I am trying to synthesise a relational database made of the following tables:
As you can see there are several tables with multiple columns and rows. When I fit the model on the whole dataset, it takes days to finish. I was wondering if there is a way to speed the process somehow, maybe from the library perspective or using other libraries that can help paralellize the computing burden.
These are the foreign keys through which the previous tables are connected:
As you can see I am using composite keys for both the primary and foreign keys, due to the fact that the SDV library does not yet support multiple keys.
