Class JacobianNumerical_F64

java.lang.Object
org.ddogleg.chainrule.JacobianNumerical_F64
All Implemented Interfaces:
JacobianNumerical<double[]>
public class JacobianNumerical_F64 extends Object implements JacobianNumerical<double[]>
Implementation of JacobianNumerical

  • Field Details

    • printPartials

      protected boolean printPartials
      print out partials before multiplying by dout

  • Constructor Details

    • JacobianNumerical_F64

      public JacobianNumerical_F64()

  • Method Details

    • configure

      public void configure(double T)
      Description copied from interface: JacobianNumerical
      Overrides default settings for computing numerical gradient.
      Specified by:
      configure in interface JacobianNumerical<double[]>
      Parameters:
      T - Sampling distance used for numerical differentiation

    • setFunction

      public void setFunction(List<CRArray<double[]>> inputs, CROperatorForward<double[]> function)
      Description copied from interface: JacobianNumerical
      Specifies the function which will be differentiated.
      Specified by:
      setFunction in interface JacobianNumerical<double[]>
      Parameters:
      inputs - Value of all inputs passed into the function
      function - The function which is to be differentiated

    • differentiate

      public void differentiate(CRMatrix<double[]> dout, List<CRMatrix<double[]>> jacInputs)
      Description copied from interface: JacobianNumerical
      Performs numerical differentiation to compute the gradients of input and parameters. The value of parameters will be the current parameters stored in the function.
      Specified by:
      differentiate in interface JacobianNumerical<double[]>
      Parameters:
      dout - (Input) Output's Jacobian, computed from next layer.
      jacInputs - (Output) Storage for input Jacobians