Re: r22862 - /trunk/gui/analyses/auto_relax_disp.py -- April 29, 2014 - 15:00

Hi Troels,

I have few points about this change:

1)  I think it would be good to put this lower down, between the "Per
model error analysis" and "Insignificance level" options.

2)  The tooltop text should be expanded to explain the negative side
of this.  Firstly this is experimental and I would not consider it
fully tested, so it would be good to include such wording.  Also
include wording such as "this experimental feature is unpublished".
This is not a perfect solution.  Setting R20 values to the minimum
R2eff/R1rho value can actually be quite dangerous if R20 >> min(R2eff)
- this will result in a completely false grid search which is not
around the real minimum.  If the user has access to a cluster then it
would be preferable to skip this option.  Some more text such as "only
use this if you know what you are doing", or "use at your own risk"
would be good as well.

3)  The text description would be better as "Set R20 to the minimum
R2eff".  All users will automatically turn on the option to "Speed up
of grid search" without thinking.  But they should only do this if
they know what they are doing, and they know how this option can fail.
 So changing the text to something more specific will make them think
twice.

Cheers,

Edward






On 25 April 2014 19:28,  <tlinnet@xxxxxxxxxxxxx> wrote:
> Author: tlinnet
> Date: Fri Apr 25 19:28:36 2014
> New Revision: 22862
>
> URL: http://svn.gna.org/viewcvs/relax?rev=22862&view=rev
> Log:
> Added True/False button to activate relax_disp.set_grid_r20_from_min_r2eff 
> in auto analyses.
>
> Feature request: #3151, (https://gna.org/support/index.php?3151) - User 
> function to set the R20 parameters in the default grid search using the 
> minimum R2eff value.
>
> Modified:
>     trunk/gui/analyses/auto_relax_disp.py
>
> Modified: trunk/gui/analyses/auto_relax_disp.py
> URL: 
> http://svn.gna.org/viewcvs/relax/trunk/gui/analyses/auto_relax_disp.py?rev=22862&r1=22861&r2=22862&view=diff
> ==============================================================================
> --- trunk/gui/analyses/auto_relax_disp.py       (original)
> +++ trunk/gui/analyses/auto_relax_disp.py       Fri Apr 25 19:28:36 2014
> @@ -360,6 +360,10 @@
>          # The grid search optimisation settings.
>          self.grid_inc = Spin_ctrl(box, self, text="Grid search 
> increments:", default=21, min=1, max=100, tooltip="This is the number of 
> increments per dimension of the grid search performed prior to numerical 
> optimisation.", width_text=self.width_text, width_button=self.width_button, 
> spacer=self.spacer_horizontal)
>
> +        # The speed up of grid search.
> +        tooltip = "True will set the grid %s values from the minimum %s/%s 
> values. This will speed up the grid search with a factor 
> GRID_INC^(Nr_spec_freq). For a CPMG experiment with two fields and standard 
> GRID_INC=21, the speed-up is a factor 441." % (r2, r2eff, r1rho)
> +        self.set_grid_r20 = Boolean_ctrl(box, self, text="Speed up of grid 
> search", default=False, tooltip=tooltip, width_text=self.width_text, 
> width_button=self.width_button, spacer=self.spacer_horizontal)
> +
>          # The MC simulation settings.
>          self.mc_sim_num = Spin_ctrl(box, self, text="Monte Carlo 
> simulation number:", default=500, min=1, max=100000, tooltip="This is the 
> number of Monte Carlo simulations performed for error propagation and 
> analysis.  For best results, at least 500 is recommended.", 
> width_text=self.width_text, width_button=self.width_button, 
> spacer=self.spacer_horizontal)
>          self.mc_sim_all_models = Boolean_ctrl(box, self, text="Per model 
> error analysis:", default=False, tooltip="A flag which if True will cause 
> Monte Carlo simulations to be performed for each individual model.  
> Otherwise Monte Carlo simulations will be reserved for the final model.", 
> width_text=self.width_text, width_button=self.width_button, 
> spacer=self.spacer_horizontal)
>
>
> _______________________________________________
> relax (http://www.nmr-relax.com)
>
> This is the relax-commits mailing list
> relax-commits@xxxxxxx
>
> 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-commits