Question

有谁知道在Caffe softmax层内发生了什么计算？

我正在使用预先训练好的网络，最后有一个softmax图层。

在测试阶段，对于图像的简单前进，倒数第二层（＆＃34; InnerProduct＆＃34;）的输出如下： -0.20095,0.39989,0.22510，-0.36796，-0.21991,0.43291，-0.22714，-0.22229，-0.08174,0.01931，-0.05791,0.21699,0.00437，-0.02350,0.02924，-0.28733,0.19157，-0.04191，-0.07360,0.30252

最后一层（＆＃34; Softmax＆＃34;）输出是以下值： 0.00000,0.44520,0.01115,0.00000,0.00000,0.89348,0.00000,0.00000,0.00002,0.00015,0.00003,0.00940,0.00011,0.00006,0.00018,0.00000,0.00550,0.00004,0.00002,0.05710

如果我在内部产品图层的输出上应用Softmax（使用外部工具，如matlab），我会得到以下值： 0.0398,0.0726,0.0610,0.0337,0.0391,0.0751,0.0388,0.0390,0.0449,0.0496,0.0460,0.0605,0.0489,0.0476,0.0501,0.0365,0.0590,0.0467,0.0452,0.0659

后者对我有意义，因为概率加起来为1.0（注意Caffe的Softmax图层值的总和> 1.0）。

显然，Caffe中的softmax层不是一个直接的Softmax操作。

（我认为它没有任何区别，但我只想提一下，我正在使用预先训练过的flickr风格网络，请参阅说明here）。

编辑：

这是proto txt中最后两层的定义。请注意，最后一层的类型是＆＃34; Softmax＆＃34;。

   Bank account1;

   if(choice.equals("n"))
      account1 = new Bank();
   else
   {
       System.out.println("Enter your initial balance :");
       double ibalance = scanner.nextDouble();
       account1 = new Bank(ibalance);
   }

Answer 1

你得到的结果很奇怪 "Softmax"图层的forward方法执行的操作是：

（注意，执行前两个步骤是为了防止计算中出现溢出）。

Caffe的Softmax层如何计算概率值？

1 个答案: