Fix a Gibbs bug with models where linking changes variable dimension #2472
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
mhauru
force-pushed
the
mhauru/gibbs-linking-fix
branch
from
00c46a6 to
23365a7
Compare
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## master #2472 +/- ##
==========================================
- Coverage 85.01% 79.96% -5.06%
==========================================
Files 21 21
Lines 1582 1582
==========================================
- Hits 1345 1265 -80
- Misses 237 317 +80 ☔ View full report in Codecov by Sentry. |
Pull Request Test Coverage Report for Build 12867839870Details
💛 - Coveralls |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Currently this PR only adds a test, the actual fix is to be implemented. The test reproduces the issue encountered in TuringLang/docs#579.
I think the fix is to just remove this assertion in DPPL: https://github.com/TuringLang/DynamicPPL.jl/blob/938a69dfabb7089f52700ef01fa3a9f2d667b7d1/src/varinfo.jl#L532
I don't see a reason why, when merging two
VarInfos, all the variables should have the same dimension in both VIs being merged. However, due to Chesterton's fence I'm uncomfortable making that change before I've understood why the assertion is there in the first place.