0
点赞
收藏
分享

微信扫一扫

tf.reduce_max用法


对于tf.reduce_max,这个函数有点奇怪,axis=0指的是计算矩阵每列的最大值,axis=1计算行最大值
与numpy 相同

import tensorflow as tf
import numpy as np


a=np.array([[2,4,5,7],[9,3,6,2]])

print('a=\n',a)



print('-'*30+'分割线'+'-'*30)
a1=tf.reduce_max(a,axis=0)
print( 'tf.reduce_max(a,axis=0)=\n',a1.numpy())

a2=np.max(a,axis=0)
print( 'np.max(a,axis=0)=\n',a2)


print('-'*30+'分割线'+'-'*30)
a1=tf.reduce_max(a,axis=1)
print( 'tf.reduce_max(a,axis=1)=\n',a1.numpy())
a2=np.max(a,axis=1)
print( 'np.max(a,axis=1)=\n',a2)

print('-'*30+'分割线'+'-'*30)
a1=tf.reduce_max(a)
print( 'tf.reduce_max(a)=\n',a1.numpy())

a2=np.max(a)
print( 'np.max(a)=\n',a2)

a=
[[2 4 5 7]
[9 3 6 2]]
------------------------------分割线------------------------------
tf.reduce_max(a,axis=0)=
[9 4 6 7]
np.max(a,axis=0)=
[9 4 6 7]
------------------------------分割线------------------------------
tf.reduce_max(a,axis=1)=
[7 9]
np.max(a,axis=1)=
[7 9]
------------------------------分割线------------------------------
tf.reduce_max(a)=
9
np.max(a)=
9


举报

相关推荐

0 条评论