[SciPy-User] [SciPy-user] Error using fmin

josef.pktd@gmai... josef.pktd@gmai...
Wed Feb 22 07:53:26 CST 2012


On Wed, Feb 22, 2012 at 8:41 AM, RDX <rishi.ou@gmail.com> wrote:
>
> I am using fmin and it returns initial guess as the final solution.
>
> def Traug (v):
>    err=0
>    Traug_RHOB = v[0] + v[1]*(z_den_s/3125.0)**v[2]
>    count =arange(size(den_s))
>    for i in count:
>        err += abs(den_s[i] - Traug_RHOB [i])
>    return err

my first guess is that fmin doesn't like the abs, you could try with a
quadratic loss instead to see whether it works and whether there are
other problems

If  den_s[i] is an array, then you could vectorize the loop

Josef

>
> vO=[1.6, 0.5, 0.35]
> t=fmin(Traug,vO )
> a=t[0]; b=t[1]; c=t[2];
>
> Code gives me a=1.6, b=0.5, c=0.35, which is my initial guess.
>
> What is wrong? Many thanks in advance.
>
> Regards,
> Rishi
>
>
> --
> View this message in context: http://old.nabble.com/Error-using-fmin-tp33370870p33370870.html
> Sent from the Scipy-User mailing list archive at Nabble.com.
>
> _______________________________________________
> SciPy-User mailing list
> SciPy-User@scipy.org
> http://mail.scipy.org/mailman/listinfo/scipy-user


More information about the SciPy-User mailing list