Scipy optimize.fmin wrong return values

Question

I am currently using scipy.optimize.fmin() function and I am experiencing a problem with it. When I look at the documentation it says:

Returns:

    xopt : ndarray
        Parameter that minimizes function.

    fopt : float
        Value of function at minimum: fopt = func(xopt).

    iter : int
        Number of iterations performed.

    funcalls : int
        Number of function calls made.

    warnflag : int
        1 : Maximum number of function evaluations made. 2 : Maximum number of iterations reached.

    allvecs : list
        Solution at each iteration.

But when I try this:

res, min = opt.fmin(optim, self._params, (param_optim, self._paramsIni, Qmes, critere_efficacite, self, codeBV, interval), maxiter=5)

I get this error:

ValueError: too many values to unpack (expected 2)

Anyone has an idea why? I mean is the documention wrong (I guess not) or am I doing something wrong? I am using scipy 0.19 and Python34

Thanks in advance.

The documentation you stated shows you the function returns six values, yet you assign it to two variables. — user707650
– user707650, Commented May 3, 2017 at 8:24
I tried res, min, a, b, c, d = opt.fmin(optim, self._params, (param_optim, self._paramsIni, Qmes, critere_efficacite, self, codeBV, interval), maxiter=5), it does not work either. — Silveris
– Silveris, Commented May 3, 2017 at 8:27
The only one that is working is res = opt.fmin(optim, self._params, (param_optim, self._paramsIni, Qmes, critere_efficacite, self, codeBV, interval), maxiter=5) but I really need this second returned value. — Silveris
– Silveris, Commented May 3, 2017 at 8:28
Unfortunately, fmin is one of those scipy functions that changes its output depending on its output. The relevant input parameter from the documentation: "full_output : bool, optional. Set to True if fopt and warnflag outputs are desired." So fmin outputs either 1 value, or 6 values, depending on the full_output input flag. — user707650
– user707650, Commented May 3, 2017 at 8:30
Thank you Evert! It solved my problem. I was wondering what was this full_output flag for, as it was not displaying anything. — Silveris
– Silveris, Commented May 3, 2017 at 8:33

Chris · Accepted Answer · 2017-05-03 08:34:53Z

3

To be slightly more exact: The function returns either a tuple of 6 values (

full_output : bool, optional Set to True if fopt and warnflag outputs are desired.

) or one (if it is left at False which is the default). If you want to have only the second value of the full output, I recommend you set full_output=True and pattern-match as suggested in the comments. Alternativly you can store the result in one tuple res = opt.fmin(<your arguments>) and then access r=res[0] min=res[1].

answered May 3, 2017 at 8:34

Chris

7107 silver badges15 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Scipy optimize.fmin wrong return values

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related