pymc-devs/pymc

Increase support for batched multivariate distributions

Open

#5,383 opened on Jan 24, 2022

View on GitHub
 (9 comments) (0 reactions) (0 assignees)Python (1,902 forks)batch import
enhancementshelp wanted

Repository metrics

Stars
 (7,926 stars)
PR merge metrics
 (Avg merge 11d) (7 merged PRs in 30d)

Description

The logp of several multivariate distributions does not work (or is not tested) for arbitrarily batched dimensions. Some cases I could confirm include:

  • MvNormal - #6897
  • MvStudentT #6897
  • KroneckerNormal
  • MatrixNormal
  • LKJCorr
  • LKJCholeskyCov
  • StickBreakingWeights (alpha) #6042
  • Wishart (may be fine to ignore) #8246
  • CARRV (maybe better to leave this one as is)

Reproducible code:

size = (4, 3)
pm.logp(pm.MvNormal.dist(mu=np.ones(2), cov=np.eye(2), size=size), np.ones((*size, 2)))
# ValueError: Invalid dimension for value: 3
pm.logp(pm.MvStudentT.dist(nu=3, mu=np.ones(2), cov=np.eye(2), size=size), np.ones((*size, 2)))
# ValueError: Invalid dimension for value: 3

Distributions that already support (and have tests for) arbitrary shapes

Contributor guide

Research direction
Inspect the logp method implementation for each listed multivariate distribution and trace the shape handling to identify where batched dimensions fail.
Tech stack
python
Domain
datamachine learningaibackend
Issue type
Bug
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
3-5 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
PythonPyMCProbability
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
55