facially-edge

(Work in progress) An edge app, part of Facially-targeted Ads Platform, that is responsible for running a neural network model for inference data based on facial features, estimate age, gender, emotions and head-pose, send this data to a web service in order to get the appropriate ads for that particular audience, and display it on the screen. The app will run on a device with a video camera attached, and internally will use Intel® Distribution of OpenVINO™ toolkit for DNN inference.

Requirements

Python 3.5+.

Note

Because Python 2.7 supports ended January 1, 2020, new projects should consider supporting Python 3 only, which is simpler than trying to support both. As a result, support for Python 2.7 in this example project has been dropped.

Windows Support

Summary: On Windows, use py instead of python3 for many of the examples in this documentation.

This package fully supports Windows, along with Linux and macOS, but Python is typically installed differently on Windows. Windows users typically access Python through the py launcher rather than a python3 link in their PATH. Within a virtual environment, all platforms operate the same and use a python link to access the Python version used in that virtual environment.

Installation

To install and run the program without development environment folloe the instruction

Create and start the virtual environment (Check section "Virtual Environments" for more details) .. code:: bash

$ python3 -m venv venv $ source venv/bin/activate

Install the requirements: $ pip -r requirements.txt

To execute the cli (example) .. code:: bash

$ cd src $ python -m facially.cli -s World

Packaging

This project is designed as a Python package, meaning that it can be bundled up and redistributed as a single compressed file.

Packaging is configured by:

setup.py
MANIFEST.in

Source Distribution

To package the project as a source distribution:

$ python3 setup.py sdist

This will generate dist/facially-0.0.1.tar.gz.

Dependencies

Dependencies are defined in:

requirements.in
requirements.txt
dev-requirements.in
dev-requirements.txt

Virtual Environments

It is best practice during development to create an isolated Python virtual environment using the venv standard library module. This will keep dependant Python packages from interfering with other Python projects on your system.

On *Nix:

$ python3 -m venv venv
$ source venv/bin/activate

On Windows cmd:

> py -m venv venv
> venv\Scripts\activate.bat

Once activated, it is good practice to update pip to the latest version.

(venv) $ pip install --upgrade pip

(Applications Only) Locking Dependencies

This project uses pip-tools to lock project dependencies and create reproducible virtual environments.

Note: Library projects should not lock their requirements.txt. Since python-blueprint also has a CLI application, this end-user application example is used to demonstrate how to lock application dependencies.

To update dependencies:

(venv) $ pip install pip-tools
(venv) $ pip-compile --upgrade
(venv) $ pip-compile --upgrade dev-requirements.in

After upgrading dependencies, run the unit tests as described in the Unit Testing section to ensure that none of the updated packages caused incompatibilities in the current project.

Syncing Virtual Environments

To cleanly install your dependencies into your virtual environment:

(venv) $ pip-sync requirements.txt dev-requirements.txt

Testing

Automated testing is performed using tox. tox will automatically create virtual environments based on tox.ini for unit testing, PEP8 style guide checking, and documentation generation.

# Install tox (only needed once).
$ python3 -m pip install tox

# Run all environments.
#   To only run a single environment, specify it like: -e pep8
$ tox

Unit Testing

Unit testing is performed with pytest. pytest has become the defacto Python unit testing framework. Some key advantages over the built in unittest module are:

Significantly less boilerplate needed for tests.
PEP8 compliant names (e.g. pytest.raises() instead of self.assertRaises()).
Vibrant ecosystem of plugins.

pytest will automatically discover and run tests by recursively searching for folders and .py files prefixed with test for any functions prefixed by test.

The tests folder is created as a Python package (i.e. there is an __init__.py file within it) because this helps pytest uniquely namespace the test files. Without this, two test files cannot be named the same, even if they are in different sub-directories.

Code coverage is provided by the pytest-cov plugin.

When running a unit test tox environment (e.g. tox, tox -e py37, etc.), a data file (e.g. .coverage.py37) containing the coverage data is generated. This file is not readable on its own, but when the coverage tox environment is run (e.g. tox or tox -e -coverage), coverage from all unit test environments is combined into a single data file and an HTML report is generated in the htmlcov folder showing each source file and which lines were executed during unit testing. Open htmlcov/index.html in a web browser to view the report. Code coverage reports help identify areas of the project that are currently not tested.

Code coverage is configured in the .coveragerc file.

Code Style Checking

PEP8 is the universally accepted style guide for Python code. PEP8 code compliance is verified using flake8. flake8 is configured in the [flake8] section of tox.ini. Three extra flake8 plugins are also included:

pep8-naming: Ensure functions, classes, and variables are named with correct casing.
flake8-quotes: Ensure that ' ' style string quoting is used consistently.
flake8-import-order: Ensure consistency in the way imports are grouped and sorted.

Project Structure

Traditionally, Python projects place the source for their packages in the root of the project structure, like:

root_folder
├── facially
│   ├── __init__.py
│   ├── cli.py
│   └── lib.py
├── tests
│   ├── __init__.py
│   └── test_generate.py
├── tox.ini
└── setup.py

However, this structure is known to have bad interactions with pytest and tox, two standard tools maintaining Python projects. The fundamental issue is that tox creates an isolated virtual environment for testing. By installing the distribution into the virtual environment, tox ensures that the tests pass even after the distribution has been packaged and installed, thereby catching any errors in packaging and installation scripts, which are common. Having the Python packages in the project root subverts this isolation for two reasons:

Calling python in the project root (for example, python -m pytest tests/) causes Python to add the current working directory (the project root) to sys.path, which Python uses to find modules. Because the source package facially is in the project root, it shadows the facially package installed in the tox environment.
Calling pytest directly anywhere that it can find the tests will also add the project root to sys.path if the tests folder is a a Python package (that is, it contains a __init__.py file). pytest adds all folders containing packages to sys.path because it imports the tests like regular Python modules.

In order to properly test the project, the source packages must not be on the Python path. To prevent this, there are three possible solutions:

Remove the __init__.py file from tests and run pytest directly as a tox command.
Remove the __init__.py file from tests and change the working directory of python -m pytest to tests.
Move the source packages to a dedicated src folder.

The dedicated src directory is the recommended solution by pytest when using tox and the solution this blueprint promotes because it is the least brittle even though it deviates from the traditional Python project structure. It results is a directory structure like:

root_folder
├── src
│   └── facially
│       ├── __init__.py
│       ├── cli.py
│       └── lib.py
├── tests
│   ├── __init__.py
│   └── test_generate.py
├── tox.ini
└── setup.py

Licensing

Licensing for the project is defined in:

LICENSE.txt
setup.py

This project uses a common permissive license, the MIT license.

computationalcore/facially-edge