search

caseypen / parallel_machine_scheduling

This repository is to solve parallel machine scheduling problems with release constraints
BSD 3-Clause "New" or "Revised" License
34 stars 5 forks source link

problem in runnig setup.py #1

Open Heliaemami opened 4 years ago

Heliaemami commented 4 years ago

Hi, I am new to programming and I am trying to use caseypen/parallel_machine_scheduling from GitHub ( " https://github.com/caseypen/parallel_machine_scheduling " ). Part of the code in the heuristic model folder is implemented in C and Python-API is wrapped with Cython package. But I have confronted these problems: I have tried both minGw32 and mingw-w64 as a cgg g++ compiler for cython on both python2.7 and 3.6 version, as well as, Visual studio c++ build tools and windows SDK compatible version considering each python versions. Unfortunately. none of them have worked properly using setup.py for building python extension (.pyd) file which was needed. Each time a specific error pops up. I wonder if I'm doing sth wrong or the code itself has issues?

P.S. when I omit these lines ( include_dirs = [numpy.get_include() ) from setup.py which was in previous versions of the code history in Github but not in the latest cleared up one, following error arises (which is when I used visual studio, however similar error arises after assigning MinGW as the main compiler) : "fatal error c1083: cannot open include file: 'numpy/arrayobject.h': no such file or directory error: command 'c:\program files (x86)\microsoft visual studio\2019\community\vc\tools\msvc\14.25.28610\bin\hostx86 \x64\cl.exe' failed with exit status 2"

caseypen commented 4 years ago

Hi Heliaemami, Thanks for concerning. Honestly, I never tried it on Windows/Mac environment. It is suggested to use the Linux system to try. Also, based on my testing, the MIP model is faster to solve the problem. Good luck!

Best regards,

Chen

Heliaemami commented 4 years ago

Dear Chen,  Thank you for your prompt reply. You have also imported some functions and classes from Jsp_bbs from the heuristic model in the MIP model for providing some search boundaries. Since I am new to programming, and a major part of my work is on windows (Revit and other Autodesk products), It would significantly speed up my work if it was a way to continue the progress on windows rather than writing the code from scratch or switching to Linux. I would like to know is it possible to, for example, run the setup.py on another system with Linux and use the generated python extension library on windows? Or you could attach it on GitHub? (Sorry, if I am asking basic questions, I am still learning. your help and support is greatly appreciated.) Sincerely,Helia Emami On Saturday, May 9, 2020, 12:55:54 AM GMT+4:30, Chen Peng notifications@github.com wrote:

Hi Heliaemami, Thanks for concerning. Honestly, I never tried it on Windows/Mac environment. It is suggested to use the Linux system to try. Also, based on my testing, the MIP model is faster to solve the problem. Good luck!

Best regards,

Chen

—

caseypen commented 4 years ago

Hi Heliaemami,

I suggest you look into the mathematic modeling - mixed integer programming for this problem (you can also see that in this repo) and then solve it with the solver, like Gurobi or CPLEX. This will be faster and the company will have better management for different operating systems.

Thanks,

Chen

On Fri, May 8, 2020 at 2:23 PM Heliaemami notifications@github.com wrote:

Dear Chen, Thank you for your prompt reply. You have also imported some functions and classes from Jsp_bbs from the heuristic model in the MIP model for providing some search boundaries. Since I am new to programming, and a major part of my work is on windows (Revit and other Autodesk products), It would significantly speed up my work if it was a way to continue the progress on windows rather than writing the code from scratch or switching to Linux. I would like to know is it possible to, for example, run the setup.py on another system with Linux and use the generated python extension library on windows? Or you could attach it on GitHub? (Sorry, if I am asking basic questions, I am still learning. your help and support is greatly appreciated.) Sincerely,Helia Emami On Saturday, May 9, 2020, 12:55:54 AM GMT+4:30, Chen Peng < notifications@github.com> wrote:

Hi Heliaemami, Thanks for concerning. Honestly, I never tried it on Windows/Mac environment. It is suggested to use the Linux system to try. Also, based on my testing, the MIP model is faster to solve the problem. Good luck!

Best regards,

Chen

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/caseypen/parallel_machine_scheduling/issues/1#issuecomment-626026772, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADSENGJ5TT47MZN7D2E5VBTRQRZ4JANCNFSM4M4OABBQ .

--

Chen Peng

Ph.D. Candidate Dept. of Mechanical and Aerospace Engineering University of California, Davis https://caseypen.github.io/