python numpy模块 universal functon reduce() 函数用法

这里简单地介绍一下​​numpy​​​模块中地​​reduce()​​​函数的用法.
代码如下：

# -*- coding: utf-8 -*-
import numpy as np


class Debug:
def __init__(self):
self.array1 = np.array([1, 2, 3, 4])
self.array2 = np.array([5, 6, 7, 8])
self.array3 = np.array([[1, 2, 3, 4], [5, 6, 7, 8]])

def mainProgram(self):
result = np.add(self.array1, self.array2)
print("The value of result is: ")
print(result)
result1 = np.add.reduce(self.array3, axis=0)
print("The value of result1 is: ")
print(result1)
result2 = np.add.reduce(self.array3, axis=1)
print("The value of result2 is: ")
print(result2)


if __name__ == '__main__':
main = Debug()
main.mainProgram()
"""
The value of result is:
[ 6 8 10 12]
The value of result1 is:
[ 6 8 10 12]
The value of result2 is:
[10 26]
"""

我们可以看到，当我们指定坐标轴为​​axis=0​​​时，​​np.add.reduce()​​​函数会将数组沿着​​y​​​轴加起来，当指定坐标轴​​axis=1​​​时，​​np.add.reduce()​​​函数会将数组沿着​​x​​​轴加起来。对于为什么是这样，可以参考​​np.repeat()的坐标轴问题​​。

​