Hi,
Ok, I'll push out both the dispersion speed and zooming grid search
(+preset parameter and optimisation user function changes) into relax
3.3.0. If you can complete the R1rho plotting branch before I go on
holiday at the end of next week, I'll include that too. Otherwise
that can be relax 3.3.1. I'm aiming to release this on Wednesday or
Thursday next week though if I can't complete the branch, it'll be 2
weeks later.
As for the R1 optimisation, the hardest part would be for you to learn
new parts of the relax infrastructure that you haven't worked with
before. But I think you'll find this very easy now. As this has
different parameters, you'd need new models for each. Just duplicate
a current one, add R1 as a parameter, create a new target function
which converts the R1 values to the correct numpy structure, and then
call the same backend lib.dispersion code. That's it.
Regards,
Edward
On 23 July 2014 16:24, Troels Emtekær Linnet <tlinnet@xxxxxxxxxxxxx> wrote:
Hi Ed.
Let it be together in 3.3.0.
I have no rush to version numbers, and I think the total upgrade is a
good surprise. :-)
I am currently working with extending the R1rho plotting, which I hope
can go into same version.
sr #3138: (https://gna.org/support/?3138) Interpolating theta through
spin-lock offset [Omega], rather than spin-lock field strength [w1]
sr #3124: (https://gna.org/support/?3124) Grace graphs production for
R1rho analysis with R2_eff as function of Omega_eff
And then I dream to fix this support request.
sr #3135: (https://gna.org/support/?3135) Optimisation of the R1
relaxation rate for the off-resonance R1rho relaxation dispersion
models.
But that would probably be a little more tricky. I will discuss this
on the mailing list later, after the graphs.
Best
Troels
2014-07-23 15:53 GMT+02:00 Edward d'Auvergne <edward@xxxxxxxxxxxxx>:
Hi Troels,
I've mentioned this a couple of times:
http://article.gmane.org/gmane.science.nmr.relax.devel/6491
http://thread.gmane.org/gmane.science.nmr.relax.scm/22089/focus=6504
But, as this is your masterpiece, I'll let you choose how it is
released to the relax user community. Would you prefer to have the
disp_spin_speed branch code released as relax 3.2.4 by itself? Or
would you prefer to include it with the specific API changes of the
zooming_grid_search branch which require the minor version number
incremented, so this will be relax 3.3.0?
Note that because of the zooming_grid_search changes, scripts for
relax <= 3.2.x will not be compatible with relax >= 3.3.0. I will
also complete the zooming_grid_search branch soon, and maybe release
relax 3.3.0 before the end of next week (unless there are
complications and the release will be in the second half of August as
I'll be on holiday).
Cheers,
Edward
Re: Release of the super fast dispersion code.