PyHelpers

An open-source toolkit for facilitating Python users' data manipulation tasks.

PyHelpers is an open-source Python package designed to streamline data (pre-)processing and manipulation tasks. It accommodates a wide range of functions and classes grounded in practical applications, making common data operations more accessible and efficient. This toolkit is particularly useful for Python learners, researchers and data scientists seeking to enhance their workflows.

The package supports handling various data types, such as geographical and textual data, allowing for flexibility for diverse data processing needs. It also simplifies data input and output operations by offering functionalities for managing many different file-like objects. In addition, PyHelpers facilitates communication with relational databases, such as PostgreSQL and Microsoft SQL Server. This capability greatly smooths data integration with database systems through efficient data storage and retrieval.

With its comprehensive suite of practical tools, PyHelpers simplifies complex data processing tasks and boosts productivity. It is ready to serve as an essential resource for effective data manipulation, management and analysis for anyone working with data in Python.

Installation

PyHelpers can be installed using uv (recommended for speed, reliability and modern dependency resolution), conda-forge (including pixi) or traditional pip.

Using uv (Recommended)

To add PyHelpers to an existing project managed by uv:

uv add pyhelpers

To include all optional features (e.g. geospatial drivers and database connectors):

uv add "pyhelpers[full]"

If you are working in an activated virtual environment:

uv pip install --upgrade pyhelpers

To install all optional features inside an active virtual environment:

uv pip install --upgrade "pyhelpers[full]"

Using pixi

To add the core PyHelpers package to a workspace using pixi:

pixi add pyhelpers

To add PyHelpers along with optional dependencies (e.g. gdal, pyarrow, fiona and so on) via pixi:

pixi add pyhelpers gdal pyarrow fiona

Alternatively, to install PyHelpers with PyPI extras inside a pixi project:

pixi add --pypi "pyhelpers[full]"

Using pip

To install PyHelpers into an active environment using pip:

pip install --upgrade pyhelpers

To install with all optional dependencies:

pip install --upgrade "pyhelpers[full]"

Using conda

To install PyHelpers into an active environment using conda (or mamba):

# Core package
conda install -c conda-forge pyhelpers

# Full suite with binary C-extensions
conda install -c conda-forge pyhelpers gdal pyarrow fiona

For detailed options, development setup and Windows troubleshooting (e.g. installing C-extension wheels), see the full Installation Guide.

Quick start

For a concise guide on how to use PyHelpers, check out the Quick Start tutorial, which includes illustrative examples for each of the subpackages.

These examples briefly demonstrate the capabilities of PyHelpers in facilitating data manipulation tasks and streamlining work processes.

Documentation

The complete PyHelpers Documentation is available in HTML and PDF formats.

It is hosted on Read the Docs, and the HTML version is also accessible via GitHub Pages. The documentation includes detailed examples, tutorials and comprehensive references to help users get the most out of PyHelpers.

Cite as

Fu, Q. (2020). PyHelpers: An open-source toolkit for facilitating Python users' data manipulation tasks. Zenodo. doi:10.5281/zenodo.4017438.

@software{Fu_PyHelpers_2020, 
    author = {Fu, Qian},
    title = {{PyHelpers: An open-source toolkit for facilitating Python users' data manipulation tasks}},
    year = {2020},
    publisher = {Zenodo},
    doi = {10.5281/zenodo.4017438},
    license = {MIT},
    url = {https://doi.org/10.5281/zenodo.4017438},
}

For specific version references, please refer to Zenodo.

License

PyHelpers is licensed under the MIT License.

Please note that this project was initially licensed under the GPLv3+ up to version 1.5.2. Starting with version 2.0.0, it has been re-licensed under the MIT License.