极客教程Python Numpy recarray.var()函数
在numpy中，数组可以有一个包含字段的数据类型，类似于电子表格中的列。一个例子是[(a, int), (b, float)] ，其中数组中的每个条目是一对(int, float)。通常情况下，这些属性使用字典查询，如arr[‘a’] 和arr[‘b’] 。
numpy.recarray.var()函数返回数组元素的方差，沿着给定的轴。
语法: numpy.recarray.var(axis=None, dtype=None, out=None, ddof=0, keepdims=False)
参数:
axis : [int or tuples of int] 我们想沿着这个轴来计算方差。否则，它将认为arr是平坦的（对所有的轴都有效）。 axis = 0意味着沿列的方差， axis = 1意味着沿行的方差。
dtype : [data-type, optional] 我们在计算方差时希望的类型。
out : [ndarray, optional] 一个储存结果的位置。
-> 如果提供，它必须有一个输入广播到的形状。
-> 如果没有提供或没有，将返回一个新分配的数组。
ddof : [int, optional] Delta自由度”：计算中使用的除数是N-ddof，其中N代表元素数。默认情况下ddof为零。
keepdims : [bool, optional] 如果设置为 “True”，被缩小的轴将作为尺寸为1的尺寸留在结果中。有了这个选项，结果将正确地针对输入阵列进行广播。
返回 : [ndarray] 如果out=None，返回一个包含方差的新数组；否则，返回一个对输出数组的引用。
代码#1：
# Python program explaining
# numpy.recarray.var() method
# importing numpy as geek
import numpy as geek
# creating input array with 2 different field
in_arr = geek.array([[(5.0, 2), (3.0, -4), (6.0, 9)],
[(9.0, 1), (5.0, 4), (-12.0, -7)]],
dtype =[('a', float), ('b', int)])
print ("Input array : ", in_arr)
# convert it to a record array,
# using arr.view(np.recarray)
rec_arr = in_arr.view(geek.recarray)
print("Record array of float: ", rec_arr.a)
print("Record array of int: ", rec_arr.b)
# applying recarray.var methods
# to float record array along axis 1
out_arr = rec_arr.a.var(axis = 1)
print ("Output array containing variance along axis 1: ", out_arr)
# applying recarray.var methods
# to float record array along axis 0
out_arr = rec_arr.a.var(axis = 0)
print ("Output array containing variance along axis 0: ", out_arr)
# applying recarray.var methods
# to float record array along default axis
out_arr = rec_arr.a.var()
print ("Output array containing variance along default axis : ", out_arr)
# applying recarray.var methods
# to int record array along axis 1
out_arr = rec_arr.b.var(axis = 1)
print ("Output array containing variance along axis 1: ", out_arr)
# applying recarray.var methods
# to int record array along axis 0
out_arr = rec_arr.b.var(axis = 0)
print ("Output array containing variance along axis 0: ", out_arr)
# applying recarray.var methods
# to int record array along default axis
out_arr = rec_arr.b.var()
print ("Output array containing variance along default axis : ", out_arr)
输出:
Input array : [[( 5., 2) ( 3., -4) ( 6., 9)]
[( 9., 1) ( 5., 4) (-12., -7)]]
Record array of float: [[ 5. 3. 6.]
[ 9. 5. -12.]]
Record array of int: [[ 2 -4 9]
[ 1 4 -7]]
Output array containing variance along axis 1: [ 1.55555556 82.88888889]
Output array containing variance along axis 0: [ 4. 1. 81.]
Output array containing variance along default axis : 46.22222222222222
Output array containing variance along axis 1: [28.22222222 21.55555556]
Output array containing variance along axis 0: [ 0.25 16. 64. ]
Output array containing variance along default axis : 27.138888888888882