Include deterministic variables in AutoDelta's sample_posterior #1584
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Outdated suggestions cannot be applied.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 attempts to fix #951, by including deterministic variables in the output of AutoDelta's sample_posterior
Changes made to AutoDelta's
sample_posterior
method:Predictive
instance using the latent samples and generate samples of only the deterministic variables with thereturn_sites
keyword. Since the Predictive instance in the newsample_posterior
method requires the model's*args
and**kwargs
, these are now passed tosample_posterior
as wellsample_posterior
methodImportant: Since the
Predictive
instance is called in the newsample_posterior
method, now the model needs to be callable without arguments, if no arguments are passed tosample_posterior
.Therefore, the following existing tests have been slightly modified, such that the model can also be called without passing data
New tests added:
sample_posterior
method contains deterministic variables (test_autodelta_capture_deterministic_variables)sample_shape
argument is used in the newsample_posterior
method (test_autodelta_sample_posterior_with_sample_shape)