powell.control {powell} | R Documentation |
Controls convergence in powell
.
powell.control(trace = 0, rhobeg = log(100)/10, rhoend = 1e-04, maxit = 1e+05, fnscale = 1, parscale = 1, ...)
trace |
an integer that determines the level of information
produced at each step of the optimizer. The value of trace
should be set to 0, 1, 2 or 3, which controls the amount of
printing. Specifically, there is no output if trace=0 and
there is output only at the return if trace=1 . Otherwise,
each new value of RHO is printed, with the best vector of variables
so far and the corresponding value of the objective
function. Further, each new value of F with its variables are output
if trace=3 . |
rhobeg |
About one tenth of the greatest expected change to a variable. |
rhoend |
Indicates the accuracy that is required in the final values of the variables. |
maxit |
The upper bound on the number of times fn is to
be called. |
fnscale |
A scalar to divide the function value
by. fnscale less than zero implies a maximization rather
than a minimization. |
parscale |
A vector to divide the parameter values by. If only
one value is given it is replicated to the length of par . |
... |
Additional arguments that will be ignored! |
rhobeg
and rhoend
must be set to the initial and
final values of a trust region radius, so both must be positive with
rhoend<=rhobeg
.
Sundar Dorai-Raj