了解平均值的类型 |设置 2

它是统计学中最重要的概念之一，是学习机器学习的关键学科。

• 几何平均值：算术平均值是集合中所有离散值的总和，几何平均值是集合中离散值的乘积。它对于一组正离散值很有用。
  
  例子 -

  Sequence = {1, 3, 9}
  
  product         = 27
  n, Total values = 3
  Harmonic Mean   = (27)^(1/3)
  

  代码 -

  # Geometric Mean 
    
  import numpy as np
    
  # discrete set of numbers
  from scipy.stats.mstats import gmean
  x = gmean([1, 3, 9])
    
  # Mean 
  print("Geometric Mean is :", x) 
  

  输出 ：

  Geometric Mean is : 3

• 谐波平均值：在计算与任何单位相关的术语的平均值时，谐波平均值起着重要作用。它是数据倒数的平均值的倒数。它用于数据中涉及反向变化的情况。
  

  例子 -

  Sequence = {1, 3, 9}
  
  sum of reciprocals = 1/1 + 1/3 + 1/9
  n, Total values    = 3
  Harmonic Mean      = 3 / (sum of reciprocals)
  

  代码 -

  # Harmonic Mean 
    
  import numpy as np
    
  # discrete set of numbers
  from scipy.stats.mstats import hmean
  x = hmean([1, 3, 9])
    
  # Mean 
  print("Harmonic Mean is :", x) 
  

  输出 ：

  Harmonic Mean is : 2.076923076923077

• 算术 (AM)、谐波 (HM) 和几何平均值 (GM) 之间的关系：
  
  例子 -

  Sequence = {1, 3, 9}
  
  sum of reciprocals = 1/1 + 1/3 + 1/9
  Sum                = 10
  Product            = 27
  n, Total values    = 3
  
  Arithmetic Mean = 4.33
  Geometric Mean  = 3 
  Harmonic Mean   = 3 / (sum of reciprocals) = 2.077