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Outdated suggestions cannot be applied.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.
JLD2 might deal better with objects, although it will not save the model function correctly (it's just a ref) but hopefully, when resuming sampling this isn't an issue anyway because it should not reuse any of that if we send in the sampler, model func, etc to the
sample
call.Fixes #151
This removes dependencies for HDF5 and MCMCChainsStorage, and adds JLD2. The save/resume functionality now just saves the entire chain with
JLD2.save_object
... this might need some further tweaking / testing as it throws a warning.A test has been added to continue sampling from a saved chain, which seems to work, but I can't guarantee it really "resets" the state (in fact I'm pretty sure it doesn't) so I'm not sure how to make the test more realistic (maybe create a new model that's exactly the same, a new sampler instance, etc?)