LensKit

LensKit is a set of Python tools for experimenting with and studying recommender systems. It provides support for training, running, and evaluating recommender algorithms in a flexible fashion suitable for research and education.

LensKit for Python (also known as LKPY) is the successor to the Java-based LensKit toolkit and a part of the LensKit project.

If you use Lenskit in published research, cite [LKPY].

[LKPY]

Michael D. Ekstrand. 2020. LensKit for Python: Next-Generation Software for Recommender Systems Experiments. In <cite>Proceedings of the 29th ACM International Conference on Information and Knowledge Management</cite> (CIKM ‘20). DOI:10.1145/3340531.3412778. arXiv:1809.03125 [cs.IR].

Throughout this documentation, we use the notation of Ekstrand and Konstan [EK19].

Resources

• Mailing list, etc.

• Source and issues on GitHub

Overview

• Install LensKit
• Getting Started
  • Setup
  • Loading Data
  • Defining Algorithms
  • Running the Evaluation
• Examples
• Release Notes

Running Experiments

• Loading Data
  • Data Loaders
  • MovieLens Data Sets
• Splitting Data
  • Row-based splitting
  • User-based splitting
  • Utility Classes
• Batch-Running Recommenders
  • Recommendation
  • Rating Prediction
  • Isolated Training
• Evaluating Recommender Output
  • Prediction Accuracy Metrics
  • Top-N Evaluation
  • Saving and Loading Outputs

Algorithms

• Algorithm Interfaces
• Algorithm Summary
• Basic and Utility Algorithms
• Ranking Methods
• Bias
• k-NN Collaborative Filtering
• Classic Matrix Factorization
• Add-On Algorithms

Tips and Tricks

• Performance Tips
  • Quick Tips
  • Controlling Parallelism
  • Other Notes
• Errors and Diagnostics
  • Logging
  • Warnings
• Algorithm Implementation Tips
  • Performance
  • Pickling and Sharing
  • Random Number Generation

Configuration and Internals

• Utility Functions
  • Data Utilities
  • Math utilities
  • Miscellaneous
• LensKit Internals
  • Model Sharing
  • Parallel Execution

Links

• References

Indices and tables

• Index

• Module Index

• Search Page

Acknowledgements

This material is based upon work supported by the National Science Foundation under Grant No. IIS 17-51278. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.