lenskit / lenskit/lkpy

Add support for metrics comparing two stages of a pipeline

Open
#767 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

evaluation
Dominant language
Python
Stars
314
Forks
77
Avg merge
4d 6m
Merged PRs (30d)
10

Description

Right now, all recommendation lists only look at the final stage.

However, we also want to support things like rank-biased overlap, which measures the difference between two stages of the pipeline.

This may be distinct from #766, as that ticket is about supporting lists from two different parameters. However, it is likely that we want to use the same machinery.

Specifically, the following:

  • The run analyzer supports pairwise metrics.
  • For pairwise metrics, a reference run can be specified. If specified, all other runs are compared to that run.
  • To support RBO, a plain Top-N run can be used as the reference run.
  • If no reference is supplied, all pairs are computed.

Contributor guide

No contributing guide indexed for this repository

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

Start by locating the run analyzer and its existing pairwise-metric support, then trace how runs and recommendation lists are selected. Define tests for an explicit reference run, a plain Top-N reference for RBO, and the no-reference case where all pairs are computed; done means each comparison mode produces the expected metrics.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Feature
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.