[SciPy-user] fsolve side affecting output

[Gerard Vermeulen]

On Mon, 26 Jul 2004 16:55:38 -0400
matt <matt at hotdispatch.com> wrote:

> optimize.fsolve returns the inputs that result in a zero for the 
> function it is passed.   It doesn't, however, throw an exception or 
> give a special return value when it dumps one of these warnings to 
> standard out.
> 
> Is there a way to supress the dumps to std out?  I still want my print 
> commands to be visible, so I can't just redirect everything that goes 
> to std out.
> 
> Does anyone know if there are plans to add an exception or success code 
> to the output of this function?
> 
> -Matt

If you do:

>>> from scipy import *
>>> help(optimize.fsolve)

you'll see that you can set full_output = 1 to get all information about
success or failure of the algorithm, which I use in Python programs. As
a side effect it suppresses the dumps to standard out:

>>> def g(x): return x*x+1
...
>>> optimize.fsolve(g, 10.0, full_output=1)
(-7.5682952661081625e-05, {'qtf': array([-1.00000001]), 'nfev': 34, 'fjac': array([       [-1.]]), 'r': array([-0.00108154]), 'fvec': array([ 1.00000001])}, 5, 'The iteration is not making good progress, as measured by the \n  improvement from the last ten iterations.')
>>>

The dumps to standard out are perfectly fine to me when I am using the
interpreter.

Gerard