Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[PDP] Add term diff features, and improve enrollment year #15

Open
wants to merge 18 commits into
base: develop
Choose a base branch
from

Conversation

bdewilde
Copy link
Member

@bdewilde bdewilde commented Oct 27, 2024

changes

  • adds a new family of cumulative features referred to as "term diffs", which compute term-over-term differences for a configurable set of columns; I tried to keep this minimal, and cover three basic indicators: “participation” (num courses taken), “progression” (num credits earned), and “performance” (avg course grade)
  • (re-)implements logic to compute start/end dates for terms and cohorts, and uses term+cohort start dates to compute a new "year of enrollment" feature that should align with how PDP computes it in other, similar fields
  • adds cleanup logic to nullify column values based on *_term_X, in the same way as already implemented for *year_Y features
  • adds a minimal types.py module, in which we can place shared types; starting out with just TermType

context

Continuing to add more info for model training, as possible. All of the year-of-enrollment stuff came out of this Slack thread

Follows PR #14

questions

  • Is the generic term diff logic correct?
  • Is the new term dt / enrollment year logic correct? Would you like me to drop the old enrollment year feature in favor of the v2?

@bdewilde bdewilde marked this pull request as ready for review October 27, 2024 15:56
Base automatically changed from pdp-improve-grades-handling to develop October 28, 2024 15:34
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant