In [1]:

    

import tensorflow as tf
import numpy as np


    

In [2]:

    

A = np.array([[[1,2,3],[0.1,0.2,0.3],[0.01,0.02,0.03]],[[11,12,13],[0.11,0.12,0.13],[0.011,0.012,0.013]],[[111,112,113],[0.111,0.112,0.113],[0.0111,0.0112,0.0113]]])
B = np.array([[[4,5,6],[0.4,0.5,0.6],[0.04,0.05,0.06]],[[14,15,16],[0.14,0.15,0.16],[0.014,0.015,0.016]],[[114,115,116],[0.114,0.115,0.116],[0.0114,0.0115,0.0116]]])
C = np.array([[[7,8,9],[0.7,0.8,0.9],[0.07,0.08,0.09]],[[17,18,19],[0.17,0.18,0.19],[0.017,0.018,0.019]],[[117,118,119],[0.117,0.118,0.119],[0.0117,0.0118,0.0119]]])


    

In [3]:

    

"""
tf.stack()

tf.stack(values, axis=0, name=’stack’)

将 a list of R 维的Tensor堆成 R+1维的Tensor。 
Given a list of length N of tensors of shape (A, B, C); 
if axis == 0 then the output tensor will have the shape (N, A, B, C)

这时 res[i,:,:,:] 就是原 list中的第 i 个 tensor
1
. if axis == 1 then the output tensor will have the shape (A, N, B, C).

这时 res[:,i,:,:] 就是原list中的第 i 个 tensor
"""
sess = tf.InteractiveSession()
A


    

    
Out[3]:





array([[[  1.00000000e+00,   2.00000000e+00,   3.00000000e+00],
        [  1.00000000e-01,   2.00000000e-01,   3.00000000e-01],
        [  1.00000000e-02,   2.00000000e-02,   3.00000000e-02]],

       [[  1.10000000e+01,   1.20000000e+01,   1.30000000e+01],
        [  1.10000000e-01,   1.20000000e-01,   1.30000000e-01],
        [  1.10000000e-02,   1.20000000e-02,   1.30000000e-02]],

       [[  1.11000000e+02,   1.12000000e+02,   1.13000000e+02],
        [  1.11000000e-01,   1.12000000e-01,   1.13000000e-01],
        [  1.11000000e-02,   1.12000000e-02,   1.13000000e-02]]])

In [4]:

    

D = tf.stack([A,B])
D1 = tf.stack([A,B],axis=1)
D2 = tf.stack([A,B],axis=2)


    

In [5]:

    

D.eval()


    

    
Out[5]:





array([[[[  1.00000000e+00,   2.00000000e+00,   3.00000000e+00],
         [  1.00000000e-01,   2.00000000e-01,   3.00000000e-01],
         [  1.00000000e-02,   2.00000000e-02,   3.00000000e-02]],

        [[  1.10000000e+01,   1.20000000e+01,   1.30000000e+01],
         [  1.10000000e-01,   1.20000000e-01,   1.30000000e-01],
         [  1.10000000e-02,   1.20000000e-02,   1.30000000e-02]],

        [[  1.11000000e+02,   1.12000000e+02,   1.13000000e+02],
         [  1.11000000e-01,   1.12000000e-01,   1.13000000e-01],
         [  1.11000000e-02,   1.12000000e-02,   1.13000000e-02]]],


       [[[  4.00000000e+00,   5.00000000e+00,   6.00000000e+00],
         [  4.00000000e-01,   5.00000000e-01,   6.00000000e-01],
         [  4.00000000e-02,   5.00000000e-02,   6.00000000e-02]],

        [[  1.40000000e+01,   1.50000000e+01,   1.60000000e+01],
         [  1.40000000e-01,   1.50000000e-01,   1.60000000e-01],
         [  1.40000000e-02,   1.50000000e-02,   1.60000000e-02]],

        [[  1.14000000e+02,   1.15000000e+02,   1.16000000e+02],
         [  1.14000000e-01,   1.15000000e-01,   1.16000000e-01],
         [  1.14000000e-02,   1.15000000e-02,   1.16000000e-02]]]])

In [6]:

    

D1.eval()


    

    
Out[6]:





array([[[[  1.00000000e+00,   2.00000000e+00,   3.00000000e+00],
         [  1.00000000e-01,   2.00000000e-01,   3.00000000e-01],
         [  1.00000000e-02,   2.00000000e-02,   3.00000000e-02]],

        [[  4.00000000e+00,   5.00000000e+00,   6.00000000e+00],
         [  4.00000000e-01,   5.00000000e-01,   6.00000000e-01],
         [  4.00000000e-02,   5.00000000e-02,   6.00000000e-02]]],


       [[[  1.10000000e+01,   1.20000000e+01,   1.30000000e+01],
         [  1.10000000e-01,   1.20000000e-01,   1.30000000e-01],
         [  1.10000000e-02,   1.20000000e-02,   1.30000000e-02]],

        [[  1.40000000e+01,   1.50000000e+01,   1.60000000e+01],
         [  1.40000000e-01,   1.50000000e-01,   1.60000000e-01],
         [  1.40000000e-02,   1.50000000e-02,   1.60000000e-02]]],


       [[[  1.11000000e+02,   1.12000000e+02,   1.13000000e+02],
         [  1.11000000e-01,   1.12000000e-01,   1.13000000e-01],
         [  1.11000000e-02,   1.12000000e-02,   1.13000000e-02]],

        [[  1.14000000e+02,   1.15000000e+02,   1.16000000e+02],
         [  1.14000000e-01,   1.15000000e-01,   1.16000000e-01],
         [  1.14000000e-02,   1.15000000e-02,   1.16000000e-02]]]])

In [7]:

    

D2.eval()


    

    
Out[7]:





array([[[[  1.00000000e+00,   2.00000000e+00,   3.00000000e+00],
         [  4.00000000e+00,   5.00000000e+00,   6.00000000e+00]],

        [[  1.00000000e-01,   2.00000000e-01,   3.00000000e-01],
         [  4.00000000e-01,   5.00000000e-01,   6.00000000e-01]],

        [[  1.00000000e-02,   2.00000000e-02,   3.00000000e-02],
         [  4.00000000e-02,   5.00000000e-02,   6.00000000e-02]]],


       [[[  1.10000000e+01,   1.20000000e+01,   1.30000000e+01],
         [  1.40000000e+01,   1.50000000e+01,   1.60000000e+01]],

        [[  1.10000000e-01,   1.20000000e-01,   1.30000000e-01],
         [  1.40000000e-01,   1.50000000e-01,   1.60000000e-01]],

        [[  1.10000000e-02,   1.20000000e-02,   1.30000000e-02],
         [  1.40000000e-02,   1.50000000e-02,   1.60000000e-02]]],


       [[[  1.11000000e+02,   1.12000000e+02,   1.13000000e+02],
         [  1.14000000e+02,   1.15000000e+02,   1.16000000e+02]],

        [[  1.11000000e-01,   1.12000000e-01,   1.13000000e-01],
         [  1.14000000e-01,   1.15000000e-01,   1.16000000e-01]],

        [[  1.11000000e-02,   1.12000000e-02,   1.13000000e-02],
         [  1.14000000e-02,   1.15000000e-02,   1.16000000e-02]]]])

In [8]:

    

X = np.array([[[1,2,3],[4,5,6],[7,8,9]],[[11,12,13],[14,15,16],[17,18,19]],[[21,22,23],[24,25,26],[27,28,29]]])
Y = np.array([[[101,102,103],[104,105,106],[107,108,109]],[[111,112,113],[114,115,116],[117,118,119]],[[121,122,123],[124,125,126],[127,128,129]]])
Z = tf.stack([X,Y]).eval()
X.shape


    

    
Out[8]:





(3, 3, 3)

In [9]:

    

Z


    

    
Out[9]:





array([[[[  1,   2,   3],
         [  4,   5,   6],
         [  7,   8,   9]],

        [[ 11,  12,  13],
         [ 14,  15,  16],
         [ 17,  18,  19]],

        [[ 21,  22,  23],
         [ 24,  25,  26],
         [ 27,  28,  29]]],


       [[[101, 102, 103],
         [104, 105, 106],
         [107, 108, 109]],

        [[111, 112, 113],
         [114, 115, 116],
         [117, 118, 119]],

        [[121, 122, 123],
         [124, 125, 126],
         [127, 128, 129]]]])

In [10]:

    

y = tf.transpose(Z,perm=[1,2,3,0]).eval()


    

In [11]:

    

y.shape


    

    
Out[11]:





(3, 3, 3, 2)

In [12]:

    

y


    

    
Out[12]:





array([[[[  1, 101],
         [  2, 102],
         [  3, 103]],

        [[  4, 104],
         [  5, 105],
         [  6, 106]],

        [[  7, 107],
         [  8, 108],
         [  9, 109]]],


       [[[ 11, 111],
         [ 12, 112],
         [ 13, 113]],

        [[ 14, 114],
         [ 15, 115],
         [ 16, 116]],

        [[ 17, 117],
         [ 18, 118],
         [ 19, 119]]],


       [[[ 21, 121],
         [ 22, 122],
         [ 23, 123]],

        [[ 24, 124],
         [ 25, 125],
         [ 26, 126]],

        [[ 27, 127],
         [ 28, 128],
         [ 29, 129]]]])

In [23]:

    

X = np.array([[[1,2,3,4],[5,6,7,8],[9,10,11,12]],[[13,14,15,16],[17,18,19,20],[21,22,23,24]]])
X.shape


    

    
Out[23]:





(2, 3, 4)

In [25]:

    

X_trans_1 = tf.transpose(X,perm=[2,0,1]).eval()
print(X_trans_1)


    

    




[[[ 1  5  9]
  [13 17 21]]

 [[ 2  6 10]
  [14 18 22]]

 [[ 3  7 11]
  [15 19 23]]

 [[ 4  8 12]
  [16 20 24]]]

In [28]:

    

X_reshape_1 = tf.reshape(X_trans_1,[4,-1]).eval()


    

In [29]:

    

X_reshape_1


    

    
Out[29]:





array([[ 1,  5,  9, 13, 17, 21],
       [ 2,  6, 10, 14, 18, 22],
       [ 3,  7, 11, 15, 19, 23],
       [ 4,  8, 12, 16, 20, 24]])

In [ ]: