ACL calculation, sampler thinned samples extraction#58
ACL calculation, sampler thinned samples extraction#58Richard-Sti wants to merge 48 commits intocdcapano:masterfrom
Conversation
|
Hello @Richard-Sti! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found: There are currently no PEP 8 issues detected in this Pull Request. Cheers! 🍻 Comment last updated at 2022-04-22 20:11:25 UTC |
|
Possible further addition is to add the Gelmen-Rubin statistic. Moreover, epsie's internal estimate of the ACL can potentially be used to introduce new sampler termination conditions until N independent samples are collected. Blobs when thinning? |
|
Added the GH diagnostic for MH sampler. Need to check what is the best way to calculate it for the PT sampler. @cdcapano please check the GH calculation. |
|
When thinning samples return the loglikegood and log prior, along with whatever the blobs are. |
Richard-Sti
force-pushed
the
acl_calculation
branch
from
96a73bc to
601d460
Compare
|
The German Rubin statistic sometimes produces values of < 1 for a PT sampler. Bug? Will fail tests. |
2 participants
PR to add a classical ACL calculation and thinning sampler's samples.