Multiple outputs

[lewisrichardson2103]

Hello,

I have tried using this with multiple outputs but the error does not reduce.

I have a network with 6 inputs, 1 hidden layer (7 nodes) and output layer of 3 outputs(a,b,c). When trying to teach the network there is no change to the error over time. When I use the same network but reduce the output to 1 node (replace with 0=a, 0.5 =b and 1 =c) then it works and the error is reduced over time.

Do you know why this would be? I believe the backpropagation does not work correctly with multiple outputs.

thank you

Lewis

[GiorgosXou]

Hello Lewis,

I have tried using this with multiple outputs but the error does not reduce.

You are right, thanks you so much for pointing this issue out, i wasn't aware of it. Now it is fixed, you can now update the library from arduino IDE , here is the release ^_& fix^1,2

thank you

❤️

_{(BTW idk if this is a coincidence but 1 day ago i commented on this video here)}