shallow neural network

Explanation for vectorized implementation

In this lecture, we study vectorized implementation. It’s simple. with vetorization, we don’t need to use for-loop for all data. we just define Data X, Weight W, (+ bias) and do the math with Numpy. To understand the below, recap the notation.

$a^{[1](i)}$ : activation result of 1st layer and i-th element(node).