Hands-On Artificial Intelligence for Cybersecurity

Implementing a simple predictor with NumPy

To fully understand the use of the dot() method of NumPy in matrix multiplication operations, we can try to implement a simple predictor from scratch, to predict future values starting from a set of multiple inputs and on the basis of relative weights, using the product between matrices and vectors:

import numpy as npdef predict(data, w):       return data.dot(w)
# w is the vector of weightsw = np.array([0.1, 0.2, 0.3]) 
# matrices as input datasetsdata1 = np.array([0.3, 1.5, 2.8]) 
data2 = np.array([0.5, 0.4, 0.9]) 
data3 = np.array([2.3, 3.1, 0.5])data_in = np.array([data1[0],data2[0],data3[0]]) print('Predicted value: $%.2f' %  predict(data_in, w) )