This repository is meant to facilitate the benchmarking of fairness aware machine learning algorithms.
The associated paper is:
A comparative study of fairness-enhancing interventions in machine learning by Sorelle A. Friedler, Carlos Scheidegger, Suresh Venkatasubramanian, Sonam Choudhary, Evan P. Hamilton, and Derek Roth. https://arxiv.org/abs/1802.04422
To install this software run:
$ pip3 install fairness
The below instructions are still in the process of being updated to work with the new pip install-able version.
To run the benchmarks:
$ from fairness.benchmark import run
$ run()
This will write out metrics for each dataset to the results/ directory.
To generate graphs and other analysis run:
$ python3 analysis.py
If you do not yet have all the packages installed, you may need to run:
$ pip install -r requirements.txt
Optional: The benchmarks rely on preprocessed versions of the datasets that have been included in the repository. If you would like to regenerate this preprocessing, run the below command before running the benchmark script:
$ python3 preprocess.py
To add new datasets or algorithms, see the instructions in the readme files in those directories.
OS-specific things
On Ubuntu
(We tested on Ubuntu 16.04, your mileage may vary)
You'll need python3-dev:
$ sudo apt-get install python3-dev
Additional analysis-specific requirements
To regenerate figures (this is messy right now. we're working on it)
Python requirements (use pip):
ggplot
System requirements:
pandoc(brew install pandocon a Mac orapt-get install pandocon Linux)- R (Mac download link: https://cran.rstudio.com/bin/macosx/R-3.4.3.pkg)
R package requirements (use install.packages):
rmarkdownstringrggplot2dplyrmagrittrcorrplotrobust
1 maintainer
Metadata
1 maintainer
Release files for fairness 0.1.8
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| fairness-0.1.8.tar.gz | 12.9 MB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| fairness-0.1.8-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 27.1 MB
Release files / fairness-0.1.8.tar.gz
| Download URL | fairness-0.1.8.tar.gz |
|---|---|
| Size | 12.9 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
7cd98a95dfbf3342ad990822afda97514bed2c6be789889db2ee2ff39f8f534d
|
|
BLAKE2b-256 checksum How to use checksums |
91017b2d3e84bfd69b73757c3dc16e4a721e36ec662e03e6c9279675d099a86f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/1.12.1 pkginfo/1.5.0.1 requests/2.21.0 setuptools/39.0.1 requests-toolbelt/0.8.0 tqdm/4.30.0 CPython/3.6.7
|
Release files / fairness-0.1.8-py3-none-any.whl
| Download URL | fairness-0.1.8-py3-none-any.whl |
|---|---|
| Size | 14.2 MB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
7f21cf0943f2a6fef7f7597eb99461bd4b2f7a26fe9810964675496cde6992e1
|
|
BLAKE2b-256 checksum How to use checksums |
f6d0038541647d46112174ae8f9d7ef256d73cfccc0668923748826a0d4cb63c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/1.12.1 pkginfo/1.5.0.1 requests/2.21.0 setuptools/39.0.1 requests-toolbelt/0.8.0 tqdm/4.30.0 CPython/3.6.7
|
