Question

我同时训练两个神经网络。当我尝试存储它们时，我收到以下错误：

tensorflow - 警告 - 序列化时遇到错误 data_preprocessing。类型不受支持，或项目类型不要在CollectionDef中匹配字段类型。＆＃39; NoneType＆＃39;对象没有属性＆＃39;名称＆＃39;

两个网络都有自己的类，下面是一些只有一个Network类的简化代码。我只想表明我使用相同的tf.Session创建了两个带有tflearn的网络。当我试图在训练过程中保存它们时，我得到了上面的错误

class Network:
    def create_network(self):
        inputs = tflearn.input_data(shape=[None, self.s_dim])
        net = tflearn.fully_connected(inputs, 400, activation='relu',bias=True)
        out = tflearn.fully_connected(net, self.a_dim, activation='tanh')
        return inputs, out

    def init(self,sess):
        ...
        self.inputs, self.out = self.create_network()
        self.loss = tflearn.mean_square(self.placeholder, self.out)
        self.optimize = tf.train.AdamOptimizer(self.learning_rate).minimize(self.loss)
        ...

class Main:
        with tf.Session() as sess:
                    network1=NN.Network(sess)
                    network2=NN.Network(sess)
                    self.saver = tf.train.Saver()
                    sess.run(tf.initialize_all_variables())
                    for i in range(steps):
                               train(network1,network2,sess)
                               self.saver.save(self.sess,'name',i)

btw：我使用的是Cuda 8，cudnn 5和tensorflow rc 0.11.0

tensorflow尝试序列化两个NN时遇到错误

0 个答案: