tensorflow_toolkit

Training Toolbox for TensorFlow

Training Toolbox for TensorFlow provides a convenient environment to train Deep Learning models and convert them using OpenVINO™ Toolkit for optimized inference.

Pre-requisites

• Ubuntu 16.04 / 18.04
• Python 3.4-3.6
• libturbojpeg
• For Python pre-requisites refer to requirements.txt
• (Optional) TensorFlow GPU pre-requisites.
• (Optional) OpenVINO™ R3 for exporting of the trained models

Quick Start Guide

Setup Training Toolbox for TensorFlow

1. Create virtual environment

cd /<path_to_working_dir>/training_toolbox/tensorflow_toolkit
bash init_venv.sh

2. Start to work

. venv/bin/activate

Tests

In virtual environment run tests:

cd /<path_to_working_dir>/training_toolbox/tensorflow_toolkit
nosetests

or if you are going to use the OpenVino toolkit:

cd /<path_to_working_dir>/training_toolbox/tensorflow_toolkit
export OPEN_VINO_DIR=<PATH_TO_OPENVINO>
nosetests

Note: if you have installed the OpenVino toolkit after creating a virtual environment then you have to recreate one to install required packages for the Model Optimizer into one.

Do not forget to update several environment variables are required to compile and run OpenVINO™ toolkit applications, for details see: https://software.intel.com/en-us/articles/OpenVINO-Install-Linux.

Models

After installation, you are ready to train your own models, evaluate them, use them for predictions.

• SSD Object Detection
• LPRNet
• Vehicle Attributes