Feng Wang

Results 200 comments of Feng Wang

@smfullman , @mdadhich , The phase in Caffe is no longer string 'TRAIN' or 'TEST'. It is 0(TRAIN) or 1(TEST) now. So just modify ``` cfg_key = str(self.phase) ``` to...

@davidshen84 I guess you can read Mandarin from your name. Here is a blog describes how to build leveldb.pyd: http://blog.csdn.net/toontong/article/details/8442995 If you cannot read it. Try this repository: https://github.com/chirino/leveldb/blob/master/WINDOWS.md

@davidshen84 I have just created a new repository https://github.com/happynear/py-leveldb-windows . Sorry that I am not familiar with cmake, so I just created a VS project. May it help.

@vfdev-5 Thanks for your work.

中间过程哪错了你指一下我才能解释啊。。

貌似第一个箭头那里确实是多了个波浪线...应该是x_j。

我看看最近修改一版更新上去吧,前边也有些地方描述得不清楚。

![image](https://user-images.githubusercontent.com/3897999/42195990-24bcb200-7eae-11e8-9390-2c1ab9586bf8.png) 应该是这个样子。

x_i是一个数字 前边一项是对分子上的x_i求导,后面一项是对分母上的||x||求导。因为每个x_j都要除||x||,而||x||里面包含了x_i,每个对\tilde{x_j}的导数都要往x_i上传,所以是个求和的形式。

$\frac{\partial L}{\partial \bar{x}}$ is the gradient passed from back-propagation. It can be in any direction.