首页 > 其他 > 详细

[Intro to Deep Learning with PyTorch -- L2 -- N9] Perceptron Trick

时间:2020-06-08 00:13:48      阅读:63      评论:0      收藏:0      [点我收藏+]

Give this:

技术分享图片

 

We have a wrongn classified point, how to move the line to come closer to the points?

技术分享图片

We apply learning rate and since wrong point is in positive area, we need to  W - learning rate * wrong cord

 

The same, if the wrong point in negivate area, we do : W + learing rate * wrong cord:

技术分享图片

 

For the second example, where the line is described by 3x1+ 4x2 - 10 = 0, if the learning rate was set to 0.1, how many times would you have to apply the perceptron trick to move the line to a position where the blue point, at (1, 1), is correctly classified?

Answer: 10 times

[Intro to Deep Learning with PyTorch -- L2 -- N9] Perceptron Trick

原文:https://www.cnblogs.com/Answer1215/p/13062878.html

(0)
(0)
   
举报
评论 一句话评论(0
关于我们 - 联系我们 - 留言反馈 - 联系我们:wmxa8@hotmail.com
© 2014 bubuko.com 版权所有
打开技术之扣,分享程序人生!