SBM implementation in the new version of bgms #32
SBM implementation in the new version of bgms #32MaartenMarsman merged 13 commits intoBayesian-Graphical-Modelling-Lab:mainfrom
Conversation
… parameters and block matrices (2) we sample the number of clusters by first calculating the probability of a sampled cluster configuration using Equation 3.7 pf Miller & Harrison (2018)
…nges K_max = no_variables + 10
MaartenMarsman
left a comment
MaartenMarsman
left a comment
There was a problem hiding this comment.
Looking good Nikola. Can you make the suggested style edits, and revert the setting of the inclusion_probability? Then, I can merge it with the main branch.
| #' number of clusters. The number of clusters is derived from the | ||
| #' cardinality of the sampled \code{allocations} vector for each iteration of | ||
| #' the MCMC sampler (see \code{save = TRUE}). | ||
| #' } |
There was a problem hiding this comment.
Please use insertCite for the references.
There was a problem hiding this comment.
I believe I already fixed that
| allocations = summarySbm$allocations, | ||
| clusters = summarySbm$no_clusters, | ||
| arguments = arguments) | ||
| } else { |
There was a problem hiding this comment.
I need to check if components has names.
There was a problem hiding this comment.
This is how it looks! In easybgm it is printed in a more user-friendly way (as I showed you).
Hi Maarten, I think I have addressed all your comments. |
MaartenMarsman
merged commit 22d3d38
into
Bayesian-Graphical-Modelling-Lab:main
2 participants
Additionally, I also tried fixing the saved prior inclusion probability in the arguments list of the output.