Skip to main content

tubesML, a package that allows for flexible ML pipelines, model validation, and model inspection

Project description

A package that allows for flexible ML pipelines, model validation, and model inspection.

The project started to productize code written for Kaggle competitions and it is mostly designed to facilitate the creation of flexible processing and modeling pipeline. Everything is built around scikit-learn transformers and methods, but it is possible to extend the concept to other ML packages.

The main characteristic that makes the pipeline components of tubesML flexible is the fact that it always preserves the pandas DataFrame structure, hence making it easy to, for example, create a feature within a pipeline and test it in a grid search.



tubesML requires:

  • Python (>= 3.7)

  • Matplotlib (>= 3.5.3)

  • Pandas (>= 1.3.5)

  • Scikit-Learn (>= 1.0.2)

  • Seaborn (>= 0.12.0)

These requirements are good for a Kaggle notebook, however the package has been developed with the following requirements

  • Python (>= 3.10)

  • Matplotlib (>= 3.6.0)

  • Pandas (>= 1.5.0)

  • Numpy (>= 1.23.1)

  • Scikit-Learn (>= 1.1.2)

  • Seaborn (>= 0.12.0)

User installation

If you already have a working installation of pandas and sklearn, the easiest way to install scikit-learn is using pip

pip install -U tubesml

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

tubesml-0.7.1.tar.gz (39.6 kB view hashes)

Uploaded source

Built Distribution

tubesml-0.7.1-py3-none-any.whl (41.0 kB view hashes)

Uploaded py3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page