lyst/lightfm

View on GitHub

A Python implementation of LightFM, a hybrid recommendation algorithm.

Stars: 5.1k
Forks: 724
Open beginner issues: 0
Indexed issues: 151

Dominant language: Python
License: Apache-2.0
Last GitHub push: Apr 30, 2023
Latest indexed: Sep 19, 2026
Contributing guide: No contributing guide
Code of conduct: No code of conduct
Beginner labels: help wanted
PR merge metrics: No merged PRs in 30d

Beginner friendly Everything

Open All

151 open issues indexed

lightfm pip install fails on pip==25.3 because of pep517 Open

Difficulty 2/5 1-3 hours Newbie friendliness 55/100

lyst/lightfm#727 · 1 reaction · Opened Nov 24, 2025
bug in lightfm's build system Open

Difficulty 2/5 1-3 hours Newbie friendliness 42/100

lyst/lightfm#725 · 1 comment · 1 reaction · Opened Oct 22, 2025
LightFM .fit() (loss) Open

Difficulty 4/5 3-5 days Newbie friendliness 30/100

lyst/lightfm#724 · 2 comments · Opened Oct 12, 2025
Code silently fails after running model.fit() Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

lyst/lightfm#720 · 4 comments · Opened Nov 28, 2024
Cannot learn from large numbers of interactions Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

lyst/lightfm#719 · Opened Nov 5, 2024
Recommend for cold start user Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

lyst/lightfm#718 · Opened Oct 26, 2024
Recommending documents to users based on keywords in the absence of interactions Open

Difficulty 4/5 3-5 days Newbie friendliness 20/100

lyst/lightfm#716 · Opened Aug 4, 2024
Recommendation model with text embedding as item feature Open

Difficulty 3/5 1-2 days Newbie friendliness 25/100

lyst/lightfm#713 · 2 comments · Opened Jun 18, 2024
ModuleNotFoundError: No module named 'lightfm._lightfm_fast_no_openmp' Open

Difficulty 4/5 3-5 days Newbie friendliness 30/100

lyst/lightfm#712 · 3 reactions · Opened May 27, 2024
Can't seem to install with with poetry Open

Difficulty 3/5 1-2 days Newbie friendliness 35/100

lyst/lightfm#711 · 3 comments · Opened May 22, 2024
Can't install on Python 3.12 Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

lyst/lightfm#709 · 11 comments · 5 reactions · Opened May 17, 2024
Creating recommendations for single users without leveraging user identity features Open

Difficulty 3/5 1-2 days Newbie friendliness 45/100

lyst/lightfm#708 · Opened Apr 3, 2024
Does calculating precision@k for one user makes sense? Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

lyst/lightfm#707 · 1 comment · Opened Feb 22, 2024
LightFM kills Jupyter Kernel when fitting model Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

lyst/lightfm#706 · Opened Jan 31, 2024
Implementing Incremental Training and Feedback loop using Lightfm Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

lyst/lightfm#705 · Opened Jan 26, 2024
Can't install on Python 3.10.6 in Ubuntu Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

lyst/lightfm#703 · 1 comment · 1 reaction · Opened Oct 27, 2023
Auc evalutation return always nan Open

Difficulty 3/5 1-2 days Newbie friendliness 38/100

lyst/lightfm#702 · Opened Oct 27, 2023
Evaluate AUC on a subset of items Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

lyst/lightfm#701 · Opened Oct 24, 2023
How to get clusters of users from trained lightfm model on implicit rating model? Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

lyst/lightfm#700 · Opened Sep 10, 2023
Is there a way to save Dataset()? Open

Difficulty 4/5 3-5 days Newbie friendliness 30/100

lyst/lightfm#698 · 1 comment · Opened Aug 23, 2023
Item and User Normalization Open

Difficulty 4/5 3-5 days Newbie friendliness 30/100

lyst/lightfm#697 · Opened Jul 10, 2023
Incorporating item and user biases in LightFM embeddings for accurate recommendations in Vespa Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

lyst/lightfm#696 · 1 comment · Opened Jul 4, 2023
Recommender System with RFM ratings to put into Interaction matrix Open

Difficulty 5/5 Over a week Newbie friendliness 15/100

lyst/lightfm#693 · 9 comments · Opened Jun 9, 2023
How to reproduce the evaluation score in the paper. Open

Difficulty 4/5 3-5 days Newbie friendliness 30/100

lyst/lightfm#692 · Opened May 24, 2023
Lightfm 1.17 - kernel dies in Jupyter, Spyder when fitting Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

lyst/lightfm#690 · 3 comments · 1 reaction · Opened May 11, 2023
Recommendation for users the new items in the dataset Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

lyst/lightfm#688 · 1 comment · 1 reaction · Opened Apr 28, 2023
Light FM throwing an installation error: AttributeError: 'dict' object has no attribute '__LIGHTFM_SETUP__' Open

Difficulty 2/5 1-3 hours Newbie friendliness 38/100

lyst/lightfm#687 · 8 comments · 11 reactions · Opened Apr 20, 2023
losses except the logistic kills the kernel Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

lyst/lightfm#684 · 3 comments · Opened Apr 12, 2023
do I need to pass user and item matrixes to predict method for known users and items in fit step Open

Difficulty 3/5 1-2 days Newbie friendliness 25/100

lyst/lightfm#683 · Opened Apr 6, 2023
Cannot get MovieLens labels' correctly Open

Difficulty 2/5 1-3 hours Newbie friendliness 52/100

lyst/lightfm#682 · 2 comments · 2 reactions · Opened Mar 23, 2023
Prediction Scores not varying its same for all test users Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

lyst/lightfm#681 · Opened Mar 14, 2023
Python 3.10 (and above) Support Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

lyst/lightfm#675 · 7 comments · Opened Feb 20, 2023
Meaning of the recommendation scores Open

Difficulty 3/5 1-2 days Newbie friendliness 25/100

lyst/lightfm#672 · Opened Dec 14, 2022
UPD: The problem installation persists with version 1.17_30.03.2023/lightfm 1.16 installation problem_11.12.2022 Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

lyst/lightfm#671 · 7 comments · 5 reactions · Opened Dec 11, 2022
How to validate the Dataset.mapping Open

Difficulty 3/5 1-2 days Newbie friendliness 25/100

lyst/lightfm#670 · 1 comment · Opened Dec 5, 2022
How to deploy model using node js Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

lyst/lightfm#669 · 1 comment · Opened Nov 22, 2022
Using LightFM for Retail Customer Recommendation based on purchased products Open

Difficulty 5/5 Over a week Newbie friendliness 15/100

lyst/lightfm#668 · 2 reactions · Opened Nov 20, 2022
Add new user/item ids or features Open

Difficulty 5/5 Over a week Newbie friendliness 30/100

lyst/lightfm#667 · 1 comment · 2 reactions · Opened Nov 6, 2022
Failed to install to macOS Monterey (MacBook Pro 2019; Intel Core i5) Open

Difficulty 3/5 1-2 days Newbie friendliness 30/100

lyst/lightfm#665 · 1 comment · 2 reactions · Opened Oct 24, 2022
LightFM rerecommending items Open

Difficulty 5/5 Over a week Newbie friendliness 20/100

lyst/lightfm#664 · Opened Oct 17, 2022
Temporal splitting Open

Difficulty 4/5 3-5 days Newbie friendliness 30/100

lyst/lightfm#663 · 1 comment · 1 reaction · Opened Oct 12, 2022
Model prediction scores are zero for all user-item pairs Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

lyst/lightfm#662 · 4 comments · Opened Oct 4, 2022
can i set different n_components (the embedding dimension) for user, item and interactions Open

Difficulty 5/5 Over a week Newbie friendliness 15/100

lyst/lightfm#659 · Opened Aug 17, 2022
It is unclear why evaluation metrics (auc_score, precision_at_k) fail when train and test data have high overlap Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

lyst/lightfm#658 · Opened Aug 15, 2022
Dividing data into train/test based on Time Open

Difficulty 4/5 3-5 days Newbie friendliness 28/100

lyst/lightfm#657 · 1 comment · Opened Aug 15, 2022
Do momentum and gradients required for predictions? Open

Difficulty 4/5 3-5 days Newbie friendliness 35/100

lyst/lightfm#656 · Opened Aug 3, 2022
[Feature request] NDCG metric for model evaluation Open

Difficulty 4/5 3-5 days Newbie friendliness 25/100

lyst/lightfm#655 · Opened Jul 30, 2022
Duplicate user-item pairs Open

Difficulty 5/5 Over a week Newbie friendliness 15/100

lyst/lightfm#654 · 1 comment · Opened Jul 28, 2022
Do I need to provide item/user features for predict() if I've already include the features in fit()? Open

Difficulty 2/5 1-3 hours Newbie friendliness 20/100

lyst/lightfm#651 · Opened Jun 27, 2022
Converting ranks into probability to calculate average precision Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

lyst/lightfm#650 · 1 comment · Opened Jun 25, 2022

lyst/lightfm

Get new issues in your inbox