![](https://img-blog.csdnimg.cn/20201014180756913.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
UFLDL
ShenYounger
这个作者很懒,什么都没留下…
展开
-
UFLDL(1)Sparse Autoencoder
1. The second term is aregularization term (also called a weight decay term) that tends todecrease the magnitude of the weights, and helpsprevent overfitting. you may also recognize this weigh原创 2017-06-14 17:53:16 · 304 阅读 · 0 评论 -
UFLDL(2)Vectorized implementation
1.Whenever possible, avoid using explicit for-loops in your code.2.Thus, for your first few programs, you might choose to first implement your algorithm without too many vectorization tricks, and原创 2017-06-14 21:55:25 · 386 阅读 · 0 评论 -
epoch vs batch vs iteration
In the neural network terminology:one epoch = one forward pass and one backward pass of all the training examplesbatch size = the number of training examples in one forward/backward pass.转载 2017-07-26 11:22:25 · 512 阅读 · 0 评论