tensorflow中的tensor值的获取:
import tensorflow as tf
#定义变量a
a=tf.Variable([[[1,2,3],[4,5,6]],[[7,8,9],[10,11,12
]]])
#定义索引
indics=[[0,0,0],[0,1,1],[0,1,2
]]
#把a中索引为indics的值取出
b=
tf.gather_nd(a,indics)
#初始化
init=
tf.global_variables_initializer()
with tf.Session() as sess:
#执行初始化
sess.run(init)
#打印结果
print(a.eval())
print(b.eval())
转载于:https://www.cnblogs.com/shixisheng/p/9353757.html
相关资源:linux下利用/proc进行进程树的打印