如何关闭Tensorflow中的辍学测试? [英] How to turn off dropout for testing in Tensorflow?

问题描述

总体上来说，我对Tensorflow和ML还是比较陌生，因此我对一个(可能)琐碎的问题表示歉意.

I am fairly new to Tensorflow and ML in general, so I hereby apologize for a (likely) trivial question.

我使用辍学技术来提高我的网络的学习率，而且看起来效果很好.然后，我想在一些数据上测试网络，看它是否可以这样工作:

I use the dropout technique to improve learning rates of my network, and it seems to work just fine. Then, I would like to test the network on some data to see if it works like this:

   def Ask(self, image):
        return self.session.run(self.model, feed_dict = {self.inputPh: image})

很明显，由于辍学仍然存在，因此每次产生不同的结果.我能想到的一种解决方案是创建两个单独的模型-一个用于训练，另一个用于实际以后使用网络，但是，这种解决方案对我来说似乎不切实际.

Obviously, it yields different results each time as the dropout is still in place. One solution I can think of is to create two separate models - one for a training and the other one for an actual later use of the network, however, such a solution seems impractical to me.

解决此问题的常用方法是什么?

What's the common approach to solving this problem?

推荐答案

最简单的方法是使用

The easiest way is to change the keep_prob parameter using a placeholder_with_default:

prob = tf.placeholder_with_default(1.0, shape=())
layer = tf.nn.dropout(layer, prob)

通过这种方式训练时，您可以像这样设置参数:

in this way when you train you can set the parameter like this:

sess.run(train_step, feed_dict={prob: 0.5})

，当您评估默认值1.0时.

and when you evaluate the default value of 1.0 is used.

这篇关于如何关闭Tensorflow中的辍学测试?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！