Initial AbstractMCMC.step should not sample#366
Merged
Conversation
should not perform any sampling of the parameters.
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
devmotion
previously approved these changes
Apr 24, 2024
yebai
approved these changes
Nov 1, 2024
Member
|
Is this good to merge? Except that we should also bump a breaking version number. |
Member
Author
|
I think there are still some tests that need to be updated, but otherwise, yeah it should be good:) |
torfjelde
added a commit
that referenced
this pull request
Jan 13, 2025
This reverts commit 32d0446.
torfjelde
added a commit
that referenced
this pull request
Jan 13, 2025
This reverts commit 32d0446.
Member
Author
|
Accidentally merged this 🤦 Will re-open |
Member
|
I wasn't prompted by @MateusMaiaDS to look into if we could make sure that the initialization is included in the chain for debugging purposes. @penelopeysm points this PR to me (I remembered I read it time ago.) There is an old issue at TuringLang/Turing.jl#1282 also related to this. Looking at main, I think the merge here was reverted because main branch doesn't contain the code. Thoughts on should we still push this through @torfjelde @yebai @devmotion ? |
This file contains hidden or 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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Generally speaking, the first
AbstractMCMC.stepshould just construct the initial transition and state, not actually perform any sampling. If it performs sampling in the initial step, it becomes increasingly difficult to test that arguments such asinitial_paramsandinitial_stateare properly respected (since we can't just compare the resultinginitial_paramswith the ones produced by the first step). Moreover, it can hide certain behaviors from the user since a traceplot will show the "starting value" as something different than what they set.This PR just makes the initial
AbstractMCMC.stepdo exactly this.Should be a quick merge. @devmotion @yebai @xukai92