tensorflow API：tf.group

tf.group

group(    *inputs,    **kwargs)

创建一个包含几个操作的op结点

当这个op结点运行完成，所有作为input的ops都被运行完成，这个操作没有返回值

例子：

tag =tf.reduce_any([True])b = tf.Variable(tf.random_normal([5,3],stddev=0.35),name="weights")  #init = tf.initialize_all_variables()sess = tf.Session()#sess.run(init)print(sess.run(b.initialized_value()))with tf.control_dependencies([tag]):    do_updates = tf.group(        tf.reduce_sum(input_tensor=b),        b.assign(b+10)        )init = tf.initialize_all_variables()sess.run(init)sess.run([do_updates])print(sess.run(tag))print(sess.run(b))

结果：
[[-0.06633449 0.83556134 0.14434727]
[-0.3425082 -0.39835241 0.51608163]
[ 0.21805006 -0.02585059 -0.04273668]
[-0.18699114 0.01443172 0.10423679]
[-0.62762541 -0.14095569 0.34139016]]
True
[[ 10.06451797 10.65713215 10.08987713]
[ 10.30089855 10.40508556 10.25630856]
[ 9.92398548 9.75419617 9.68528366]
[ 10.20005417 10.23743248 10.3180685 ]
[ 10.57752132 10.41219616 9.6665287 ]]