使用nd4j的S型导数

时间:2018-12-27 10:10:18

标签: java python scala numpy nd4j

我执行的Sigmoid效果很好,但sigmoidDerivative的结果与nd4j中的Sigmoid相同。 Transforms.sigmoidDerivative(x)Transforms.sigmoidDerivative(x, true)有什么区别?

INDArray x = Nd4j.create(new double[] { 0.1812, 0.1235, 0.8466 });
System.out.println(x);
System.out.println(Transforms.sigmoid(x));
System.out.println(Transforms.sigmoidDerivative(x));
System.out.println(Transforms.sigmoidDerivative(x, true));

给出输出:

[[    0.1812,    0.1235,    0.8466]]
[[    0.5452,    0.5308,    0.6999]]
[[    0.5452,    0.5308,    0.6999]]
[[    0.2480,    0.2490,    0.2101]]

与python的numpy比较:

>>> def sigmoid(x):
...     return 1.0 / (1 + np.exp(-x))
... 
>>> def sigmoid_derivative(x):
...     a = sigmoid(x)
...     return a * (1.0 - a)

>>> x = np.array([    0.1812,    0.1235,    0.8466])
>>> sigmoid(x)
array([0.54517646, 0.53083582, 0.69985343])
>>> sigmoid_derivative(x)
array([0.24795909, 0.24904915, 0.21005861])

Nd4j pom:

<dependency>
     <groupId>org.nd4j</groupId>
     <artifactId>nd4j-native-platform</artifactId>
     <version>1.0.0-beta3</version>
</dependency>

1 个答案:

答案 0 :(得分:1)

您是对的,Transforms.sigmoidDerivative(x)Transforms.sigmoidDerivative(x, true)应该给出相同的结果,这是dl4j中的错误。正确的行为具有后一种方法。我已经提交了pull request来解决它。