Neural Network: N number of numeric inputs and M number of Numeric ouptus

Hello,

I have been searching all over the internet for an answer to this and the most likely reason I haven't found a solution is perhaps I'm missunderstanding something. My problem is this: I have 8 (or, N) features (namely, some properties of neurons) that correlate to 2 (or, M) outputs (namely, x and y positions of the hand). I am looking for a mutli-output regression technique and have researched a bunch of things but I think a neural network is the way to go. However I can't figure out how to make such a network in matlab- there are examples in the documentation of 2 outputs for Image data, however I need to be using numeric imputs. Can anyone point me in the right direction here, I've spent a few days and I cant find anything in the Matlab documentation that resembles what i'm looking for. My sanity would be extremely grateful for your help!

Answer 1

[Shiva Kalyan Diwakaruni]

Hi,

You can follow the below links which contains the example of training network with numeric features.

https://www.mathworks.com/help/deeplearning/ref/trainnetwork.html#mw_fcd828cc-de80-4922-8cc2-806c07deb31a

https://www.mathworks.com/help/deeplearning/deep-learning-with-time-series-sequences-and-text.html

hope it helps,

thanks.