Getting Started on Windows

  • To run examples, go to our quickstart.

  • While our Windows install is not always easy, Deeplearning4j is one of the few open-source deep learning projects that actually cares about trying to support the Windows community. Please see the Windows section of our ND4J page for more instructions.

  • Install MinGW 32 bits even if you have a 64-bit computer (the download button is on the upper right), and then download the Prebuilt dynamic libraries using Mingw.

  • Install Lapack. (Lapack will ask if you have Intel compilers. You do not.)

  • Lapack offers the alternative of VS Studio Solution. You’ll also want to look at the documentation for Basic Linear Algebra Subprograms (BLAS).

  • Alternatively, you can bypass MinGW and copy the Blas dll files to a folder in your PATH. For example, the path to the MinGW bin folder is: /usr/x86_64-w64-mingw32/sys-root/mingw/bin. To read more about the PATH variable in Windows, please read the top answer on this StackOverflow page.

  • Cygwin is not supported. You must install DL4J from DOS Windows.

  • Running this file, WindowsInfo.bat, can help debug your Windows install. Here’s one example of its output that shows what to expect. First download it, then open a command window / terminal. cd to the directory to which it was dowloaded. Enter WindowsInfo and hit enter. To copy its output, right click on command window -> select all -> hit enter. Output is then on clipboard.

For OpenBlas (see below) on Windows, download this file. Extract to somewhere such as C:/BLAS. Add that directory to your system’s PATH environment variable.


To make sure the native libs on the nd4j-native backend work, you need /opt/OpenBLAS/lib on the system path. After that, enter these commands in the prompt

		sudo cp
		sudo cp

We added this so that Spark would work with OpenBlas.

If OpenBlas is not working correctly, follow these steps:

  • Remove Openblas if you installed it.
  • Run sudo apt-get remove libopenblas-base
  • Download the development version of OpenBLAS
  • git clone git://
  • cd OpenBLAS
  • make FC=gfortran
  • sudo make PREFIX=/usr/local/ install
  • As a last step, restart your IDE.

  • For a complete review of requirements, see our getting started page.