@LensKitRS

Open-source Python tools for #recsys experiments.

lenskit.org
Joined August 2011
  • Tweets118
  • Following31
  • Followers496
Today, we're presenting the next generation of LensKit: Python tools for #recsys experiments. #recsys2018 #reveal lenskit.org/
4
1
17
New LensKit just dropped - 0.14.0 is released, Conda packages should be available soon: github.com/lenskit/lkpy/rele…

Release Cleaning Up · lenskit/lkpy

This release has a lot of cleanups and refactorings. Highlights: ⚠️ MultiEval's deprecation has been upgraded to removal ⚠️ ImplicitMF's default has changed to no longer use rating values,...

github.com
2
2
7
Does your paper at #RecSys2021 or one of the workshops use LensKit (either version)? Please let us know - I would love to include it on lenskit.org/research/.
1
1
We finally have an M1 Mac Mini, have run tests on the relevant packages, and are pleased that LensKit works natively on Apple Silicon. If you want to try it out, we recommend installing Python, LensKit, and dependencies from Miniforge. github.com/conda-forge/minif…

GitHub - conda-forge/miniforge: A conda-forge distribution.

A conda-forge distribution. Contribute to conda-forge/miniforge development by creating an account on GitHub.

github.com
Cut a new quick bugfix release today - 0.13.1 is now on PyPI and conda-forge. This just cleans up some docs and fixes a bug in Bias (this bug doesn't affect correctness, just makes a documented feature work instead of erroring)
1
LensKit 0.13.0 is now available. This includes some significant bug fixes (including an algorithmic correctness fix for biased matrix factorizers), so please upgrade! github.com/lenskit/lkpy/rele…

Release LensKit 0.13.0 - critical bugs fixed · lenskit/lkpy

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 wer...

github.com
2
1
3
We are considering adopting NEP29 to govern LensKit's supported Python versions going forward. If you have input, it's in the Google Groups thread: groups.google.com/g/lenskit-…
1
If we adopt NEP29, the next release will no longer support Python 3.6, so if you use RHEL8 or an old Ubuntu LTS you would need to install a different Python (e.g. from Anaconda) to use the current version of LensKit.
We've now released LensKit 0.12, releases on PyPI and conda-forge. Pulled out a major complexifier into its own package, distribution & install are easier to maintain now. Also, much-improved TensorFlow performance. github.com/lenskit/lkpy/rele…

Release No Longer Candide · lenskit/lkpy

This version of LensKit splits out the CSR routines into a separate CSR package, allowing LensKit to be a pure Python package. This also makes a major change to TensorFlow BPR, using popularity-wei...

github.com
2
LensKit is now in conda-forge: anaconda.org/conda-forge/len… Should make installation even easier now!
1
1
"Going" to #CIKM2020? The Python version of LensKit, our toolkit for #recsys research, will be in the Thursday morning "Search" Research Track session. Preprint: arxiv.org/abs/1809.03125

LensKit for Python: Next-Generation Software for Recommender...

LensKit is an open-source toolkit for building, researching, and learning about recommender systems. First released in 2010 as a Java framework, it has supported diverse published research,...

arxiv.org
2
6
Does your #recsys2020 paper or demo use LensKit, either the Python or Java version? Let us know so we can add it to the list!
1
LensKit Rec. Toolkit retweeted
MovieLens@movielens18 Sep 2020
If you have a moment check out episode 16 of Crash Course AI: "Let’s make a movie recommendation system" They show how you can use the MovieLens datasets, and @LensKitRS (the same algorithm library we use) to build a movie recommender for yourself! youtu.be/iaIW3CO4rcY
3
2
Pleased to announce that the paper on LensKit for Python will appear in the Resource Track for #CIKM2020. Preprint here: arxiv.org/abs/1809.03125 (also, LensKit users, this will be the one to cite once we have a DOI.)

LensKit for Python: Next-Generation Software for Recommender...

LensKit is an open-source toolkit for building, researching, and learning about recommender systems. First released in 2010 as a Java framework, it has supported diverse published research,...

arxiv.org
3
12
Pleased to announce the release of LensKit 0.9.0! #recsys github.com/lenskit/lkpy/rele…

Release Gone in 60 Seconds · lenskit/lkpy

This release has some performance and improvements, including full Python 3.8, Pandas 1.0, and Numba 0.49 testing. This is the last release we expect to use JobLib to parallelize batch prediction a...

github.com
3
2
LensKit for Python 0.8.4 is now available in the usual places (PyPI, Conda). 0.8 brought a bunch of perf improvements, and 0.8.4 fixes some dependency compatibility issues. github.com/lenskit/lkpy/rele…

Release Will It Blend? · lenskit/lkpy

This release cleans up dependency problems to make it easier to reliably install LensKit. We remove some unused utility code that had compatibility problems. Remove CSR.sort_values - we were no l...

github.com
1
We've updated the LensKit for Python paper to reflect what's happened since initial release, like adopting SciKit-style APIs: arxiv.org/abs/1809.03125

LensKit for Python: Next-Generation Software for Recommender...

LensKit is an open-source toolkit for building, researching, and learning about recommender systems. First released in 2010 as a Java framework, it has supported diverse published research,...

arxiv.org
1
LensKit Rec. Toolkit retweeted
PIReT @ Boise State@PIReTship15 Mar 2018
Infrastructure for some of our #recsys research, presented by @Kimura_Tian at @BoiseStateRCD #RCDays.
2
6
Reminder: if you use LensKit in published research, please cite the paper: lenskit.org/research/ This makes it easier for us to track LensKit usage and demonstrate its impact.
1
LensKit Rec. Toolkit retweeted
csrtales@csr_tales29 Dec 2017
The last #csrtale of 2017 comes from @mdekstrand who tells us about yak shaving and alpaca shaving :) medium.com/csr-tales/csr-tal…

CSR Tale #4: Yak Shaving with Michael Ekstrand

Our final CSRTale for 2017 comes from Michael Ekstrand, an assistant professor in the Department of Computer Science at Boise State…

medium.com
3
1
3
Load more