Should I then make a grid search for all values, and do a new analysis?
2014-06-17 11:08 GMT+02:00 Edward d'Auvergne <[email protected]>: > Ah, that is probably because of a bug you fixed > (https://gna.org/bugs/?21954). The results were generated before this > bug was fixed, so now they will be slightly different as the > test_suite/shared_data/dispersion/Hansen/base_pipe.bz2 and > test_suite/shared_data/dispersion/Hansen/r2eff_pipe.bz2 files have > been updated. From the log file > test_suite/shared_data/dispersion/Hansen/relax_results/relax_disp.log, > you can see that the results are not too different. So just change > the values at the start of the test to the final optimised values, and > then change the values in the assertAlmostEqual() checks to the final > values. I'll update the precision of these checks after running this > on different OSes and 32-bit vs. 64-bit systems. > > You can probably see that this is not ideal, but the only 2 other > programs that support this 3-site model are Art Palmer's CPMGFit, and > Michael Bieri's NESSY. But CPMGFit has the same R20 values for all > fields, so we cannot compare. And the NESSY model number 4, which is > this model, is not fully implemented and functional. So we have > nothing to compare to. I have created synthetic data in > test_suite/shared_data/dispersion/lm63_3site which was to be used for > a system test, but I don't know why that is not in a test :S There is > even a relax script at > test_suite/shared_data/dispersion/lm63_3site/relax_results/relax_disp.py > which can be converted into a single model optimisation. I must have > been so busy with the paper drafting that I simply forgot to convert > this to a system test. > > Regards, > > Edward > > On 17 June 2014 09:18, <[email protected]> wrote: > > Author: tlinnet > > Date: Tue Jun 17 09:18:24 2014 > > New Revision: 24009 > > > > URL: http://svn.gna.org/viewcvs/relax?rev=24009&view=rev > > Log: > > Added systemtest for model LM63 3 site. > > > > According to results folder in: > > test_suite/shared_data/dispersion/Hansen/relax_results/LM63 3-site > > > > This should pass. > > > > But i doesnt? > > > > Modified: > > trunk/test_suite/system_tests/relax_disp.py > > > > Modified: trunk/test_suite/system_tests/relax_disp.py > > URL: > http://svn.gna.org/viewcvs/relax/trunk/test_suite/system_tests/relax_disp.py?rev=24009&r1=24008&r2=24009&view=diff > > > ============================================================================== > > --- trunk/test_suite/system_tests/relax_disp.py (original) > > +++ trunk/test_suite/system_tests/relax_disp.py Tue Jun 17 09:18:24 2014 > > @@ -2492,6 +2492,70 @@ > > self.assertAlmostEqual(spin71.phi_ex, 0.0553787828347638) > > self.assertAlmostEqual(spin71.kex/1000, 2781.72293906248/1000) > > self.assertAlmostEqual(spin71.chi2, 17.0776399916287) > > + > > + > > + def test_hansen_cpmg_data_to_lm63_3site(self): > > + """Optimisation of Dr. Flemming Hansen's CPMG data to the LM63 > dispersion model. > > + > > + This uses the data from Dr. Flemming Hansen's paper at > http://dx.doi.org/10.1021/jp074793o. This is CPMG data with a fixed > relaxation time period. > > + """ > > + > > + # Base data setup. > > + self.setup_hansen_cpmg_data(model='LM63 3-site') > > + > > + # Alias the spins. > > + spin70 = return_spin(":70") > > + spin71 = return_spin(":71") > > + > > + # The R20 keys. > > + r20_key1 = generate_r20_key(exp_type=EXP_TYPE_CPMG_SQ, > frq=500e6) > > + r20_key2 = generate_r20_key(exp_type=EXP_TYPE_CPMG_SQ, > frq=800e6) > > + > > + ## Set the initial parameter values. > > + spin70.r2 = {r20_key1: 7.570370921220954, r20_key2: > 8.694446951909107} > > + spin70.phi_ex_B = 0.14872003058250227 > > + spin70.phi_ex_C = 0.1319419923472704 > > + spin70.kB = 4103.672910444741 > > + spin70.kC = 7029.001690726248 > > + spin71.r2 = {r20_key1: 5.1347793381636, r20_key2: > 7.156573986051575} > > + spin71.phi_ex_B = 0.04013553485505605 > > + spin71.phi_ex_C = 0.020050748406928887 > > + spin71.kB = 4045.3007136121364 > > + spin71.kC = 3586.38798270774 > > + > > + #self.interpreter.relax_disp.r20_from_min_r2eff(force=False) > > + #self.interpreter.grid_search(lower=None, upper=None, inc=41, > constraints=True, verbosity=1) > > + > > + # Low precision optimisation. > > + self.interpreter.minimise(min_algor='simplex', > line_search=None, hessian_mod=None, hessian_type=None, func_tol=1e-25, > grad_tol=None, max_iter=10000000, constraints=True, scaling=True, > verbosity=1) > > + > > + # Printout. > > + print("\n\nOptimised parameters:\n") > > + print("%-20s %-20s %-20s" % ("Parameter", "Value (:70)", "Value > (:71)")) > > + print("%-20s %20.15g %20.15g" % ("R2 (500 MHz)", > spin70.r2[r20_key1], spin71.r2[r20_key1])) > > + print("%-20s %20.15g %20.15g" % ("R2 (800 MHz)", > spin70.r2[r20_key2], spin71.r2[r20_key2])) > > + print("%-20s %20.15g %20.15g" % ("phi_ex_B", spin70.phi_ex_B, > spin71.phi_ex_B)) > > + print("%-20s %20.15g %20.15g" % ("phi_ex_C", spin70.phi_ex_C, > spin71.phi_ex_C)) > > + print("%-20s %20.15g %20.15g" % ("kB", spin70.kB, spin71.kB)) > > + print("%-20s %20.15g %20.15g" % ("kC", spin70.kC, spin71.kC)) > > + print("%-20s %20.15g %20.15g\n" % ("chi2", spin70.chi2, > spin71.chi2)) > > + > > + # Checks for residue :70. > > + #self.assertAlmostEqual(spin70.r2[r20_key1], 6.74362294539099) > > + #self.assertAlmostEqual(spin70.r2[r20_key2], 6.57406797067481, > 6) > > + self.assertAlmostEqual(spin70.phi_ex_B, 0.148720030582502) > > + self.assertAlmostEqual(spin70.phi_ex_C, 0.13194199234727) > > + self.assertAlmostEqual(spin70.kB/1000, 4103.67291044474/1000, 6) > > + self.assertAlmostEqual(spin70.kC/1000, 7029.00169072625/1000, 6) > > + #self.assertAlmostEqual(spin70.chi2, 363.534044873483) > > + > > + # Checks for residue :71. > > + #self.assertAlmostEqual(spin71.r2[r20_key1], 5.00776657712558) > > + self.assertAlmostEqual(spin71.phi_ex_B, 0.040135534855056) > > + self.assertAlmostEqual(spin71.phi_ex_C, 0.0200507484069289) > > + self.assertAlmostEqual(spin71.kB/1000, 4045.30071361214/1000) > > + self.assertAlmostEqual(spin71.kC/1000, 3586.38798270774/1000) > > + #self.assertAlmostEqual(spin71.chi2, 17.0776399916287) > > > > > > def test_hansen_cpmg_data_to_ns_cpmg_2site_3D(self): > > > > > > _______________________________________________ > > 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
