[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Octave-bug-tracker] [bug #40101] fminunc outputs - gradient and hessian

From: jose
Subject: [Octave-bug-tracker] [bug #40101] fminunc outputs - gradient and hessian not always calculated in the solutio
Date: Mon, 03 Feb 2014 16:18:14 +0000
User-agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Iron/29.0.1600.1 Chrome/29.0.1600.1 Safari/537.36

Follow-up Comment #1, bug #40101 (project octave):

Here is a Mercurial changeset (fminunc.diff) equivalent to the proposed
fminunc2.m with an additional modification that fixes a new problem introduced
in fminunc2.m itself. This new problem affected the cases in which the
gradient function was supplied by the user as a row vector.

An easier way to reproduce the original problem (compared to session.txt) is
as follows:
Before applying the changeset

octave:1> l=@(x) exp(sum(x.^2));
octave:2> [x,fval,info,output,grad,hess]=fminunc(l,[1 1]);
octave:3> info
info =  3
octave:4> grad
grad =
octave:5> __fdjac__(l,x,l(x),[1 1],1)
ans =
  -2.4385e-09   2.4568e-09
octave:6> hess
hess =
   1.49999   0.50000
   0.50000   1.50001

After applying the changeset 

octave:7> [x,fval,info,output,grad,hess]=fminunc(l,[1 1]);
octave:8> info
info =  3
octave:9> grad
grad =
  -2.4385e-09   2.4568e-09
octave:10> hess
hess =
   1.50920   0.49988
   0.49988   1.49081

(file #30447)

Additional Item Attachment:

File name: fminunc.diff                   Size:1 KB


Reply to this item at:


  Message sent via/by Savannah

reply via email to

[Prev in Thread] Current Thread [Next in Thread]