Skip to content

Commit 5fd88bb

Browse files
itsrainingdataitsrainingdata
committed
Merge branch 'dev'
2 parents 6a1141b + 914fe66 commit 5fd88bb

File tree

5 files changed

+119
-9
lines changed

5 files changed

+119
-9
lines changed

‎.Rbuildignore

+4-3
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,6 @@
1-
^.*\.Rproj$
2-
^\.Rproj\.user$
1+
### For some reason Hadley's regex's from http://r-pkgs.had.co.nz/package.html are being ignored, so do this manually
2+
^cran-comments\.md$
33
^README\.Rmd$
4-
^README-*\.png$
4+
^ccdrAlgorithm\.Rproj$
5+
^\.Rproj\.user$
56
^\.travis\.yml$

‎DESCRIPTION

+4-4
Original file line numberDiff line numberDiff line change
@@ -1,14 +1,14 @@
11
Package: ccdrAlgorithm
22
Title: CCDr Algorithm for Learning Sparse Gaussian Bayesian Networks
3-
Version: 0.0.0.9000
4-
Date: 2016-04-05
3+
Version: 0.0.1
4+
Date: 2016-08-08
55
Authors@R: person("Bryon", "Aragam", email = "[email protected]", role = c("aut", "cre"))
66
Maintainer: Bryon Aragam <[email protected]>
7-
Description: Implementation of the CCDr algorithm described in Aragam and Zhou (2015).
7+
Description: Implementation of the CCDr (Concave penalized Coordinate Descent with reparametrization) structure learning algorithm as described in Aragam and Zhou (2015) <http://www.jmlr.org/papers/v16/aragam15a.html>. This is a fast, score-based method for learning Bayesian networks that uses sparse regularization and block-cyclic coordinate descent.
88
Depends:
99
R (>= 3.2.3)
1010
Imports:
11-
sparsebnUtils,
11+
sparsebnUtils (>= 0.0.2),
1212
Rcpp (>= 0.11.0)
1313
LinkingTo: Rcpp
1414
Suggests:

‎README.Rmd

+32-1
Original file line numberDiff line numberDiff line change
@@ -16,5 +16,36 @@ knitr::opts_chunk$set(
1616

1717
[![Travis-CI Build Status](https://travis-ci.org/itsrainingdata/ccdrAlgorithm.svg?branch=master)](https://travis-ci.org/itsrainingdata/ccdrAlgorithm)
1818

19-
This package is currently under construction. If you are looking for the source code to the CCDr algorithm, please go to https://github.com/itsrainingdata/ccdr.
19+
`ccdrAlgorithm` implements the CCDr structure learning algorithm described in \[[1](#references)\]. Based on observational data, this algorithm estimates the structure of a Bayesian network (aka edges in a DAG) using penalized maximum likelihood based on L1 or concave (MCP) regularization.
2020

21+
Presently, this package consists of a single method that implements the main algorithm; more functionality will be provided in the future. To generate data from a given Bayesian network and/or simulate random networks, the following R packages are recommended:
22+
23+
- `bnlearn`: [bnlearn on CRAN](https://cran.r-project.org/package=bnlearn), [www.bnlearn.com](http://www.bnlearn.com)
24+
- `pcalg`: [pcalg on CRAN](https://cran.r-project.org/package=pcalg)
25+
- `igraph`: [igraph on CRAN](https://cran.r-project.org/package=igraph), [http://igraph.org/r/](http://igraph.org/r/)
26+
27+
## Overview
28+
29+
The main method is `ccdr.run`, which runs the CCDr structure learning algorithm as described in \[[1](#references)\].
30+
31+
## Installation
32+
33+
You can install:
34+
35+
* the latest CRAN version with
36+
37+
```R
38+
install.packages("ccdrAlgorithm")
39+
````
40+
41+
* the latest development version from GitHub with
42+
43+
```R
44+
devtools::install_github(c("itsrainingdata/sparsebnUtils/dev", "itsrainingdata/ccdrAlgorithm/dev"))
45+
```
46+
47+
## References
48+
49+
\[1\] Aragam, B. and Zhou, Q. (2015). [Concave penalized estimation of sparse Gaussian Bayesian networks.](http://jmlr.org/papers/v16/aragam15a.html) _The Journal of Machine Learning Research_. 16(Nov):2273−2328.
50+
51+
\[2\] Fu, F. and Zhou, Q. (2013). [Learning sparse causal Gaussian networks with experimental intervention: Regularization and coordinate descent.](http://amstat.tandfonline.com/doi/abs/10.1080/01621459.2012.754359) Journal of the American Statistical Association, 108: 288-300.

‎README.md

+36-1
Original file line numberDiff line numberDiff line change
@@ -3,4 +3,39 @@ ccdrAlgorithm
33

44
[![Travis-CI Build Status](https://travis-ci.org/itsrainingdata/ccdrAlgorithm.svg?branch=master)](https://travis-ci.org/itsrainingdata/ccdrAlgorithm)
55

6-
This package is currently under construction. If you are looking for the source code to the CCDr algorithm, please go to <https://github.com/itsrainingdata/ccdr>.
6+
`ccdrAlgorithm` implements the CCDr structure learning algorithm described in \[[1](#references)\]. Based on observational data, this algorithm estimates the structure of a Bayesian network (aka edges in a DAG) using penalized maximum likelihood based on L1 or concave (MCP) regularization.
7+
8+
Presently, this package consists of a single method that implements the main algorithm; more functionality will be provided in the future. To generate data from a given Bayesian network and/or simulate random networks, the following R packages are recommended:
9+
10+
- `bnlearn`: [bnlearn on CRAN](https://cran.r-project.org/package=bnlearn), [www.bnlearn.com](http://www.bnlearn.com)
11+
- `pcalg`: [pcalg on CRAN](https://cran.r-project.org/package=pcalg)
12+
- `igraph`: [igraph on CRAN](https://cran.r-project.org/package=igraph), <http://igraph.org/r/>
13+
14+
Overview
15+
--------
16+
17+
The main method is `ccdr.run`, which runs the CCDr structure learning algorithm as described in \[[1](#references)\].
18+
19+
Installation
20+
------------
21+
22+
You can install:
23+
24+
- the latest CRAN version with
25+
26+
``` r
27+
install.packages("ccdrAlgorithm")
28+
```
29+
30+
- the latest development version from GitHub with
31+
32+
``` r
33+
devtools::install_github(c("itsrainingdata/sparsebnUtils/dev", "itsrainingdata/ccdrAlgorithm/dev"))
34+
```
35+
36+
References
37+
----------
38+
39+
\[1\] Aragam, B. and Zhou, Q. (2015). [Concave penalized estimation of sparse Gaussian Bayesian networks.](http://jmlr.org/papers/v16/aragam15a.html) *The Journal of Machine Learning Research*. 16(Nov):2273−2328.
40+
41+
\[2\] Fu, F. and Zhou, Q. (2013). [Learning sparse causal Gaussian networks with experimental intervention: Regularization and coordinate descent.](http://amstat.tandfonline.com/doi/abs/10.1080/01621459.2012.754359) Journal of the American Statistical Association, 108: 288-300.

‎cran-comments.md

+43
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,43 @@
1+
## Test environments
2+
* local OS X install, R 3.3.1
3+
* ubuntu 12.04 (travis-ci), R 3.2.4
4+
* win-builder (devel and release)
5+
6+
## R CMD check results
7+
There were no ERRORs or WARNINGs.
8+
9+
There was 1 NOTE:
10+
11+
* checking CRAN incoming feasibility ... NOTE
12+
Maintainer: ‘Bryon Aragam <[email protected]>’
13+
14+
New submission
15+
16+
This is the first version of this package that has been submitted to CRAN.
17+
18+
## Dependencies
19+
20+
CHECK has been run on all dependencies and passed.
21+
22+
## Re-submission note
23+
24+
- Pls explain CCDr, and provide a DOI for the reference.
25+
26+
Pls write these as
27+
28+
Authors (year) <DOI:.....>
29+
30+
(with no space after 'DOI:').
31+
32+
The description has been updated with a more detailed description of the CCDr algorithm, and the citation has been removed entirely.
33+
34+
- Again, we would prefer if you explained the acronym, and ideally
35+
provided a reference with a DOI. I see that this is
36+
37+
Concave penalized Coordinate Descent with reparametrization (CCDr)
38+
39+
and the reference has URL
40+
<http://www.jmlr.org/papers/v16/aragam15a.html>: if you cannot find a
41+
DOI, then pls use the URL.
42+
43+
URL has been added (no DOI available) and acronym is now explained.

0 commit comments

Comments
 (0)