Install ONNX Runtime for Radeon GPUs

Install ONNX Runtime for Radeon GPUs#

Refer to this section to install ONNX via the PIP installation method.

Overview#

Ensure that the following prerequisite installations are successful before proceeding to install ONNX Runtime for use with ROCm™ on Radeon™ GPUs.

Prerequisites#

  • Radeon Software for Linux (with ROCm) is installed.

  • MIGraphX is installed. This enables ONNX Runtime to build the correct MIGraphX EP.

  • The half library is installed.

    NOTE The half library is installed with MIGraphX. If the half library install is not verified, use the following install command:

    $ sudo apt install half

  • If the prerequisite installations are successful, proceed to install ONNX Runtime.

    NOTE Unless adding custom features, use the pre-built Python wheel files provided in the PIP installation method.

Install ONNX Runtime#

Important!

  • Use the provided pre-built Python wheel files from the PIP installation method, unless adding custom features.

  • The wheel file contains the MIGraphX and ROCm Execution Providers (EP). Refer to Install MIGraphX for ONNX RT for more information.

  • Refer to ONNX Runtime Documentation for additional information on ONNX Runtime topics.

  • See ONNX Runtime Tutorials to try out real applications and tutorials on how to get started.

AMD recommends the PIP install method to create an ONNX Runtime environment when working with ROCm for machine learning development.

To install via PIP,

Enter this command to download and install the ONNX Runtime wheel.

pip3 install https://repo.radeon.com/rocm/manylinux/rocm-rel-6.0.2/onnxruntime_rocm-inference-1.17.0-cp310-cp310-linux_x86_64.whl

Next, verify your ONNX Runtime installation.

Option B: Build from source for your environment, followed by local wheel file installation (Advanced)

Use this method for advanced customization usecases. This requires the user install the desired ROCm and MIGraphX versions, and creation of softlink prior to starting the build.

NOTE The build time typically takes ~45 minutes.

Prerequisites to build ONNX from source

To build from source,

  1. Verify that MIGraphX is installed with the half library.

    $ dpkg -l | grep migraphx
$ dpkg -l | grep half

    Expected result:

    root@aus-navi3x-02:/workspace/AMDMIGraphX# dpkg -l | grep migraphx
ii  migraphx                      2.9.0                             amd64        AMD's graph optimizer
ii  migraphx-dev                  2.9.0                             amd64        AMD's graph optimizer
ii  migraphx-tests                2.9.0                             amd64        AMD's graph opt


$ dpkg -l | grep half
ii  half                          1.12.0.60000-91~20.04             amd64        HALF-PRECISION FLOATING POINT LIBRARY

    NOTE Versions may vary between ROCm builds and installed versions of MIGraphX, but the desired result is the same.

    Enter the following install command if the half library is not installed with MIGraphX:

    $ dpkg -l | grep migraphx
$ dpkg -l | grep half

  2. Run rocm-smi to ensure there’s an Onnxruntime supported GPU in the system.

    $ rocm-smi

    Expected result:

    ======================================= ROCm System Management Interface =======================================
================================================= Concise Info =================================================
Device  [Model : Revision]    Temp    Power  Partitions      SCLK     MCLK   Fan     Perf  PwrCap  VRAM%  GPU%
        Name (20 chars)       (Edge)  (Avg)  (Mem, Compute)
================================================================================================================
0       [0x0e0d : 0x00]       32.0°C  73.0W  N/A, N/A        1526Mhz  96Mhz  31.76%  auto  241.0W    0%   50%
        0x7448
================================================================================================================
============================================= End of ROCm SMI Log ==============================================

  3. Clone the Onnxruntime repository into the root directory.

    cd /
git clone https://github.com/microsoft/onnxruntime.git

  4. Run the build and install script from the MIGraphX project root directory.

    cd ~
git clone https://github.com/ROCm/AMDMIGraphX.git
cd AMDMIGraphX
sudo tools/build_and_test_onnxrt.sh

    This builds Onnxruntime and adds ROCm and MIGraphX EP support to the Onnxruntime interface.

  5. Install ONNX Runtime once the build is completed and unit tests have run on the binary.

    $ pip3 install /onnxruntime/build/Linux/Release/dist/*.whl

Next, verify your ONNX Runtime installation.

Verify ONNX Runtime installation#

Confirm if ONNX Runtime is correctly installed.

$ python3
$ import onnxruntime as ort
  ort.get_available_providers()

Expected result: The following EPs are displayed.

>>> ort.get_available_providers()
['MIGraphXExecutionProvider', 'ROCMExecutionProvider', 'CPUExecutionProvider']

Environment set-up is complete, and the system is ready for use with ONNX Runtime to work with machine learning models, and algorithms.

Contents