From b2f659b1b5ae4414057e2160d440c18cd5e284af Mon Sep 17 00:00:00 2001 From: vrakesh Date: Thu, 20 Sep 2018 11:53:48 -0700 Subject: [PATCH] Update and modify Windows docs 1. Removed sudo from not necessaary locations, since it is a windows doc. 2. Added CUDNN setup information link(s). --- docs/install/windows_setup.md | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/docs/install/windows_setup.md b/docs/install/windows_setup.md index 01cd46658a7b..87fd1cc07d8f 100755 --- a/docs/install/windows_setup.md +++ b/docs/install/windows_setup.md @@ -77,6 +77,7 @@ When using supported NVIDIA GPU hardware, inference and training can be vastly f The following steps will setup MXNet with CUDA. cuDNN can be enabled only when building from source. 1. Install [Microsoft Visual Studio 2017](https://www.visualstudio.com/downloads/) or [Microsoft Visual Studio 2015](https://www.visualstudio.com/vs/older-downloads/). 1. Download and install [NVIDIA CUDA](https://developer.nvidia.com/cuda-downloads?target_os=Windows&target_arch=x86_64&target_version=10&target_type=exelocal). CUDA versions 9.2 or 9.0 are recommended. Some [issues with CUDA 9.1](https://github.com/apache/incubator-mxnet/labels/CUDA) have been identified in the past. +1. Download and install [NVIDIA_CUDA_DNN](https://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#install-windows) 1. Install MXNet with CUDA support with pip: ```bash @@ -93,6 +94,7 @@ The following steps will setup MXNet with CUDA and MKL. 1. Install [Microsoft Visual Studio 2017](https://www.visualstudio.com/downloads/) or [Microsoft Visual Studio 2015](https://www.visualstudio.com/vs/older-downloads/). 1. Download and install [Intel MKL](https://software.intel.com/en-us/mkl/choose-download/windows) (registration required). 1. Download and install [NVIDIA CUDA](https://developer.nvidia.com/cuda-downloads?target_os=Windows&target_arch=x86_64&target_version=10&target_type=exelocal). +1. Download and install [NVIDIA_CUDA_DNN](https://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html#install-windows) 1. Install MXNet with MKL support with pip: ```bash @@ -163,7 +165,6 @@ cd C:\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 -DCUDA_ARCH_LIST=Common -DCUDA_TOOLSET=9.2 -DCUDNN_INCLUDE=C:\cuda\include -DCUDNN_LIBRARY=C:\cuda\lib\x64\cudnn.lib "C:\incubator-mxnet" ``` -**Note**: you may add to the cmake compilation options the compiler version to use with: `-T version=14.11` 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 @@ -216,7 +217,7 @@ These steps are required after building from source. If you already installed MX ```bash # Assuming you are in root mxnet source code folder cd python - sudo python setup.py install + python setup.py install ``` Done! We have installed MXNet with Python interface.