Coursera, Machine Learning, Neural Networks: Representation - week4/5

Neural Network

Motivations

想要拟合一条曲线,在feature 很多的情况下,feature的组合也很多,在现实中不适用,比如在computer vision问题中feature就太多了.

  Coursera, Machine Learning, Neural Networks: Representation - week4/5

  Coursera, Machine Learning, Neural Networks: Representation - week4/5

Applications

  Coursera, Machine Learning, Neural Networks: Representation - week4/5

  Coursera, Machine Learning, Neural Networks: Representation - week4/5

cost function and BP

   Coursera, Machine Learning, Neural Networks: Representation - week4/5
           
Coursera, Machine Learning, Neural Networks: Representation - week4/5
 
  Coursera, Machine Learning, Neural Networks: Representation - week4/5

  Coursera, Machine Learning, Neural Networks: Representation - week4/5

  Coursera, Machine Learning, Neural Networks: Representation - week4/5

  Coursera, Machine Learning, Neural Networks: Representation - week4/5

  

Gradient Checking

  Coursera, Machine Learning, Neural Networks: Representation - week4/5

  Coursera, Machine Learning, Neural Networks: Representation - week4/5

 
  1. we can use Gradient checking to check if the backpropagation algorithm is working correctly.
  2. need randomly initialize theta
上一篇:ios项目绕过证书访问https程序


下一篇:Handler dispatch failed; nested exception is java.lang.NoClassDefFoundError: org/springframework/util/PatternMatchUtils