Python – tensorflow.concat()

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。

concat()用于沿一维连接张量。

Syntax: tensorflow.concat( values, axis, name )

Parameter:

• values: It is a tensor or list of tensor.
• axis: It is 0-D tensor which represents dimension to concatenate.
• name(optional): It defines the name for the operation.

Returns: It returns the concatenated Tensor.

示例 1：

Python3

# Importing the library
import tensorflow as tf
  
# Initializing the input tensor
t1 = [[[1, 2], [3, 4]], [[5, 6], [7, 8]]]
t2 = [[[7, 4], [8, 4]], [[2, 10], [15, 11]]]
  
  
# Printing the input tensor
print('t1: ', t1)
print('t2: ', t2)
  
# Calculating result
res = tf.concat([t1, t2], 2)
  
# Printing the result
print('Result: ', res)

输出：

t1:  [[[1, 2], [3, 4]], [[5, 6], [7, 8]]]
t2:  [[[7, 4], [8, 4]], [[2, 10], [15, 11]]]
Result:  tf.Tensor(
[[[ 1  2  7  4]
  [ 3  4  8  4]]

 [[ 5  6  2 10]
  [ 7  8 15 11]]], shape=(2, 2, 4), dtype=int32)
  
  

示例 2：

Python3

# Importing the library
import tensorflow as tf
  
# Initializing the input tensor
t1 = [[[1, 2], [3, 4]], [[5, 6], [7, 8]]]
t2 = [[[7, 4], [8, 4]], [[2, 10], [15, 11]]]
  
  
# Printing the input tensor
print('t1: ', t1)
print('t2: ', t2)
  
# Calculating result
res = tf.concat([t1, t2], 1)
  
# Printing the result
print('Result: ', res)

输出：

t1:  [[[1, 2], [3, 4]], [[5, 6], [7, 8]]]
t2:  [[[7, 4], [8, 4]], [[2, 10], [15, 11]]]
Result:  tf.Tensor(
[[[ 1  2]
  [ 3  4]
  [ 7  4]
  [ 8  4]]

 [[ 5  6]
  [ 7  8]
  [ 2 10]
  [15 11]]], shape=(2, 4, 2), dtype=int32)