Skip to content

oentaryorj/smu.softeng.crossact

Repository files navigation

crossact

Documentation Status

Python Boilerplate contains all the boilerplate you need to create a Python package.

Datasets

Code

  • KEEN MODEL
    • Running the code to save the data
      • We need to create three files for each Github or StackOverflow dataset (i.e., interactions_non_activities.pickle, i_features_non_activities.pickle, and u_features_CO.pickle)
        • interactions_non_activities.pickle: interaction between users and items
        • i_features_non_activities.pickle: item features
        • u_features_CO.pickle: user features
    • Training KEEN model
      • We have two steps to train KEEN model:
        • Training interaction model (See step 1a in the KEEN model)
        • Training threshold model (See step 1b in the KEEN model)
  • ACT MODEL
    • Running the code to save the data
      • We need to create two files for each Github or StackOverflow dataset (i.e., interactions_with_first_activities_for_ACT.pickle, interactions_with_second_activities_for_ACT.pickle)
        • interactions_with_first_activities_for_ACT.pickle: interaction between users and items for the first activity
        • interactions_with_second_activities_for_ACT.pickle: interaction between users and items for the second activity
    • Training ACT model
      • We have two steps to train ACT model:
        • Training interaction model (See in the ACT model)
        • Training threshold model (See in the ACT model)

Credits

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.

Releases

No releases published

Packages

No packages published

Languages