[SciPy-user] L-BFGS in scipy

Nils Wagner nwagner at iam.uni-stuttgart.de
Wed Sep 13 07:47:28 CDT 2006

Robert Kern wrote:
> Nils Wagner wrote:
>> Hi all,
>> Has someone implemented the limited memory BFGS method in scipy ?
> Yes. scipy.optimize.fmin_l_bfgs_b(). Please grep for these things.
Thank you Robert.
If bounds=None we have an unconstraint version.
Thus fmin_l_bfgs_b is also an unconstrained optimizer. I missed that.
Maybe fmin_l_bfgs_b should also be added to the list of general-purpose
optimization routines

help (optimize) yields

     A collection of general-purpose optimization routines.

       fmin        --  Nelder-Mead Simplex algorithm
                         (uses only function calls)
       fmin_powell --  Powell's (modified) level set method (uses only
                         function calls)
       fmin_cg     --  Non-linear (Polak-Ribiere) conjugate gradient
                         (can use function and gradient).
       fmin_bfgs   --  Quasi-Newton method
                         (can use function and gradient)
       fmin_ncg    --  Line-search Newton Conjugate Gradient (can use
                         function, gradient and Hessian).
       leastsq     --  Minimize the sum of squares of M equations in
                         N unknowns given a starting estimate.

Constrained Optimizers (multivariate)

       fmin_l_bfgs_b -- Zhu, Byrd, and Nocedal's L-BFGS-B constrained
                          (if you use this please quote their papers --
see help)

and I disregard fmin_l_bfgs_b because it is given in the section
Constrained Optimizers.

Sorry for the noise.



More information about the SciPy-user mailing list