relax SVN archive

Fixes for the dispersion specific _back_calc() method.

This method still has a long way to go before it is of any use.

The relaxation dispersion target function class back_calc variable now matches the values variable.

Instead of being a temporary structure which is overwritten for each spin and each exponential
curve, the structure now matches the dimensions of the values variable and hence is persistent per
function call. This allows external code to access the structure - for example for data back
calculation in the relaxation dispersion specific analysis module.

Increased the speed of the Relax_disp.test_exp_fit system test by using a smaller grid search.

Fix for the relaxation dispersion grid search.

The lower and upper bounds are no longer continually scaled with each optimisation.

Fixes for the dispersion specific _assemble_param_vector() method.

The R2eff and I0 spin dictionary structures are now checked for their keys before pulling the value
out.

Merged revisions 19374 via svnmerge from
svn+ssh://bugman@.../svn/relax/trunk

........
r19374 | bugman | 2013-04-04 22:28:57 +0200 (Thu, 04 Apr 2013) | 5 lines

Fix for the analysis specific API common method _data_init_spin().

The data types are now correctly checked - they are not strings but types.
........

Fix for the analysis specific API common method _data_init_spin().

The data types are now correctly checked - they are not strings but types.

The relaxation dispersion specific _disassemble_param_vector() method is now functional.

This should allow the minimise user function to complete.

The 'exp_fit' relaxation dispersion model now uses the minfx.grid sparseness argument.

This is used to skip all parts of the grid search belonging to a different exponential curve or
different spin. If the number of curves is N and the number of spins M, the grid size decreases
from inc**(2*N*M) to (inc**2)*N*M. For lots of spins and curves, this is a huge decrease.

Fix for the func_exp_fit() relaxation dispersion target function.

The parameter index was not correctly calculated.