---

Welcome to PockyML!

This project is named Pocky Machine Learning. It's a lightweight open source C++ machine learning library.
I'm a novice in ML, but I think I love it. So I will try my best to write this tool. Meanwhile the Chocolate bars of Pocky is so delicious.

Algorithm

1. K-means

	//instantiate the model
	kmeans_clusters mykmean;
	mykmean.input_data(test_data);
	//train and set the number of clusters is 3
	mykmean.train(3);
	//display the result
	mykmean.display();

2. GMM (Gaussian Mixture Model, Unfinished)

3. KNN (K-Nearest Neighbor)

	//instantiate the model
	knn_model myknn;
	myknn.input_data(test_data, label);
	//set k=3,
	result = myknn.classification(try_data,3);

4. SVM (Suport Vector Machine, based on LibSVM)

5. ANN (Artifical Neural Network)

	//instantiate the model
	ann_model myANN(data,lab);
	//set the num of hidden nodes
	myANN.init(10);
	//set other parameters, learning_rate, iteration_times, 
	myANN.parameter_setting(0.02, 80000, 0.000000001);
	//train and classify
	myANN.train();
	test_lab = myANN.classification(data);

6. PCA (Principal Component Analysis)

	//instantiate the model
	PCA_model myPCA(test_data,row_num,col_num);
	//computer the eigvalue and eigvector
	myPCA.pca_solve_eig();
	myPCA.print();

7. NB (Naive Bayesian)

	//instantiate the model
	nb_model myNB(data,lab);
	//train and classify
	myNB.train();
	test_lab = myNB.classification(data);

8. LR (Linear Regression)

	//instantiate the model
	lr_model myLR(x,y);
	//initialise the model
	myLR.init();
	//train hte model
	myLR.train();
	//regression
	test_y = myLR.prediction(test_y);

9. C4.5 (Deccision Tree, Unfinished)

10. Adaboost (Baesd on C4.5 Unfinished)

How to use it

	//Makefile
	make
	
	./ann_test
	./pca_test
	./nb_test
	./knn_test
	./kmean_test
	
	//Clean the object file and others
	make clean

Author

Issac (issac.sacrifice@gmail.com,from Zhejiang University)