最后,为了可视化这些模型,我强烈推荐将 R 绑定到优秀的 Graphviz 库; R 绑定收集在 R 包 RGraphviz 中。 RGraphviz 在 CRAN 上不可用,但在 Bioconductor 上可用;它取决于 GraphViz。
Yes, there are quite a few actually.
On CRAN, look at the gRaphical Models Task View. Under each header is a listing of R Packages subsumed under that subject header--the first being Representation, manipulation and display of graphs.
Scroll down this page to the last three section headers:
Miscellaneous: Model search, specialized types of models etc.,
Bayesian Networks/Probabilistic expert systems; and
BUGS models, just below it.
Under these three headers are a total of 16 R Packages (seven, three and six packages, respectively). Within these nine, you should have no trouble finding a couple most suited for your project.
The only one I have personally used is bnlearn, a Package for Bayesian network structure learning. This is not my field, so I recall that the Package documentation (Manual and Vignette) are excellent and include working code examples.
Finally, for visualizing these models, I recommend highly the R bindings to the excellent Graphviz Library; the R bindings are gathered in the R Package, RGraphviz. RGraphviz is not available on CRAN but rather on Bioconductor; it depends on GraphViz.
实际上还没有。道格的回答本质上是具有误导性的。如果有人正在寻找所请求模型的非 R 实现,我可以提供此链接。
Not yet, actually. Doug's answer is inherently misleading. If someone's looking for non-R implementation of the requested models, I may provide this link.
是的,实际上有很多。
在 CRAN 上,查看 图形模型任务视图。
每个标题下都有一个包含在该主题标题下的 R 包列表——第一个是图形的表示、操作和显示。
向下滚动此页面到最后三个部分标题:
其他:模型搜索、专门类型的模型等。、
贝叶斯网络/概率专家系统< /em>;和
BUGS 模型,就在它的下面。
这三个标题下总共有 16 R 包(分别为七个、三个和六个包)。在这九个中,您应该可以毫不费力地找到最适合您的项目的几个。
我个人使用过的唯一一个是 bnlearn,一个用于贝叶斯网络结构学习的软件包。这不是我的领域,所以我记得包文档(手册和 Vignette)非常出色,并且包含工作代码示例。
最后,为了可视化这些模型,我强烈推荐将 R 绑定到优秀的 Graphviz 库; R 绑定收集在 R 包 RGraphviz 中。 RGraphviz 在 CRAN 上不可用,但在 Bioconductor 上可用;它取决于 GraphViz。
Yes, there are quite a few actually.
On CRAN, look at the gRaphical Models Task View.
Under each header is a listing of R Packages subsumed under that subject header--the first being Representation, manipulation and display of graphs.
Scroll down this page to the last three section headers:
Miscellaneous: Model search, specialized types of models etc.,
Bayesian Networks/Probabilistic expert systems; and
BUGS models, just below it.
Under these three headers are a total of 16 R Packages (seven, three and six packages, respectively). Within these nine, you should have no trouble finding a couple most suited for your project.
The only one I have personally used is bnlearn, a Package for Bayesian network structure learning. This is not my field, so I recall that the Package documentation (Manual and Vignette) are excellent and include working code examples.
Finally, for visualizing these models, I recommend highly the R bindings to the excellent Graphviz Library; the R bindings are gathered in the R Package, RGraphviz. RGraphviz is not available on CRAN but rather on Bioconductor; it depends on GraphViz.