Tensorflow分布：如何让两名工人在拖车上

Question

我想让一名工人上“10.222.152.100”而另一名工人上“10.222.152.101”。 这是worker1.py：

worker_1="10.222.152.100:2222"
worker_2="10.222.152.101:2222"
worker_host=[worker_1,worker_2]
cluster_spec=tf.train.clusterSpec({"FirstJob":worker_host})
server=tf.train.Server(cluster_spec,job_name="FirstJob",task_index=0)
server.join()

这是worker2.py：

worker_1="10.222.152.100:2222"
worker_2="10.222.152.101:2222"
worker_host=[worker_1,worker_2]
cluster_spec=tf.train.clusterSpec({"FirstJob":worker_host})
server=tf.train.Server(cluster_spec,job_name="FirstJob",task_index=1)
server.join()

我希望让工人1和火车上的损失在woeker2上，Client.py是线性回归：

w=tf.Variable(0.0,name="W")
b=tf.Variable(0.0,name="b")

with tf.devic("job:FirstJob/task:0/"):
    loss=tf.square(Y-tf.mul(X,w)-b)
with tf.devic("job:FirstJob/task:1/"):
    train=tf.train.GradientDescentOptimizer(0.01).minimize(loss)
with tf.Session(target) as sess:
    init=tf.initialize_all_variables()
    sess.run(init)
    for i in range(100):
        for (x,y) in zip (Train_X,Train_Y):
            sess.run(train,feed_dic={X:x,Y:y})

我的IP是“10.222.152.100”，当target =“10.222.152.100：2222”时，它打印出来：

Could not satisfy explicit device specification '/job:FirstJob/task:1'

当target =“10.222.152.101:2222”时，它会打印：

tensorflow.python.framework.errors.UnimplementedError

希望你的帮助＆gt;＆gt;＆gt;