Using OpenMatrix in Atom on Windows

  1. After installing, invoke OpenMatrix in Atom using the run_atom_oml.bat file provided in the scripts/installation folder.
  2. From the Atom OML menu, select Start OML.
  3. In the OML command window, enter any OML command. See the Reference Guide for more information on commands.
  4. You may need to adapt the run_atom_oml.bat file to specific paths on your machine. Specifically, OpenMatrix offers a bridge between OML and Python. This bridge is disabled by default.

Enable the bridge between OML and Python

  1. Install/build Python 3.5.2 , Numpy 1.14.5.
  2. Enable omlpythonbridge toolbox loading.
    1. Add the command loadtoolbox omlpythonbridge in the file <INSTALLATION_HOME>\OpenMatrix\scripts\loadtoolboxes.oml.
  3. In file <INSTALLATION_HOME>\OpenMatrix\scripts\run_atom_oml.bat, adapt the .bat file to specific paths on your machine.
    1. Set the environment variable OML_PYTHONHOME to point to the Python home.
      set OML_PYTHONHOME=C:/python/python3.5.2
    2. Add the Python DLL (python35.dll) location to the environment variable "PATH”
      set path= C:/python/python3.5.2;%path%
    Note: The omlpythonbridge toolbox is built using Python binaries, which are built with Visual Studio 2015 (win64) Update 2, Intel 2016Update 2 Fortran and C++ compilers. It is advised to use Python binaries built using the same compiler versions.
    Note: The binaries are built using Visual Studio 2015 (win64) Update 2, Intel 2016Update 2 FortraOpenMatrix and C++ compilers with the following compilation options.
    cl -nologo -MD -O2 -FC -DWIN32 -DOS_WIN -D_WIN64
    cl -nologo -MD -O2 -W3 -wd4996 -wd4251 -GR -EHsc -FC -DWIN32 -DOS_WIN -D_WIN64
    Link Options
    link -nologo -manifest:no
    Note: Verified using Python 3.5.2 binaries installed from Installed Numpy, Scipy, Matplotlib using pip.

    Python.exe -m pip install numpy==1.14.5

    Python.exe -m pip install scipy==1.1.0

    Python.exe -m pip install matplotlib==2.2.2