Uses of Class
org.ddogleg.chainrule.CRArray

Packages that use CRArray

Package	Description
org.ddogleg.chainrule
org.ddogleg.chainrule.ops
org.ddogleg.chainrule.sym

Uses of CRArray in org.ddogleg.chainrule

Methods in org.ddogleg.chainrule that return CRArray

Modifier and Type	Method	Description
static CRArray<double[]>	CRArray.arrayD(double... values)
CRArray<A>	CRArray.copy()	Creates a copy of the original array and this wrapper.
static <T> CRArray<T>	FactoryCRArray.create(int length, Class<T> type)
CRArray<T>	MathCRMatrix.createArray(int length)	Allocates a new array with the specified length
CRArray<double[]>	MathCRMatrixD.createArray(int length)
static <T> CRArray<T>	FactoryCRArray.random(Random rand, int length, Class<T> type)
CRArray<T>	MathCRMatrix.random(Random rand, double min, double max, int length)	Creates a random array
CRArray<double[]>	MathCRMatrixD.random(Random rand, double min, double max, int length)
CRArray<A>	CRArray.reserve(int length)	Ensure that the array has at least this many elements in it
CRArray<A>	CRArray.setOffset(int offset)	Assigns a new value to the offset and returns a reference to 'this' allowing operations to be chained.
CRArray<A>	CRArray.wrap(A array, int offset)

Methods in org.ddogleg.chainrule with parameters of type CRArray

Modifier and Type	Method	Description
void	JacobianSerialChain.backwardsF(int chainID, CRArray<T> input)	Same as JacobianSerialChain.backwards(int), except that it runs a forward pass to get the inputs then runs the backwards pass to compute Jacobians.
void	GradientNumerical_F64.differentiate(CRArray<double[]> dout, List<CRArray<double[]>> gradInput)
void	GradientNumerical.differentiate(CRArray<T> dout, List<CRArray<T>> gradInput)	Performs numerical differentiation to compute the gradients of input and parameters.
void	CROperatorForward.forward(CRArray<T> output)	Performs forward pass of the function.
void	JacobianSerialChain.Block.forward(CRArray<T> output)	Calls the forward function and handles bookkeeping
void	JacobianSerialChain.forward(int chainID, CRArray<T> input, @Nullable CRArray<T> output)	Performs the chain in the forward directly while saving all the intermediate inputs in JacobianSerialChain.savedInputs array
void	CROperatorBackwardsGrad.gradient(CRArray<T> dout, List<CRArray<T>> gradientInputs)	Computes the derivatives of all the inputs and parameters to this function .
JacobianChainGraph<T>.Forward	JacobianChainGraph.Forward.input(int chainID, CRArray<T> array)	Provides an input for the specified chain.
boolean	CRArray.isEquals(CRArray<?> compare, int length, double tol)	Checks to see if the two arrays are equals to each other to within the specified tolerance and over the specified number of elements
default void	CROperatorForward.normalizeAfterStep(int which, CRArray<T> params)	This function is called after an iteration and will normalize an input's values.
void	JacobianChainGraph.Forward.run(@Nullable CRArray<T> output)	Runs every block in the order worked out by JacobianChainGraph.GraphConstructor.fixate().
void	CROperatorForward.setInput(int which, CRArray<T> input)	Specifies the value of an input's vector.
void	MathCRMatrix.zero(CRArray<T> m, int length)	Fills all elements in the array with zero
void	MathCRMatrixD.zero(CRArray<double[]> m, int length)

Method parameters in org.ddogleg.chainrule with type arguments of type CRArray

Modifier and Type	Method	Description
void	GradientNumerical_F64.differentiate(CRArray<double[]> dout, List<CRArray<double[]>> gradInput)
void	GradientNumerical.differentiate(CRArray<T> dout, List<CRArray<T>> gradInput)	Performs numerical differentiation to compute the gradients of input and parameters.
void	CROperatorBackwardsGrad.gradient(CRArray<T> dout, List<CRArray<T>> gradientInputs)	Computes the derivatives of all the inputs and parameters to this function .
void	GradientNumerical_F64.setFunction(List<CRArray<double[]>> inputs, CROperatorForward<double[]> function)
void	GradientNumerical.setFunction(List<CRArray<T>> inputs, CROperatorForward<T> function)	Sets the function which will be differentiated and other parameters.
void	JacobianNumerical_F64.setFunction(List<CRArray<double[]>> inputs, CROperatorForward<double[]> function)
void	JacobianNumerical.setFunction(List<CRArray<T>> inputs, CROperatorForward<T> function)	Specifies the function which will be differentiated.

Uses of CRArray in org.ddogleg.chainrule.ops

Fields in org.ddogleg.chainrule.ops declared as CRArray

Modifier and Type	Field	Description
protected final CRArray<double[]>[]	CROperatorBilinearND.knots	Where each knot's polynomials live.

Methods in org.ddogleg.chainrule.ops with parameters of type CRArray

Modifier and Type	Method	Description
void	CRErrorHuberNorm.forward(CRArray<double[]> output)
void	CRErrorVectorSW.forward(CRArray<double[]> output)
void	CRLossHuber.forward(CRArray<double[]> output)
void	CROperatorBilinearND.forward(CRArray<double[]> output)
void	CROperatorClampLower.forward(CRArray<double[]> output)
void	CROperatorConcat.forward(CRArray<double[]> output)
void	CROperatorDivide.forward(CRArray<double[]> output)
void	CROperatorDivideVS.forward(CRArray<double[]> output)
void	CROperatorDivideVSDamped.forward(CRArray<double[]> output)
void	CROperatorGridBilinearWeights.forward(CRArray<double[]> output)
void	CROperatorMeanVector.forward(CRArray<double[]> output)
void	CROperatorMult.forward(CRArray<double[]> output)
void	CROperatorMultVS.forward(CRArray<double[]> output)
void	CROperatorPlus.forward(CRArray<double[]> output)
void	CROperatorPlusAB.forward(CRArray<double[]> output)
void	CROperatorRemapParams.forward(CRArray<double[]> output)
void	CRPenaltyL2.forward(CRArray<double[]> output)
void	CRErrorHuberNorm.setInput(int which, CRArray<double[]> input)
void	CRErrorVectorSW.setInput(int which, CRArray<double[]> input)
void	CRLossHuber.setInput(int which, CRArray<double[]> input)
void	CROperatorBilinearND.setInput(int which, CRArray<double[]> input)
void	CROperatorClampLower.setInput(int which, CRArray<double[]> src)
void	CROperatorConcat.setInput(int which, CRArray<double[]> input)
void	CROperatorDivide.setInput(int which, CRArray<double[]> input)
void	CROperatorDivideVS.setInput(int which, CRArray<double[]> input)
void	CROperatorDivideVSDamped.setInput(int which, CRArray<double[]> input)
void	CROperatorGridBilinearWeights.setInput(int which, CRArray<double[]> input)
void	CROperatorMeanVector.setInput(int which, CRArray<double[]> input)
void	CROperatorMult.setInput(int which, CRArray<double[]> input)
void	CROperatorMultVS.setInput(int which, CRArray<double[]> input)
void	CROperatorPlus.setInput(int which, CRArray<double[]> input)
void	CROperatorPlusAB.setInput(int which, CRArray<double[]> input)
void	CROperatorRemapParams.setInput(int which, CRArray<double[]> input)
void	CRPenaltyL2.setInput(int which, CRArray<double[]> input)

Uses of CRArray in org.ddogleg.chainrule.sym

Methods in org.ddogleg.chainrule.sym that return CRArray

Modifier and Type	Method	Description
CRArray<double[]>	SymFunction.feed(Sym terminal)	Binds the terminal to a freshly allocated array of its declared size.
CRArray<double[]>	SymFunction.feed(Sym terminal, double[] data)	Binds the terminal to the caller's own storage.
CRArray<double[]>	SymFunction.terminal(Sym node)	The handle backing a terminal, creating it on first use.

Methods in org.ddogleg.chainrule.sym with parameters of type CRArray

Modifier and Type	Method	Description
void	SymFunction.run(CRArray<double[]> out)	Runs the forward pass with the current terminal values and writes the residual rows at the output's offset