Numpy中矩阵乘法和点乘的区别

Numpy中矩阵乘法和点乘的区别

---

import numpy as np

a = np.array([[1,2],
              [3,4]])
b = np.array([[5,6],
              [7,8]])

print('a*b:\n', a*b)  # element_wise 一一对应乘积
print('np.matmul:\n', np.matmul(a, b))  # 矩阵乘法
print('np.multiply:\n', np.multiply(a, b))  # element_wise 一一对应乘积
print('np.dot:\n', np.dot(a, b))  # 矩阵乘法

输出：

a*b:
 [[ 5 12]
 [21 32]]
np.matmul:
 [[19 22]
 [43 50]]
np.multiply:
 [[ 5 12]
 [21 32]]
np.dot:
 [[19 22]
 [43 50]]