DL : Numpy for Vector Manipulation in Neural Network

Part 1.2 of Deep Learning Specialization

Computation time : Explicit for-loop vs. Numpy function

You should avoid using explicit for-loop for better computation time !!! (link)

A*B, np.multiply (element-wise multiplication) vs. np.dot (vector multiplication)

element-wise multiplication

vector multiplication

Broadcasting

Broadcasting is simply a set of rules for applying functions (e.g., addition, subtraction, multiplication, etc.) on arrays of different sizes. ( link )

np.sum & axis

np.reshape

reshape image input for training (flatten)

Vector shape (m,) vs. (m,n)

Related Articles

• Neural Network — Part 1.1
• Neural Network — Part 1.2
• Neural Network — Part 2
• Neural Network — Part 3
• CNN — Part 4.1
• CNN — Part 4.2
• RNN — Part 5.1
• RNN — Part 5.2

Reference

Deep Learning Specialization: Neural Networks and Deep Learning (Coursera) (Youtube)