Python Numpy MaskedArray.median()函数
numpy.MaskedArray.median()函数用于计算被屏蔽的数组沿指定轴线的中位数,它返回数组元素的中位数。
语法: numpy.ma.median(arr, axis=None, out=None, overwrite_input=False, keepdims=False)
参数:
arr : [ ndarray ] 输入屏蔽的数组。
axis : [ int, optional] 计算中位数的轴。默认情况下(无)是在扁平化的数组上计算中值。
dtype : [dtype, optional] 返回数组的类型,以及与元素相乘的累积器的类型。
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。
overwrite_input : [bool, optional] 如果为真,则允许使用输入数组的内存进行计算。输入数组将被调用median修改。当你不需要保留输入数组的内容时,这将节省内存。将输入数组视为未定义的,但它可能会被完全或部分排序。默认值是False。注意,如果overwrite_input为True,而输入的数组还不是ndarray,则会产生一个错误。
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 : [median_along_axis, ndarray] 返回一个保存结果的新数组,除非指定out,在这种情况下,返回对out的引用。
代码#1:
# Python program explaining
# numpy.MaskedArray.median() method
# importing numpy as geek
# and numpy.ma module as ma
import numpy as geek
import numpy.ma as ma
# creating input array
in_arr = geek.array([[1, 2], [ 3, -1], [ 5, -3]])
print ("Input array : ", in_arr)
# Now we are creating a masked array.
# by making entry as invalid.
mask_arr = ma.masked_array(in_arr, mask =[[1, 0], [ 1, 0], [ 0, 0]])
print ("Masked array : ", mask_arr)
# applying MaskedArray.median
# methods to masked array
out_arr = ma.median(mask_arr)
print ("median of masked array along default axis : ", out_arr)
输出:
Input array : [[ 1 2]
[ 3 -1]
[ 5 -3]]
Masked array : [[-- 2]
[-- -1]
[5 -3]]
median of masked array along default axis : 0.5
代码#2:
# Python program explaining
# numpy.MaskedArray.median() method
# importing numpy as geek
# and numpy.ma module as ma
import numpy as geek
import numpy.ma as ma
# creating input array
in_arr = geek.array([[1, 0, 3], [ 4, 1, 6]])
print ("Input array : ", in_arr)
# Now we are creating a masked array.
# by making one entry as invalid.
mask_arr = ma.masked_array(in_arr, mask =[[ 0, 0, 0], [ 0, 0, 1]])
print ("Masked array : ", mask_arr)
# applying MaskedArray.median methods
# to masked array
out_arr1 = ma.median(mask_arr, axis = 0)
print ("median of masked array along 0 axis : ", out_arr1)
out_arr2 = ma.median(mask_arr, axis = 1)
print ("median of masked array along 1 axis : ", out_arr2)
输出:
Input array : [[1 0 3]
[4 1 6]]
Masked array : [[1 0 3]
[4 1 --]]
median of masked array along 0 axis : [2.5 0.5 3.0]
median of masked array along 1 axis : [1.0 2.5]