Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

写得真是很不错,看得赏心悦目 #2

Open
daiab opened this issue Apr 12, 2017 · 6 comments
Open

写得真是很不错,看得赏心悦目 #2

daiab opened this issue Apr 12, 2017 · 6 comments

Comments

@daiab
Copy link

daiab commented Apr 12, 2017

好久没看写得这么规范的代码了,竟然IDE上几乎一个格式or拼写warning都没有,看得心情超好

@ictxiangxin
Copy link
Owner

@daiab 谢谢,就是宽度的标准没有遵照,我觉得以现在显示器的分辨率已经不需要严格遵照这种古老的规定了。

@daiab
Copy link
Author

daiab commented Apr 13, 2017

我不太喜欢宽度的限定,哈哈哈,一行能搞定的内容非得分两行看着难受。 你的水平真的很高,挺完整的python实现。不知道你是怎么考虑的,感觉看着相对tensorflow更像mxnet的风格

@zealseeker
Copy link

@daiab (不知是否扯远了。)之前我还不知有MXNET这个框架,感谢提出,观察了下里面教程非常好,比如http://mxnet.io/architecture/program_model.html 描述了Symbolic Programs,我才想通了之前以前无法理解的Tensor的概念~
个人感觉从用法上此框架更像TensorFlow,不过我猜测真正的计算引擎和TensorFlow、MXNET都不一样吧,毕竟他们都是C++写的。

@daiab
Copy link
Author

daiab commented Apr 13, 2017

@ictxiangxin 能提点小想法吗?Network.add()方法如果能return self,这样定义网络好像可以方便点

@ictxiangxin
Copy link
Owner

@daiab 很好,我在想一个共享机制,在add() return时应当返回一个self的copy,这样达到复用网络的目的,之后会仔细想一想这个怎么搞。

@daiab
Copy link
Author

daiab commented Apr 13, 2017

@ictxiangxin 你说的复用是说这种场景吗?比如:有多个loss,多个loss共享同一层网络。之前看到这个人写的caffe转tensorflow模型的代码,他自己封装了一层,感觉挺优雅的,比如这个code。他这个没有考虑网络多个分支的情况,不过安装他的风格,扩展起来也挺简单的

ictxiangxin pushed a commit that referenced this issue Apr 14, 2017
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants