![]() It eliminates round-off error on the zero entries of the eigenvectors. Here's a way using ZeroTest that works with two of the three alternatives. Since most algorithms call other such algorithms, the errors compound and a good algorithm will produce much greater errors. For instance, for a good algorithm for a function, the best result in double precision ( $MachinePrecision in Mathematica) should be expected to have a rounding error up to 0.5 * $MachineEpsilon, which is around 10^-16. So getting familiar with round-off error is helpful in such cases. It is faster and uses less memory than exact methods. Sometimes working in floating-point is advisable.
0 Comments
Leave a Reply. |