Team Patiently Matching

Members:

• Christopher Pan
• Drake Lin
• Enya Xing

Patient Matching

Set up instructions

1. Open Jupyter notebook written in Python
2. Download libraries in first cell of notebook
3. Change file destinations to appropriate destinations
4. Read through notebook instructions (important for using other test files as input)

Proof of Concept Steps

1. Read in Patient Matching Data files and/or other files to be used for training/testing
2. Run notebook which will train model
3. Notebook will add predicted groups column to the input file and write out to output file destination & name
4. more detailed info in notebook

The web app is not fully functional so we won't use it for testing. In the future, we will have a website that asks for a csv file input and outputs a csv file with the GroupIDs, built using HTML, Django, and Python.

Contact info

christopher.pan@berkeley.edu