Document
CNN accelerated by cuda.
The start-of-art result's of popular datasets
- Test on mnist and get 99.76%, after voting(99.82%) (best 99.79%)
- Test on cifar-10 and get 81.38% (best 90%)
- Test on cifar-100 and get 51.13% (best 65%)
- Use DropConnnect to train the NetWork
- Support checkpoint, the program will save the best test result and save the network weight in the file "Result/checkPoint.txt", If the program exit accidentally, you can continue the program form this checkpoint.
- Translate the data set of mnist, including scale, rotate, distortion.
- The log will be saved in the file "Result/log.txt".
- In the convolutional layers, you can chose combine feature maps, according to "notes on Convolutional Neural NetWorks"
Depend on opencv and cuda
You can compile the code on windows or linux.
###SDK path
- linux: /usr/local/cuda/samples/common/inc/ (For include file "helper_cuda")
- windows: X:/Program Files (x86) /NVIDIA Corporation/CUDA Samples/v6.5/common/inc (For include file "helper_cuda")
###Library search path(-L)
- linux: /usr/local/lib/
- windows: X:/Program Files/opencv/vs2010/install/x86/cv10/lib (Depend on situation)
###libraries(-l)
- opencv_core
- opencv_highgui
- opencv_imgproc
- opencv_imgcodecs (need for opencv3.0)
- cublas
- curand
###GPU compute
- capability 2.0
###Windows
- Install vs2010.
- Download and install opencv-2.4 or other higher versions
- Download and install cuda-5.0 or other higher versions
- When you create a new project using VS2010, You can find NVIDIA-CUDA project template, create a cuda-project.
- Add the "include path" and "lib path" to the project
###Linux
- Install opencv and cuda
- Start the nsight from cuda
- Create an 'empty cuda' project and import the clone code
- Add the "include path" and "lib path" to the project
Config
- Author :zhxfl
- Mail :zhxfl@mail.ustc.edu.cn
- Welcome for any suggest!!