Use-cases

This repository can be used for the problems pertaining to allocation problems, specifically which involve preference ordering(read stable matching problem from both parties. The algorithm employed is a tailored version of Gale-Shapley algorithm to cater many-one matchings.

The common use-cases for these could be for Project allocation to students, Courses allocation, team formations, etc.

This was used to allocate projects to the students during the summer'22 for the projects offered by Science and Technology Council, IIT Kanpur.

Running

Before running please read through the instructions!

python allot.py

• allot.py is the only file you need to run. All the other files are utilities.
  It produces two important files, one which contains (Roll Number, Project Alloted) pairs, and one which contains fraudulent pairs, which may arise from a student being allotted a project which was not in his preferences or if a student had not filled the preference form altogether. The other files are common statistics regarding the allocation.
• config.py has some parameters-
  • Number of preferences a student is allowed to fill
  • The path to master_responses
• Dependencies: pandas

Common debugging:

• Duplicate Roll Number found...: There is a duplicate entry (of a roll number) in master_responses. Some suggestions-
  • Keep the form editable to ensure that the same person does not fill the form twice
  • It might be the case that someone made a typo while filling the form. Take suitable actions
• Check the project names, most probably there is an inconsistency
• A student/project not getting an allotment is completely possible and is not erroneous. No bugs here, only functionalities

🦇 Credits 🦇

• Aditya Tanwar
• Prem Bharwani