Skip to content

wakanawakana/win-darknet

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Darknet Implement Win32/64

Supports

  • yolo V2
  • use OPENCV option
  • use GPU option
  • use CUDNN option

Tested pattern

  • Win32 (VS2013)
  • Win32 (OPENCV & VS2013)
  • Win32 (GPU(Cuda6.5) & OPENCV & VS2013)
  • x64 (VS2013)
  • x64 (OPENCV & VS2013)
  • x64 (GPU(Cuda6.5) & OPENCV & VS2013)
  • x64 (GPU(Cuda8.0) & CUDNN & OPENCV & VS2013)

What Modify

  • I made it possible to run with only CPU even in GPU build.
  • Implementation of functions missing in Windows.
  • Extended for recall function, Calcutate Precision, Recall, Parameterized threashold, store predict images
  • Extended for demo function, add video writing

For more information see the Darknet project website.

"yolov2wgpu.bat" is a test batch of yolov 2.
Download the weits file to the TOP hierarchy and use it.
Surely predictions images will be generated.

Packages

No packages published

Languages

  • C 91.2%
  • Cuda 7.9%
  • Python 0.3%
  • C++ 0.3%
  • Makefile 0.2%
  • Shell 0.1%