Introduction

All previous neural net studies has been transformed from my own neural network codes to tiny-dnn which is c++14 based open source. Please make sure that all applications here are tested on mac os environment.

Cloning and build

git clone --recursive https://github.com/supertigim/deep-learning-2nd.git  
  
// to make tiny-dnn the latest version  
git submodule foreach git pull origin master   

Go to sub folders like cnn, and xor-problem, and read the README.md to build each application.

Changes in tiny-DNN

Don't change unless any improvement happens in your environment

Uncomment defines in config.h

#define CNN_USE_SSE  
#define CNN_USE_OMP  
#define CNN_USE_GCD  

Random generator in random.h

class random_generator {  
public:  
	static random_generator &get_instance() {  
	static random_generator instance;  
	return instance;  
}  

std::mt19937 &operator()() {   
	set_seed(rd_());  // ADD!!!
	return gen_;  
}    

void set_seed(unsigned int seed) { gen_.seed(seed); }  

private:  
	// avoid gen_(0) for MSVC known issue  
	// https://connect.microsoft.com/VisualStudio/feedback/details/776456  
	random_generator() : gen_(3) {}  
	std::mt19937 gen_;  
	std::random_device rd_;	// ADD!!!  
};

Reference

1.tiny-dnn online manual
2.comparison between neural network libraries
3.Solving XOR problem using tiny-DNN
4.딥러닝 활용 추천 시스템 개발
5.구글 AI Experiments
6.AI Duet이라는 들은 음악에 맞춰 자동 연주
7.Caffe로 구현한 breakout with DQN
8.python으로 구현한 self driving car
9.8번 설명 사이트
9.Caffe c++ 스터디

license

MIT