Generalizability of experimental studies

This repository is the official implementation of Generalizability of Experimental Studies.

The genexpy Python module contains several functions to efficiently work on rankings, probability distributions of rankings, and kernels for rankings.

The demos contain the generalizability analysis performed in the paper on categorical encoders and Large Language Models.

Usage

As specified in setup.py, genexpy requires Python 3.11.8, which you can get here. After that, you can install genexpy from terminal in one of the following ways.

Download and install from the anonymous repository

(1) Download the repo and extract the files to a directory named genexpy; (2) navigate to genexpy; (3) create a virtual environment venv; (4) activate venv (the command is for Windows only); (5) install genexpy.

cd genexpy                                                           
python -m venv venv                                       
venv\Scripts\activate.bat                                 
pip install .                                                

Installation via pip (not working for the anonymous repo)

To ensure anonymity, the module is yet to be published: this installation method is not working.

pip install genexpy

Installation via git (not working for the anonymous repo)

(1) Clone the repo; (2) navigate to the cloned repo; (3) install the module.

git clone https://github.com/DrCohomology/genexpy.git     
cd genexpy                                                           
pip install .                                             

Cloning without installation (not working for the anonymous repo)

(1) Clone the repo; (2) navigate to the cloned repo; (3) create a virtual environment venv; (4) activate venv; (5) install the requirements.

git clone https://github.com/DrCohomology/genexpy.git     
cd genexpy
python -m venv venv                                       
venv\Scripts\activate.bat                                 
pip install -r requirements.txt