View on GitHub ROCm_Logo

ROCm, a New Era in Open GPU Computing

Platform for GPU-Enabled HPC and Ultrascale Computing

Deep Learning on ROCm

Announcing our new Foundation for Deep Learning acceleration MIOpen 1.0 which introduces support for Convolution Neural Network (CNN) acceleration — built to run on top of the ROCm software stack!

This release includes the following:

Porting from cuDNN to MIOpen

The porting guide highlights the key differences between the current cuDNN and MIOpen APIs.

The ROCm 1.6 has prebuilt packages for MIOpen

Install the ROCm MIOpen implementation (assuming you already have the ‘rocm’ and ‘rocm-opencl-dev” package installed):

For just OpenCL development

sudo apt-get install miopengemm miopen-opencl

For HIP development

sudo apt-get install miopengemm miopen-hip

Or you can build from source code

Deep Learning Framework support for ROCm

Framework Status MIOpen Enabled Upstreamed   Current Repository
Caffe       Public     Yes                          https://github.com/ROCmSoftwarePlatform/hipCaffe
Tensorflow   Development Yes             CLA in Progress Notes: Working on NCCL and XLA enablement, Running
Caffe2       Upstreaming Yes CLA in Progress https://github.com/ROCmSoftwarePlatform/caffe2
Torch HIP   Upstreaming Development In process https://github.com/ROCmSoftwarePlatform/cutorch_hip
HIPnn        Upstreaming Development                    https://github.com/ROCmSoftwarePlatform/cunn_hip
PyTorch      Development Development    
MxNet       Development Development                 https://github.com/ROCmSoftwarePlatform/mxnet  
CNTK       Development Development                              

Tutorials

hipCaffe