rubenkerkhofs

from fairlearn.postprocessing import ThresholdOptimizer

reduction = ThresholdOptimizer(estimator = classifier)
reduction.fit(df, target, sensitive_features = gender)
prediction_threshold = reduction.predict(df, sensitive_features = gender)

rubenkerkhofs

FairlearnDashboard(y_true = target,
                   y_pred = {"prediction_original" : prediction,
                             "prediction_dp": prediction_dp,
                             "prediction_eo": prediction_eo},
                   sensitive_features = gender,
                   sensitive_feature_names = ["gender"])

rubenkerkhofs

eo = EqualizedOdds()
reduction = ExponentiatedGradient(classifier, eo)

reduction.fit(df, target, sensitive_features=gender)
prediction_eo = reduction.predict(df)

rubenkerkhofs

from fairlearn.reductions import ExponentiatedGradient, DemographicParity, EqualizedOdds

classifier = DecisionTreeClassifier(min_samples_leaf=10, max_depth=4)
dp = DemographicParity()
reduction = ExponentiatedGradient(classifier, dp)

reduction.fit(df, target, sensitive_features=gender)
prediction_dp = reduction.predict(df)

rubenkerkhofs

from fairlearn.widget import FairlearnDashboard

FairlearnDashboard(y_true = target,
                   y_pred = prediction,
                   sensitive_features = gender,
                   sensitive_feature_names = ["gender"])

rubenkerkhofs

from fairlearn.metrics import demographic_parity_difference, equalized_odds_difference, demographic_parity_ratio

dpd = demographic_parity_difference(target, prediction, sensitive_features = gender)
eod = equalized_odds_difference(target, prediction, sensitive_features = gender)
dpr = demographic_parity_ratio(target, prediction, sensitive_features = gender)

print("Demographic parity difference: {}".format(round(dpd, 2)))
print("Equalized odds difference: {}".format(round(eod, 2)))
print("Demographic parity ratio: {}".format(round(dpr, 2)))

rubenkerkhofs

from fairlearn.metrics import group_summary
from sklearn.metrics import accuracy_score

group_summary(accuracy_score , target, prediction, sensitive_features = gender)

rubenkerkhofs

from sklearn.tree import DecisionTreeClassifier

classifier = DecisionTreeClassifier(min_samples_leaf=10, max_depth=4) # parameters have not been tuned
classifier.fit(df, target)

# Note that we are predicting using the same data as we used for training, this is just for the sake of example
# Never do this in real life
prediction = classifier.predict(df)

rubenkerkhofs

race = df.pop("race") # Pop function drops and assigns at the same time
gender = df.pop("gender")

rubenkerkhofs

df.loc[:, "target"] = df.income.apply(lambda x: int(x == ">50K")) # int(): Fairlearn dashboard requires integer target
df = df.drop("income", axis = 1) # drop income variable to avoid perfect prediction