python 中的 numpy.inner()
numpy.inner(arr1, arr2)
: 计算两个数组的内积。
Parameters :
arr1, arr2 : array to be evaluated.
Return: Inner product of the two arrays.
代码#1 :
# Python Program illustrating
# numpy.inner() method
import numpy as geek
# Scalars
product = geek.inner(5, 4)
print("inner Product of scalar values : ", product)
# 1D array
vector_a = 2 + 3j
vector_b = 4 + 5j
product = geek.inner(vector_a, vector_b)
print("inner Product : ", product)
输出:
inner Product of scalar values : 20
inner Product : (-7+22j)
代码#2 : 正常矩阵乘法
# Python Program illustrating
# numpy.inner() method
import numpy as geek
# 1D array
vector_a = geek.array([[1, 4], [5, 6]])
vector_b = geek.array([[2, 4], [5, 2]])
product = geek.inner(vector_a, vector_b)
print("inner Product : \n", product)
product = geek.inner(vector_b, vector_a)
print("\ninner Product : \n", product)
输出:
inner Product :
[[18 13]
[34 37]]
inner Product :
[[18 34]
[13 37]]