Unbox is a debugging workspace for ML & Data Science. Unbox combines and builds upon SOTA techniques in explainability, model and dataset versioning, synthetic data generation, data-centric testing and much more to form a powerful, unified platform for model development.
π Join our Slack community! We'd love to meet you and help you get started with Unbox!
This is the official Python library for interacting with the Unbox platform. Navigate here for a quickstart guide and for in-depth tutorials.
This library's primary function is to enable you to easily package your models and datasets and add them to your Unbox account.
Install with PyPI (pip)
pip install --upgrade unboxapi
or install with Anaconda (conda)
conda install unboxapi --channel conda-forge
The official documentation for this Python library can be found here.
All contributions, bug reports, bug fixes, documentation improvements, enhancements, and ideas are welcome! Just send us a message on Slack.