Machine Learning Neural Network example

[igpema]

Hello,

at the Bim Seminar in Uni Stuttgart we are trying to play with Machine Learning concepts in Dynamo but we dont really get how it is working out. We downloaded the examples from provingGround Lunchbox but when we try to apply to our issues we dont really get any clear result.

Our exercise consist on a neihbourhood with plots, and within these plots some houses. We calculate in dynamo how many houses with their areas there is withing each plot and when we draw a new plot, we would like the machine tell us how many houses and areas should have this new plot. We tried NetworkNeuralNetwork and networks.RestrictedBotzmann from LunchBox package, but we get nowhere. Maybe we are not using the right nodes, our main purpose, to learn how to use the data around to tell to a particular plot how it should be, in this case amount of houses and surface.

Here the area:

Here the two nodes I am trying to use for this purpose

and attached is also the dyn file.
Umgebung_ ML.dyn (46.8 KB)

thanks in advance!

[igpema]

I found this video

[Durmus_Cesur]

Hi,

You can look at this topic.

Lunchbox Machine Learning

So, I got close, but it’s not quite working? [Capture] This ‘works’ but only if the number of rooms is less than my ‘test’ rooms, which is obviously not ideal… [image] Also the outputs aren’t right…! I tried plugging everything in the other way round and playing with the various other inputs but couldn’t get it working. I realise I could use IF statements for this. The inspiration came from John Pierson and Nate Miller here… https://www.youtube.com/watch?v=e6WglyUB-7A though it’s all in…

[stepanste]

I see, you did it a long time ago, but wouldn’t you give me a revit file? I also want ro try