12.4 Forward Propagation

Forward propagation is how a network makes a prediction. The input values are fed into the first layer, each neuron computes its weighted sum and activation, and the results flow forward layer by layer until the output layer produces the final prediction.

It is like an Ethiopian injera production line: teff flour enters, passes through mixing, fermenting, and baking stations in order, and a finished injera comes out the end. Each station transforms what the previous one handed it.