{"id":20063168,"url":"https://github.com/alan-turing-institute/rpsmf","last_synced_at":"2025-05-05T17:32:27.347Z","repository":{"id":54545529,"uuid":"292842707","full_name":"alan-turing-institute/rPSMF","owner":"alan-turing-institute","description":"Code for Probabilistic Sequential Matrix Factorization","archived":false,"fork":false,"pushed_at":"2021-04-27T20:41:53.000Z","size":21466,"stargazers_count":15,"open_issues_count":1,"forks_count":5,"subscribers_count":6,"default_branch":"master","last_synced_at":"2025-04-09T01:51:20.982Z","etag":null,"topics":["bayesian-methods","matrix-factorization","online-algorithms","probabilistic-models","reproducible-research","reproducible-science","sequential-data","time-series","time-series-analysis","time-series-forecasting"],"latest_commit_sha":null,"homepage":"http://proceedings.mlr.press/v130/akyildiz21a.html","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/alan-turing-institute.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2020-09-04T12:29:50.000Z","updated_at":"2024-01-22T00:11:05.000Z","dependencies_parsed_at":"2022-08-13T19:20:10.641Z","dependency_job_id":null,"html_url":"https://github.com/alan-turing-institute/rPSMF","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/alan-turing-institute%2FrPSMF","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/alan-turing-institute%2FrPSMF/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/alan-turing-institute%2FrPSMF/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/alan-turing-institute%2FrPSMF/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/alan-turing-institute","download_url":"https://codeload.github.com/alan-turing-institute/rPSMF/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":252542375,"owners_count":21764956,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["bayesian-methods","matrix-factorization","online-algorithms","probabilistic-models","reproducible-research","reproducible-science","sequential-data","time-series","time-series-analysis","time-series-forecasting"],"created_at":"2024-11-13T13:40:49.173Z","updated_at":"2025-05-05T17:32:22.329Z","avatar_url":"https://github.com/alan-turing-institute.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Probabilistic Sequential Matrix Factorization\n\nThis repository contains the code to reproduce the experiments in:\n\n[**Akyildiz, van den Burg, Damoulas, Steel - Probabilistic Sequential Matrix \nFactorization (2021)**](http://proceedings.mlr.press/v130/akyildiz21a.html)\n\nWork that uses the methods described in the paper or the code in this \nrepository should cite the paper, for instance using the following BibTeX \nentry:\n\n```bib\n@inproceedings{akyildiz2021probabilistic,\n    title={Probabilistic Sequential Matrix Factorization},\n    author={{\\\"O}mer Deniz Akyildiz and Gerrit J. J. {van den Burg} and Theodoros Damoulas and Mark F. J. Steel},\n    booktitle={Proceedings of the 24th International Conference on Artificial Intelligence and Statistics},\n    editor={Banerjee, Arindam and Fukumizu, Kenji},\n    pages={3484--3492},\n    volume={130},\n    year={2021},\n    series={Proceedings of Machine Learning Research},\n    publisher={PMLR},\n}\n```\n\nIf you encounter a problem when using this repository or simply want to ask a \nquestion, please don't hesitate to [open an issue on \nGitHub](https://github.com/alan-turing-institute/rPSMF) or send an email to \n``odakyildiz at turing dot ac dot uk`` and/or ``gertjanvandenburg at gmail dot \ncom``.\n\n## Introduction\n\nOur Probabilistic Sequential Matrix Factorization (PSMF) method allows you to \nmodel high-dimensional timeseries data that exhibits non-stationary dynamics. \nWe also propose a robust variant of the model, called rPSMF, that handles \nmodel misspecification and outliers.\n\nSee [the paper](http://proceedings.mlr.press/v130/akyildiz21a.html) for \nfurther details.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg width=\"40%\" src=\"./.github/rpsmf.png\" alt=\"Illustration of fitting \n  rPSMF to a multidimensional time series that contains outliers\"\u003e\n  \u003cbr\u003e\n  \u003cbr\u003e\n  \u003cspan\u003e\u003ci\u003eIllustration of using rPSMF to model a 20-dimensional time series \n  with non-linear dynamics and t-distributed outliers. Blue lines are the \n  observed data and yellow lines are unobserved future data. The red dashed \n  line shows the predictions from our model.\n  \u003c/i\u003e\u003c/span\u003e\n\u003c/p\u003e\n\n## Usage\n\nThe code in this repository is organized by experiment. The \n[Convergence](./Convergence) directory contains the code for the convergence \nexperiment in the appendix of the paper.\n\nThe accompanying Makefile can be used to reproduce the experiments by simply \nrunning\n\n```bash\n$ make ExperimentSynthetic   # or ExperimentBeijing, or ExperimentImpute, or ExperimentRecursive\n```\n\nThese experiments will be run through a Python virtual environment that will \nbe automatically created with the required dependencies. For the imputation \nexperiment the results are captured in LaTeX tables that are automatically \ngenerated as well.\n\nThe code for the change point detection Experiment and the convergence \nexperiment are written in Matlab, so require a Matlab installation to \nreproduce. See the readme files in the corresponding directories for more \ninformation.\n\n## Notes\n\nThe code is licensed under the MIT license unless otherwise noted, see the \n[LICENSE](./LICENSE) file for further details.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Falan-turing-institute%2Frpsmf","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Falan-turing-institute%2Frpsmf","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Falan-turing-institute%2Frpsmf/lists"}