Stan development repository. The master branch contains the current release. The develop branch contains the latest stable development. See the Developer Process Wiki for details.
The laplace service method added in #3148 produces a CSV file where the final two columns are log_p and log_q. This has two issues:
All other services have "algorithmic parameters" (e.g. lp__ for sampling, log_g__ for ADVI) output before the model's parameters
These names are not suffixed with the usual two underscores, which means a user can write a model with a parameter named log_p and receive a confounding CSV header which many libraries will refuse to parse:
The laplace service method added in #3148 produces a CSV file where the final two columns are
log_p
andlog_q
. This has two issues:lp__
for sampling,log_g__
for ADVI) output before the model's parameterslog_p
and receive a confounding CSV header which many libraries will refuse to parse:Current Version:
v2.32.0