Java实现深度神经网络：Sigmoid层

Java实现深度神经网络：Sigmoid层

使用Java实现

Sigmoid.java

/**
 * Sigmoid层
 * 
 * @author SXC 2020年9月11日 下午12:56:31
 */
public class Sigmoid {
	public Sigmoid() {

	}

	public double[] out(double[] x) {
		double[] out = new double[x.length];
		for (int i = 0; i < x.length; i++) {
			out[i] = sigmoidF(x[i]);
		}
		return out;
	}

	private double sigmoidF(double x) {
		return 1f / (1f + Math.pow(Math.E, -1 * x));
	}
}

测试代码

public class t {

	public static void main(String[] args) {
        double[] x= {5,7,3};
        Sigmoid sigmoid=new Sigmoid();
        double[] out1=sigmoid.out(x);
        System.out.println(out1[0]+"  " +out1[1]+"  " +out1[2]);
	}
}

运行结果

0.9933071490757153  0.9990889488055994  0.9525741268224331

使用Pytorch实现

import torch as t
input=t.tensor([5,7,3],dtype=t.float32)
sig=t.nn.Sigmoid()
print(sig(input))

运行结果

H:\ProgramData\Anaconda3\python.exe D:/PycharmProjects/untitled/t.py
tensor([0.933, 0.9991, 0.9526])

Process finished with exit code 0

结果相同，达到目标