Python – tensorflow.math.multiply()

TensorFlow 是由 Google 设计的开源Python库，用于开发机器学习模型和深度学习神经网络。 multiply()用于逐元素查找 x*y。它支持广播。

Syntax: tf.math.multiply(x, y, name)

Parameter:

• x: It’s the input tensor. Allowed dtype for this tensor are bfloat16, half, float32, float64, uint8, int8, uint16, int16, int32, int64, complex64, complex128.
• y: It’s the input tensor of same dtype as x.
• name(optional): It’s defines the name for the operation.

Returns:
It returns a tensor of same dtype as x.

示例 1：

Python3

# Importing the library
import tensorflow as tf
  
# Initializing the input tensor
a = tf.constant([.2, .5, .7, 1], dtype = tf.float64)
b = tf.constant([.1, .3, 1, 5], dtype = tf.float64)
  
# Printing the input tensor
print('a: ', a)
print('b: ', b)
  
# Calculating result
res = tf.math.multiply(x = a, y = b)
  
# Printing the result
print('Result: ', res)

输出：

a:  tf.Tensor([0.2 0.5 0.7 1. ], shape=(4, ), dtype=float64)
b:  tf.Tensor([0.1 0.3 1.  5. ], shape=(4, ), dtype=float64)
Result:  tf.Tensor([0.02 0.15 0.7  5.  ], shape=(4, ), dtype=float64)

示例 2：复数乘法

Python3

# importing the library
import tensorflow as tf
  
# Initializing the input tensor
a = tf.constant([-2 + 3j, -5 + 4j, 7 + 2j, 1 + 7j], dtype = tf.complex128)
b = tf.constant([-1 + 2j, -6 + 8j, 8 + 2j, 0 + 1j], dtype = tf.complex128)
  
# Printing the input tensor
print('a: ', a)
print('b: ', b)
  
# Calculating result
res = tf.math.multiply(x = a, y = b)
  
# Printing the result
print('Result: ', res)

输出：

a:  tf.Tensor([-2.+3.j -5.+4.j  7.+2.j  1.+7.j], shape=(4, ), dtype=complex128)
b:  tf.Tensor([-1.+2.j -6.+8.j  8.+2.j  0.+1.j], shape=(4, ), dtype=complex128)
Result:  tf.Tensor([-4. -7.j -2.-64.j 52.+30.j -7. +1.j], shape=(4, ), dtype=complex128)