Hi, Would you be able to submit a bug report about this? The link is https://gna.org/bugs/?func=additem&group=relax. It would be great if you could come up with a script with a tiny, absolutely minimal data set which triggers the bug, and attach these as well. There is definitely a problem, but I am unable to do much with the current info. Actually, one thing to try would be to change "inc=[16, 16, 16]" to "inc=[1, 16, 16]".
Cheers, Edward On 11 April 2011 12:34, Tiago Pais <[email protected]> wrote: > Hi Edward, > I have been trying several combinations but the grid search does not appear > to be doing its job. > I have used the following command in the script: > "grid_search(lower=[0.111, 0, -0.0], upper=[0.111, 1.00, 200], inc=[16, 16, > 16], verbosity=7)" > > But I keep seeing this during the calculations: > "Min f: 3474.82 > > k: 3860 xk: [ 0.111, 0.066667, -0] fk: nan > > Increment: array([ 5, 16, 16]) > Params: array([ 0.111, 1. , Inf]) > Min params: array([ 0.111 , 0.06666667, -0. ]) > f: nan > > Min f: 3474.82 > > k: 3861 xk: [ 0.111, 0.066667, -0] fk: nan > > Increment: array([ 6, 16, 16]) > Params: array([ 0.111, 1. , Inf]) > Min params: array([ 0.111 , 0.06666667, -0. ]) > f: nan > " > > And at the end the tauS is zero for all spins and the S2s is always close to > 1. It appears that the space is not being adequately created... > > Any suggestion? > Thanks > Tiago P > > > -----Original Message----- > From: [email protected] [mailto:[email protected]] On > Behalf Of Edward d'Auvergne > Sent: quinta-feira, 7 de Abril de 2011 12:15 > To: Tiago Pais > Cc: [email protected] > Subject: Re: How to set S2f to 0.111? - SOLVED with 1.3.8 > > Ah, right. I did fix that bug! I thought so :) Ok, if you have any > problems, don't hesitate to ask. > > Bye, > > Edward > > > > On 7 April 2011 13:09, Tiago Pais <[email protected]> wrote: >> I have now also running the 1.3.8 version and the problem seems to have >> disappeared. Took me some time to use this version because the column >> numbering has changed, now columns start at number one! Important! ;-) >> >> I hope now this goes well. >> >> Thanks once more for your valuable help. >> Cheers >> Tiago >> >> -----Original Message----- >> From: [email protected] [mailto:[email protected]] On >> Behalf Of Edward d'Auvergne >> Sent: quinta-feira, 7 de Abril de 2011 10:46 >> To: Tiago Pais >> Cc: [email protected] >> Subject: Re: How to set S2f to 0.111? >> >> Hi, >> >> I've looked up that line in the specific_fns/model_free/mf_minimise.py >> file, and it looks like you are using an older version of relax. >> Could you copy the output of: >> >> $ relax --info >> >> This looks like an ancient bug that was eliminated a long, long time ago. >> >> Cheers, >> >> Edward >> >> >> >> >> On 6 April 2011 19:51, Tiago Pais <[email protected]> wrote: >>> I used the expression like this: >>> grid_search(lower=[0.111, 0, 0], upper=[0.111, 2, 1E12], inc=[1, 11, 11]) >>> >>> But I get the following error: >>> Unconstrained grid search size: 121 (constraints may decrease this size). >>> >>> Traceback (most recent call last): >>> File "/usr/local/bin/relax", line 418, in <module> >>> Relax() >>> File "/usr/local/bin/relax", line 127, in __init__ >>> self.interpreter.run(self.script_file) >>> File "/usr/local/relax/prompt/interpreter.py", line 276, in run >>> return run_script(intro=self.__intro_string, local=self.local, >>> script_file=script_file, quit=self.__quit_flag, >>> show_script=self.__show_script, >> raise_relax_error=self.__raise_relax_error) >>> File "/usr/local/relax/prompt/interpreter.py", line 537, in run_script >>> return console.interact(intro, local, script_file, quit, >>> show_script=show_script, raise_relax_error=raise_relax_error) >>> File "/usr/local/relax/prompt/interpreter.py", line 433, in >>> interact_script >>> execfile(script_file, local) >>> File "mf_multimodel_m5S2f0111.py", line 82, in <module> >>> grid_search(lower=[0.111, 0, 0], upper=[0.111, 1, 11], inc=[1, 11, > 11]) >>> File "/usr/local/relax/prompt/minimisation.py", line 156, in grid_search >>> minimise.grid_search(lower=lower, upper=upper, inc=inc, >>> constraints=constraints, verbosity=verbosity) >>> File "/usr/local/relax/generic_fns/minimise.py", line 191, in > grid_search >>> grid_search(lower=lower, upper=upper, inc=inc, > constraints=constraints, >>> verbosity=verbosity) >>> File "/usr/local/relax/specific_fns/model_free/mf_minimise.py", line > 479, >>> in grid_search >>> self.minimise(min_algor='grid', lower=lower, upper=upper, inc=inc, >>> constraints=constraints, verbosity=verbosity, sim_index=sim_index) >>> File "/usr/local/relax/specific_fns/model_free/mf_minimise.py", line > 922, >>> in minimise >>> min_options = self.grid_search_config(num_params, spin=spin, >>> lower=lower, upper=upper, inc=inc, scaling_matrix=scaling_matrix) >>> File "/usr/local/relax/specific_fns/model_free/mf_minimise.py", line > 554, >>> in grid_search_config >>> self.grid_search_config_fin(min_options, lower, upper, scaling_matrix) >>> File "/usr/local/relax/specific_fns/model_free/mf_minimise.py", line > 631, >>> in grid_search_config_fin >>> for i in xrange(n): >>> NameError: global name 'n' is not defined >>> >>> >>> Any ideas? >>> >>> Cheers >>> Tiago >>> >>> -----Original Message----- >>> From: [email protected] [mailto:[email protected]] On >>> Behalf Of Edward d'Auvergne >>> Sent: quarta-feira, 6 de Abril de 2011 18:05 >>> To: Tiago Pais >>> Cc: [email protected] >>> Subject: Re: How to set S2f to 0.111? >>> >>> Hi, >>> >>> These match the parameters as listed in the help for the >>> model_free.select_model() user function. The grid search arrays match >>> the parameter arrays. >>> >>> Regards, >>> >>> Edward >>> >>> >>> On 6 April 2011 18:54, Tiago Pais <[email protected]> wrote: >>>> Ok, >>>> Im sorry to insist but I am probably missing something here. How do I >>> state >>>> which bound goes with which parameter? That is, which position in the >>> array >>>> bound corresponds to the S2f parameter? >>>> Regards. >>>> TP >>>> >>>> -----Original Message----- >>>> From: [email protected] [mailto:[email protected]] On >>>> Behalf Of Edward d'Auvergne >>>> Sent: quarta-feira, 6 de Abril de 2011 17:42 >>>> To: Tiago Pais >>>> Cc: [email protected] >>>> Subject: Re: How to set S2f to 0.111? >>>> >>>> Hi, >>>> >>>> That is roughly correct. Just set the lower and upper bound to 0.111 >>>> and set the number of increments for that parameter to 1. Then it is >>>> like a grid search without S2f being present, or being fixed to 0.111. >>>> >>>> Regards, >>>> >>>> Edward >>>> >>>> >>>> >>>> On 6 April 2011 18:38, Tiago Pais <[email protected]> wrote: >>>>> Hi Ed, >>>>> Thanks for the quick reply. >>>>> Let's see if I understand your suggestion. For example, the >>>> mf_multimodel.py >>>>> sample script performs a grid search minimization using the grid_search >>>>> function. So if I insert an array of upper and lower bounds for the >>>>> parameters in the model I can constrain the interval of possible >> results, >>>>> correct? If so, what is the sequence for the parameters in the array? >>>>> TP >>>>> >>>>> >>>>> >>>>> -----Original Message----- >>>>> From: Edward d'Auvergne [mailto:[email protected]] >>>>> Sent: quarta-feira, 6 de Abril de 2011 17:13 >>>>> To: Tiago Pais >>>>> Cc: [email protected] >>>>> Subject: Re: How to set S2f to 0.111? >>>>> >>>>> Hi, >>>>> >>>>> Unfortunately, there is no ability to fix one of the model-free >>>>> parameters during optimisation in relax yet. This is something that >>>>> has been requested one or two times over the years. This is >>>>> complicated as the gradients and Hessians (first and second partial >>>>> derivates) need to be modified with this in mind, as the curvature of >>>>> the space changes. The fixed parameter is not part of the gradient or >>>>> Hessian, and has to be removed. This really complicates the target >>>>> function, gradient, and Hessian and, due to the low demand, has not >>>>> been implemented. >>>>> >>>>> Are you able to optimise this parameter and see if you get 0.111? And >>>>> are you able to extract the timescale of this process (the methyl >>>>> group rotamer jumps)? From memory, someone has optimsed these models >>>>> using a grid search. This coarse optimisation can be performed in >>>>> relax as you are able to specify the fine details of the grid search. >>>>> It is slower, but will do the job. You could even implement a zooming >>>>> grid search if your scripting skills are up to it to be able to >>>>> relatively quickly find the parameters to high precision. Anyway, I >>>>> hope this info helps. >>>>> >>>>> Regards, >>>>> >>>>> Edward >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> On 6 April 2011 17:39, Tiago Pais <[email protected]> wrote: >>>>>> Hi, >>>>>> >>>>>> I would like to make an analysis of side-chain relaxation data with > the >>>>> S2f >>>>>> parameter (in model 5) set to 0.111. >>>>>> >>>>>> Is it possible to do this with RELAX? How? >>>>>> >>>>>> Thanks in advance. >>>>>> >>>>>> Cheers >>>>>> >>>>>> Tiago P >>>>>> >>>>>> >>>>>> >>>>>> ************************************** >>>>>> >>>>>> Tiago Pais, PhD. student >>>>>> >>>>>> Cell Physiology & NMR >>>>>> >>>>>> Instituto de Tecnologia Química e Biológica-Oeiras >>>>>> >>>>>> >>>>>> >>>>>> __________ Information from ESET NOD32 Antivirus, version of virus >>>>> signature >>>>>> database 6019 (20110406) __________ >>>>>> >>>>>> The message was checked by ESET NOD32 Antivirus. >>>>>> >>>>>> http://www.eset.com >>>>>> >>>>>> _______________________________________________ >>>>>> relax (http://nmr-relax.com) >>>>>> >>>>>> This is the relax-users mailing list >>>>>> [email protected] >>>>>> >>>>>> To unsubscribe from this list, get a password >>>>>> reminder, or change your subscription options, >>>>>> visit the list information page at >>>>>> https://mail.gna.org/listinfo/relax-users >>>>>> >>>>>> >>>>> >>>>> __________ Information from ESET NOD32 Antivirus, version of virus >>>> signature >>>>> database 6019 (20110406) __________ >>>>> >>>>> The message was checked by ESET NOD32 Antivirus. >>>>> >>>>> http://www.eset.com >>>>> >>>>> >>>>> >>>>> >>>>> __________ Information from ESET NOD32 Antivirus, version of virus >>>> signature >>>>> database 6019 (20110406) __________ >>>>> >>>>> The message was checked by ESET NOD32 Antivirus. >>>>> >>>>> http://www.eset.com >>>>> >>>>> >>>>> >>>> >>>> __________ Information from ESET NOD32 Antivirus, version of virus >>> signature >>>> database 6019 (20110406) __________ >>>> >>>> The message was checked by ESET NOD32 Antivirus. >>>> >>>> http://www.eset.com >>>> >>>> >>>> >>>> >>>> __________ Information from ESET NOD32 Antivirus, version of virus >>> signature >>>> database 6019 (20110406) __________ >>>> >>>> The message was checked by ESET NOD32 Antivirus. >>>> >>>> http://www.eset.com >>>> >>>> >>>> >>> >>> __________ Information from ESET NOD32 Antivirus, version of virus >> signature >>> database 6019 (20110406) __________ >>> >>> The message was checked by ESET NOD32 Antivirus. >>> >>> http://www.eset.com >>> >>> >>> >>> >>> __________ Information from ESET NOD32 Antivirus, version of virus >> signature >>> database 6019 (20110406) __________ >>> >>> The message was checked by ESET NOD32 Antivirus. >>> >>> http://www.eset.com >>> >>> >>> >> >> __________ Information from ESET NOD32 Antivirus, version of virus > signature >> database 6021 (20110407) __________ >> >> The message was checked by ESET NOD32 Antivirus. >> >> http://www.eset.com >> >> >> >> >> __________ Information from ESET NOD32 Antivirus, version of virus > signature >> database 6021 (20110407) __________ >> >> The message was checked by ESET NOD32 Antivirus. >> >> http://www.eset.com >> >> >> > > __________ Information from ESET NOD32 Antivirus, version of virus signature > database 6021 (20110407) __________ > > The message was checked by ESET NOD32 Antivirus. > > http://www.eset.com > > > > > __________ Information from ESET NOD32 Antivirus, version of virus signature > database 6031 (20110410) __________ > > The message was checked by ESET NOD32 Antivirus. > > http://www.eset.com > > > _______________________________________________ relax (http://nmr-relax.com) This is the relax-users mailing list [email protected] To unsubscribe from this list, get a password reminder, or change your subscription options, visit the list information page at https://mail.gna.org/listinfo/relax-users
