def test_fetch_adult():
"""Tests Adult Income dataset shapes with various options."""
adult = fetch_adult()
assert len(adult) == 3
assert adult.X.shape == (45222, 13)
assert fetch_adult(dropna=False).X.shape == (48842, 13)
assert fetch_adult(numeric_only=True).X.shape == (48842, 7)
import numpy as np
from sklearn.model_selection import GridSearchCV
from sklearn.metrics import accuracy_score
import tensorflow as tf
from aif360.datasets import AdultDataset
from aif360.sklearn.datasets import fetch_adult
from aif360.algorithms.inprocessing import AdversarialDebiasing as OldAdversarialDebiasing
from aif360.sklearn.inprocessing import AdversarialDebiasing
X, y, sample_weight = fetch_adult(numeric_only=True)
adult = AdultDataset(instance_weights_name='fnlwgt',
categorical_features=[],
features_to_keep=[
'age', 'education-num', 'capital-gain',
'capital-loss', 'hours-per-week'
],
features_to_drop=[])
def test_adv_debias_old_reproduce():
"""Test that the old AdversarialDebiasing is reproducible."""
sess = tf.Session()
old_adv_deb = OldAdversarialDebiasing(unprivileged_groups=[{
'sex': 0
}],
privileged_groups=[{
'sex': 1
}],
scope_name='old_classifier',
sess=sess,
import numpy as np
import pandas as pd
from sklearn.model_selection import train_test_split
from aif360.sklearn.datasets import fetch_adult
from aif360.sklearn.metrics import statistical_parity_difference
from aif360.sklearn.preprocessing import fairadapt
X, y, sample_weight = fetch_adult(
dropcols=['education', 'capital-gain', 'capital-loss', 'relationship'])
X = X[0:5000]
y = y[0:5000]
(X_train, X_test, y_train, y_test) = train_test_split(X,
y,
train_size=0.8,
random_state=1234567)
def test_fairadapt_adult():
"""Test that FairAdapt works when applied to Adult dataset."""
train_df = pd.concat([X_train, y_train], axis=1)
adj_mat = pd.DataFrame(np.zeros(
(len(train_df.columns), len(train_df.columns)), dtype=int),
index=train_df.columns.values,
columns=train_df.columns.values)
# Construct the adjacency matrix of the causal graph
adj_mat.loc[["sex", "age", "native-country"], [
"marital-status", "education-num", "workclass", "hours-per-week",
"occupation", "annual-income"
]] = 1