A series of classes to match mentors and mentees
Project description
Mentor Match
This is a package to help match mentees and mentors. It's specifically designed for a volunteer programme I support, but you could probably extend or alter it to suit whatever you're doing.
It uses this implementation of Munkres to find the most effective pairings. The Munkres algorithm works on a grid of scores.
Scoring
Full details of how the matches are calculated can be read in the code itself. Customisable configurations are on the roadmap but are not planned for any upcoming releases.
Installation
You can install this project with python -m pip install mentor-match
Use
To use this library, first install it (see above). You may need to munge your data for the system to be happy with it. Use the example CSV file as guides for your mentor and mentee data, then put them together in the same folder.
The software will run three matching exercises. Participants who don't match in the first round are more heavily weighted in the next round. The aim is to improve the experience for everyone.
The weightings are as follows:
property | First run | Second run | Third run |
---|---|---|---|
mentee's target profession is the same as mentor's profession | 4 | 4 | 0 |
grade is one or two grades different | 3 | 3 | 3 |
bonus for anyone's who's not got a match yet | 0 | 50 | 100 |
Here is a snippet that outlines a minimal use in a Python project:
from matching import process
data_folder = "Documents/mentoring-data"
mentors, mentees = process.conduct_matching_from_file(data_folder)
output_folder = data_folder / "output"
process.create_mailing_list(mentors, output_folder)
process.create_mailing_list(mentees, output_folder)
This creates a mailing list according to a set template, ready for processing by your favourite/enterprise mandated email solution
Alternatively, you can run this software from the command line as follows
python -m matching /path/to/participant/data
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for mentor_match-5.1.3-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 2050cd5d963f2fe8c49e44cd77cacc88217ade24836582f2d8a9ed392652084d |
|
MD5 | 6bd62602a79f41ca1d1cad3fd9d24b98 |
|
BLAKE2b-256 | 0f6a40340655ac3205ea402dd98f83c9d1d0b33a97d5b741550c77f44f4df370 |