1
$\begingroup$

am trying to build a air speed prediction system,

usecase - given air speed for a particular city in the US, I want to know the air speed inside the city within the buildings, with the help of Computational Fluid Dynamics (CFD) modeling.

I have training dataset which consists of input as airspeeds of a city and I have 10,000 outputs i.e airspeed surrounding a building for that one single airspeed.

Input - airspeed = 20 mph
Output - 
(4.3302 0.00073217 2.4997)
(4.33021 0.00128552 2.49909)
(4.33101 0.00219388 2.49989)
(4.33124 0.00192743 2.49877)
(4.33261 0.0035236 2.50044)
(4.33293 0.00313777 2.49831)
(4.33453 0.0055888 2.50033)
(4.33481 0.0054796 2.4982)
(4.33676 0.00871265 2.49948)
(4.33573 0.00764592 2.49836)
(4.33811 0.0126655 2.50087)
(4.33453 0.0089677 2.49504)
(4.32326 0.0101457 2.50085)
(4.29871 0.0193168 2.51788)
(4.27203 0.0277074 2.5316)
(4.24884 0.067923 2.52527)
(4.23519 0.0522209 2.52181)
(4.21154 0.135795 2.48838) 
and so on upto 10,000

My approach - I think multi output regression is used to solve this but I am not sure how do I start this, how do I split the output? how do I train one input and 10,000 output? how do I split the dataset to train?

If multi output regression is not the right approach, what other choice do I have?

$\endgroup$

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.