机器学习-sklearn模块数据预处理

1.数据标准化，使数据满足高斯分布

preprocessing.scale()

函数

import numpy as np
from sklearn import preprocessing
from scipy.stats import anderson

rain = np.load('rain.npy')
rain =  * rain
rain[rain < ] = /

scaled = preprocessing.scale(rain)
print("rain mean",scaled.mean())
print("rain variance",scaled.var())
print("anderson rain",anderson(scaled))
           

2.使数据所有样本数值缩放到（-1，1）之间

方法一：

from sklearn import preprocessing

X = [[ , -,  ],
     [ ,  ,  ],
     [ ,  , -]]
X_normalized = preprocessing.normalize(X, norm='l2')

>>> X_normalized                                      
array([[ ..., -...,  ...],
       [   ...,    ...,    ...],
       [   ...,  ..., -...]])
           

方法二：

XY = [[ 1., -1.,  2.],
     [ 2.,  0.,  0.],
      [ 0.,  1., -1.]]
normalizer = preprocessing.Normalizer().fit(XY)
print(normalizer.transform(XY))
           

3.二值化数据

>>> X = [[ , -,  ],
...      [ ,  ,  ],
...      [ ,  , -]]

>>> binarizer = preprocessing.Binarizer().fit(X)  # fit does nothing
>>> binarizer
Binarizer(copy=True, threshold=)

>>> binarizer.transform(X)
array([[ ,  ,  ],
       [ ,  ,  ],
       [ ,  ,  ]])
           

有阈值

>>> binarizer = preprocessing.Binarizer(threshold=)
>>> binarizer.transform(X)
array([[ 0.,  0.,  1.],
       [ 1.,  0.,  0.],
       [ 0.,  0.,  0.]])
           

4.标签二值化

>>> lb = preprocessing.LabelBinarizer()  
>>> lb.fit([, , , , ])  
LabelBinarizer(neg_label=, pos_label=)  
>>> lb.classes_  
array([, , , ])  
>>> lb.transform([, ])  
array([[1, 0, 0, 0],  
       [0, 0, 0, 1]])  
           

多标签显示

>>> lb.fit_transform([(, ), (,)]) #(,)实例中就包含两个label  
array([[1, 1, 0],  
       [0, 0, 1]])  
>>> lb.classes_  
array([, , ])