Concepts
When we count layers without taking into account the input payer.
Logistic Regression is a one layer neural network.
Notation
L = Number of layers.
x[l] = Input features.
n[l] = Number of units in the lager l.
a[l] = Activation in layer l.
a[l] = g[l] (z[l]).
w[l] = Weight for z[l].
b[l] = Bia for z[l].
$ y_hat $ = Predicted values.
In this case n[0] = 2,
Forward Propagation
x: z[l] = w[l] + b[l]
a[l] = g
