Skip to content

Latest commit

 

History

History
50 lines (37 loc) · 1.39 KB

openvino-setup-ubuntu.md

File metadata and controls

50 lines (37 loc) · 1.39 KB

OpenVINO Install on Ubuntu 20.04

Step 1: Install NEO driver

Install the Intel Graphics Compute Runtime for oneAPI Level Zero and OpenCL™ Driver for running deep learning inference with OpenVINO in Intel iGPU/dGPU.

wget https://raw.githubusercontent.com/openvinotoolkit/openvino/master/scripts/install_dependencies/install_NEO_OCL_driver.sh

chmod +x ./install_NEO_OCL_driver.sh

sudo ./install_NEO_OCL_driver.sh

sudo usermod -aG video $USER
sudo usermod -aG render $USER

# Verify install by running:
clinfo

Step 2: Install OpenVINO

  • Instal OpenVINO in a python virtual environment. See official instructions HERE.
  • Python3.8 is recommended.
python3 -m venv openvino_env
source openvino_env/bin/activate

python -m pip install --upgrade pip
pip install openvino-dev[onnx,tensorflow2,pytorch]

Step 3: Benchmark Sample model

  • Download googlenet-v1-tf model from OpenVINO Model Zoo (OMZ)
omz_downloader --name googlenet-v1-tf
  • Convert googlenet-v1-tf model to OpenVINO IR
omz_converter --name googlenet-v1-tf
benchmark_app \
-m public/googlenet-v1-tf/FP32/googlenet-v1-tf.xml \
-hint throughput