sklearn的特征处理

coding=utf-8

“”"
author:lei
function:
“”"

from sklearn.preprocessing import MinMaxScaler # 归一化
from sklearn.preprocessing import StandardScaler # 标准化
from sklearn.preprocessing import Imputer # 数据预处理
import random
import numpy as np

num_list = [[random.randint(10, 40) for i in range(4)] for j in range(4)]
print(num_list)

def mm():
“”"
归一化处理 13
归一化对异常点的处理不太好，最大值最小值可能出现比较大的波动
鲁棒性较差，知识和传统精确小数据场景
:return:
“”"
mm = MinMaxScaler(feature_range=(2, 3))

data = mm.fit_transform(num_list)

print(data)

return None

def stc():
“”“标准化缩放”""

std = StandardScaler()

data = std.fit_transform(num_list)

print(data)

return None

def im():
“”"
缺失值处理
“”"
im = Imputer(missing_values=“NaN”, strategy=“mean”, axis=0) # 填补的方式：平均值 0是列，1是行

data = im.fit_transform([[1, 2], [np.nan, 3], [7, 6]])

print(data)

return None

if name == ‘main’:
# mm()
# stc()
im()