0

I am using the lpsolve55 library in python to find some linear programming solutions. The input data that is loaded needs to have up to 14 digits accuracy, but when the lp solver writes the problem to a file the accuracy is only 11 digits. How can I increase the accuracy?

11 digit accuracy in the output file real accuracy of 14 digits

CyberHunter
  • 69
  • 7
  • How are you creating this file? I don't think `lpsolve` writes anything to file. That's almost certainly being handled by your program. It's up to you to do the formatting the way you want it. However, it's extremely unlikely that your input data actually has 14 digits of precision. There are very few natural processes that result in measurements that accurate. – Tim Roberts Mar 31 '21 at 22:21
  • I guess you would need to change the source code of the routine that writes the lp file. Having said that, this precision is likely enormous overkill. The numerical linear algebra used inside lpsolve is not capable of handling and maintaining that type of precision. – Erwin Kalvelagen Apr 01 '21 at 06:49
  • @TimRoberts the input data has 14 digit accuracy, this is not the problem, there is the command lpsolve('write_lp', lp, filename) that you can use to output the lp problem in an lp file that is readable from the lpsolve IDE. – CyberHunter Apr 02 '21 at 09:33
  • @ErwinKalvelagen if I need to change the subroutine this is going too far away thanks anyway – CyberHunter Apr 02 '21 at 09:34

0 Answers0