11.数据归一化

import numpy as np
import matplotlib.pyplot as plt

最值归一化

x = np.random.randint(0,100,size=100)
np.mean(x),np.std(x)

(50.16, 28.943641788828167)

x1 = (x - np.min(x))/(np.max(x)-np.min(x))
np.mean(x1), np.std(x1)

(0.5018181818181818, 0.2771724533959654)

X = np.random.randint(0,100,(50,2))
X[:10,:]

array([[60., 72.],
       [50., 77.],
       [39., 84.],
       [46., 35.],
       [76.,  8.],
       [48., 78.],
       [50., 45.],
       [41., 57.],
       [61.,  4.],
       [27.,  0.]])

X = np.array(X,dtype=float)
X[:10,:]

array([[60., 72.],
       [50., 77.],
       [39., 84.],
       [46., 35.],
       [76.,  8.],
       [48., 78.],
       [50., 45.],
       [41., 57.],
       [61.,  4.],
       [27.,  0.]])

X[:,0] = (X[:,0]-np.min(X[:,0]))/(np.max(X[:,0])-np.min(X[:,0]))
X[:,1] = (X[:,1]-np.min(X[:,1]))/(np.max(X[:,1])-np.min(X[:,1]))
X[:10,:]

array([[0.6185567 , 0.72727273],
       [0.51546392, 0.77777778],
       [0.40206186, 0.84848485],
       [0.4742268 , 0.35353535],
       [0.78350515, 0.08080808],
       [0.49484536, 0.78787879],
       [0.51546392, 0.45454545],
       [0.42268041, 0.57575758],
       [0.62886598, 0.04040404],
       [0.27835052, 0.        ]])

plt.scatter(X[:,0],X[:,1])

均值方差归一化 Standardization

X1 = np.random.randint(0,100,(50,2))
X1 = np.array(X1,dtype=float)
X1[:,0] = (X1[:,0]-np.mean(X1[:,0]))/(np.std(X1[:,0]))
X1[:,1] = (X1[:,1]-np.mean(X1[:,1]))/(np.std(X1[:,1]))
plt.scatter(X1[:,0],X1[:,1])

X1[:10,:]

array([[ 0.51447692,  0.00322813],
       [-0.74357475,  0.32604148],
       [-0.41250852, -0.31958522],
       [-0.67736151, -0.0774752 ],
       [ 1.07728951,  0.20498648],
       [-0.97532111,  1.41553654],
       [-0.41250852, -0.84415691],
       [-1.14085423, -1.73189362],
       [ 0.18341069, -0.03712354],
       [-1.1077476 , -0.76345357]])

np.min(X1[:,0])

-1.7698800621130713

np.max(X1[:,1])

1.7787015556184822

np.max(X1)

1.7787015556184822