Neural networks – a redux

The fundamental understanding that neural networks are mathematical expressions leads to really simple and easy implementations of neural networks. Recall from the previous chapter that a neural network can be written like this:

func affine(weights [][]float64, inputs []float64) []float64 {
  return activation(matVecMul(weights, inputs))
}

If we rewrite the code as a mathematical equation, we can write a neural network like this:

A side note: is the same as .

We can simply write it out using Gorgonia, like this:

import (
  G "gorgonia.org/gorgonia"
)

var Float tensor.Float = tensor.Float64
func main() {
  g := G.NewGraph()
  x := G.NewMatrix(g, Float, G.WithName("x"), G.WithShape(N, 728))
  w := G.NewMatrix(g, Float, G.WithName("w"), G.WithShape(728, 800), 
       G.WithInit(G.Uniform(1.0)))
  b := G.NewMatrix(g, Float, G.WithName...