初始模型有两个softmax输出吗？ [英] Does the Inception Model have two softmax outputs?
问题描述
此图像显示了Inception v3模型：
The Inception v3 model is shown in this image:
图片来自此博客文章：
https：// research.googleblog.com/2016/03/train-your-own-image-classifier-with.html
似乎有两种Softmax分类输出。为什么？
It seems that there are two Softmax classification outputs. Why is that?
在TensorFlow示例中使用哪个文件作为输出张量，在该文件中名称为'softmax：0'？
Which one is used in the TensorFlow example as the output tensor with the name 'softmax:0' in this file?
Inception v3的学术论文模型似乎没有Inception模型的图片：
The academic paper for the Inception v3 model doesn't seem to have this image of the Inception model:
http://arxiv.org/pdf/1512.00567v3.pdf
我试图了解为什么会有这两个分支
I'm trying to understand why there are these two branches of the network with seemingly two different softmax-outputs.
感谢您的澄清！
推荐答案
您引用的关于辅助分类器的纸张的第4节。这些是添加到网络较低层的分类器，可通过减轻消失的梯度问题和加速收敛来改善训练。为了在经过训练的网络上进行推理，您应该使用模型中的主分类器 softmax：0 和 NOT 辅助分类器 auxiliary_softmax：0 。
Section 4 of the paper you cite is about auxiliary classifiers. These are classifiers added to the lower levels of the network, that improve training by mitigating the vanishing gradients problem and speedup convergence. For running inference on a trained network, you should use the main classifier, called softmax:0 in the model, and NOT the auxiliary classifier, called auxiliary_softmax:0.
这篇关于初始模型有两个softmax输出吗？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！
