Hi Edward. I currently do not aim to implement minimisation with BFGS or similar. This is only for estimating the error.
But I think I have it. dw is returned in rad/s, and I need to scale i back to ppm. Best Troels 2014-09-02 9:24 GMT+02:00 Edward d'Auvergne <[email protected]>: > Hi Troels, > > If you are interested in speed for these functions - something which > may not be worth your effort right now - you should minimise > repetitive calculations. For example, you have: > > dw*tcp = 5 times, > k_AB / dw = 2 times, > sin(dw*tcp) = 2 times. > > If you pre-calculate these in the calling function and then add them > as new arguments, the code will not look as clean but it will be much > faster. This is done in the model-free code, but instead of arguments > I have a special data container holding all the structures which is > then passed into each function, gradient, and Hessian function. > > Regards, > > Edward > > > > On 1 September 2014 20:51, <[email protected]> wrote: >> Author: tlinnet >> Date: Mon Sep 1 20:51:45 2014 >> New Revision: 25513 >> >> URL: http://svn.gna.org/viewcvs/relax?rev=25513&view=rev >> Log: >> Added partial derivatives for model TSMFK01 and the Jacobian function. >> >> task #7824(https://gna.org/task/index.php?7824): Model parameter ERROR >> estimation from Jacobian and Co-variance matrix of dispersion models. >> >> Modified: >> branches/est_par_error/lib/dispersion/tsmfk01.py >> >> Modified: branches/est_par_error/lib/dispersion/tsmfk01.py >> URL: >> http://svn.gna.org/viewcvs/relax/branches/est_par_error/lib/dispersion/tsmfk01.py?rev=25513&r1=25512&r2=25513&view=diff >> ============================================================================== >> --- branches/est_par_error/lib/dispersion/tsmfk01.py (original) >> +++ branches/est_par_error/lib/dispersion/tsmfk01.py Mon Sep 1 20:51:45 >> 2014 >> @@ -66,7 +66,7 @@ >> """ >> >> # Python module imports. >> -from numpy import fabs, min, sin, isfinite, sum >> +from numpy import array, cos, fabs, min, sin, isfinite, ones, sum, transpose >> from numpy.ma import fix_invalid, masked_where >> >> >> @@ -129,3 +129,72 @@ >> if not isfinite(sum(back_calc)): >> # Replaces nan, inf, etc. with fill value. >> fix_invalid(back_calc, copy=False, fill_value=1e100) >> + >> + >> +def r2eff_TSMFK01_jacobian(r20a=None, dw=None, k_AB=None, tcp=None): >> + """The Jacobian matrix of TSMFK01. >> + >> + @keyword r20a: The R20 parameter value of state A (R2 with no >> exchange). >> + @type r20a: numpy float array of rank [NE][NS][NM][NO][ND] >> + @keyword dw: The chemical exchange difference between states >> A and B in rad/s. >> + @type dw: numpy float array of rank [NE][NS][NM][NO][ND] >> + @keyword k_AB: The k_AB parameter value (the forward exchange >> rate in rad/s). >> + @type k_AB: float >> + @keyword tcp: The tau_CPMG times (1 / 4.nu1). >> + @type tcp: numpy float array of rank [NE][NS][NM][NO][ND] >> + """ >> + >> + # Get the partial derivatives. >> + get_d_f_d_r20a = d_f_d_r20a(r20a=r20a, dw=dw, k_AB=k_AB, tcp=tcp) >> + get_d_f_d_dw = d_f_d_dw(r20a=r20a, dw=dw, k_AB=k_AB, tcp=tcp) >> + get_d_f_d_k_AB = d_f_d_k_AB(r20a=r20a, dw=dw, k_AB=k_AB, tcp=tcp) >> + >> + return transpose(array( [get_d_f_d_r20a , get_d_f_d_dw, get_d_f_d_k_AB] >> ) ) >> + >> + >> +def d_f_d_r20a(r20a=None, dw=None, k_AB=None, tcp=None): >> + """Partial derivative with respect to r20a. >> + >> + @keyword r20a: The R20 parameter value of state A (R2 with no >> exchange). >> + @type r20a: numpy float array of rank [NE][NS][NM][NO][ND] >> + @keyword dw: The chemical exchange difference between states >> A and B in rad/s. >> + @type dw: numpy float array of rank [NE][NS][NM][NO][ND] >> + @keyword k_AB: The k_AB parameter value (the forward exchange >> rate in rad/s). >> + @type k_AB: float >> + @keyword tcp: The tau_CPMG times (1 / 4.nu1). >> + @type tcp: numpy float array of rank [NE][NS][NM][NO][ND] >> + """ >> + >> + return ones(dw.shape) >> + >> + >> +def d_f_d_dw(r20a=None, dw=None, k_AB=None, tcp=None): >> + """Partial derivative with respect to dw. >> + >> + @keyword r20a: The R20 parameter value of state A (R2 with no >> exchange). >> + @type r20a: numpy float array of rank [NE][NS][NM][NO][ND] >> + @keyword dw: The chemical exchange difference between states >> A and B in rad/s. >> + @type dw: numpy float array of rank [NE][NS][NM][NO][ND] >> + @keyword k_AB: The k_AB parameter value (the forward exchange >> rate in rad/s). >> + @type k_AB: float >> + @keyword tcp: The tau_CPMG times (1 / 4.nu1). >> + @type tcp: numpy float array of rank [NE][NS][NM][NO][ND] >> + """ >> + >> + return -k_AB * cos( dw * tcp) / dw + k_AB * sin(dw * tcp) / ( dw**2 * >> tcp) >> + >> + >> +def d_f_d_k_AB(r20a=None, dw=None, k_AB=None, tcp=None): >> + """Partial derivative with respect to k_AB. >> + >> + @keyword r20a: The R20 parameter value of state A (R2 with no >> exchange). >> + @type r20a: numpy float array of rank [NE][NS][NM][NO][ND] >> + @keyword dw: The chemical exchange difference between states >> A and B in rad/s. >> + @type dw: numpy float array of rank [NE][NS][NM][NO][ND] >> + @keyword k_AB: The k_AB parameter value (the forward exchange >> rate in rad/s). >> + @type k_AB: float >> + @keyword tcp: The tau_CPMG times (1 / 4.nu1). >> + @type tcp: numpy float array of rank [NE][NS][NM][NO][ND] >> + """ >> + >> + return 1. - sin( dw * tcp) / (dw * tcp) >> >> >> _______________________________________________ >> relax (http://www.nmr-relax.com) >> >> This is the relax-commits 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-commits > > _______________________________________________ > relax (http://www.nmr-relax.com) > > This is the relax-devel 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-devel _______________________________________________ relax (http://www.nmr-relax.com) This is the relax-devel 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-devel
