Package: poismf 0.4.0-3
poismf: Factorization of Sparse Counts Matrices Through Poisson Likelihood
Creates a non-negative low-rank approximate factorization of a sparse counts matrix by maximizing Poisson likelihood with L1/L2 regularization (e.g. for implicit-feedback recommender systems or bag-of-words-based topic modeling) (Cortes, (2018) <arxiv:1811.01908>), which usually leads to very sparse user and item factors (over 90% zero-valued). Similar to hierarchical Poisson factorization (HPF), but follows an optimization-based approach with regularization instead of a hierarchical prior, and is fit through gradient-based methods instead of variational inference.
Authors:
poismf_0.4.0-3.tar.gz
poismf_0.4.0-3.zip(r-4.5)poismf_0.4.0-3.zip(r-4.4)poismf_0.4.0-3.zip(r-4.3)
poismf_0.4.0-3.tgz(r-4.4-x86_64)poismf_0.4.0-3.tgz(r-4.4-arm64)poismf_0.4.0-3.tgz(r-4.3-x86_64)poismf_0.4.0-3.tgz(r-4.3-arm64)
poismf_0.4.0-3.tar.gz(r-4.5-noble)poismf_0.4.0-3.tar.gz(r-4.4-noble)
poismf_0.4.0-3.tgz(r-4.4-emscripten)poismf_0.4.0-3.tgz(r-4.3-emscripten)
poismf.pdf |poismf.html✨
poismf/json (API)
| # Install 'poismf' in R: |
| install.packages('poismf', repos = c('https://david-cortes.r-universe.dev', 'https://cloud.r-project.org')) |
Bug tracker:https://github.com/david-cortes/poismf/issues
implicit-feedbackpoisson-factorization
Last updated 3 months agofrom:6a218d6781. Checks:OK: 4 NOTE: 5. Indexed: yes.
| Target | Result | Date |
|---|---|---|
| Doc / Vignettes | OK | Sep 16 2024 |
| R-4.5-win-x86_64 | NOTE | Sep 16 2024 |
| R-4.5-linux-x86_64 | NOTE | Sep 16 2024 |
| R-4.4-win-x86_64 | NOTE | Sep 16 2024 |
| R-4.4-mac-x86_64 | NOTE | Sep 16 2024 |
| R-4.4-mac-aarch64 | NOTE | Sep 16 2024 |
| R-4.3-win-x86_64 | OK | Sep 16 2024 |
| R-4.3-mac-x86_64 | OK | Sep 16 2024 |
| R-4.3-mac-aarch64 | OK | Sep 16 2024 |
Exports:factorsfactors.singleget.factor.matricesget.model.mappingspoismfpoismf_unsafepredict.poismfprint.poismfsummary.poismftopNtopN.new
Readme and manuals
Help Manual
| Help page | Topics |
|---|---|
| Determine latent factors for new rows/users | factors |
| Get latent factors for a new user given her item counts | factors.single |
| Extract Latent Factor Matrices | get.factor.matrices |
| Extract user/row and item/column mappings from Poisson model. | get.model.mappings |
| Factorization of Sparse Counts Matrices through Poisson Likelihood | poismf |
| Poisson factorization with no input casting | poismf_unsafe |
| Predict expected count for new row(user) and column(item) combinations | predict.poismf |
| Get information about poismf object | print.poismf |
| Get information about poismf object | summary.poismf |
| Rank top-N highest-predicted items for an existing user | topN |
| Rank top-N highest-predicted items for a new user | topN.new |
