search

donaldRwilliams / BGGM

Bayesian Gaussian Graphical Models
https://donaldrwilliams.github.io/BGGM/
GNU General Public License v2.0
54 stars 14 forks source link

State of the field #7

Closed paulgovan closed 4 years ago

paulgovan commented 4 years ago

This issue is for github.com/openjournals/joss-reviews/issues/2111.

"Do the authors describe how this software compares to other commonly-used packages?"

It's pretty clear that the software is useful, but it's not really clear how this software compares to and is better in some way than other software that's currently available. In addition, the background on existing software seems incomplete, as there are numerous packages for building Bayesian graphical models.

donaldRwilliams commented 4 years ago

We are happy to include this. To our knowledge, for GGMs in particular (and not Bayesian networks), there are only two R packages (BDgraph and beam). We are not aware of others, but can take another look.

donaldRwilliams commented 4 years ago

We added a note to the README. To our knowledge, there is no R package that includes all the algorithms and methods.

jayrobwilliams commented 4 years ago

The updated README and paper note that there is no R package that implements all of these methods. Are there other R packages that implement different subsets of these methods? If so, they would be important to mention here. Is the new c++ implementation faster than these packages? This would also be good to mention.

donaldRwilliams commented 4 years ago

OK. I will add this

jayrobwilliams commented 4 years ago

I am satisfied with the way these issues have been addressed.

donaldRwilliams commented 4 years ago

To address these comments, we followed the suggestions of the reviewers to include a section comparing to other software.