Leon-trainee/cnn

基于Java实现CNN，并附MNIST和语音（MFCC特征）性别识别示例。

JavaCNN

基于Java实现CNN

构建CNN

LayerBuilder builder = new LayerBuilder();
builder.addLayer(Layer.buildInputLayer(new Size(28, 28)));
builder.addLayer(Layer.buildConvLayer(6, new Size(5, 5)));
builder.addLayer(Layer.buildSampLayer(new Size(2, 2)));
builder.addLayer(Layer.buildConvLayer(12, new Size(5, 5)));
builder.addLayer(Layer.buildSampLayer(new Size(2, 2)));
builder.addLayer(Layer.buildOutputLayer(10));
CNN cnn = new CNN(builder, 50);

运行MNIST数据集

String fileName = "data/train.format";
Dataset dataset = Dataset.load(fileName, ",", 784);
cnn.train(dataset, 100);
Dataset testset = Dataset.load("data/test.format", ",", -1);
cnn.predict(testset, "data/test.predict");

计算精度可以达到97.8%。