting this pull request, the copyright holder is agreeing to license the submitted work under the following licenses:

• Code: BSD 3-clause (https://opensource.org/licenses/BSD-3-Clause)
• Documentation: CC-BY 4.0 (https://creativecommons.org/licenses/by/4.0/)

[mitzimorris]

@WardBrian - Ubuntu tests failing because of nodejs warning?

[run stan tests (ubuntu-20.04, 3.8)](https://github.com/stan-dev/stan/actions/runs/10059471447/job/27804774061#step:4:1737)
Process completed with exit code 136.
[run stan fvar-var model tests (ubuntu-20.04, 3.8)](https://github.com/stan-dev/stan/actions/runs/10059471447/job/27804773722)
The following actions uses Node.js version which is deprecated and will be forced to run on node20: actions/checkout@v3. For more info: https://github.blog/changelog/2024-03-07-github-actions-all-actions-will-run-on-node20-instead-of-node16-by-default/
run stan tests (ubuntu-20.04, 3.8)
The following actions uses Node.js version which is deprecated and will be forced to run on node20: actions/checkout@v3. For more info: https://github.blog/changelog/2024-03-07-github-actions-all-actions-will-run-on-node20-instead-of-node16-by-default/

the offending test uses the Stan CSV reader - but works on Mac. is there something cached that needs cleanup or????

[WardBrian]

The node warning is not the source of the error, its

[==========] Running 1 test from 1 test suite.
[----------] Global test environment set-up.
[----------] 1 test from ServicesLaplaceJacobian
[ RUN      ] ServicesLaplaceJacobian.laplace_jacobian_adjust
Floating point exception (core dumped)
test/unit/services/optimize/laplace_jacobian_test --gtest_output="xml:test/unit/services/optimize/laplace_jacobian_test.xml" failed
exit now (07/23/24 13:21:14 UTC)

There is also a separate failure in Jenkins:

[ RUN      ] StanIoStanCsvReader.skip_warmup
unknown file: Failure
C++ exception with description "Error with header of input file in parse" thrown in the test body.
[  FAILED  ] StanIoStanCsvReader.skip_warmup (1 ms)
[ RUN      ] StanIoStanCsvReader.missing_data
[       OK ] StanIoStanCsvReader.missing_data (10 ms)
[----------] 11 tests from StanIoStanCsvReader (69 ms total)
[----------] Global test environment tear-down
[==========] 11 tests from 1 test suite ran. (69 ms total)
[  PASSED  ] 10 tests.
[  FAILED  ] 1 test, listed below:
[  FAILED  ] StanIoStanCsvReader.skip_warmup
 1 FAILED TEST
test/unit/io/stan_csv_reader_test --gtest_output="xml:test/unit/io/stan_csv_reader_test.xml" failed

[WardBrian]

Looking at the laplace test, it is setting up a writer without a comment prefix
stan::callbacks::stream_writer sample_writer(sample1_ss, "");

and then later feeding that back into the csv_reader, which is now expecting to find #s, maybe?

[mitzimorris]

test/unit/io/stan_csv_reader_test --gtest_output="xml:test/unit/io/stan_csv_reader_test.xml" failed

I just fixed that one - checked in a test file.

[stan-buildbot]

---

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
arma/arma.stan	0.35	0.33	1.03	3.01% faster
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.01	0.01	1.01	1.1% faster
gp_regr/gen_gp_data.stan	0.02	0.02	1.06	6.09% faster
gp_regr/gp_regr.stan	0.1	0.1	0.98	-1.67% slower
sir/sir.stan	72.55	68.8	1.05	5.17% faster
irt_2pl/irt_2pl.stan	4.44	3.96	1.12	10.85% faster
eight_schools/eight_schools.stan	0.06	0.06	1.06	6.07% faster
pkpd/sim_one_comp_mm_elim_abs.stan	0.26	0.24	1.11	10.09% faster
pkpd/one_comp_mm_elim_abs.stan	20.14	18.44	1.09	8.42% faster
garch/garch.stan	0.47	0.41	1.14	12.13% faster
low_dim_gauss_mix/low_dim_gauss_mix.stan	2.86	2.6	1.1	9.0% faster
arK/arK.stan	1.81	1.72	1.05	4.79% faster
gp_pois_regr/gp_pois_regr.stan	2.85	2.69	1.06	5.49% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	9.23	8.37	1.1	9.32% faster
performance.compilation	196.43	187.85	1.05	4.37% faster
Mean result: 1.0686199247172619

---

Jenkins Console Log
Blue Ocean
Commit hash: c83a397be67c59ba135871db25c53717e78d796e

---

Machine information

No LSB modules are available. Distributor ID: Ubuntu Description: Ubuntu 20.04.3 LTS Release: 20.04 Codename: focal

CPU:
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
Address sizes: 46 bits physical, 48 bits virtual
CPU(s): 80
On-line CPU(s) list: 0-79
Thread(s) per core: 2
Core(s) per socket: 20
Socket(s): 2
NUMA node(s): 2
Vendor ID: GenuineIntel
CPU family: 6
Model: 85
Model name: Intel(R) Xeon(R) Gold 6148 CPU @ 2.40GHz
Stepping: 4
CPU MHz: 2400.000
CPU max MHz: 3700.0000
CPU min MHz: 1000.0000
BogoMIPS: 4800.00
Virtualization: VT-x
L1d cache: 1.3 MiB
L1i cache: 1.3 MiB
L2 cache: 40 MiB
L3 cache: 55 MiB
NUMA node0 CPU(s): 0,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46,48,50,52,54,56,58,60,62,64,66,68,70,72,74,76,78
NUMA node1 CPU(s): 1,3,5,7,9,11,13,15,17,19,21,23,25,27,29,31,33,35,37,39,41,43,45,47,49,51,53,55,57,59,61,63,65,67,69,71,73,75,77,79
Vulnerability Gather data sampling: Mitigation; Microcode
Vulnerability Itlb multihit: KVM: Mitigation: VMX disabled
Vulnerability L1tf: Mitigation; PTE Inversion; VMX conditional cache flushes, SMT vulnerable
Vulnerability Mds: Mitigation; Clear CPU buffers; SMT vulnerable
Vulnerability Meltdown: Mitigation; PTI
Vulnerability Mmio stale data: Mitigation; Clear CPU buffers; SMT vulnerable
Vulnerability Retbleed: Mitigation; IBRS
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Mitigation; IBRS, IBPB conditional, STIBP conditional, RSB filling, PBRSB-eIBRS Not affected
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Mitigation; Clear CPU buffers; SMT vulnerable
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cdp_l3 invpcid_single pti intel_ppin ssbd mba ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm cqm mpx rdt_a avx512f avx512dq rdseed adx smap clflushopt clwb intel_pt avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req pku ospke md_clear flush_l1d arch_capabilities

G++:
g++ (Ubuntu 9.4.0-1ubuntu1~20.04) 9.4.0
Copyright (C) 2019 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

Clang:
clang version 10.0.0-4ubuntu1
Target: x86_64-pc-linux-gnu
Thread model: posix
InstalledDir: /usr/bin

[mitzimorris]

that upstream test failure is because now "fixed params" outputs aren't parsing.
the hack would be to make sure that comment block at beginning of CSV file always contains a line "# algorithm = fixed_param". will file issue on CmdStan.

update: issue exists since 3 years: stan-dev/cmdstan#1029

[WardBrian]

I am still not sure I follow what the goal of this PR is -- is there a bug it is fixing?

[mitzimorris]

This will make the returned stan_csv object correctly capture the adaptation information in the case that the Stan CSV file has saved warmup draws - that's technically a bug, no?

It will also ensure that the stan_csv object's samples array contains only post-warmup draws, which, given the logic in the parse function, appears to be what is intended. Without this fix, when the StanCSV output has saved warmup draws, then the returned samples array contains a bunch of draws that aren't used anywhere, which is a waste of memory.

It will also simplify the logic in the stan/mcmc/chains - which I how I ended up going down this rabbit hole.

[WardBrian]

Ah, I missed how this would fix reading adaptation. That is nice to have.

How positive are we that nobody uses the warmup draws read in by this class? I noticed this class is imported at least once in RStan

[mitzimorris]

How positive are we that nobody uses the warmup draws read in by this class? I noticed this class is imported at least once in RStan

Imported, but no longer used? All the diagnostics were rewritten circa 2019 when the improved Rhat paper came out.

[mitzimorris]

I filed stan-dev/cmdstan#1286 on CmdStan which would make deciding when to skip warmup more straightforward.

[mitzimorris]

fixed by #3311