Data Skeptic cover image

Water Demand Forecasting

Data Skeptic

00:00

How Many Quantiles Do You Use?

It was actually five quantiles. One can have as many values as he wants if he defines the quampal levels while training. And also, i think that this an important point, because these approaches can be fully automatic and they can be applied to large statusets. The computational requirements are fordable. So this is another reason why they can be applications for operational purposes. For example, there is the nopelansiorin, which tells us that among ap falgrims, there is no way to know which of them will perform well for one particular problem. That means that the results of cas study are not generalizable general so if we want to make a proper assessment

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app