Tensorflow重用神经网络

Question

我是tensorflow的新手，我一直在训练一个简单的神经网络，但是一旦训练完成，我就不知道如何重用NN来获得输入的输出。

def train_neural_network(x,y,aDataTrain,aTargetTrain,aDataTest,aTargetTest):
    batch_size = 500
    prediction = neural_network_model(x,len(aDataTrain[0]))
    cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=prediction,labels=y))
    optimizer = tf.train.AdamOptimizer().minimize(cost)
    hm_epochs = 1

    saver = tf.train.Saver()
    with tf.Session() as sess:
        sess.run(tf.global_variables_initializer())

        for epoch in range(hm_epochs):
            epoch_loss = 0
            i = 0
            while i < len(aDataTrain):
                start = i
                end = i + batch_size
                batch_x = np.array(aDataTrain[start:end])
                batch_y = np.array(aTargetTrain[start:end])

                _,c = sess.run([optimizer,cost],feed_dict={x:batch_x,y:batch_y})
                epoch_loss += c
                i += batch_size
            print ("Epoch", epoch, "completed out of", hm_epochs, "loss", epoch_loss)

        correct =tf.equal(tf.argmax(prediction,1), tf.argmax(y,1))

        accurracy = tf.reduce_mean(tf.cast(correct,'float'))
        finalAcc = accurracy.eval({x:aDataTest,y:aTargetTest})
        saver.save(sess, 'model/model.ckpt')

    print("Accuracy:",finalAcc)

因此，一旦我保存了模型并尝试恢复它，我就不知道如何继续从"input_data“中获得NN的输出。

def execute_neural_network(x,y,aDataTrain,aTargetTrain,aDataTest,aTargetTest):
    batch_size = 1
    y_pred = []

    prediction = neural_network_model(x,len(aDataTrain[0]))
    cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=prediction,labels=y))
    optimizer = tf.train.AdamOptimizer().minimize(cost)

    input_data = [5.0, 3.0, 1.0, 5.0, 6.0, 5.0, 2.0, 4.0, 7.0, 6.0, 3.0, 3.0, 3.0, 3.0, 3.0, 4.0, 2.0, 3.0, 3.0, 3.0, 3.0, 3.0, 2.0, 3.0, 2.0, 3.0, 2.0, 3.0, 3.0, 4.0, 3.0, 3.0, 2.0, 4.0, 3.0, 3.0, 2.0, 4.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 61.0, 21.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 75.0, 3.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 6.0, 35.0, 11.0, 10.0, 33.0, 24.0, 6.0, 2.0, 2.0, 3.0, 4.0, 3.0, 3.0, 8.0, 6.0, 5.0, 6.0, 5.0, 8.0, 9.0, 13.0, 7.0, 25.0, 11.0, 2.0, 2.0, 2.0, 2.0, 2.0]

    saver = tf.train.Saver()
    with tf.Session() as sess:
        saver.restore(sess, 'model/model.ckpt')
        #Get neural network output from input_data

Answer 1

假设您以某种方式创建了图/网络的模型：

with tf.Session() as sess:
    #do other stuff
    predictionOp = tf.argmax(py_x, 1)

saver.save(sess, 'model') 

其中predictionOp是变量，它是网络的输出。

您可以在之后添加类似这样的内容：tf.add_to_collection("predictionOp", predictionOp)，以便为predictionOp指定一个更容易查找的名称。然后，您可以使用以下命令重新加载模型并获得预测结果：

with tf.Session() as sess:
    new_saver = tf.train.import_meta_graph('model.meta')
    new_saver.restore(sess, 'model')
    predictionOp = tf.get_collection("predictionOp")[0]

    #get the prediction
    prediction = sess.run(predictionOp, feed_dict={"x:0": input_data})

有关更多信息，请查看tensorflow documentation和here，了解有关基础知识的更多信息。此外，还有一些其他线程处理类似的问题，如this和this one。