Install DL Workbench from the Intel® Distribution for OpenVINO™ Toolkit Package

You can install the DL Workbench by installing the OpenVINO™ toolkit package that contains the tool.

NOTE: This installation option works on Linux systems only. The recommended installation option is installing from Docker Hub.

Installation Prerequisites

Install OpenVINO™ Package

Download the Intel® Distribution of OpenVINO™ toolkit for Linux* OS from the download page (registration required).

Follow the instructions to Install the Intel® Distribution of OpenVINO™ toolkit for Linux*.

NOTE: Do not remove the archive you used to install the OpenVINO™ toolkit package, because the archive is required for running the DL Workbench.

Install Docker* CE

To install Docker CE for Linux, follow the steps described in the Docker documentation (registration required).

Run DL Workbench

Follow the steps below:

  1. Open a terminal in the DL Workbench folder. The path to the folder is /<path_to_installed_package>/deployment_tools/tools/workbench.
  2. Build the DL Workbench image specifying the path to the downloaded archive with the OpenVINO™ toolkit package:
    ./build_docker.sh -PACKAGE_PATH <path-to-archive>

    NOTE: To see the list of available arguments, run the following command:

    ./build_docker.sh --help
    This command sets up the environment for the DL Workbench Docker* image and builds the image.
  3. Run the DL Workbench Docker container with the following command:
    ./start_workbench.sh -IMAGE_NAME workbench

    NOTE: To see the list of available arguments, run the command below:

    ./start_workbench.sh --help
  4. Once the http://127.0.0.1:5665 link followed by a generated token appears in the console log, open it in your browser. See Enter DL Workbench for details. The DL Workbench Get Started page appears:
    Get_Started_Page.png
    You have successfully installed the OpenVINO™ DL Workbench. Now you can move on to Work with Models and Sample Datasets.

Proxy Settings

If you are behind a corporate proxy, set environment variables with the the following command before running the DL Workbench:

export https_proxy=<https_proxy>
export http_proxy=<http_proxy>
export no_proxy=<no_proxy>