Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Appearance settings

Minimal runtime core of Caffe cut from origin caffe, Forward only, GPU support and Memory efficiency.

Notifications You must be signed in to change notification settings

LilySys/tiny-caffe

Open more actions menu

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

轻量级caffe前向推理框架

本项目为对happynear版本的caffe进行精简化:
1、去除冗余依赖boost、hdf5、lmdb、levedb等,简化以后caffe前向框架更加小型化,并且保持了原生caffe调用习惯(相比于mini-caffe)
2、去除训练用到的loss计算以及solver源码
3、调整修改内存管理机制(去除boost)

小赞:
	相比于mini-caffe,该精简化版本的caffe更加原生态,函数调用和原版的caffe没有区别,同时也支持mkl编译,多GPU模式编译(nccl)
	提供的cmake可以在ubuntu下编译,同时也可以在windows下编译使用。

注意:
	windows下编译,blob.hpp、io.hpp、net.hpp、common.hpp、layer.hpp、caffe.pb.h的文件开始加入:
1、编译时:
		#define LIBCAFFE __declspec(dllexport)
2、调用时:
		#ifdef CAFFE_EXPORTS
		#define LIBCAFFE __declspec(dllexport)
		#else
		#define LIBCAFFE __declspec(dllimport)
		#endif
		
这是一个坑,一开始编译时加入第二种宏定义方式,总是报错。最后不得已使用这种方式。ubuntu下不存在这些问题!

如果需要添加新的层,需要重新生成caffe.pb.h和caffe.pb.cc,protobuf库自由选择,但是版本一定要一致。

About

Minimal runtime core of Caffe cut from origin caffe, Forward only, GPU support and Memory efficiency.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published
Morty Proxy This is a proxified and sanitized view of the page, visit original site.