3/31 LAB - HackMD

--- title: 3/31 LAB tags: mess description: OpenVINO and Intel Movidius (AI Accelerator) --- # Install Intel® OpenVINO™ toolkit ## register & download reference website： https://software.intel.com/content/www/us/en/develop/tools/openvino-toolkit/choose-download.html ![](https://i.imgur.com/JGNGoDF.png) ![](https://i.imgur.com/oFCDq6R.png) reference website： https://docs.openvinotoolkit.org/latest/openvino_docs_install_guides_installing_openvino_linux.html ``` tar -xvzf l_openvino_toolkit_p_<version>.tgz sudo ./install.sh cd /opt/intel/openvino_2020.3.341/install_dependencies sudo -E ./install_openvino_dependencies.sh source /opt/intel/openvino_2020.3.341/bin/setupvars.sh ``` ## download model ``` cd /opt/intel/openvino_2020.3.341/deployment_tools/tools/model_downloader/ python3 info_dumper.py --print_all sudo python3 ./downloader.py --name squeezenet1.1 --output_dir ~/models ``` ## Runs the Model Optimizer to convert the model to the IR Note that install_prerequisites scripts ``` aaeon@aaeon:/opt/intel/openvino_2020.3.341/deployment_tools/model_optimizer/install_prerequisites$ sudo ./install_prerequisites_caffe.sh ``` ``` cd /opt/intel/openvino_2020.3.341/deployment_tools/model_optimizer python3 ./mo.py --input_model ~/models/public/squeezenet1.1/squeezenet1.1.caffemodel --data_type FP32 --output_dir . ``` ## Builds the Image Classification Sample Async application. ``` cd /opt/intel/openvino_2020.3.341/deployment_tools/inference_engine/samples/cpp/ ./build_samples.sh cd ~/inference_engine_cpp_samples_build/intel64/Release ``` ## run inference with Accelerator - Runs the compiled sample with the car.png image located in the demo directory ![](https://i.imgur.com/ZUCMyXX.png) [Note] There are lots of demo scripts here ``` /opt/intel/openvino_2020.3.341/deployment_tools/demo ```