Crate liblbfgs_sys
source · [−]Structs
L-BFGS optimization parameters. Call lbfgs_parameter_init() function to initialize parameters to the default values.
Constants
The minimization process has been canceled.
A logic error occurred; alternatively, the interval of uncertainty became too small.
The current search direction increases the objective function value.
A logic error (negative line-search step) occurred.
Invalid parameter lbfgs_parameter_t::delta specified.
Invalid parameter lbfgs_parameter_t::epsilon specified.
Invalid parameter lbfgs_parameter_t::ftol specified.
Invalid parameter lbfgs_parameter_t::gtol specified.
Invalid parameter lbfgs_parameter_t::linesearch specified.
Invalid parameter lbfgs_parameter_t::max_linesearch specified.
Invalid parameter lbfgs_parameter_t::max_step specified.
Invalid parameter lbfgs_parameter_t::max_step specified.
Invalid number of variables specified.
Invalid number of variables (for SSE) specified.
Invalid parameter lbfgs_parameter_t::orthantwise_c specified.
Invalid parameter lbfgs_parameter_t::orthantwise_end specified.
Invalid parameter lbfgs_parameter_t::orthantwise_start specified.
Invalid parameter lbfgs_parameter_t::past specified.
Invalid parameter lbfgs_parameter_t::wolfe specified.
Invalid parameter lbfgs_parameter_t::xtol specified.
The array x must be aligned to 16 (for SSE).
Logic error.
The algorithm routine reaches the maximum number of iterations.
The line-search routine reaches the maximum number of evaluations.
The line-search step became larger than lbfgs_parameter_t::max_step.
The line-search step became smaller than lbfgs_parameter_t::min_step.
The line-search step went out of the interval of uncertainty.
Insufficient memory.
A rounding error occurred; alternatively, no line-search step satisfies the sufficient decrease and curvature conditions.
Unknown error.
Relative width of the interval of uncertainty is at most lbfgs_parameter_t::xtol.
The initial variables already minimize the objective function.
L-BFGS reaches convergence.
The backtracking method with the defualt (regular Wolfe) condition.
Backtracking method with the Armijo condition. The backtracking method finds the step length such that it satisfies the sufficient decrease (Armijo) condition,
Backtracking method with strong Wolfe condition. The backtracking method finds the step length such that it satisfies both the Armijo condition (LBFGS_LINESEARCH_BACKTRACKING_ARMIJO) and the following condition,
Backtracking method with regular Wolfe condition. The backtracking method finds the step length such that it satisfies both the Armijo condition (LBFGS_LINESEARCH_BACKTRACKING_ARMIJO) and the curvature condition,
The default algorithm (MoreThuente method).
MoreThuente method proposd by More and Thuente.
L-BFGS reaches convergence.
L-BFGS reaches convergence.
Functions
Free an array of variables.
Allocate an array for variables.
Initialize L-BFGS parameters to the default values.
Get string description of an lbfgs() return code.
Type Definitions
Return values of lbfgs().
Line search algorithms.
Callback interface to provide objective function and gradient evaluations.
Callback interface to receive the progress of the optimization process.