Is it necessary, useful to model constraints?

frictionlessdata / pilot-dm4t

Pilot project with DM4T

http://www.cs.bath.ac.uk/dm4t/index.shtml

1 stars 1 forks source link

Is it necessary, useful to model constraints? #8

Closed danfowler closed 7 years ago

danfowler commented 7 years ago

Currently, we could model the following constraints:

[ ] Each appliance can only draw up 4000 watts, so no reading should be higher than this number.

pwalsh commented 7 years ago

@danfowler can this now be closed, or? I'd say, yes, absolutely have a constraint for this.

danfowler commented 7 years ago

I was coming at this from the perspective of a "dirty" dataset that has already been cleaned to remove these high wattage readings, but, yes, even in the cleaned dataset, especially if it gets updated, it makes sense to model and adds context to the dataset.