Open kavir1698 opened 4 years ago
The package doesn't currently have that implemented. I think it would be fairly easy to add "exclude" and "include" edges to particular algorithms. It could be an optional argument to Distributions.fit
for particular algorithms like greedy hill climbing. Those lists could be just vectors of node pairs. It should be fairly straightforward to implement. PRs are definitely welcome!
Is it possible to add constraints to structure learning? so that the user can pre-exclude some connections based on expert knowledge. That would be a functionality similar to
tabu_edges
in CausalNex.