Quickstart¶
Installation¶
pip install pyfomo
Dependencies¶
The requirements for fomo
are listed in environment.yml
.
Note on pymoo If you are working in linux and get a warning about pymoo, is recommended that you manually install it from the github repo rather than pip:
git clone https://github.com/anyoptimization/pymoo
cd pymoo
make compile
pip install .
Development¶
To install a development version, do the following:
git clone https://github.com/cavalab/fomo
cd fomo
pip install .
Basic Usage¶
Here is an example of training a fair classifier on the adult income prediction dataset.
from fomo import FomoClassifier
from pmlb import pmlb
dataset = pmlb.fetch_data('adult')
X = dataset.drop('target',axis=1)
y = dataset['target']
groups = ['race','sex']
est = FomoClassifier()
est.fit(X,y, protected_features=groups)