A Python 3 package for learning Bayesian Networks (DAGs) from data. Official implementation of the paper "DAGMA: Learning DAGs via M-matrices and a Log-Determinant Acyclicity Characterization"
Yes. For example, if you want to restrict node 5, you would have to fix the 5th row of W to zero in the optimization.
This is simpler to do for linear models, for MLPs it would get more tricky but I think is doable.
Yes. For example, if you want to restrict node 5, you would have to fix the 5th row of W to zero in the optimization. This is simpler to do for linear models, for MLPs it would get more tricky but I think is doable.