Crow Tutorial1: Neural Sensing and Control in Grasshopper

Supervised learning is a branch of machine learning in which a mapping function is programmed into a neural network by exposing examples to the network (training).
This tutorial shows how to use the plug in Crow to classify certain environmental states (based on sensor data) and relate them to a parametric model, hence a neural network classifies sensor states and maps them to respective parametric outputs.
All the files can be found here:
github.com/Hei...
...and here:
www.food4rhino....

[@mat1330]

i am not getting flexhopper for download

[@hammadharoon1871]

Is it possible to get metrics like accuracy or validation loss?

[@benjaminf.3760]

Hi Hammad, the component outputs the MSE of the training set after training. Getting the MSE of new test samples isn't implemented in a certain component, but can relatively easily be plugged together in GH... Cheers

[@hammadharoon1871]

@@benjaminf.3760 Oh okay thanks!!