There are two possible situations when we want to learn a network from a database. Either, the structure of the network is known, for instance from a domain expert, or the structure of the network is unknown. We here describe learning in both situations. Other problems might also arise, there might be variables that are hidden - that causes events or are caused by other events, but are not present in the database. This situation will not be handled here. More information on learning networks from data can be found in [19].