Bayesian Network Model Size Calculation

Question

How do you calculate the (storage) size of a Bayesia Network model with n nodes/features/columns/attributes and k arcs? The model may be generated through pymc or pomegranate.

Answer 1

So, the following could partially answer my question. From a theorethical, the bayesian network model may be stored as adjacency matrix which scales linear with regards to the arcs by factor 2.

Additionally, the probability of ocurrance need to be stored, this can be - depending on the precision - a double. Also, the calculation needs to contain the network meta data and node details, whereas the latter also scales linear to the number of nodes.