Package minimise :: Module generic_line_search :: Class generic_line_search
[hide private]
[frames] | no frames]

Class generic_line_search

source code


Instance Methods [hide private]
 
__init__(self)
Class containing non-specific line search algorithm code.
source code
 
new_param_func(self)
Pure Newton minimisation.
source code
Method Details [hide private]

new_param_func(self)

source code 
Pure Newton minimisation.

Function options
~~~~~~~~~~~~~~~~

func                    - The function to minimise.
dfunc                   - The function which returns the gradient vector.
d2func                  - The function which returns the hessian matrix.
x0                      - The initial parameter vector.
args                    - The tuple of arguments to supply to the functions func, dfunc, and d2func.
tol                     - The cutoff value used to terminate minimisation by comparison to the difference in function values between iterations.
maxiter                 - The maximum number of iterations.
full_output             - A flag specifying what should be returned (see below).
print_flag              - A flag specifying how much information should be printed to standard output during minimisation:

The print flag corresponds to:
        0 - No output.
        1 - Minimal output.
        2 - Full output.


Returned objects
~~~~~~~~~~~~~~~~

If full_output=0, then only the minimised parameter vector is returned.
If full_output=1, then the minimised parameter vector, function value at the minimum, number of iterations, and a warning flag are returned.
The warning flag corresponds to:
        0 - Minimisation terminated successfully.
        1 - Maximum number of iterations have been reached.


Internal variables
~~~~~~~~~~~~~~~~~~

k       - The iteration number.
xk      - Parameter vector at iteration number k.
fk      - Function value at xk.
fk_last - Function value at xk-1.
dfk     - Gradient vector at xk.
d2fk    - Hessian matrix at xk.
pk      - Descent direction of the iteration number k.