This matches our torch06.py logic. We want to learn a non-linear pattern: y = |x|.

Inputs: [-2, -1, 0, 1, 2] | Targets: [2, 1, 0, 1, 2]
This requires a Hidden Layer and a ReLU activation function to learn the "V" shape.
Click the button to train, then edit the code to try adding more neurons to the hidden layer!

...