Automunge: A Tabular Data Preprocessing Platform


Modern paradigms of machine learning libraries often have prerequisites that data is provided in a numerically encoded form with all valid entries. Automunge is an open source python library for preparing tabular data for machine learning. Through application raw data is transformed into a form suitable for the direct application of machine learning. Subsequent data, such as may be intended to generate predictions from a trained model, can then be consistently prepared on the same basis.


What’s more Automunge may serve as a platform for assembling data pipelines. A library of feature engineering transforms is available, and through simple assignment columns may be subject to feature engineering transforms, or in some cases even sets of feature engineering transforms. Missing data infill may be handled in a sophisticated manner using machine learning models automatically trained on the data.

  Date and Time




  • Date: 06 Oct 2020
  • Time: 06:30 PM to 07:45 PM
  • All times are US/Eastern
  • Add_To_Calendar_icon Add Event to Calendar
If you are not a robot, please complete the ReCAPTCHA to display virtual attendance info.
  • Orlando, Florida
  • United States
  • Starts 24 September 2020 04:00 PM
  • Ends 06 October 2020 07:15 PM
  • All times are US/Eastern
  • No Admission Charge


Nicholas Teague


Automunge: A Tabular Data Preprocessing Platform


Nicholas Teague is the founder of Automunge, an open source platform for preparing tabular data for machine learning. He has been working full time on development for over a year and is interested in feedback from users. Teague also regularly publishes essays on a range of subjects like machine learning, quantum computing, and entrepreneurship, and has collected his writings in an online journal titled “From the Diaries of John Henry”. Teague is a licensed professional engineer with an extensive background in the renewable energy field.



This presentation will offer a 30-45 minutes introductory presentation to the Automunge library followed by a walkthrough of a demonstration Jupyter notebook, with remaining time available for audience Q&A.