sciPy stats . scoratpercentile()函数| Python

原文:https://www . geeksforgeeks . org/scipy-stats-scoratpercentile-function-python/

**scipy.stats.scoreatpercentile(a, score, kind='rank')**函数帮助我们计算输入数组给定百分位的得分。

百分位数= 50 的分数是中位数。如果期望的分位数位于两个数据点之间,我们根据插值的值在它们之间进行插值。

参数: arr:【array _ like】输入数组。 per:【array _ like】我们需要分数的百分位数。 极限:【元组】计算百分位数的下限和上限。 轴:【int】轴,我们需要沿着这个轴计算分数。

结果:相对于数组元素的百分位得分。

代码#1:

# scoreatpercentile
from scipy import stats
import numpy as np 

# 1D array  
arr = [20, 2, 7, 1, 7, 7, 34, 3]

print("arr : ", arr)  

print ("\nScore at 50th percentile : ", 
       stats.scoreatpercentile(arr, 50))

print ("\nScore at 90th percentile : ", 
       stats.scoreatpercentile(arr, 90))

print ("\nScore at 10th percentile : ", 
       stats.scoreatpercentile(arr, 10))

print ("\nScore at 100th percentile : ", 
       stats.scoreatpercentile(arr, 100))

print ("\nScore at 30th percentile : ", 
       stats.scoreatpercentile(arr, 30))

Output:

arr :  [20, 2, 7, 1, 7, 7, 34, 3]

Score at 50th percentile :  7.0

Score at 90th percentile :  24.2

Score at 10th percentile :  1.7

Score at 100th percentile :  34.0

Score at 30th percentile :  3.4

代码#2:

# scoreatpercentile
from scipy import stats
import numpy as np 

arr = [[14, 17, 12, 33, 44],   
       [15, 6, 27, 8, 19],  
       [23, 2, 54, 1, 4, ]] 

print("arr : ", arr)  

print ("\nScore at 50th percentile : ", 
       stats.scoreatpercentile(arr, 50))

print ("\nScore at 50th percentile : ", 
       stats.scoreatpercentile(arr, 50, axis = 1))

print ("\nScore at 50th percentile : ", 
       stats.scoreatpercentile(arr, 50, axis = 0))

Output:

arr :  [[14, 17, 12, 33, 44], [15, 6, 27, 8, 19], [23, 2, 54, 1, 4]]

Score at 50th percentile :  15.0

Score at 50th percentile :  [ 17\.  15\.   4.]

Score at 50th percentile :  [ 15\.   6\.  27\.   8\.  19.]