dssg/education-college-public

Name: education-college-public

Owner: Data Science for Social Good

Description: A 2015 DSSG project assisting school networks to increase the proportion of their alumni who graduate college

Created: 2015-11-04 16:49:20.0

Updated: 2016-02-18 21:43:40.0

Pushed: 2015-11-04 16:54:04.0

Homepage: null

Size: 232

Language: Python

GitHub Committers

User	Most Recent Commit	# Commits

Other Committers

User	Email	Most Recent Commit	# Commits

README

Predicting College Persistence

This is a 2015 Data Science For Social Good summer project focused on increase the proportion of their alumni who graduate college. We are working with several charter school partners to help target their alumni intervention efforts by predicting which students are at risk of not persisting.

Team Members

• Benedict Kuester
• Michael Stepner
• Masha Westerlund
• Anushka Anand
• Alan Fritzler

To run this code

run_project.sh does all the configuration and verifies that all requirements are met to run our code. It is the best starting point for getting our code running.

Before starting, you will need to copy config.sh.example to config.sh. Then open config.sh in a text editor and fill in the settings.

If you run the shell script without arguments, it will simply verify that the configuration is complete and the required Python packages are installed.

The script also accepts two arguments (alone or combined):

• --etl runs our ETL pipeline and loads all data into an empty Postgres database.
• --modeling runs our modeling pipeline, which runs one or many models and logs the results. By default this runs the model config in code/modeling/configs/default.yaml. But you can also specify another config file using --modeling=PATH_TO_CONFIG.

Repo Contents

• run_project.sh: a shell script that safely runs our code
  • verifies all dependencies are installed
  • loads and applies config from config.sh
  • sets required PYTHONPATH
  • runs ETL or Modeling
• requirements.txt: lists Python package dependencies required to run all the code
• code: contains all analysis code
  • etl: code for cleaning and uploading raw data to databases
  • modeling: code for generating features, fitting and evaluating models
  • util: utility modules that are used throughout the codebase
  • visualizations: code for generating model figures and other visualizations

Copyright and License

Code is copyright 2015 Data Science for Social Good Summer Fellowship and released under the MIT license.