Hi Andras, To push changes into the source code repository, you'd have to become a relax developer. If you are interested, I'd recommend first reading of the 'open source infrastructure' and 'relax development' chapters of the manual. These chapters outline the procedure in full detail. Do you have experience with CVS, subversion, git, or other source code revision control systems?
The first step to have changes made to relax is to upload either files or patches to a relax task, and ask for it to be included or applied. For example I could take the script you have just attached to the task in the 'saturation_recovery.tar.gaz' file (https://gna.org/task/?7415, https://gna.org/task/download.php?file_id=22997), and place this into the relax system test directories. However, if you feel adventurous, you could try to create your own system test from this! I could then incorporate your self designed test and add this to relax. Creating this test is rather straight forward: - Firstly you will need a copy of the relax trunk in your home directory that you can modify yourself. The /sw/lib/ version is no good for this. - Copy your script to the new file test_suite/system_tests/scripts/relax_fit_saturation_recovery.py. Add this to the repository by typing: $ svn add test_suite/system_tests/scripts/relax_fit_saturation_recovery.py You can then see the status of the relax source code by typing: $ svn st which will show your newly added script. - The next step would be to add a copyright notice. Open any relax *.py file and copy the notice from the top into your script. Remove all authors and place your name and the year there. You could also include the link to your Gna! profile in brackets after your name (https://gna.org/users/andras). - Then to create the test itself. I would suggest that this is called Relax_fit.test_saturation_recovery. Open the test_suite/system_tests/relax_fit.py in a text editor (preferably something such as gvim or emacs where source code is recognised, and coloured, and the programs help you with the code). Copy the 11 lines of the test_zooming_grid_search() method (http://www.nmr-relax.com/api/3.3/test_suite.system_tests.relax_fit-pysrc.html#Relax_fit.test_zooming_grid_search) and paste and rename it to test_saturation_recovery(). Be sure that alphabetical order is preserved. Change the script name to your script, and the chi-squared value to 0.0, and the I0 and Rx values to your values. As you are changing this file, add a new line to the copyright notice with your name and the year. - Try running the test with: $ relax -s Relax_fit.test_saturation_recovery - This will probably not run straight away, as you need to fix the paths in the script. Open your script and the test_suite/system_tests/scripts/relax_fit_zooming_grid.py script side-by-side (run 'svn up' now as I just simplified this reference script for you). Note the use of the relax status object (used to find where relax is installed) and the os.sep Python module to support all operating systems, as '/' in Windows is '\' (fortunately Macs don't use ':' as a folder separator any more). After this the test should run and be complete. Yet it will fail due to the chi-squared and parameter check, and because the saturation-recovery equations are not implemented yet. - Finally prepare a patch by typing: $ svn diff > patch This will contain all the changes, including the new script file. Upload the 'patch' file to the task, remembering to add a comment describing the patch. You may even wish to create your own commit message - see the development chapter. - Code review. I'll look at the patch and make suggestions for improvements. Changes can be made and a new patch created and uploaded. - Once the patch is in a good state, I can apply it to the relax trunk and permanently commit the change to the repository. This is essentially the starting procedure for becoming a developer. Many more details are given in the development chapter. The reason I'm mentioning all of this now is because it's 7 pm Friday evening here and I won't have a chance to look at this until next week. For me, creating this test is less than 5 min of work. If you have the interest, feel free to try it out yourself. Regards, Edward P. S. As for deleting files, this is not possible. The infrastructure that relax uses, and open source infrastructure in general, is deliberately designed so that everything is permanent. On 28 November 2014 at 18:38, Boeszoermenyi, Andras <andras_boeszoerme...@hms.harvard.edu> wrote: > Hi, > > great, this really helped. > > The script runs now for exponential fitting 'exp' and throws an error for > inversion recovery fitting 'inv'. > > I uploaded a new saturation recovery folder: > > file #22997: saturation_recovery.tar.gaz > > Is there a way to remove the old one? And is there a way for me to push > changes through svn? > > Best, > Andras > ____________________________ > From: edward.dauver...@gmail.com [edward.dauver...@gmail.com] On Behalf Of > Edward d'Auvergne [edw...@nmr-relax.com] > Sent: Friday, November 28, 2014 12:16 PM > To: Boeszoermenyi, Andras > Cc: relax-devel@gna.org; Sébastien M > Subject: Re: [sr #3195] Fitting of saturation recovery experiment > > Hi, > > Ok, then you're almost there. You'll need to obtain a copy of minfx > (https://gna.org/projects/minfx/) and bmrblib > (https://gna.org/projects/bmrblib/). These used to be relax packages, > but they have been spun out into their own for others to take > advantage of. For this, type: > > $ cd /sw/lib/relax-trunk/ > $ svn co http://svn.gna.org/svn/minfx/tags/1.0.11/minfx > $ svn co http://svn.gna.org/svn/bmrblib/tags/1.0.3/bmrblib > > This is performed when a new relax release is made so that minfx and > bmrblib are bundled with relax, but this needs to be performed > manually when using the most cutting edge code. Note that it is much > better to place the relax-trunk into your home directory as this will > be constantly modified by the relax developers. Then, as a normal > user (without sudo), you can obtain the latest code by typing: > > $ svn up > > Regards, > > Edward > > > P. S. The command for showing the PATH environmental variable is > "echo $PATH" rather than just "$PATH". > > > > > > > > On 28 November 2014 at 18:07, Boeszoermenyi, Andras > <andras_boeszoerme...@hms.harvard.edu> wrote: >> Hi, >> >> so I aliased relax to the trunk relax and can start it now, but here is the >> problem: >> >> File "/sw/lib/relax-trunk/relax", line 4, in <module> >> import relax >> File "/sw/lib/relax-trunk/relax.py", line 37, in <module> >> import dep_check >> File "/sw/lib/relax-trunk/dep_check.py", line 90, in <module> >> if not minfx.__version__ == 'trunk' and >> version_comparison(minfx.__version__, min_version) == -1: >> AttributeError: 'module' object has no attribute '__version__' >> >> which relax now gives: >> >> which relax >> relax: aliased to /sw/lib/relax-trunk/relax >> >> and $PATH: >> >> /sw/bin:/sw/sbin:.:/Applications/NMRpipe/nmrbin.mac:/Applications/NMRpipe/com:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/Applications/xplor-nih-2.33/bin:/opt/X11/bin: >> Command not found. >> >> >> The locations are: >> >> /sw/lib/relax-py27/relax_gui_mode.py >> /sw/lib/relax-py27/relax_gui_mode.pyo >> /sw/lib/relax-trunk/relax_gui_mode.py >> >> I installed relax previously with fink. Unfortunately, fink does't support >> yosemite yet, so I can't install newer versions this way, except for the app. >> >> Best regards, >> Andras >> ________________________________________ >> From: edward.dauver...@gmail.com [edward.dauver...@gmail.com] On Behalf Of >> Edward d'Auvergne [edw...@nmr-relax.com] >> Sent: Friday, November 28, 2014 11:46 AM >> To: Boeszoermenyi, Andras >> Cc: relax-devel@gna.org; Sébastien M >> Subject: Re: [sr #3195] Fitting of saturation recovery experiment >> >> Hi, >> >> Hmmm, this might be difficult to work out due to the number of >> possibilities (Fink, Homebrew, framework installs, standard installs, >> etc.). So I'd need a bit more information. Which method did you use >> to obtain the code for the relax trunk? Do you have Xcode, Scons, >> subversion, Python, etc. installed all via DMG packages? How did you >> install relax to start with? Are you the administrator for your >> computer? To find the different relax installations, maybe type: >> >> $ sudo /usr/libexec/locate.updatedb >> $ locate relax_gui_mode.py >> >> Could you copy and paste the output of this into a mail? This should >> find all places where relax is located (the relax_gui_mode.py script >> is simply a unique file name). Then to find out where the default >> relax is, type: >> >> $ which relax >> >> This will be the version that runs when you type 'relax'. If you have >> the relax trunk in the ~/relax-trunk directory (where ~ is your home >> directory), then to run this version type: >> >> $ ~/relax-trunk/relax >> >> Does this work for you? You can sometimes override the version >> installed in /usr/local/ by creating a 'bin' directory and linking the >> main relax file: >> >> $ cd ~ >> $ mkdir bin >> $ cd bin >> $ ln -s ~/relax-trunk/relax >> $ cd >> $ which relax >> >> Replace "~/relax-trunk" with any path you wish. Normally ~/bin will >> come before /usr/local in your $PATH environmental variable. To be >> sure, could you tell me what you see when you type: >> >> $ echo $PATH >> >> Note that in all of the above commands, that the '$ ' is the terminal >> or prompt and that you should not type this. >> >> Regards, >> >> Edward >> >> >> >> >> >> >> >> >> On 28 November 2014 at 17:30, Boeszoermenyi, Andras >> <andras_boeszoerme...@hms.harvard.edu> wrote: >>> Hi, >>> >>> would you happen to know where all these things install themselfes to. I am >>> at a complete loss. >>> >>> Best, >>> Andras >>> ________________________________________ >>> From: edward.dauver...@gmail.com [edward.dauver...@gmail.com] On Behalf Of >>> Edward d'Auvergne [edw...@nmr-relax.com] >>> Sent: Friday, November 28, 2014 9:53 AM >>> To: Boeszoermenyi, Andras >>> Cc: relax-devel@gna.org; Sébastien M >>> Subject: Re: [sr #3195] Fitting of saturation recovery experiment >>> >>> Hi, >>> >>> Somehow the relax-trunk version is not located on the system path or >>> it doesn't have priority over the installed 3.1.7 version. You could >>> supply the full path to override this, uninstall the 3.1.7 version, or >>> change the order of the directories in the $PATH environmental >>> variable. >>> >>> Regards, >>> >>> Edward >>> >>> >>> >>> On 28 November 2014 at 15:47, Boeszoermenyi, Andras >>> <andras_boeszoerme...@hms.harvard.edu> wrote: >>>> Well I installed the relax-trunk version, but I am not sure if that is >>>> what I am running. >>>> >>>> When I type relax -i >>>> >>>> I get this header: >>>> >>>> relax 3.1.7 >>>> >>>> Molecular dynamics by NMR data analysis >>>> >>>> Copyright (C) 2001-2006 Edward d'Auvergne >>>> Copyright (C) 2006-2014 the relax development team >>>> >>>> This is free software which you are welcome to modify and redistribute >>>> under the conditions of the >>>> GNU General Public License (GPL). This program, including all modules, is >>>> licensed under the GPL >>>> and comes with absolutely no warranty. For details type 'GPL' within the >>>> relax prompt. >>>> >>>> Assistance in using the relax prompt and scripting interface can be >>>> accessed by typing 'help' within >>>> the prompt. >>>> >>>> Processor fabric: Uni-processor. >>>> >>>> Best, >>>> Andras >>>> >>>> ________________________________________ >>>> From: edward.dauver...@gmail.com [edward.dauver...@gmail.com] On Behalf Of >>>> Edward d'Auvergne [edw...@nmr-relax.com] >>>> Sent: Friday, November 28, 2014 9:44 AM >>>> To: Boeszoermenyi, Andras >>>> Cc: relax-devel@gna.org; Sébastien M >>>> Subject: Re: [sr #3195] Fitting of saturation recovery experiment >>>> >>>> Hi, >>>> >>>> What do you see when you run: >>>> >>>> $ relax -i >>>> >>>> Are you using the latest relax trunk version? Or the 3.3.3 version? >>>> I only see this on versions 3.2.3 or earlier, in relax 3.3.0 and later >>>> this works. >>>> >>>> Regards, >>>> >>>> Edward >>>> >>>> >>>> >>>> On 28 November 2014 at 15:40, Boeszoermenyi, Andras >>>> <andras_boeszoerme...@hms.harvard.edu> wrote: >>>>> That's exactly what I see. >>>>> >>>>> But you seem to be getting that error too, or am I mixing something up >>>>> now? >>>>> >>>>> Best, >>>>> Andras >>>>> ________________________________________ >>>>> From: edward.dauver...@gmail.com [edward.dauver...@gmail.com] On Behalf >>>>> Of Edward d'Auvergne [edw...@nmr-relax.com] >>>>> Sent: Friday, November 28, 2014 9:34 AM >>>>> To: Boeszoermenyi, Andras >>>>> Cc: relax-devel@gna.org; Sébastien M >>>>> Subject: Re: [sr #3195] Fitting of saturation recovery experiment >>>>> >>>>> Hi Andras, >>>>> >>>>> That message is what happens when you execute the minimise.grid_search >>>>> user function on an old relax version. Could you check if you are >>>>> running the latest: >>>>> >>>>> $ relax -i >>>>> >>>>> This is what I see with the last 3.1 relax release: >>>>> >>>>> """ >>>>> relax 3.1.7 >>>>> >>>>> Molecular dynamics by NMR data analysis >>>>> >>>>> Copyright (C) 2001-2006 Edward d'Auvergne >>>>> Copyright (C) 2006-2014 the relax development >>>>> team >>>>> >>>>> This is free software which you are welcome to modify and redistribute >>>>> under the conditions of the >>>>> GNU General Public License (GPL). This program, including all >>>>> modules, is licensed under the GPL >>>>> and comes with absolutely no warranty. For details type 'GPL' within >>>>> the relax prompt. >>>>> >>>>> Assistance in using the relax prompt and scripting interface can be >>>>> accessed by typing 'help' within >>>>> the prompt. >>>>> >>>>> Processor fabric: Uni-processor. >>>>> >>>>> relax> minimise.grid_search() >>>>> Traceback (most recent call last): >>>>> File "<console>", line 1, in <module> >>>>> AttributeError: 'Uf_object' object has no attribute 'grid_search' >>>>> """ >>>>> >>>>> Regards, >>>>> >>>>> Edward >>>>> >>>>> >>>>> >>>>> >>>>> On 28 November 2014 at 15:18, Boeszoermenyi, Andras >>>>> <andras_boeszoerme...@hms.harvard.edu> wrote: >>>>>> Hi Edward, >>>>>> >>>>>> I upgraded everything to newest. >>>>>> >>>>>> I tried to updated the script too, but when I run it I get this error >>>>>> message: >>>>>> >>>>>> File "/sw/lib/relax-py27/multi/processor.py", line 494, in run >>>>>> self.callback.init_master(self) >>>>>> File "/sw/lib/relax-py27/multi/__init__.py", line 318, in >>>>>> default_init_master >>>>>> self.master.run() >>>>>> File "/sw/lib/relax-py27/relax.py", line 194, in run >>>>>> self.interpreter.run(self.script_file) >>>>>> File "/sw/lib/relax-py27/prompt/interpreter.py", line 275, in run >>>>>> return run_script(intro=self.__intro_string, local=locals(), >>>>>> script_file=script_file, show_script=self.__show_script, >>>>>> raise_relax_error=self.__raise_relax_error) >>>>>> File "/sw/lib/relax-py27/prompt/interpreter.py", line 578, in >>>>>> run_script >>>>>> return console.interact(intro, local, script_file, >>>>>> show_script=show_script, raise_relax_error=raise_relax_error) >>>>>> File "/sw/lib/relax-py27/prompt/interpreter.py", line 477, in >>>>>> interact_script >>>>>> exec_script(script_file, local) >>>>>> File "/sw/lib/relax-py27/prompt/interpreter.py", line 352, in >>>>>> exec_script >>>>>> runpy.run_module(module, globals) >>>>>> File "/sw/lib/python2.7/runpy.py", line 180, in run_module >>>>>> fname, loader, pkg_name) >>>>>> File "/sw/lib/python2.7/runpy.py", line 72, in _run_code >>>>>> exec code in run_globals >>>>>> File >>>>>> "/Users/andrasboeszoermenyi/relax-trunk/test_suite/shared_data/curve_fitting/saturation_recovery/relax_sym.py", >>>>>> line 106, in <module> >>>>>> minimise.grid_search(inc=11) >>>>>> AttributeError: 'Uf_object' object has no attribute 'grid_search' >>>>>> >>>>>> Not sure if that could be related to my recent upgrade to Yosemite, but >>>>>> I think it shouldn't. >>>>>> >>>>>> I uploaded the project as: >>>>>> >>>>>> file #22996: saturation_recovery.tar.gz >>>>>> >>>>>> Best, >>>>>> Andras >>>>>> >>>>>> ________________________________________ >>>>>> From: edward.dauver...@gmail.com [edward.dauver...@gmail.com] On Behalf >>>>>> Of Edward d'Auvergne [edw...@nmr-relax.com] >>>>>> Sent: Thursday, November 27, 2014 1:16 PM >>>>>> To: Boeszoermenyi, Andras >>>>>> Cc: relax-devel@gna.org; Sébastien M >>>>>> Subject: Re: [sr #3195] Fitting of saturation recovery experiment >>>>>> >>>>>> Hi Andras, >>>>>> >>>>>> I have now included your files in relax >>>>>> (http://article.gmane.org/gmane.science.nmr.relax.scm/24542). If you >>>>>> have the trunk source code copy of relax and type: >>>>>> >>>>>> $ svn up >>>>>> >>>>>> You will see the files appear in the >>>>>> test_suite/shared_data/curve_fitting/saturation_recovery directory. >>>>>> If you could update your script for the latest relax version, I could >>>>>> add it to the test_suite/system_tests/scripts/ directory. This would >>>>>> then be very simple to set up as a system test by duplicating and >>>>>> modifying this test: >>>>>> >>>>>> http://www.nmr-relax.com/api/3.3/test_suite.system_tests.relax_fit-pysrc.html#Relax_fit.test_zooming_grid_search >>>>>> >>>>>> Cheers, >>>>>> >>>>>> Edward >>>>>> >>>>>> >>>>>> On 27 November 2014 at 19:06, Edward d'Auvergne <edw...@nmr-relax.com> >>>>>> wrote: >>>>>>> Hi Andras, >>>>>>> >>>>>>> >>>>>>>> ok I uploaded a tar file with synthetic peaks for one spin system: >>>>>>>> >>>>>>>> file #22989 >>>>>>> >>>>>>> Cheers! The permanent link to the file is >>>>>>> http://gna.org/task/download.php?file_id=22989, and this is attached >>>>>>> to the task #7415 (http://gna.org/task/?7415). >>>>>>> >>>>>>> >>>>>>>> The saturation recovery formula is: >>>>>>>> >>>>>>>> I0*(1 - exp(−R1*t)) >>>>>>>> >>>>>>>> The parameters I used were >>>>>>>> >>>>>>>> I0 = 1000000000000000.00 >>>>>>>> >>>>>>>> and >>>>>>>> >>>>>>>> R1 = 0.5 >>>>>>> >>>>>>> I'm wondering if this equation is correct. Should this not be: >>>>>>> >>>>>>> I(t) = I_inf*(1 - exp(−R1*t)) >>>>>>> >>>>>>> Is the magnitisation not returning to the steady-state of I_inf? I >>>>>>> guess that in this experiment I0 == I_inf. >>>>>>> >>>>>>> >>>>>>>> Obviously, the same numbers also work for the inversion recovery >>>>>>>> experiment: >>>>>>>> >>>>>>>> I(t) = I∞ − I0*exp(−R1*t) >>>>>>>> >>>>>>>> with I∞ set to 1000000000000000.00 as well. Not sure how much that >>>>>>>> helps though. >>>>>>> >>>>>>> If I0 is set to something less than I_inf, this would give different >>>>>>> I(t) values. Such a data set could be used to implement the inversion >>>>>>> recovery experiment. >>>>>>> >>>>>>> >>>>>>>> If "inv" is not implemented, then that would explain the weird results >>>>>>>> :) >>>>>>> >>>>>>> More of a warning is probably required. Or a synthetic data set, >>>>>>> relax script, and system test created, and then this equation finally >>>>>>> implemented. >>>>>>> >>>>>>> >>>>>>>> Unfortunately, I have no primary reference for the saturation recovery >>>>>>>> experiment either. >>>>>>> >>>>>>> It's a pity. The ancient primary references for all these basic and >>>>>>> old experiment types would be great for adding to the relax manual and >>>>>>> user function documentation. I can see that lots of people discuss >>>>>>> this older method, but I also don't see any references. >>>>>>> >>>>>>> I was wondering if you were able to update to the latest version of >>>>>>> relax (3.3.3 at http://www.nmr-relax.com/download.html#Mac_OS_X), or >>>>>>> even better to use the relax trunk source code >>>>>>> (http://thread.gmane.org/gmane.science.nmr.relax.devel/3693/focus=7348). >>>>>>> Could you try to update the scripts to run with these versions? The >>>>>>> newest relax versions will tell you how the user functions have been >>>>>>> renamed. >>>>>>> >>>>>>> Cheers, >>>>>>> >>>>>>> Edward _______________________________________________ relax (http://www.nmr-relax.com) This is the relax-devel mailing list relax-devel@gna.org 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-devel
