# openvino_training_extensions

**Repository Path**: smartwell/openvino_training_extensions

## Basic Information

- **Project Name**: openvino_training_extensions
- **Description**: Trainable models and NN optimization tools
- **Primary Language**: Python
- **License**: Apache-2.0
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2019-08-09
- **Last Updated**: 2020-12-19

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Training Toolbox for TensorFlow

Training Toolbox for TensorFlow provides a convenient environment to train
Deep Learning models and convert them using [OpenVINO™
Toolkit](https://software.intel.com/en-us/openvino-toolkit) for optimized
inference.

## Pre-requisites

- Ubuntu 16.04 / 18.04
- Python 3.4-3.6
- [libturbojpeg](https://github.com/ajkxyz/jpeg4py)
- For Python pre-requisites refer to `requirements.txt`
- (Optional) [TensorFlow GPU pre-requisites](https://www.tensorflow.org/install/gpu).
- (Optional) [OpenVINO™ R3](https://software.intel.com/en-us/openvino-toolkit)
  for exporting of the trained models

# Quick Start Guide

## Setup Training Toolbox for TensorFlow

1. Clone repository in the working directory

  ```
  cd /<path_to_working_dir>
  git clone https://github.com/opencv/training_toolbox_tensorflow.git
  cd training_toolbox_tensorflow
  git submodule update --init --recursive
  ```

2. Install prerequisites

  ```
  sudo apt-get install libturbojpeg python3-tk python3-pip virtualenv 2to3
  ```

3. Create virtual environment
<a name="create_venv"></a>

  ```
  bash tools/init_venv.sh
  ```

4. Start to work

  ```
  cd /<path_to_working_dir>/training_toolbox_tensorflow
  . venv/bin/activate
  ```

## Tests
In virtual environment run tests:

```
cd training_toolbox
nosetests
```

or if you are going to use the OpenVino toolkit:

```
cd training_toolbox
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](#create_venv)
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](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](training_toolbox/ssd_detector/README.md)