You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository was archived by the owner on Jun 4, 2024. It is now read-only.
mdpham edited this page Aug 26, 2016
·
7 revisions
#Large Benchmark Dataset
##Description
This dataset has been provided by bnlearn repository (an R package for Bayesian network learning and inference). This dataset is used for benchmarking the PGMLab application and to show the speed at which the tool can perform learning and inference on a typical large dataset.
###Files
Large benchmark dataset can be found in the folder "data/munin-dataset/"
####Pairwise interaction file
The pairwise interaction file can be use to create a logical factorgraph using either the R or C interface. It is used to let the PGMLab know which nodes to index when performing learning and inference as well.
In order to generate a learnt factorgraph you will need to use the logical factorgraph that you created from the pairwise interaction file, in combination with one of the three observed data files.