search

erdogant / bnlearn

Python library for learning the graphical structure of Bayesian networks, parameter learning, inference and sampling methods.
https://erdogant.github.io/bnlearn
Other
463 stars 45 forks source link

Sparse Structure Learning for Massive Datasets #59

Open arainboldt opened 1 year ago

arainboldt commented 1 year ago

Hi,

Great package. I really enjoy using it!

I'm working on a project with datasets where the number of variables is 3000+ for which a bayesian network would be ideal choice to model the data. Unfortunately, as we all know, structure learning for Bayes Nets is exponential and consequently, unfeasible for such a wide dataset.

In my research, I've come across this paper which looks very encouraging as a method for addressing the exponential complexity of structure learning.

Is there a straightforward way to implement the parent constraint method described in the paper using the existing structure learning framework in the package?

Any insights or direction is much appreciated

Thanks!

Andrew

erdogant commented 1 year ago

Thanks for your question. Did you already find a solution for this?