Hi Troels, This is not the correct approach. These higher dimensions must be missing from the gradient, Hessian, Jacobian, and covariance matrix. If N is the number of parameters and M is the number of input data sets (R2eff/R1rho/R1), then these structures must absolutely have the following dimensions:
- function = 1, - gradient = N, - Hessian = NxN, - Jacobian = NxM, - covariance matrix = NxN. If they don't have these exact dimensions, then they cannot be called by those names. You have no choice. Their dimensionality matches that of the input parameter vector! You have to loose all of the NE, NS, NM, NO, and ND dimensions in all of these structures. Regards, Edward On 2 September 2014 10:29, <[email protected]> wrote: > Author: tlinnet > Date: Tue Sep 2 10:29:48 2014 > New Revision: 25530 > > URL: http://svn.gna.org/viewcvs/relax?rev=25530&view=rev > Log: > Added comments to co-variance module, for explanation of data dimensionality. > > 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/statistics.py > > Modified: branches/est_par_error/lib/statistics.py > URL: > http://svn.gna.org/viewcvs/relax/branches/est_par_error/lib/statistics.py?rev=25530&r1=25529&r2=25530&view=diff > ============================================================================== > --- branches/est_par_error/lib/statistics.py (original) > +++ branches/est_par_error/lib/statistics.py Tue Sep 2 10:29:48 2014 > @@ -229,6 +229,11 @@ > # Get the expected shape of the higher dimensional column numpy array. > if len(weights.shape) == 2: > # Extract shapes from data. > + # NE: Number of experiments. > + # NS: Number of spins. > + # NM: Number of spectrometer frequencies. > + # NO: Maximum number of offsets. > + # ND: Number of dispersion(data) points. > NE, NS, NM, NO, ND = 1, 1, 1, 1, weights.shape[-1] > > # Make a eye matrix, with Shape [ND][ND] > > > _______________________________________________ > 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
