From 31852491d42ceee5e4a7f0ae63433c46ab1e6ebb Mon Sep 17 00:00:00 2001 From: stevenj Date: Thu, 28 Aug 2008 23:27:25 -0400 Subject: [PATCH] clarification darcs-hash:20080829032725-c8de0-5cdef0ccef6ecb04d94ea3db377878fa61d4b17b.gz --- api/nlopt_minimize.3 | 11 ++++++----- api/nlopt_minimize_constrained.3 | 28 +++++++++++++++++----------- 2 files changed, 23 insertions(+), 16 deletions(-) diff --git a/api/nlopt_minimize.3 b/api/nlopt_minimize.3 index 0111933..7dbb8a3 100644 --- a/api/nlopt_minimize.3 +++ b/api/nlopt_minimize.3 @@ -253,13 +253,14 @@ works (both for simple bound constraints via .I lb and .I ub -as well as nonlinear constraints as described above). +as well as nonlinear constraints via the crude technique of returning ++Inf when the constraints are violated, as explained above). .TP .B NLOPT_LN_PRAXIS Local (L) derivative-free (N) optimization using the principal-axis method, based on code by Richard Brent. Designed for unconstrained -optimization, although bound constraints are supported too (via a -potentially inefficient method). +optimization, although bound constraints are supported too (via the +inefficient method of returning +Inf when the constraints are violated). .TP .B NLOPT_LD_LBFGS Local (L) gradient-based (D) optimization using the limited-memory BFGS @@ -291,7 +292,7 @@ other variants of this algorithm: (same without restarting or preconditioning). .TP .B NLOPT_GN_CRS2_LM -Global (G) derivative-free (N) optimization using controlled random +Global (G) derivative-free (N) optimization using the controlled random search (CRS2) algorithm of Price, with the "local mutation" (LM) modification suggested by Kaelo and Ali. .TP @@ -314,7 +315,7 @@ pseudo-random numbers (instead of an LDS) as in the original MLSL algorithm. .B NLOPT_LD_MMA Local (L) gradient-based (D) optimization using the method of moving asymptotes (MMA), or rather a refined version of the algorithm as -published by Svanberg (2002). (NLopt uses an independent free +published by Svanberg (2002). (NLopt uses an independent free-software/open-source implementation of Svanberg's algorithm.) The .B NLOPT_LD_MMA algorithm supports both bound-constrained and unconstrained optimization, diff --git a/api/nlopt_minimize_constrained.3 b/api/nlopt_minimize_constrained.3 index e26455b..b2b2158 100644 --- a/api/nlopt_minimize_constrained.3 +++ b/api/nlopt_minimize_constrained.3 @@ -78,11 +78,16 @@ require the gradient (derivatives) of the function to be supplied via .IR f , and other algorithms do not require derivatives. Some of the algorithms attempt to find a global minimum within the given bounds, -and others find only a local minimum. Some of the algorithms can handle -nonlinear constraints, but most of the algorithms only handle the +and others find only a local minimum. Most of the algorithms only handle the case where .I m -is zero (no explicit nonlinear constraints). +is zero (no explicit nonlinear constraints); the only algorithms that +currently support positive +.I m +are +.B NLOPT_LD_MMA +and +.BR NLOPT_LN_COBYLA . .PP The .B nlopt_minimize_constrained @@ -244,11 +249,11 @@ and is used to distinguish between different constraints. .sp In particular, the constraint function .I fc -will be called +will be called (at most) .I m times for each .IR x , -and the i-th constraint (0 <= i <= +and the i-th constraint (0 <= i < .IR m ) will be passed an .I fc_datum @@ -269,7 +274,7 @@ parameter. Then, your .I fc function would be called .I m -times for each point, and be passed data[0] through data[m-1] in sequence. +times for each point, and be passed &data[0] through &data[m-1] in sequence. .SH ALGORITHMS The .I algorithm @@ -350,13 +355,14 @@ works (both for simple bound constraints via .I lb and .I ub -as well as nonlinear constraints as described above). +as well as nonlinear constraints via the crude technique of returning ++Inf when the constraints are violated, as explained above). .TP .B NLOPT_LN_PRAXIS Local (L) derivative-free (N) optimization using the principal-axis method, based on code by Richard Brent. Designed for unconstrained -optimization, although bound constraints are supported too (via a -potentially inefficient method). +optimization, although bound constraints are supported too (via the +inefficient method of returning +Inf when the constraints are violated). .TP .B NLOPT_LD_LBFGS Local (L) gradient-based (D) optimization using the limited-memory BFGS @@ -388,7 +394,7 @@ other variants of this algorithm: (same without restarting or preconditioning). .TP .B NLOPT_GN_CRS2_LM -Global (G) derivative-free (N) optimization using controlled random +Global (G) derivative-free (N) optimization using the controlled random search (CRS2) algorithm of Price, with the "local mutation" (LM) modification suggested by Kaelo and Ali. .TP @@ -411,7 +417,7 @@ pseudo-random numbers (instead of an LDS) as in the original MLSL algorithm. .B NLOPT_LD_MMA Local (L) gradient-based (D) optimization using the method of moving asymptotes (MMA), or rather a refined version of the algorithm as -published by Svanberg (2002). (NLopt uses an independent free +published by Svanberg (2002). (NLopt uses an independent free-software/open-source implementation of Svanberg's algorithm.) The .B NLOPT_LD_MMA algorithm supports both bound-constrained and unconstrained optimization, -- 2.30.2