{"id":16276442,"url":"https://github.com/hendersontrent/tsgp","last_synced_at":"2026-01-17T22:51:44.537Z","repository":{"id":238109227,"uuid":"795887965","full_name":"hendersontrent/tsgp","owner":"hendersontrent","description":"Simple Time Series Modelling Using Gaussian Processes","archived":false,"fork":false,"pushed_at":"2024-05-05T23:21:53.000Z","size":8657,"stargazers_count":0,"open_issues_count":4,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-02-08T05:15:01.575Z","etag":null,"topics":["bayesian-inference","bayesian-statistics","gaussian-processes","machine-learning","statistics","time-series"],"latest_commit_sha":null,"homepage":"https://hendersontrent.github.io/tsgp/","language":"R","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/hendersontrent.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-05-04T10:57:36.000Z","updated_at":"2024-05-05T23:21:57.000Z","dependencies_parsed_at":"2024-05-04T11:50:53.417Z","dependency_job_id":null,"html_url":"https://github.com/hendersontrent/tsgp","commit_stats":null,"previous_names":["hendersontrent/tsgp"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hendersontrent%2Ftsgp","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hendersontrent%2Ftsgp/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hendersontrent%2Ftsgp/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/hendersontrent%2Ftsgp/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/hendersontrent","download_url":"https://codeload.github.com/hendersontrent/tsgp/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246828500,"owners_count":20840474,"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-inference","bayesian-statistics","gaussian-processes","machine-learning","statistics","time-series"],"created_at":"2024-10-10T18:48:23.671Z","updated_at":"2026-01-17T22:51:44.511Z","avatar_url":"https://github.com/hendersontrent.png","language":"R","funding_links":[],"categories":[],"sub_categories":[],"readme":"\n# tsgp \u003cimg src=\"man/figures/logo.png\" align=\"right\" width=\"120\" /\u003e\n\nSimple, lightweight R package for fitting, visualising, and predicting\ntime-series data using Gaussian processes.\n\n## Installation\n\nYou can install the development version of `tsgp` from GitHub using the\nfollowing:\n\n``` r\ndevtools::install_github(\"hendersontrent/tsgp\")\n```\n\n## Premise\n\n`tsgp` implements the functionality presented in a [recent\ntutorial](https://hendersontrent.github.io/posts/2024/05/gaussian-process-time-series/)\nfor modelling time-series data with Gaussian processes. Currently, only\nthe univariate setting is supported. `tsgp` works on a [structural time\nseries](https://www.sciencedirect.com/science/article/abs/pii/S0169716105800458)\nperspective. That is, by decomposing a time series into its constituent\nstatistical parts (e.g., trend, seasonality, noise), one can model each\ncomponent independently before combining them to form the complete\npicture of temporal dynamics. This is not only intuitive, but it is also\nhighly transparent—meaning that intelligent and justifiable modelling\ndecisions must be made in order to appropriately capture the data\ngenerating process.\n\n`tsgp` is extremely lightweight in both its dependencies and\ncomputational approach. If you are seeking a more rigorous or flexible\napproach to using GPs for time-series analysis, please look into\n[`Stan`](https://mc-stan.org),\n[`GPy`](https://gpy.readthedocs.io/en/deploy/),\n[`GauPro`](https://github.com/CollinErickson/GauPro), [Tensorflow\nProbability](https://www.tensorflow.org/probability), or\n[`GaussianProcesses.jl`](https://github.com/STOR-i/GaussianProcesses.jl).\n\n## Functionality\n\nCurrently, `tsgp` supports the following covariance functions (kernels):\n\n- Exponentiated quadratic (squared exponential)\n- Rational quadratic\n- Periodic\n- Linear\n\n`tsgp` flexibly enables composite kernels (either through addition or\nmultiplication) to be constructed and is actively encouraged to\nappropriately model complex temporal dynamics.\n\n`tsgp` also includes functions for computing and visualising draws from\nGaussian process priors and posteriors, visualising covariance matrices,\nand plotting predictions.\n\n## Performance\n\n`tsgp` is *extremely* fast at what it does. Well, as fast as it can be\ngiven the computation time involved in computing a GP posterior. `tsgp`\nimplements well-known methods for efficiency and stability, such as\nusing the Cholesky factorisation instead of computing a matrix inverse\ndirectly. A full model with trend, seasonality, and noise can be\ncalculated on a time series of $T = 1000$ time points in a few seconds.\nThis makes it an ideal tool for iterative model building and principled\ntime-series exploration.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhendersontrent%2Ftsgp","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fhendersontrent%2Ftsgp","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fhendersontrent%2Ftsgp/lists"}