From: Cameron McBride Date: 2007-11-21T05:11:09+09:00 Subject: Re: [Matrix] eigenvalues, eigenvectors in Ruby ??? On Nov 20, 2007 2:35 PM, Une Bévue wrote: > > Am I wrong in thinking that you can use that library without being > > forced to calculate in Floats, i.e. can one use Rationals, Sqrts, > > continued fractions, whatever as matrix entries ? > > yes i had this prob with gsl, my matrices (coming from chemistry then > having only integer numbers...) print out with -0.000e+00 for 0 and > 1.000e+00 for 1 ...) To be clear, the printing accuracy isn't the machine accuracy. The -0.000e+00 is just the specific format that the GSL::Matrix object uses for printing. Also, the default matrix in GSL is a double (float) based one. You can create integer matricies. Look at the manual. Cameron