Skip to content

A neural nets implementation only based on jdk that does not depend on any other libraries.

License

Notifications You must be signed in to change notification settings

xuejianbest/NeuralNetwork

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

NeuralNetwor in Java

A neural nets implementation only based on jdk that does not depend on any other libraries.


使用MNIST数据集进行训练和测试。

训练的模型为单隐藏层,输入为(28*28)/(2*2)个像素(我对原始图片进行了简单压缩:紧邻的四个像素只取左上角的一个像素), 隐藏层30个神经元,输出层10个神经元。

模型在测试集上的识别准确率为:93.32%(对测试集前5000和后5000样本的识别准确率分别为90.90%和95.74%)

模型信息已保存在model/model.txt

红色为识别错误的样本参考: 识别测试

About

A neural nets implementation only based on jdk that does not depend on any other libraries.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages