On 1/11/2012 9:25 PM, marco atzeri wrote:
> On 1/11/2012 8:21 PM, Olaf Till wrote:
>
>>
>> In installation of optim, the following lines of PKG_ADD:
>>
>> __all_opts__ ("nonlin_residmin");
>> __all_opts__ ("residmin_stat");
>>
>> are automatically executed to register the names of optimset-options
>> of nonlin_residmin and residmin_stat with Octave. __all_opts__ will
>> call optimset with the respective function names as argument, which
>> in turn calls the functions with the calls
>>
>> nonlin_residmin ("defaults");
>> residmin_stat ("defaults");
>>
>> to retrieve their arguments. A failure of this call is caught with
>> try-catch and produces the error message which you have seen as a
>> warning, since __all_opts__ converts it to a warning with try-catch.
>> That is why I said the above calls must be failing on your system.
>> Have you tried it?
>>
>> Olaf
>>
> I was not aware of such mechanism.
> Here the results:
>
> octave:4>nonlin_residmin ("defaults")
> ans =
>
> scalar structure containing the fields:
>
> param_config = [](0x0)
> param_order = [](0x0)
> param_dims = [](0x0)
> f_inequc_pstruct = 0
> f_equc_pstruct = 0
> f_pstruct = 0
> df_inequc_pstruct = 0
> df_equc_pstruct = 0
> dfdp_pstruct = 0
> lbound = [](0x0)
> ubound = [](0x0)
> dfdp = [](0x0)
> cpiv = @cpiv_bard
> max_fract_change = [](0x0)
> fract_prec = [](0x0)
> diffp = [](0x0)
> diff_onesided = [](0x0)
> complex_step_derivative = 0
> complex_step_derivative_inequc = 0
> complex_step_derivative_equc = 0
> cstep = 1.0000e-20
> fixed = [](0x0)
> inequc = [](0x0)
> equc = [](0x0)
> weights = [](0x0)
> TolFun = 1.0000e-04
> MaxIter = [](0x0)
> Display = off
> Algorithm = lm_svd_feasible
> plot_cmd =
>
> @(f) 0
>
> debug = 0
> lm_svd_feasible_alt_s = 0
>
> octave:5> residmin_stat ("defaults")
> ans =
>
> scalar structure containing the fields:
>
> param_config = [](0x0)
> param_order = [](0x0)
> param_dims = [](0x0)
> f_pstruct = 0
> dfdp_pstruct = 0
> dfdp = [](0x0)
> diffp = [](0x0)
> diff_onesided = [](0x0)
> complex_step_derivative = 0
> cstep = 1.0000e-20
> fixed = [](0x0)
> weights = [](0x0)
> residuals = [](0x0)
> covd = [](0x0)
> objf = [](0x0)
> ret_dfdp = 0
> ret_covd = 0
> ret_covp = 0
> ret_corp = 0
>
> octave:7> __all_opts__ ("nonlin_residmin")
> ans =
> {
> [1,1] = Algorithm
> [1,2] = AutoScaling
> [1,3] = complex_step_derivative
> [1,4] = complex_step_derivative_equc
> [1,5] = complex_step_derivative_inequc
> [1,6] = ComplexEqn
> [1,7] = covd
> [1,8] = cpiv
> [1,9] = cstep
> [1,10] = debug
> [1,11] = df_equc_pstruct
> [1,12] = df_inequc_pstruct
> [1,13] = dfdp
> [1,14] = dfdp_pstruct
> [1,15] = diff_onesided
> [1,16] = diffp
> [1,17] = Display
> [1,18] = equc
> [1,19] = f_equc_pstruct
> [1,20] = f_inequc_pstruct
> [1,21] = f_pstruct
> [1,22] = FinDiffType
> [1,23] = fixed
> [1,24] = fract_prec
> [1,25] = FunValCheck
> [1,26] = GradObj
> [1,27] = inequc
> [1,28] = Jacobian
> [1,29] = lbound
> [1,30] = lm_svd_feasible_alt_s
> [1,31] = max_fract_change
> [1,32] = MaxFunEvals
> [1,33] = MaxIter
> [1,34] = objf
> [1,35] = OutputFcn
> [1,36] = param_config
> [1,37] = param_dims
> [1,38] = param_order
> [1,39] = plot_cmd
> [1,40] = residuals
> [1,41] = ret_corp
> [1,42] = ret_covd
> [1,43] = ret_covp
> [1,44] = ret_dfdp
> [1,45] = SearchDirections
> [1,46] = TolFun
> [1,47] = TolX
> [1,48] = TypicalX
> [1,49] = ubound
> [1,50] = Updating
> [1,51] = weights
> }
>
> octave:9> __all_opts__ ("residmin_stat")
> ans =
> {
> [1,1] = Algorithm
> [1,2] = AutoScaling
> [1,3] = complex_step_derivative
> [1,4] = complex_step_derivative_equc
> [1,5] = complex_step_derivative_inequc
> [1,6] = ComplexEqn
> [1,7] = covd
> [1,8] = cpiv
> [1,9] = cstep
> [1,10] = debug
> [1,11] = df_equc_pstruct
> [1,12] = df_inequc_pstruct
> [1,13] = dfdp
> [1,14] = dfdp_pstruct
> [1,15] = diff_onesided
> [1,16] = diffp
> [1,17] = Display
> [1,18] = equc
> [1,19] = f_equc_pstruct
> [1,20] = f_inequc_pstruct
> [1,21] = f_pstruct
> [1,22] = FinDiffType
> [1,23] = fixed
> [1,24] = fract_prec
> [1,25] = FunValCheck
> [1,26] = GradObj
> [1,27] = inequc
> [1,28] = Jacobian
> [1,29] = lbound
> [1,30] = lm_svd_feasible_alt_s
> [1,31] = max_fract_change
> [1,32] = MaxFunEvals
> [1,33] = MaxIter
> [1,34] = objf
> [1,35] = OutputFcn
> [1,36] = param_config
> [1,37] = param_dims
> [1,38] = param_order
> [1,39] = plot_cmd
> [1,40] = residuals
> [1,41] = ret_corp
> [1,42] = ret_covd
> [1,43] = ret_covp
> [1,44] = ret_dfdp
> [1,45] = SearchDirections
> [1,46] = TolFun
> [1,47] = TolX
> [1,48] = TypicalX
> [1,49] = ubound
> [1,50] = Updating
> [1,51] = weights
> }
>
>
> they seems to work
>
> Regards
> Marco

I presume there is a precedence/timing issue,
if the package is not loaded, I have the warning

octave:4> __all_opts__ ("residmin_stat");
octave:5> pkg unload optim
octave:6> __all_opts__ ("residmin_stat");
warning: optimset: no defaults for function `residmin_stat'

so I can ignore the warning when installing

Regards
Marco

------------------------------------------------------------------------------
Ridiculously easy VDI. With Citrix VDI-in-a-Box, you don't need a complex
infrastructure or vast IT resources to deliver seamless, secure access to
virtual desktops. With this all-in-one solution, easily deploy virtual 
desktops for less than the cost of PCs and save 60% on VDI infrastructure 
costs. Try it free! http://p.sf.net/sfu/Citrix-VDIinabox
_______________________________________________
Octave-dev mailing list
Octave-dev@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/octave-dev

Reply via email to