lmcinnes / lmcinnes/pynndescent

Rebuilding new Pyn descent(numba 0.60, python 3.12) object from existing object(python 3.9))

Open
#252 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
970
Forks
107
PR merge metrics
No merged PRs in 30d

Description

I have a pynn descent object saved using Python 3.9 with Numba 0.56. Now, when I try to load it in Python 3.12, it fails due to the Numba version change, and I no longer have the original data used to create the object.

To resolve this, I plan to create a new object in Python 3.12 by loading the _raw_data parameter in the order specified by np.argsort(_vertex_order). Then, I'll save this new object in Python 3.9 and recreate it in Python 3.12.

Would it work? Any unknowns that i need to be careful of?

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

The issue names a saved pynn descent object and its _raw_data and _vertex_order parameters, but no files, tests, or entry points. Start by examining how those parameters are serialized and restored, then verify whether the proposed ordering survives the Python and Numba version change. Done would require a confirmed migration path or a reproducible compatibility fix.

Written by the indexing model from the issue text.

Assessment

Tech stack
numpy, python
Domain
machine-learning
Issue type
Bug
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.