Overview | Quick start: bivariate Gaussian mixture | Isotropic fit | Diagonal fit | Full precision matrix fit | Model comparison via Bayes factors | Evaluating the fitted density | Online learning | References
Package: epkde 0.1.0
epkde: Bayesian Bandwidth Selection for Multivariate KDE via Expectation Propagation
Implements the approximate Bayesian method for bandwidth selection in multivariate kernel density estimation (KDE) proposed in Filippone & Sanguinetti (2011) <doi:10.1016/j.csda.2011.05.023>. The method uses the Expectation Propagation (EP) algorithm to approximate the posterior distribution of the inverse bandwidth (precision matrix) under a leave-one-out cross-validated likelihood. Three covariance structures are supported: isotropic (scalar precision), diagonal, and full precision matrix. Online Bayesian updating is supported for the isotropic case. The approximate posterior can be used for bandwidth selection, model comparison (via the model evidence / Bayes factor), and online learning.
Authors:
epkde_0.1.0.tar.gz
epkde_0.1.0.zip(r-4.7-any)epkde_0.1.0.zip(r-4.6-any)epkde_0.1.0.zip(r-4.5-any)
epkde_0.1.0.tgz(r-4.6-any)epkde_0.1.0.tgz(r-4.5-any)
epkde_0.1.0.tar.gz(r-4.7-any)epkde_0.1.0.tar.gz(r-4.6-any)
epkde_0.1.0.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION
card.svg |card.png
epkde/json (API)
| # Install 'epkde' in R: |
| install.packages('epkde', repos = c('https://mauriziofilippone.r-universe.dev', 'https://cloud.r-project.org')) |
Bug tracker:https://github.com/mauriziofilippone/epkde/issues
Last updated from:23e76bece7. Checks:7 NOTE, 2 OK. Indexed: yes.
| Target | Result | Time | Files | Syslog |
|---|---|---|---|---|
| linux-devel | NOTE | 173 | ||
| source / vignettes | OK | 151 | ||
| linux-release | NOTE | 104 | ||
| macos-release | NOTE | 157 | ||
| macos-oldrel | NOTE | 192 | ||
| windows-devel | NOTE | 70 | ||
| windows-release | NOTE | 60 | ||
| windows-oldrel | NOTE | 64 | ||
| wasm-release | OK | 78 |
Exports:ep_kde_diagonalep_kde_fullep_kde_isotropicep_kde_onlinekde_predictmodel_evidence
Dependencies:
Readme and manuals
Help Manual
| Help page | Topics |
|---|---|
| EP inference for diagonal KDE bandwidth | ep_kde_diagonal |
| EP inference for full KDE bandwidth (precision matrix) | ep_kde_full |
| EP inference for isotropic KDE bandwidth | ep_kde_isotropic |
| Online EP update for isotropic KDE bandwidth | ep_kde_online |
| Evaluate kernel density estimate at test points | kde_predict |
| Extract the log model evidence for model comparison | model_evidence |