From bfd9012e80d440a2ab46da2eb1064d7871734860 Mon Sep 17 00:00:00 2001 From: stu1130 Date: Fri, 28 Sep 2018 11:04:34 -0700 Subject: [PATCH 01/12] update the installation document --- docs/install/windows_setup.md | 36 ++++++++++++++++++++++++++--------- 1 file changed, 27 insertions(+), 9 deletions(-) diff --git a/docs/install/windows_setup.md b/docs/install/windows_setup.md index 009679d39e2f..c5a9686ff55d 100755 --- a/docs/install/windows_setup.md +++ b/docs/install/windows_setup.md @@ -137,14 +137,14 @@ To build and install MXNet yourself using [VS2017](https://www.visualstudio.com/ ``` 1. Download and install [CMake](https://cmake.org/download) if it is not already installed. [CMake v3.12.2](https://cmake.org/files/v3.12/cmake-3.12.2-win64-x64.msi) has been tested with MXNet. 1. Download and run the [OpenCV](https://sourceforge.net/projects/opencvlibrary/files/opencv-win/3.4.1/opencv-3.4.1-vc14_vc15.exe/download) package. There are more recent versions of OpenCV, so please create an issue/PR to update this info if you validate one of these later versions. -1. This will unzip several files. You can place them in another directory if you wish. -1. Set the environment variable `OpenCV_DIR` to point to the OpenCV build directory that you just unzipped (e.g., `OpenCV_DIR = C:\utils\opencv\build`). +1. This will unzip several files. You can place them in another directory if you wish. We will use `C:\utils` as our default path. +1. Set the environment variable `OpenCV_DIR` to point to the OpenCV build directory that you just unzipped. Start ```cmd``` and type `set OpenCV_DIR=C:\utils\opencv\build`. 1. If you don’t have the Intel Math Kernel Library (MKL) installed, you can install it and follow the [MKLDNN_README](https://github.com/apache/incubator-mxnet/blob/master/MKLDNN_README.md) from here, or you can use OpenBLAS. These instructions will assume you're using OpenBLAS. 1. Download the [OpenBlas](https://sourceforge.net/projects/openblas/files/v0.2.19/OpenBLAS-v0.2.19-Win64-int32.zip/download) package. Later versions of OpenBLAS are available, but you would need to build from source. v0.2.19 is the most recent version that ships with binaries. Contributions of more recent binaries would be appreciated. 1. Unzip the file. You can place the unzipped files and folders in another directory if you wish. -1. Set the environment variable `OpenBLAS_HOME` to point to the OpenBLAS directory that contains the `include` and `lib` directories (e.g., `OpenBLAS_HOME = C:\utils\OpenBLAS`). -1. Download and install [CUDA](https://developer.nvidia.com/cuda-downloads?target_os=Windows&target_arch=x86_64&target_version=10&target_type=exelocal). If you already had CUDA, then installed VS2017, you should reinstall CUDA now so that you get the CUDA toolkit components for VS2017 integration. -1. Download and install cuDNN. To get access to the download link, register as an NVIDIA community user. Then Follow the [link](http://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#install-windows) to install the cuDNN. +1. Set the environment variable `OpenBLAS_HOME` to point to the OpenBLAS directory that contains the `include` and `lib` directories and type `set OpenBLAS_HOME=C:\utils\OpenBLAS` on the command prompt. +1. Download and install [CUDA](https://developer.nvidia.com/cuda-downloads?target_os=Windows&target_arch=x86_64&target_version=10&target_type=exelocal). If you already had CUDA, then installed VS2017, you should reinstall CUDA now so that you get the CUDA toolkit components for VS2017 integration. Note that the lastest CUDA version supported by MXNet is 9.2. You can find it on the `Legacy Releases`. +1. Download and install cuDNN. To get access to the download link, register as an NVIDIA community user. Then Follow the [link](http://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#install-windows) to install the cuDNN and put it into ```C:\cuda```. 1. Download and install [git](https://git-for-windows.github.io/) if you haven't already. After you have installed all of the required dependencies, build the MXNet source code: @@ -158,12 +158,12 @@ git clone https://github.com/apache/incubator-mxnet.git --recursive 3. Verify that the `DCUDNN_INCLUDE` and `DCUDNN_LIBRARY` environment variables are pointing to the `include` folder and `cudnn.lib` file of your CUDA installed location, and `C:\incubator-mxnet` is the location of the source code you just cloned in the previous step. 4. Create a build dir using the following command and go to the directory, for example: ``` -mkdir C:\build -cd C:\build +mkdir C:\incubator-mxnet\build +cd C:\incubator-mxnet\build ``` 5. Compile the MXNet source code with `cmake` by using following command: ``` -cmake -G "Visual Studio 15 2017 Win64" -T cuda=9.2,host=x64 -DUSE_CUDA=1 -DUSE_CUDNN=1 -DUSE_NVRTC=1 -DUSE_OPENCV=1 -DUSE_OPENMP=1 -DUSE_BLAS=open -DUSE_LAPACK=1 -DUSE_DIST_KVSTORE=0 -DCUDA_ARCH_LIST=Common -DCUDA_TOOLSET=9.2 -DCUDNN_INCLUDE=C:\cuda\include -DCUDNN_LIBRARY=C:\cuda\lib\x64\cudnn.lib "C:\incubator-mxnet" +cmake -G "Visual Studio 15 2017 Win64" -T cuda=9.2,host=x64 -DUSE_CUDA=1 -DUSE_CUDNN=1 -DUSE_NVRTC=1 -DUSE_OPENCV=1 -DUSE_OPENMP=1 -DUSE_BLAS=open -DUSE_LAPACK=1 -DUSE_DIST_KVSTORE=0 -DUSE_MKL_IF_AVAILABLE=OFF -DCUDA_ARCH_LIST=Common -DCUDA_TOOLSET=9.2 -DCUDNN_INCLUDE=C:\cuda\include -DCUDNN_LIBRARY=C:\cuda\lib\x64\cudnn.lib "C:\incubator-mxnet" ``` 6. After the CMake successfully completed, compile the the MXNet source code by using following command: ``` @@ -212,7 +212,25 @@ These steps are required after building from source. If you already installed MX 1. Install ```Python``` using windows installer available [here](https://www.python.org/downloads/release/python-2712/). 2. Install ```Numpy``` using windows installer available [here](https://scipy.org/index.html). -3. Next, we install Python package interface for MXNet. You can find the Python interface package for [MXNet on GitHub](https://github.com/dmlc/mxnet/tree/master/python/mxnet). +3. Start ```cmd``` and create a folder named ```common```(```mkdir C:\common```) +4. Download the [mingw64_dll.zip](https://sourceforge.net/projects/openblas/files/v0.2.12/mingw64_dll.zip/download), unzip and copy three libraries (.dll files) that openblas.dll depends on to ```C:\common```. +5. Copy the required .dll file to ```C:\common``` and make sure following libraries (.dll files) in the folder. +``` +libgcc_s_seh-1.dll (in mingw64_dll) +libgfortran-3.dll (in mingw64_dll) +libquadmath-0.dll (in mingw64_dll) +libopenblas.dll (in OpenBlas folder you download) +opencv_world341.dll (in OpenCV folder you download) +kernel32.dll +vcomp140.dll +nvcuda.dll (only if you use CUDA) +``` +6. Add ```C:\common``` to Environment Variables. +* Type ```control sysdm.cpl``` on ```cmp``` +* Select the **Advanced tab** and click **Environment Variables** +* Double click the **Path** and click **New** +* Add ```C:\common``` and click OK +7. Next, we install Python package interface for MXNet. You can find the Python interface package for [MXNet on GitHub](https://github.com/dmlc/mxnet/tree/master/python/mxnet). ```bash # Assuming you are in root mxnet source code folder From a758d615de4e48f28e4c43f85a00a6dcf3297eef Mon Sep 17 00:00:00 2001 From: stu1130 Date: Fri, 28 Sep 2018 11:18:42 -0700 Subject: [PATCH 02/12] fix minor text --- docs/install/windows_setup.md | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/docs/install/windows_setup.md b/docs/install/windows_setup.md index c5a9686ff55d..2626ab3dfb67 100755 --- a/docs/install/windows_setup.md +++ b/docs/install/windows_setup.md @@ -142,9 +142,9 @@ To build and install MXNet yourself using [VS2017](https://www.visualstudio.com/ 1. If you don’t have the Intel Math Kernel Library (MKL) installed, you can install it and follow the [MKLDNN_README](https://github.com/apache/incubator-mxnet/blob/master/MKLDNN_README.md) from here, or you can use OpenBLAS. These instructions will assume you're using OpenBLAS. 1. Download the [OpenBlas](https://sourceforge.net/projects/openblas/files/v0.2.19/OpenBLAS-v0.2.19-Win64-int32.zip/download) package. Later versions of OpenBLAS are available, but you would need to build from source. v0.2.19 is the most recent version that ships with binaries. Contributions of more recent binaries would be appreciated. 1. Unzip the file. You can place the unzipped files and folders in another directory if you wish. -1. Set the environment variable `OpenBLAS_HOME` to point to the OpenBLAS directory that contains the `include` and `lib` directories and type `set OpenBLAS_HOME=C:\utils\OpenBLAS` on the command prompt. +1. Set the environment variable `OpenBLAS_HOME` to point to the OpenBLAS directory that contains the `include` and `lib` directories and type `set OpenBLAS_HOME=C:\utils\OpenBLAS` on the command prompt(```cmd```). 1. Download and install [CUDA](https://developer.nvidia.com/cuda-downloads?target_os=Windows&target_arch=x86_64&target_version=10&target_type=exelocal). If you already had CUDA, then installed VS2017, you should reinstall CUDA now so that you get the CUDA toolkit components for VS2017 integration. Note that the lastest CUDA version supported by MXNet is 9.2. You can find it on the `Legacy Releases`. -1. Download and install cuDNN. To get access to the download link, register as an NVIDIA community user. Then Follow the [link](http://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#install-windows) to install the cuDNN and put it into ```C:\cuda```. +1. Download and install cuDNN. To get access to the download link, register as an NVIDIA community user. Then Follow the [link](http://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#install-windows) to install the cuDNN and put those libraries into ```C:\cuda```. 1. Download and install [git](https://git-for-windows.github.io/) if you haven't already. After you have installed all of the required dependencies, build the MXNet source code: @@ -226,10 +226,10 @@ vcomp140.dll nvcuda.dll (only if you use CUDA) ``` 6. Add ```C:\common``` to Environment Variables. -* Type ```control sysdm.cpl``` on ```cmp``` -* Select the **Advanced tab** and click **Environment Variables** -* Double click the **Path** and click **New** -* Add ```C:\common``` and click OK + * Type ```control sysdm.cpl``` on ```cmp``` + * Select the **Advanced tab** and click **Environment Variables** + * Double click the **Path** and click **New** + * Add ```C:\common``` and click OK 7. Next, we install Python package interface for MXNet. You can find the Python interface package for [MXNet on GitHub](https://github.com/dmlc/mxnet/tree/master/python/mxnet). ```bash From 69d535847d3dbbfc6a4c59329eb20950bb3f7788 Mon Sep 17 00:00:00 2001 From: stu1130 Date: Fri, 28 Sep 2018 14:18:57 -0700 Subject: [PATCH 03/12] update the rename process --- docs/install/windows_setup.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/install/windows_setup.md b/docs/install/windows_setup.md index 2626ab3dfb67..eee65c36b396 100755 --- a/docs/install/windows_setup.md +++ b/docs/install/windows_setup.md @@ -141,7 +141,7 @@ To build and install MXNet yourself using [VS2017](https://www.visualstudio.com/ 1. Set the environment variable `OpenCV_DIR` to point to the OpenCV build directory that you just unzipped. Start ```cmd``` and type `set OpenCV_DIR=C:\utils\opencv\build`. 1. If you don’t have the Intel Math Kernel Library (MKL) installed, you can install it and follow the [MKLDNN_README](https://github.com/apache/incubator-mxnet/blob/master/MKLDNN_README.md) from here, or you can use OpenBLAS. These instructions will assume you're using OpenBLAS. 1. Download the [OpenBlas](https://sourceforge.net/projects/openblas/files/v0.2.19/OpenBLAS-v0.2.19-Win64-int32.zip/download) package. Later versions of OpenBLAS are available, but you would need to build from source. v0.2.19 is the most recent version that ships with binaries. Contributions of more recent binaries would be appreciated. -1. Unzip the file. You can place the unzipped files and folders in another directory if you wish. +1. Unzip the file, rename it to ```OpenBLAS``` and put it under `C:\utils`. You can place the unzipped files and folders in another directory if you wish. 1. Set the environment variable `OpenBLAS_HOME` to point to the OpenBLAS directory that contains the `include` and `lib` directories and type `set OpenBLAS_HOME=C:\utils\OpenBLAS` on the command prompt(```cmd```). 1. Download and install [CUDA](https://developer.nvidia.com/cuda-downloads?target_os=Windows&target_arch=x86_64&target_version=10&target_type=exelocal). If you already had CUDA, then installed VS2017, you should reinstall CUDA now so that you get the CUDA toolkit components for VS2017 integration. Note that the lastest CUDA version supported by MXNet is 9.2. You can find it on the `Legacy Releases`. 1. Download and install cuDNN. To get access to the download link, register as an NVIDIA community user. Then Follow the [link](http://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#install-windows) to install the cuDNN and put those libraries into ```C:\cuda```. From daec3f225b201d7eab16a5c02186a89fd02c243e Mon Sep 17 00:00:00 2001 From: stu1130 Date: Fri, 28 Sep 2018 14:28:32 -0700 Subject: [PATCH 04/12] fix wording --- docs/install/windows_setup.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/install/windows_setup.md b/docs/install/windows_setup.md index eee65c36b396..760adeadd2f5 100755 --- a/docs/install/windows_setup.md +++ b/docs/install/windows_setup.md @@ -144,7 +144,7 @@ To build and install MXNet yourself using [VS2017](https://www.visualstudio.com/ 1. Unzip the file, rename it to ```OpenBLAS``` and put it under `C:\utils`. You can place the unzipped files and folders in another directory if you wish. 1. Set the environment variable `OpenBLAS_HOME` to point to the OpenBLAS directory that contains the `include` and `lib` directories and type `set OpenBLAS_HOME=C:\utils\OpenBLAS` on the command prompt(```cmd```). 1. Download and install [CUDA](https://developer.nvidia.com/cuda-downloads?target_os=Windows&target_arch=x86_64&target_version=10&target_type=exelocal). If you already had CUDA, then installed VS2017, you should reinstall CUDA now so that you get the CUDA toolkit components for VS2017 integration. Note that the lastest CUDA version supported by MXNet is 9.2. You can find it on the `Legacy Releases`. -1. Download and install cuDNN. To get access to the download link, register as an NVIDIA community user. Then Follow the [link](http://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#install-windows) to install the cuDNN and put those libraries into ```C:\cuda```. +1. Download and install cuDNN. To get access to the download link, register as an NVIDIA community user. Then follow the [link](http://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#install-windows) to install the cuDNN and put those libraries into ```C:\cuda```. 1. Download and install [git](https://git-for-windows.github.io/) if you haven't already. After you have installed all of the required dependencies, build the MXNet source code: From 2bb79b0785d41c9275ee0388ddee6f02ef514d7e Mon Sep 17 00:00:00 2001 From: stu1130 Date: Fri, 28 Sep 2018 15:13:10 -0700 Subject: [PATCH 05/12] remind users. of env and vs version --- docs/install/windows_setup.md | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/docs/install/windows_setup.md b/docs/install/windows_setup.md index 760adeadd2f5..b06b7f93869a 100755 --- a/docs/install/windows_setup.md +++ b/docs/install/windows_setup.md @@ -165,6 +165,7 @@ cd C:\incubator-mxnet\build ``` cmake -G "Visual Studio 15 2017 Win64" -T cuda=9.2,host=x64 -DUSE_CUDA=1 -DUSE_CUDNN=1 -DUSE_NVRTC=1 -DUSE_OPENCV=1 -DUSE_OPENMP=1 -DUSE_BLAS=open -DUSE_LAPACK=1 -DUSE_DIST_KVSTORE=0 -DUSE_MKL_IF_AVAILABLE=OFF -DCUDA_ARCH_LIST=Common -DCUDA_TOOLSET=9.2 -DCUDNN_INCLUDE=C:\cuda\include -DCUDNN_LIBRARY=C:\cuda\lib\x64\cudnn.lib "C:\incubator-mxnet" ``` +* Make sure you set the environment variables correctly (OpenBLAS_HOME, OpenCV_DIR) and change the version of the Visual studio 2017 to v14.11 before enter above command. 6. After the CMake successfully completed, compile the the MXNet source code by using following command: ``` msbuild mxnet.sln /p:Configuration=Release;Platform=x64 /maxcpucount @@ -221,7 +222,7 @@ libgfortran-3.dll (in mingw64_dll) libquadmath-0.dll (in mingw64_dll) libopenblas.dll (in OpenBlas folder you download) opencv_world341.dll (in OpenCV folder you download) -kernel32.dll +kernel32.dll (in C:\Windows\System32) vcomp140.dll nvcuda.dll (only if you use CUDA) ``` From f8ce81c051b6cde023babc0df9a24a8e477b3a88 Mon Sep 17 00:00:00 2001 From: stu1130 Date: Fri, 28 Sep 2018 16:19:17 -0700 Subject: [PATCH 06/12] leave only the required dll --- docs/install/windows_setup.md | 3 --- 1 file changed, 3 deletions(-) diff --git a/docs/install/windows_setup.md b/docs/install/windows_setup.md index b06b7f93869a..8e0a71dad495 100755 --- a/docs/install/windows_setup.md +++ b/docs/install/windows_setup.md @@ -222,9 +222,6 @@ libgfortran-3.dll (in mingw64_dll) libquadmath-0.dll (in mingw64_dll) libopenblas.dll (in OpenBlas folder you download) opencv_world341.dll (in OpenCV folder you download) -kernel32.dll (in C:\Windows\System32) -vcomp140.dll -nvcuda.dll (only if you use CUDA) ``` 6. Add ```C:\common``` to Environment Variables. * Type ```control sysdm.cpl``` on ```cmp``` From 5692c4f6817a835acd3848e41e1f2d7f7264c2d7 Mon Sep 17 00:00:00 2001 From: stu1130 Date: Mon, 1 Oct 2018 13:49:48 -0700 Subject: [PATCH 07/12] fix the link --- docs/install/windows_setup.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/docs/install/windows_setup.md b/docs/install/windows_setup.md index 8e0a71dad495..17e5b124488c 100755 --- a/docs/install/windows_setup.md +++ b/docs/install/windows_setup.md @@ -163,7 +163,7 @@ cd C:\incubator-mxnet\build ``` 5. Compile the MXNet source code with `cmake` by using following command: ``` -cmake -G "Visual Studio 15 2017 Win64" -T cuda=9.2,host=x64 -DUSE_CUDA=1 -DUSE_CUDNN=1 -DUSE_NVRTC=1 -DUSE_OPENCV=1 -DUSE_OPENMP=1 -DUSE_BLAS=open -DUSE_LAPACK=1 -DUSE_DIST_KVSTORE=0 -DUSE_MKL_IF_AVAILABLE=OFF -DCUDA_ARCH_LIST=Common -DCUDA_TOOLSET=9.2 -DCUDNN_INCLUDE=C:\cuda\include -DCUDNN_LIBRARY=C:\cuda\lib\x64\cudnn.lib "C:\incubator-mxnet" +cmake -G "Visual Studio 15 2017 Win64" -T cuda=9.2,host=x64 -DUSE_CUDA=1 -DUSE_CUDNN=1 -DUSE_NVRTC=1 -DUSE_OPENCV=1 -DUSE_OPENMP=1 -DUSE_BLAS=open -DUSE_LAPACK=1 -DUSE_DIST_KVSTORE=0 -DCUDA_ARCH_LIST=Common -DCUDA_TOOLSET=9.2 -DCUDNN_INCLUDE=C:\cuda\include -DCUDNN_LIBRARY=C:\cuda\lib\x64\cudnn.lib "C:\incubator-mxnet" ``` * Make sure you set the environment variables correctly (OpenBLAS_HOME, OpenCV_DIR) and change the version of the Visual studio 2017 to v14.11 before enter above command. 6. After the CMake successfully completed, compile the the MXNet source code by using following command: @@ -207,7 +207,7 @@ We have installed MXNet core library. Next, we will install MXNet interface pack - [Julia](#install-the-mxnet-package-for-julia) - **Scala** is not yet available for Windows -## Install MXNet for Python +## Install the MXNet Package for Python These steps are required after building from source. If you already installed MXNet by using pip, you do not need to do these steps to use MXNet with Python. @@ -228,7 +228,7 @@ opencv_world341.dll (in OpenCV folder you download) * Select the **Advanced tab** and click **Environment Variables** * Double click the **Path** and click **New** * Add ```C:\common``` and click OK -7. Next, we install Python package interface for MXNet. You can find the Python interface package for [MXNet on GitHub](https://github.com/dmlc/mxnet/tree/master/python/mxnet). +7. Finally, we install Python package interface for MXNet. You can find the Python interface package for [MXNet on GitHub](https://github.com/dmlc/mxnet/tree/master/python/mxnet). ```bash # Assuming you are in root mxnet source code folder From bcc2eadc1479491c40aa3ef5f7da802e0f688867 Mon Sep 17 00:00:00 2001 From: stu1130 Date: Mon, 1 Oct 2018 13:57:17 -0700 Subject: [PATCH 08/12] update the R anchor --- docs/install/windows_setup.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/install/windows_setup.md b/docs/install/windows_setup.md index 17e5b124488c..efa1f57b8c4e 100755 --- a/docs/install/windows_setup.md +++ b/docs/install/windows_setup.md @@ -11,7 +11,7 @@ The following describes how to install with pip for computers with CPUs, Intel C - [Build from Source](#build-from-source) - Install MXNet with a Programming Language API - [Python](#install-the-mxnet-package-for-python) - - [R](#install-mxnet-package-for-r) + - [R](#install-the-mxnet-package-for-r) - [Julia](#install-the-mxnet-package-for-julia) From c7d189577fd739c818808845c00cde73a5c924fa Mon Sep 17 00:00:00 2001 From: stu1130 Date: Tue, 2 Oct 2018 14:47:26 -0700 Subject: [PATCH 09/12] refine the description of step 7 --- docs/install/windows_setup.md | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/docs/install/windows_setup.md b/docs/install/windows_setup.md index efa1f57b8c4e..5889aa2b1c6e 100755 --- a/docs/install/windows_setup.md +++ b/docs/install/windows_setup.md @@ -228,8 +228,7 @@ opencv_world341.dll (in OpenCV folder you download) * Select the **Advanced tab** and click **Environment Variables** * Double click the **Path** and click **New** * Add ```C:\common``` and click OK -7. Finally, we install Python package interface for MXNet. You can find the Python interface package for [MXNet on GitHub](https://github.com/dmlc/mxnet/tree/master/python/mxnet). - +7. Use setup.py to insatll the package ```bash # Assuming you are in root mxnet source code folder cd python From 0e1eb01dbd2eb78a747846d89d1cd2cff80562a9 Mon Sep 17 00:00:00 2001 From: stu1130 Date: Tue, 2 Oct 2018 14:49:23 -0700 Subject: [PATCH 10/12] add missing . --- docs/install/windows_setup.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/install/windows_setup.md b/docs/install/windows_setup.md index 5889aa2b1c6e..586ac9c6a032 100755 --- a/docs/install/windows_setup.md +++ b/docs/install/windows_setup.md @@ -228,7 +228,7 @@ opencv_world341.dll (in OpenCV folder you download) * Select the **Advanced tab** and click **Environment Variables** * Double click the **Path** and click **New** * Add ```C:\common``` and click OK -7. Use setup.py to insatll the package +7. Use setup.py to insatll the package. ```bash # Assuming you are in root mxnet source code folder cd python From bc48137ff33f005f0f499d9668b97c055f09cfd5 Mon Sep 17 00:00:00 2001 From: stu1130 Date: Tue, 2 Oct 2018 15:18:28 -0700 Subject: [PATCH 11/12] fix spelling --- docs/install/windows_setup.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/install/windows_setup.md b/docs/install/windows_setup.md index 586ac9c6a032..48d82760b373 100755 --- a/docs/install/windows_setup.md +++ b/docs/install/windows_setup.md @@ -228,7 +228,7 @@ opencv_world341.dll (in OpenCV folder you download) * Select the **Advanced tab** and click **Environment Variables** * Double click the **Path** and click **New** * Add ```C:\common``` and click OK -7. Use setup.py to insatll the package. +7. Use setup.py to install the package. ```bash # Assuming you are in root mxnet source code folder cd python From 7a43d0d9777aaddf35576aa4517f67a1047a0813 Mon Sep 17 00:00:00 2001 From: stu1130 Date: Thu, 4 Oct 2018 16:31:15 -0700 Subject: [PATCH 12/12] update links and fix wording --- docs/install/windows_setup.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/install/windows_setup.md b/docs/install/windows_setup.md index 48d82760b373..8e612454a7e1 100755 --- a/docs/install/windows_setup.md +++ b/docs/install/windows_setup.md @@ -137,13 +137,13 @@ To build and install MXNet yourself using [VS2017](https://www.visualstudio.com/ ``` 1. Download and install [CMake](https://cmake.org/download) if it is not already installed. [CMake v3.12.2](https://cmake.org/files/v3.12/cmake-3.12.2-win64-x64.msi) has been tested with MXNet. 1. Download and run the [OpenCV](https://sourceforge.net/projects/opencvlibrary/files/opencv-win/3.4.1/opencv-3.4.1-vc14_vc15.exe/download) package. There are more recent versions of OpenCV, so please create an issue/PR to update this info if you validate one of these later versions. -1. This will unzip several files. You can place them in another directory if you wish. We will use `C:\utils` as our default path. +1. This will unzip several files. You can place them in another directory if you wish. We will use `C:\utils`(```mkdir C:\utils```) as our default path. 1. Set the environment variable `OpenCV_DIR` to point to the OpenCV build directory that you just unzipped. Start ```cmd``` and type `set OpenCV_DIR=C:\utils\opencv\build`. 1. If you don’t have the Intel Math Kernel Library (MKL) installed, you can install it and follow the [MKLDNN_README](https://github.com/apache/incubator-mxnet/blob/master/MKLDNN_README.md) from here, or you can use OpenBLAS. These instructions will assume you're using OpenBLAS. 1. Download the [OpenBlas](https://sourceforge.net/projects/openblas/files/v0.2.19/OpenBLAS-v0.2.19-Win64-int32.zip/download) package. Later versions of OpenBLAS are available, but you would need to build from source. v0.2.19 is the most recent version that ships with binaries. Contributions of more recent binaries would be appreciated. 1. Unzip the file, rename it to ```OpenBLAS``` and put it under `C:\utils`. You can place the unzipped files and folders in another directory if you wish. 1. Set the environment variable `OpenBLAS_HOME` to point to the OpenBLAS directory that contains the `include` and `lib` directories and type `set OpenBLAS_HOME=C:\utils\OpenBLAS` on the command prompt(```cmd```). -1. Download and install [CUDA](https://developer.nvidia.com/cuda-downloads?target_os=Windows&target_arch=x86_64&target_version=10&target_type=exelocal). If you already had CUDA, then installed VS2017, you should reinstall CUDA now so that you get the CUDA toolkit components for VS2017 integration. Note that the lastest CUDA version supported by MXNet is 9.2. You can find it on the `Legacy Releases`. +1. Download and install [CUDA](https://developer.nvidia.com/cuda-downloads?target_os=Windows&target_arch=x86_64&target_version=10&target_type=exelocal). If you already had CUDA, then installed VS2017, you should reinstall CUDA now so that you get the CUDA toolkit components for VS2017 integration. Note that the latest CUDA version supported by MXNet is [9.2](https://developer.nvidia.com/cuda-92-download-archive). You might also want to find other CUDA verion on the [Legacy Releases](https://developer.nvidia.com/cuda-toolkit-archive). 1. Download and install cuDNN. To get access to the download link, register as an NVIDIA community user. Then follow the [link](http://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#install-windows) to install the cuDNN and put those libraries into ```C:\cuda```. 1. Download and install [git](https://git-for-windows.github.io/) if you haven't already.