CXXNET (spelled as: C plus plus net) is a neural network toolkit build on mshadow(https://github.com/tqchen/mshadow). It is yet another implementation of (convolutional) neural network. It is in C++, with about 1000 lines of network layer implementations, easily configuration via config file, and can get the state of art performance.
Creater: Tianqi Chen and Bing Xu
Documentation and Tutorial: https://github.com/antinucleon/cxxnet/wiki
- Small but sharp knife: the core part of the implementation is less than 2000 lines, and easily extendible.
- cxxnet is build with mshadow, a tensor template library for unified CPU/GPU computation. All the functions are only implemented once, as a result.
- Speed: On Bing Xu’s EVGA GeForce 780 GTX with 2304 CUDA cores, cxxnet archived 211 images per second in training on ImageNet data with Alex Krizhevsky’s deep network structure.
- Common Requirement: NVIDIA CUDA with cuBLAS, cuRAND and cudaRT; OpenCV; mshadow (will be downloaded by using build.sh)
- MKL version: Intel MKL directly run
build.sh - If you don’t have MKL, using
build.sh blas=1to build with CBLAS- Depending your version of CBLAS(ATLAS, etc.), you may need to change -lblas to -lcblas in Makefile