tf.transpose

来源：互联网发布：mysql 获取上一年月份编辑：程序博客网时间：2024/05/22 03:32

 tf.transpose(X,perm = [2,0,1])就是对数组维数进行转换，X[i][j][k]=Y[k][i][j]import tensorflow as tfimport numpy  as npinput_x = [    [        [1, 2, 3, 4],        [5, 6, 7, 8],        [9, 10, 11, 12]    ],    [        [13, 14, 15, 16],        [17, 18, 19, 20],        [21, 22, 23, 24]    ]]print ("input_x shape:",np.shape(input_x))result = tf.transpose(input_x, perm=[2, 0, 1])with tf.Session() as sess:    print("input_x",sess.run(result))input_y=np.zeros((4,2,3),dtype=int)for i in range(4):    for j in range(2):        for k in range(3):            input_y[i][j][k]=input_x[j][k][i]print("input_y",input_y)

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

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

阅读全文

0 0