Python| TensorFlow abs() 方法

Tensorflow 是谷歌开发的开源机器学习库。它的应用之一是开发深度神经网络。

模块tensorflow.math为许多基本的数学运算提供支持。函数tf.abs() [别名tf.math.abs ] 为 Tensorflow 中的绝对函数提供支持。它期望复数形式的输入为$$a+bi$$或浮点数。输入类型是张量，如果输入包含多个元素，则计算元素绝对值。

对于复数$$a+bi$$ ，绝对值计算为$\sqrt{a^2+b^2}$ .
对于浮点数$$a$$ ，绝对值计算为$$a if $a>=0, else -a. $$

Syntax: tf.abs(x, name=None) or tf.math.abs(x, name=None)

Parameters:
x: A Tensor or SparseTensor of type float16, float32, float64, int32, int64, complex64 or complex128.
name (optional): The name for the operation.

Return type: A Tensor or SparseTensor with the same size and type as that of x with absolute values. For complex64 or complex128 input, the returned Tensor will be of type float32 or float64, respectively.

代码 #1：对于浮点数

Python3

# Importing the Tensorflow library
import tensorflow as tf
  
# A constant vector of size 5
a = tf.constant([-0.5, -0.1, 0, 0.1, 0.5], dtype = tf.float32)
  
# Applying the abs function and
# storing the result in 'b'
b = tf.abs(a, name ='abs')
  
# Initiating a Tensorflow session
with tf.Session() as sess:
    print('Input type:', a)
    print('Input:', sess.run(a))
    print('Return type:', b)
    print('Output:', sess.run(b))

输出：

Input type: Tensor("Const:0", shape=(5, ), dtype=float32)
Input : [-0.5 -0.1  0.   0.1  0.5]
Return Type : Tensor("abs:0", shape=(5, ), dtype=float32)
Output : [0.5 0.1 0.  0.1 0.5]

代码 #2：可视化

Python3

# Importing the Tensorflow library
import tensorflow as tf
  
# Importing the NumPy library
import numpy as np
  
# Importing the matplotlib.pyplot function
import matplotlib.pyplot as plt
  
# A vector of size 11 with values from -5 to 5
a = np.linspace(-5, 5, 11)
  
# Applying the absolute function and
# storing the result in 'b'
b = tf.abs(a, name ='abs')
  
# Initiating a Tensorflow session
with tf.Session() as sess:
    print('Input:', a)
    print('Output:', sess.run(b))
    plt.plot(a, sess.run(b), color = 'red', marker = "o") 
    plt.title("tensorflow.abs") 
    plt.xlabel("X") 
    plt.ylabel("Y") 
  
    plt.show()

输出：

Input: [-5. -4. -3. -2. -1.  0.  1.  2.  3.  4.  5.]
Output: [5. 4. 3. 2. 1. 0. 1. 2. 3. 4. 5.]

代码 #3：复数

Python3

# Importing the Tensorflow library
import tensorflow as tf
  
# A constant vector of size 2
a = tf.constant([[-2.25 + 4.75j], [-3.25 + 5.75j]],
                              dtype = tf.complex64)
  
# Applying the abs function and
# storing the result in 'b'
b = tf.abs(a, name ='abs')
  
# Initiating a Tensorflow session
with tf.Session() as sess:
    print('Input type:', a)
    print('Input:', sess.run(a))
    print('Return type:', b)
    print('Output:', sess.run(b))

输出：

Input type: Tensor("Const_1:0", shape=(2, 1), dtype=complex64)
Input : [[-2.25+4.75j] [-3.25+5.75j]]
Return Type : Tensor("abs_1:0", shape=(2, 1), dtype=float32)
Output : [[5.255949 ] [6.6049223]]