Robert Dodier <robert.dodier at gmail.com> writes:
> On 2012-12-14, Rupert Swarbrick <rswarbrick at gmail.com> wrote:
>
>> If not, I would really like to find a way to hook into the simplifier
>
> matchdeclare (mm, suitable_matrix_p);
> tellsimp (exp (mm), my_matrix_exp (mm));
>
Ah, it's easier than I feared! Cool.
> I didn't see any examples of matrix exponential calculations in the test
> script -- did I miss something?
The only example that actually checks the exponential is giving sensible
answers is the one using "diag_mat_errors" (since exp(diag_matrix(...))
exponentiates the old-fashioned way). I would like to improve this!
To prove the code at least works after a fashion:
(%i19) ward_matrix_exponential (matrix ([1.0, 0], [0, 1.0]));
[ 2.718281828459045 0.0 ]
(%o19) [ ]
[ 0.0 2.718281828459045 ]
(%i20) ward_matrix_exponential (matrix ([1b0, 0], [0, 1b0]));
[ 2.718281828459045b0 0.0b0 ]
(%o20) [ ]
[ 0.0b0 2.718281828459045b0 ]
(%i21) float(ward_matrix_exponential (
matrix ([1,-2,3,-4], [-5,-6,-7,-8],
[-9,-10,-11,-12], [-13,14,-15,-16b0])));
[ 3.470106840541206 ] [ - 23.10259917295292 ]
[ ] [ ]
[ .5451828980115822 ] [ - 1.952572569199982 ]
(%o21) Col 1 = [ ] Col 2 = [ ]
[ - 2.984728521786326 ] [ 8.393306130644127 ]
[ ] [ ]
[ - 0.00880516207293516 ] [ 9.14360740721301 ]
[ 15.10464093068505 ] [ - .8014955229469515 ]
[ ] [ ]
[ 1.321907172063737 ] [ - .1485406279134143 ]
Col 3 = [ ] Col 4 = [ ]
[ - 5.797476414263982 ] [ .8439341087562101 ]
[ ] [ ]
[ - 5.732874998799312 ] [ - .1203037903101915 ]
(The float call is because otherwise one scaling routine takes out an
exact power of %e, which makes the answer look rather odd)
No, I don't have error estimates for the last one. Yet...
Rupert
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 315 bytes
Desc: not available
URL: <http://www.math.utexas.edu/pipermail/maxima/attachments/20121214/53319ef1/attachment.pgp>;