spdeaniso
Science Score: 36.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
○CITATION.cff file
-
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
✓Academic publication links
Links to: arxiv.org -
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (4.9%) to scientific vocabulary
Repository
Basic Info
- Host: GitHub
- Owner: inlabru-org
- Language: R
- Default Branch: main
- Size: 545 MB
Statistics
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
SPDEaniso: Tool for working with anisotropic fields using SPDEs
This package can be used to simulate from and perform Bayesian inference on anisotropic Gaussian fields using stochastic partial differential equations (SPDEs). The package is based on the fmesher package. Anisotropy is implemented through a paretrization of the anisotropic SPDE
math
(\kappa^2-\nabla\cdot \mathbf{H}({\mathbf{v}})\nabla)u=\kappa\sigma\mathcal{W}.
The parameters are $\kappa >0$, a two dimensional vector $v$ and $\sigma>0$. These parameters control the length scale and anisotropy, respectively.
Spatially varying parameters $\kappa(x),v(x)$ are supported for simulation. For Bayesian simulation we consider a linear, noisy observation process
math
\mathbf{y} = \mathbf{A}\mathbf{u} + \mathbf{\epsilon},
where $\mathbf{A}$ is an observation matrix and $\mathbf{\epsilon}\sim\mathcal{N}(0,\sigma{\mathbf{\epsilon}}^2\mathbf{I})$ is a vector of independent Gaussian noise. The package supports Bayesian inference on ```math \theta:=(\kappa, \mathbf{v}, \sigma,\sigma{\mathbf{\epsilon}}) , ``` for spatially constant parameters. Here penalized complexity priors are used both for the anisotropy parameters (see...) and the noise parameters.
Owner
- Name: inlabru development organization
- Login: inlabru-org
- Kind: organization
- Email: finn.lindgren@gmail.com
- Repositories: 3
- Profile: https://github.com/inlabru-org
GitHub Events
Total
Last Year
Dependencies
- R >= 3.6 depends
- fmesher * depends
- methods * depends
- Matrix * imports
- lifecycle * imports
- stats * imports
- ggplot2 * suggests
- gsl * suggests
- inlabru >= 2.8.0 suggests
- knitr * suggests
- rmarkdown * suggests
- splancs * suggests
- testthat >= 3.0.0 suggests