patch for eigenvector to consider



Robert Dodier a ?crit :
> (%i2) [vals, vecs] : eigenvectors (M1);
> (%o2) [[[0, 2], [2, 2]], [[[1, 0, 0, 0]],
>                                    [[0, 0, 1, 0], [0, 0, 0, 1]]]]
>   
>
> I'm inclined to apply the patch because without it, the output
> is pretty much incomprehensible when there are multiple
> eigenvectors per eigenvalue. Yes / no / maybe ??
>
> Thanks for your comments.
I find a list of triples 
[[val1,mul1,[list of vecs1]],...,[valk,mulk,[list of vecsk]]] 
easier to read (each vector being "near" its eigenvalue), but it may be slightly more difficult to use (extracting the list of all eigenvectors is quite frequent).

Eric Reyssat