Skip to content

Latest commit

 

History

History
 
 

dl-workbench

README.md

OpenVINO™ Deep Learning Workbench

Table of Contents

Introduction

Deep Learning Workbench is an official OpenVINO™ graphical user interface designed to make the production of pre-trained deep learning models significantly easier.

The DL Workbench is an official UI environment of the OpenVINO™ toolkit that enables you to:

  • Learn what neural networks are, how they work, and how to analyze their architectures and performance.
  • Get familiar with the OpenVINO™ ecosystem and its main components without installing it on your system.
  • Measure and interpret model performance.
  • Analyze the quality of your model and visualize output.
  • Optimize your model and prepare it for deployment on the target system.

In the DL Workbench, you can use the following OpenVINO™ toolkit components:

Component Description
Open Model Zoo Get access to the collection of high-quality pre-trained deep learning public and Intel-trained models trained to resolve a variety of different tasks.
Model Optimizer Optimize and transform models trained in supported frameworks to the IR format.
Supported frameworks include TensorFlow*, Caffe*, Kaldi*, MXNet*, and ONNX* format.
Benchmark Tool Estimate deep learning model inference performance on supported devices.
Accuracy Checker Evaluate the accuracy of a model by collecting one or several metric values.
Post-Training Optimization Tool Optimize pre-trained models with lowering the precision of a model from floating-point precision(FP32 or FP16) to integer precision (INT8), without the need to retrain or fine-tune models.

System Requirements

The complete list of recommended requirements is available in the documentation.

To successfully run the DL Workbench with Python Starter, install Python 3.6 or higher.

Prerequisite Linux* Windows* macOS*
Operating system Ubuntu* 18.04 Windows* 10 macOS* 10.15 Catalina
Available RAM space 8 GB** 8 GB** 8 GB**
Available storage space 10 GB + space for imported artifacts 10 GB + space for imported artifacts 10 GB + space for imported artifacts
Docker* Docker CE 18.06.1 Docker Desktop 2.3.0.3 Docker CE 18.06.1

Windows*, Linux* and MacOS* support CPU targets. GPU, Intel® Neural Compute Stick 2 and Intel® Vision Accelerator Design with Intel® Movidius™ VPUs are supported only for Linux*.

Install DL Workbench

Install the DL Workbench Starter

LEGAL NOTICE: Your use of this software and any required dependent software (the “Software Package”) is subject to the terms and conditions of the Apache 2.0 License.

This section describes how to start the DL Workbench using the Python wrapper, which works on Linux OS*, macOS* and Windows*.

For additional details, such as prerequisites, security, and troubleshooting, see OpenVINO DL Workbench documentation.

Step 1. Set Up Python Virtual Environment

To avoid dependency conflicts, use a virtual environment. Skip this step only if you do want to install all dependencies globally.

Create virtual environment by executing the following commands in your terminal:

  • On Linux and MacOS:
python3 -m pip install --user virtualenv
python3 -m venv venv
  • On Windows:
py -m pip install --user virtualenv
py -m venv venv

Step 2. Activate Virtual Environment

  • On Linux and MacOS:
source venv/bin/activate
  • On Windows:
venv\Scripts\activate

Step 3. Update PIP to the Latest Version

Run the command below:

python -m pip install --upgrade pip

Step 4. Install the Python Wrapper

pip install -U openvino-workbench

Step 5. Verify the Installation

To verify that the package is properly installed, run the command below:

openvino-workbench --help

You will see the help message for the starting package if installation finished successfully.

Use the DL Workbench Starter

To start the latest available version of the DL Workbench, execute the following command:

openvino-workbench --image openvino/workbench:2021.3

You can see the list of available arguments with the following command:

openvino-workbench --help

Refer to the documentation for additional information.

Use web-form to build a starting command

This section describes how to start the DL Workbench using the plain Docker* command, which works on Linux OS*, macOS* and Windows*.

For additional details, such as prerequisites, security, and troubleshooting, see OpenVINO DL Workbench documentation.

Step 1. Navigate to the web-form website

Step 2. Select the necessary capabilities

  • Go through the web-form interface and select the most suitable capabilities for you.

Step 3. Start DL Workbench

  • Copy the resulting command and execute it in your terminal.

DL Workbench Docker Image

Pre-built DL Workbench Docker image is available on Docker Hub*.

Additional Resources

* Other names and brands may be claimed as the property of others.