Make initial_state as a keyword argument
#557
Merged
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
sunxd3
approved these changes
Nov 9, 2023
There was a problem hiding this comment.
I had the same solution, but unsure of the dynamics of the AbstractMCMC interface changes, so maybe wait for @devmotion?
| @@ -87,9 +87,9 @@ function AbstractMCMC.sample( | |||
| N::Integer; | |||
| chain_type=default_chain_type(sampler), | |||
| resume_from=nothing, | |||
There was a problem hiding this comment.
Do you know where resume_from is documented?
devmotion
approved these changes
Nov 9, 2023
There was a problem hiding this comment.
keeping it backwards compatible for now.
AFAICT we have to tag a breaking release of DynamicPPL anyway, regardless of initial_state/resume_from since init_params does not work anymore.
initial_state and resume_from are a bit different though, initial_state is supposed to be an actual state/transition of the sampler whereas currently for resume_from we support Chains objects (which potentially store the final state of the sampler as metadata). I'm not sure though if maybe we should just demand that users extract the state manually (possibly with some helper function) in the future.
I think there was some reason for why I did not add it as a keyword argument but right now I don't remember and I don't see an immediate problem.
yebai
merged commit 3687ebe
into
compathelper/new_version/2023-10-26-00-09-06-954-01573151329
github-merge-queue bot
pushed a commit
that referenced
this pull request
Nov 9, 2023
…t) (#551) * CompatHelper: bump compat for AbstractMCMC to 5, (keep existing compat) * CompatHelper: bump compat for AbstractMCMC to 5 for package test, (keep existing compat) (#552) Co-authored-by: CompatHelper Julia <[email protected]> * Update to AbstractMCMC 5 * Apply suggestions from code review Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> * Update sampler.jl * CompatHelper: bump compat for AbstractMCMC to 5 for package test, (keep existing compat) (#553) * Fix for `rand` + replace overloads of `rand` with `rand_prior_true` for testing models (#541) * preserve context from model in `rand` * replace rand overloads in TestUtils with definitions of rand_prior_true so we can properly test rand * removed NamedTuple from signature of TestUtils.rand_prior_true * updated references to previous overloads of rand to now use rand_prior_true * test rand for DEMO_MODELS * formatting * fixed tests for rand for DEMO_MODELS * fixed linkning tests * added missing impl of rand_prior_true for demo_static_transformation * formatting * fixed rand_prior_true for demo_static_transformation * bump minor version as this will be breaking * bump patch version * fixed old usage of rand * Update test/varinfo.jl Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> * fixed another usage of rand --------- Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> * Remove `tonamedtuple` (#547) * Remove dependencies to `tonamedtuple` * Remove `tonamedtuple`s * Minor version bump --------- Co-authored-by: Hong Ge <[email protected]> * CompatHelper: bump compat for AbstractMCMC to 5 for package test, (keep existing compat) --------- Co-authored-by: Tor Erlend Fjelde <[email protected]> Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: Xianda Sun <[email protected]> Co-authored-by: Hong Ge <[email protected]> Co-authored-by: CompatHelper Julia <[email protected]> * bump AbstractPPL version to 0.7 * Update AbstractPPL test dependency * add `Random.AbstractRNG` * Update sampler.jl (#557) * Update src/sampler.jl Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> --------- Co-authored-by: CompatHelper Julia <[email protected]> Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: David Widmann <[email protected]> Co-authored-by: Tor Erlend Fjelde <[email protected]> Co-authored-by: Xianda Sun <[email protected]> Co-authored-by: Hong Ge <[email protected]> Co-authored-by: Xianda Sun <[email protected]>
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.
@devmotion, maybe this will do the trick? The advantage is it allows one to override the
initial_stateargument while keeping it backwards compatible for now. We can removeresume_fromin the next breaking release.