Installation

Python editing

You can run Project Hadron in your favorite code editor, Jupyter notebook, Google Colab, or anywhere else you write Python.

Python version

We recommend using the latest version of Python. Project Hadron supports Python 3.8 and newer.

Package installation

The best way to install the component packages is directly from the Python Package Index using pip.

The component package is discovery-capability and pip installed with:

$ python -m pip install discovery-capability

if you want to upgrade your current version then using pip install upgrade with:

$ python -m pip install -U discovery-capability

This will also install or update dependent third party packages. The dependencies are limited to Python, PyArrow and related Data manipulation tooling such as Pandas, Numpy, scipy, scikit-learn and visual packages matplotlib and seaborn, and thus have a limited footprint and non-disruptive installation in a data processing environment.

dependencies

Project Hadron is written in pure Python and depends on a few key Python packages. These should be installed as parge of the pip installation but for troubleshooting, they are:

pyarrow, provides a cross-language, in-memory columnar data representation
pandas, versatile data manipulation and analysis toolset
numpy, core library for numerical and mathematical operations
scipy, scientific computing library that extends the functionality of NumPy
scikit-learn, machine learning library with a focus on predictive data analysis
matplotlib, versatile 2D plotting library
seaborn, statistical data visualization library built on top of Matplotlib

In case of any trouble related to these dependencies, please refer to their respective installation instructions.

Virtual environments

Use a virtual environment to manage the dependencies for your project, both in development and in production.

What problem does a virtual environment solve? The more Python projects you have, the more likely it is that you need to work with different versions of Python libraries, or even Python itself. Newer versions of libraries for one project can break compatibility in another project.

Virtual environments are independent groups of Python libraries, one for each project. Packages installed for one project will not affect other projects or the operating system’s packages.

Python comes bundled with the venv module to create virtual environments.

create an environment

Create a project folder and a .venv folder within:

macOS/Linux

$ mkdir myproject
$ cd myproject
$ python3 -m venv .venv

Windows

> mkdir myproject
> cd myproject
> py -3 -m venv .venv

activate the environment

Before you work on your project, activate the corresponding environment:

macOS/Linux

$ . .venv/bin/activate

Windows

> .venv\Scripts\activate

Your shell prompt will change to show the name of the activated environment.

Viewing examples

As said before, you can run Project Hadron in your favorite code editor, but to help any tutorial documentation, you can find examples of the code on GitHub as Jupyter notebooks.

Jupyter Notebooks and Jupyter Labs are available for installation via the Python Package Index using pip.

We recommend installing JupyterLab with pip:

$ pip install jupyterlab

Once installed, launch JupyterLab with:

$ jupyter lab

For more information on JupyterLab go to the Jupyter documentation

Release process

Versions to be released will govern and describe how the discovery-capability produces a new release.

To find the current version of discovery-capability, from your terminal run:

$ python -c "import ds_capability; print(ds_capability.__version__)"

major releases

A major release will include breaking changes. When it is versioned, it will be versioned as vX.0.0. For example, if the previous release was v10.2.7 the next version will be v11.0.0.

Breaking changes are changes that break backwards compatibility with prior versions. The majority of changes to the dependant core abstraction will result in a major release.

Project Hadron is committed to providing a good user experience and as such, committed to preserving backwards compatibility as much as possible. Major releases will be infrequent and will need strong justifications before they are considered.

minor releases

A minor release will include additional methods, classes, or noticeable modifications to code in a backward-compatable manner which may include miscellaneous bug fixes. If the previous version released was v10.2.7 a minor release would be versioned as v10.3.0.

Minor releases will be backwards compatible with releases that have the same major version number. In other words, all versions that would start with v10. should be compatible with each other.

patch releases

A patch release include small and encapsulated code changes that do not directly effect a major or minor release, for example changing round(... to np.around(..., and patch bug fixes that can’t wait to be released before a major or minor release. If the previous version released v10.2.7 the patch release would be versioned as v10.2.8.