A Python package for pre-processing nuMoM2b data with configuration files.

Documentation Status Build Status Code Coverage Config file image.

This is one component within a wider research project for predicting adverse events over the course of a woman’s pregnancy. This package serves the dual role of assisting with pre-processing tasks and for producing reproducible partitions of the data based on configuration files.

Maintained by Alexander L. Hayes, a Ph.D. student with the Indiana University ProHealth Group working on the Precision Health Initiative (φ). Alexander can be reached at

Getting Started

Pointers for getting nuMoM2b_preprocessing working on your local machine.


High-level overview of how this project is organized.

nuMoM2b_preprocessing architecture.

Writing Configuration Files

Configuration files define which variables should be used and how they should be aggregated. This is a worked example for writing a configuration file.

  "csv_path": "../FullData/numom_data/",
  "target": {
      "name": "Ancillary/Pregnancy_outcomes.csv",
      "variables": ["PublicID", "oDM"]
  "files": [
      "name": "Screening_Admin_Visits/Visit1.csv",
      "variables": ["PublicID", "V1BA01_KG", "V1BA01_LB"]

Indices and tables