Skip to content

Conversation

penelopeysm
Copy link
Member

@penelopeysm penelopeysm commented May 21, 2025

Important

This PR is basically complete in terms of adding the models. However, unfortunately, this is blocked by DynamicPPL 0.37 as there are spurious errors in the AD accuracy testing. These were fixed in TuringLang/DynamicPPL.jl#964 but have yet to be released.

This PR adds the models benchmarked in https://arxiv.org/pdf/2002.02702.

Some code was not very reproducible so had to be updated.

Closes #21

@penelopeysm penelopeysm force-pushed the py/dppl-models branch 2 times, most recently from 0db8abf to 39487cc Compare May 22, 2025 01:23
@penelopeysm penelopeysm changed the title [WIP] Add DPPL models Add DPPL models Jun 19, 2025
@penelopeysm penelopeysm marked this pull request as ready for review June 19, 2025 23:52
@penelopeysm penelopeysm marked this pull request as draft July 22, 2025 01:08
@penelopeysm penelopeysm marked this pull request as ready for review August 14, 2025 14:00
@penelopeysm penelopeysm merged commit 0db6027 into main Aug 14, 2025
53 checks passed
@penelopeysm penelopeysm deleted the py/dppl-models branch August 14, 2025 16:11
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Benchmark results are likely dominated by overheads instead of actual computation
1 participant