LensKit 0.13.0 - critical bugs fixed

[mdekstrand]

We're pleased to release LensKit 0.13!

Major Fixes

This release includes two critical fixes, for which everyone should upgrade:

• The Bias model's transform and inverse_transform methods were incorrect (Fix incorrect user bias transformation #265). These bugs did not affect Bias when used as a predictor or a recommender, but they did affect any model using Bias as a normalization step, namely the biased matrix factorizers (since version 0.11, when this API was added).
• Previous versions of LensKit did not clean up temporary files (or, on Python 3.8 and later, shared memory resources) when running parallel evaluation processes.

Future Changes

This release deprecates two sets of APIs that will be removed in LensKit 0.14:

• MultiEval (Remove MultiEval #254) - it doesn't work well for realistic projects, and simple evaluations are easy enough to write in a loop, so we will be removing MultiEval to reduce our maintenance burden going forward.
• RNG seed management APIs - these are replaced by seedbank. In 0.13, the APIs are kept as compatibility shims for their SeedBank replacements, but we will remove them in 0.14 in favor of directly calling seedbank.

We haven't yet adopted any formal deprecation policies for LensKit, but my current tentative plan is to use this next-release cadence for nontrivial removals while we're still releasing 0.x versions; once we decide to bump to 4.x, we will use semantic versioning on all public APIs, and thus deprecations will not be enforced until the next major release.

What’s Changed

• Remove fastparquet import (Remove fastparquet import #266) @mdekstrand
• Fix incorrect user bias transformation (Fix incorrect user bias transformation #265) @mdekstrand
• Revise dependency specifications (Revise dependency specifications #264) @mdekstrand
• Update deprecation notices (Update deprecation notices #263) @mdekstrand
• Detect problems with runtime environments (Detect problems with runtime environments #248) @mdekstrand
• Add use_ratings option to ImplicitMF (Add use_ratings option to ImplicitMF #245) @mdekstrand
• Add 'k' support to top-N metrics (Add 'k' support to top-N metrics #247) @mdekstrand
• Further Top-N optimization updates (Further Top-N optimization updates #242) @mdekstrand
• Free shared memory in parallel (Free shared memory in parallel #243) @mdekstrand
• Optimize top-N analysis (Optimize top-N analysis #237) @mdekstrand
• Add PlackettLuce stochastic ranking algorithm (Add PlackettLuce stochastic ranking algorithm #241) @mdekstrand
• Add PopScore algorithm for popularity-based scoring (Add PopScore algorithm for popularity-based scoring #240) @mdekstrand
• Refactor ranking into a 'ranking' module (Refactor ranking into a 'ranking' module #239) @mdekstrand
• Enable tests on Python 3.9 (Enable tests on Python 3.9 #234) @mdekstrand
• Deprecate MultiEval (Deprecate MultiEval #238) @mdekstrand
• Add more logging output to parallelism code (Add more logging output to parallelism code #236) @mdekstrand
• Add convenience prediction accuracy functions (Add convenience prediction accuracy functions #235) @mdekstrand

---

This discussion was created from the release LensKit 0.13.0 - critical bugs fixed.