TensorFlow不同交叉熵计算方式

最新推荐文章于 2024-05-22 16:05:58 发布

yqtaowhu

最新推荐文章于 2024-05-22 16:05:58 发布

阅读量2.1k

点赞数

分类专栏： Tensorflow/Caffe

本文链接：https://blog.youkuaiyun.com/taoyanqi8932/article/details/77601188

版权

Tensorflow/Caffe 专栏收录该内容

11 篇文章

订阅专栏

import tensorflow as tf  
#our NN's output  
logits=tf.constant([[1.0,3.0,2.0],[3.0,2.0,1.0],[1.0,2.0,3.0]])  
#step1:do softmax  
y=tf.nn.softmax(logits)  
#true label  
y_=tf.constant([[0.0,1.0,0.0],[1.0,0.0,0.0],[0.0,0.0,1.0]])  
#step2:do cross_entropy  
cross_entropy = -tf.reduce_sum(y_*tf.log(y))  
#do cross_entropy just one step  
cross_entropy2 = tf.reduce_sum(tf.nn.softmax_cross_entropy_with_logits(logits = logits,labels=y_))#dont forget tf.reduce_sum()!!  
lab = tf.constant([1,0,2])
cross_entropy3 = tf.reduce_sum(tf.nn.sparse_softmax_cross_entropy_with_logits(logits=logits,labels=lab))
with tf.Session() as sess:  
    softmax=sess.run(y)  
    c_e = sess.run(cross_entropy)  
    c_e2 = sess.run(cross_entropy2)  
    print("step1:softmax result=")  
    print(softmax)  
    print("step2:cross_entropy result=")  
    print(c_e)  
    print("Function(softmax_cross_entropy_with_logits) result=")  
    print(c_e2)  
    c_e3 = sess.run(cross_entropy3)
    print("using sparse cross_entropy")
    print(c_e3)