Preserve the identities of value/observed variables #248
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
brandonwillard
added
important
brandonwillard
force-pushed
the
preserve-valued-var-identity
branch
2 times, most recently
from
241058c
to
43bb99b
Compare
brandonwillard
force-pushed
the
preserve-valued-var-identity
branch
from
43bb99b
to
a05f707
Compare
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Outdated suggestions cannot be applied.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.
This PR changes the IR generation process so that value/observed variables are never rewritten and their identities are necessarily preserved in the log-probability results. Basically, we create dummy variables for them during IR rewriting and replace them with the value variables at the end (i.e. when log-probabilities are constructed). Most of the changes have to do with the special case of transformations, which demand the ability to change transformed value variables after IR rewriting.