如何使用张量流从CNN层提取激活? [英] How to extract activation from CNN layers using tensorflow?

问题描述

我想从完全连接的层中提取神经激活.在Caffe中，我就是这样 net.blobs [layer_name] .data

I want to extract the neural activation from the Fully connected layers. In Caffe i was doing like this net.blobs[layer_name].data

我如何在tensorflow中做同样的事情?

How can i do the same in tensorflow?

推荐答案

您应该使用会话对象来获取存储在张量中的值.尽量不要忘记将占位符张量的值作为feed_dict传递.

You should use session object to get values stored in tensors. Try not to forget to pass values of placeholder tensors as feed_dict.

sess = tf.InteractiveSesssion()

full_connected = ....
value_of_fully_connected = sess.run(fully_connected,feed_dict={your_placeholders_and_values)

这篇关于如何使用张量流从CNN层提取激活?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！