Pandas 统计函数corr

相关系数(correlation coefficients)

计算列与列之间的相关系数 (不包括NA/null),返回相关系数矩阵

DataFrame.corr( method='pearson', min_periods=1) → ’DataFrame’

参数:

  • method:{‘pearson’, ‘kendall’, ‘spearman’} or callable
    • pearson:标准相关系数
    • kendall:Kendall Tau相关系数
    • spearman:Spearman等级相关
    • callable:可输入两个1d ndarray来调用

返回值:DataFrame

Examples

allDf = pd.DataFrame({
    'x':[0,1,2,4,7,10],
    'y':[0,3,2,4,5,7],
    's':[0,1,2,3,4,5],
    'c':[5,4,3,2,1,0]
},index = ['p1','p2','p3','p4','p5','p6'])

allDf

结果:

    x    y    s    c
p1    0    0    0    5
p2    1    3    1    4
p3    2    2    2    3
p4    4    4    3    2
p5    7    5    4    1
p6    10    7    5    0
allDf.corr()

结果:

image-20200711145148964
allDf.corr()['x']
image-20200711145222334

参考

Update time: 2020-07-12

results matching ""

    No results matching ""