Package edu.mines.jtk.opt

Code for Gauss-Newton and conjugate-gradient inversion and optimization.

See:
Description

Interface Summary

BrentMinFinder.Function	A function f(x) of one variable x.
BrentZeroFinder.Function	A function f(x) of one variable x.
LinearTransform	Define methods applying a linear transform and its transpose
LineSearch.Function	The function to be minimized.
Quadratic	Define a second-order quadratic operation on a Vector 0.5 x'Hx + b'x where H is a positive semidefinite quadratic and b is a linear gradient.
ScalarSolver.Function	Implement a function of one variable to be minimized
Transform	Implement a non-linear transform and its linearizations for a non-linear optimization.
Vect	Implement a vector supporting linear vector-space methods Test your implementation with VectUtil.test().
VectConst	Vector operations that do not change the state of the vector
VectContainer	Treat an indexed collection of Vects as a new Vect.

Class Summary

ArrayVect1	Implements a Vect by wrapping an array of doubles.
ArrayVect1f	Implements a Vect by wrapping an array of floats.
ArrayVect1fs	Implements a Vect as a collection of ArrayVect1f's of fixed size.
ArrayVect2	Implement a Vect as a two dimensional array of doubles.
ArrayVect2f	Implement a Vect as a two dimensional array of floats.
ArrayVect3f	Implement a Vect as a three-dimensional array of floats.
BrentMinFinder	Brent's algorithm for finding the minimum of a function of one variable.
BrentZeroFinder	Brent's algorithm for finding a zero of a function of one variable.
CoordinateTransform	Find a best linear combination of input coordinates to fit output coordinates.
GaussNewtonSolver	Solve least-squares inverse of a non-linear Transform.
LinearTransformWrapper	Wrap a LinearTransform as a non-linear Transform, by ignoring reference model.
LineSearch	Searches along a line for a minimum of a continuously differentiable function of one or more variables.
LineSearch.Result	The result of a line search.
QuadraticSolver	Minimize a simple quadratic objective function.
ScalarSolver	Search a single variable for a value that minimizes a function
ScalarVect	Implements a Vect by wrapping a single double
TransformQuadratic	For a linearized transform, implement the Gauss-Newton quadratic approximation of a damped least-squares objective function.
VectArray	A VectContainer implemented as an array, with sequential indices.
VectMap	A VectContainer implemented as an unsynchronized Map.
VectUtil	Implements convenience methods for Vect.

Package edu.mines.jtk.opt Description

Code for Gauss-Newton and conjugate-gradient inversion and optimization. See documents in doc/opt_package/ for a description of the design and API.