Bugfix/3301 stan csv reader skip warmup#3302
Conversation
|
@WardBrian - Ubuntu tests failing because of nodejs warning? the offending test uses the Stan CSV reader - but works on Mac. is there something cached that needs cleanup or???? |
|
The node warning is not the source of the error, its There is also a separate failure in Jenkins: |
|
Looking at the laplace test, it is setting up a writer without a comment prefix and then later feeding that back into the csv_reader, which is now expecting to find |
I just fixed that one - checked in a test file. |
Jenkins Console Log Machine informationNo LSB modules are available. Distributor ID: Ubuntu Description: Ubuntu 20.04.3 LTS Release: 20.04 Codename: focalCPU: G++: Clang: |
|
that upstream test failure is because now "fixed params" outputs aren't parsing. update: issue exists since 3 years: stan-dev/cmdstan#1029 |
|
I am still not sure I follow what the goal of this PR is -- is there a bug it is fixing? |
|
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 It will also simplify the logic in the |
|
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 |
Imported, but no longer used? All the diagnostics were rewritten circa 2019 when the improved Rhat paper came out. |
|
I filed stan-dev/cmdstan#1286 on CmdStan which would make deciding when to skip warmup more straightforward. |
|
fixed by #3311 |
Submission Checklist
./runTests.py src/test/unitmake cpplintSummary
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: