relax SVN archive

Bug fix for the change to the numpy.sum() function.

In Numeric the axis keyword argument defaults to 0. In numpy it defaults to None!

Set the max number of iterations to 0 in the 5-state xz system test.

Setting the parameters to the exact solution gives a chi-squared value of 11.92!!! Something is
badly wrong@!

Bug fix for the 5-state xz system test.

The parameter alpha was being set twice (and beta was not being set).

Implemented the N-state specific is_spin_param() method.

Merged revisions 4886-4888 via svnmerge from
svn+ssh://bugman@.../svn/relax/1.3

........
r4886 | bugman | 2008-01-25 14:13:57 +0100 (Fri, 25 Jan 2008) | 9 lines

Bug fix for the generic_fns.value.set() function.

The differentiation between a spin specific parameter and a global parameter was not working for
specific analyses with no spin specific parameters. The return_data_name() function was being used
to differentiate, which makes no sense. So instead the is_spin_param() function has been introduced
to allow the specific code to answer the question explicitly. Note that this will temporarily break
many things.
........
r4887 | bugman | 2008-01-25 15:31:19 +0100 (Fri, 25 Jan 2008) | 5 lines

Implemented the is_spin_param() base class method.

This function always returns true.
........
r4888 | bugman | 2008-01-25 15:56:06 +0100 (Fri, 25 Jan 2008) | 3 lines

Implemented the model-free specific is_spin_param() method.
........

Implemented the model-free specific is_spin_param() method.

Implemented the is_spin_param() base class method.

This function always returns true.

Bug fix for the generic_fns.value.set() function.

The differentiation between a spin specific parameter and a global parameter was not working for
specific analyses with no spin specific parameters. The return_data_name() function was being used
to differentiate, which makes no sense. So instead the is_spin_param() function has been introduced
to allow the specific code to answer the question explicitly. Note that this will temporarily break
many things.

Changed the 5-state xz system test to start at the true parameter values.

This should speed things up considerably in this test.

Bug fix for the 5-state xz-plane system test.

The reduced tensors were incorrect. They were the tensors for the first state, not the averaged
tensors! They have been replaced by the correct ones and the number of states set to 5.