BayOpt

Bayesian Methods for the calibration and control of storm water networks

Performance

1. System Scale Control Example - done

Scalability

1. 11 Basin Example TODO: Re run them to remove errors TODO: Plot heat maps

Uncertanity

1. Decide on the which path to take.

There is an other way to do it too, but we can explore it later.

For the final thing, I need to have a plot with more uncertanity. Also most important double check everything to ensure that HETRO GP implementation is right.

The thing is, when we do this, we need to force the GP to take more samples. or we wont get the accurate uncertanity. But the GP might not do it. So that is the drawback.