from sklearn.datasets import fetch_california_housing
from sklearn.model_selection import train_test_split
from sklearn.pipeline import make_pipeline
from sklearn.preprocessing import QuantileTransformer
from sklearn.neural_network import MLPRegressor
from sklearn.inspection import PartialDependenceDisplay

X, y = fetch_california_housing(as_frame=True, return_X_y=True)
y -= y.mean()

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=0)

est = make_pipeline(
QuantileTransformer(),
MLPRegressor(
hidden_layer_sizes=(30, 15),
learning_rate_init=0.01,
early_stopping=True,
random_state=0,
),
)
est.fit(X_train, y_train)

common_params = {
"n_jobs": 2,
"grid_resolution": 10,
"centered": True,
"random_state": 0,
}

display = PartialDependenceDisplay.from_estimator(
est,
X_train,
features=["MedInc", "AveOccup", "HouseAge"],
kind="average",
**common_params,
)

display.figure_.suptitle("centered=True")
display.plot(centered=False)
_ = display.figure_.suptitle("centered=False")