[GitHub] [incubator-mxnet] Rmwak commented on issue #14479: Unable to load mxnet-cu100 on Windows 10 after install

[GitBox <gi...@apache.org>]

Rmwak commented on issue #14479: Unable to load mxnet-cu100 on Windows 10 after install
URL: https://github.com/apache/incubator-mxnet/issues/14479#issuecomment-475213396
 
 
   Hello zachgk.
   
   The requirements in that link are:
   
   > Install with CUDA Support
   > The following steps will setup MXNet with CUDA. cuDNN can be enabled only when building from source.
   > 
   > Install Microsoft Visual Studio 2017 or Microsoft Visual Studio 2015.
   
   Microsoft Visual Studio 2017 is in my system
   
   > Download and install NVIDIA CUDA. CUDA versions 9.2 or 9.0 are recommended. Some issues with CUDA 9.1 have been identified in the past.
   
   I have CUDA 10.1 installed in the system, for the mxnet-cu100 version
   
   > Download and install NVIDIA_CUDA_DNN
   
   Already in place, version 7.5
   ```
   #define CUDNN_MAJOR 7
   #define CUDNN_MINOR 5
   #define CUDNN_PATCHLEVEL 0
   ```
   
   I'll add that I have installed and working in my system both pytorch(from pip) and tensorflow(from conda) in their gpu versions. For pytorch I had to install the nightly version since they had an issue with the windows version in the latest build.
   
   Not sure which .dll may mxnet be looking for : /
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services